TRAPPED CHARGE DATING (ESR, TL, OSL)

Trapped charge dating techniques - electron spin resonance, commonly abbreviated to ESR, thermoluminescence (TL) and optically stimulated luminescence (OSL) - have played a major role in the establishment of chronologies in archaeology, particularly in the pre-radiocarbon time range. In archaeological applications, ESR is mostly used for dating tooth enamel, TL for sediments and burnt flint and the latest method, OSL, for sediments. TL and ESR have been instrumental in documenting the early evolution of modern humans in Israel (Valladas et al. 1988, Mercier et al. 1995, Grün and Stringer 1991). OSL allows the age estimation of sediments that were only exposed to dim sunlight for short time periods. At present, OSL dating is not yet routinely applied in archaeological studies but one can expect that this technique will lead to a breakthrough of knowledge similar to the revolution achieved by radiocarbon dating in the 1950s and 1960s. Recent reviews on ESR dating in archaeology were published by Rink (1997) and Grün (1997, 2000a-c), and on luminescence dating by Aitken (1997, 1998) and Roberts (1997).

Basic principles
Trapped charge dating (TCD) comprises a family of dating methods which are based on the same principle, namely the time-dependent accumulation of electrons and holes in the crystal lattice of certain common minerals (Figure 1). The minerals are acting as natural radiation dosimeters. When a mineral is formed or reset, all electrons are in the ground state. Naturally occurring radioactive isotopes (U, Th, and K) emit a variety of rays which ionize atoms. Negatively charged electrons are knocked off atoms in the ground state (valence band) and transferred to a higher energy state (conduction band); positively charged holes remain near the valence band. After a short time of diffusion, most electrons recombine with holes, thus returning the mineral back to its original state. However, all natural minerals contain defect sites (e.g., lattice defects, interstitial atoms, etc.) at which electrons and holes can be trapped. The trapped electrons and holes form paramagnetic centres which can be detected with an ESR spectrometer, giving rise to a characteristic ESR signal. For the measurement of a luminescence signal, the trapped electrons have to be either thermally (by heating) or optically (by light exposure) activated. The electrons return to the conduction band and most of them will recombine with the holes. If such holes are luminescence centres, light emission (luminescence) is observed.

Figure 1. The basis for trapped charge dating (from Grün 1997). Left: ionizing radiation knocks off negative electrons from atoms, which are transferred to the conduction band; positively charged holes are left near the valence band. Most electrons recombine with holes, but some are trapped in the crystal lattice. These trapped electrons can be directly measured by ESR. Right: subsequent heating (TL) or light exposure (OSL) releases the electrons and light emission (luminescence) is observed. Ea = activation energy or trap depth.

A TCD age, T, is derived from the simple relationship:

(1)

If the dose rate, D(t) or , is constant, Equation (1) is reduced to:

(2)

The determination of the DE value is the actual ESR/TL/OSL part of the dating procedure. The dose rate is calculated from an analysis of the radioactive elements in the sample and its surroundings. The concentrations of the radioactive elements are converted into dose rates by published tables (see below and Table 1). The determination of the radioactivity that influences the sample is rather complex and has to be carefully evaluated.

Figure 3. The equivalent dose (DE) value results from fitting the data points of the dose response with an exponential function and extrapolation to I0.

Table 1: Dose rates calculations for the U and Th decay chains and K (Adamiec and Aitken 1998).

Concentration of radioactive elements
1 ppm 238U + 235U	2780	146	113	µGy/a
1 ppm 232Th	732	27	48	µGy/a
1 % K		782	243	µGy/a

ESR measurement

In an ESR spectrometer (Figure 4 and 5), the sample is placed into a microwave cavity which is located in a strong external magnetic field. A paramagnetic centre (i.e., the trapped electron or hole) has a magnetic moment, and this has the same orientation as the magnetic field. In resonance, the magnetic moment is flipped into the opposite direction by absorption of microwave energy which is conducted to the sample from a microwave generator. The amount of absorbed microwave energy is directly proportional to the number of paramagnetic centres, and, in the end, to the age of the sample.

Figure 4: The new Bruker Elexys ESR spectrometer

Figure 5: The Bruker ECS 104 ESR spectrometer

Figure 6: A typical ESR spectrum of tooth enamel. Solid line: natural sample, dotted line: irradiated sample.

Luminescence measurement

The difference between TL and OSL is the method of re-activating the trapped electrons. For TL the sample is heated, leading to light emission from the sample when electrons recombine with luminescence centres (Figure 1). Figures 7 and 8 show the equipment used for TL and OSL measurements. A mineral component of the sample is deposited on a disk which is placed on a heater for TL measurements (Figure 7). At elevated temperatures, electrons are evicted from the traps (see Figure 1) and upon recombining with luminescence centres, the sample emits light. The photons are converted into electric pulses with a photomultiplier. The light emission is then plotted versus the heating temperature, resulting in a glow curve. The configuration for OSL measurements is very similar (Figure 8). Light in a narrow frequency range (green light for measuring quartz, red for feldspars) is shone onto the sample. This activates the electrons in those traps that are light sensitive. When the electrons combine with luminescence centres, the sample emits light. The colour filters in front of the photo-multiplier are used to eliminate quantitatively the light emitted from the light source. The light emission is plotted against the time elapsed after the light source was switched on, resulting in a shine-down curve. In most cases, the light emission used for dating lies in the ultra violet range, but other colours have also been investigated (Krbetschek et al. 1997). For more details on the basic principles of luminescence see McKeever and Chen (1997), for instrumentation Bøtter-Jensen (1997) and for OSL measurements Aitken (1998).

Figure 7. The basic instrumental components for thermoluminescence measurements (after Aitken 1985, 1998). (A) Schematic view (B) The RISO TL reader

Figure 8. The basic instrumental components for OSL measurements (after Aitken 1985, 1998). (A) Schematic view (B) The ELSEC OSL reader

Determination of the dose value, DE

In order to provide reliable results the measured TCD signal must have the following properties;
(i) when the sample is reset it contains an initial signal that can be either experimentally determined or assumed to be zero,
(ii) the signal intensity grows in proportion to the dose received,
(iii) the signals must have a stability time which is at least one order of magnitude higher than the age of the sample,
(iv) the number of traps is constant or changes in a predictable manner,
(v) recrystallization, crystal growth or phase transitions must not have occurred,
(vi) the signals should not show anomalous fading, and
(vii) the signal is not influenced by sample preparation (grinding, exposure to laboratory light, etc.).

The term equivalent dose stems from the fact that the laboratory procedures utilize mono-energetic b or g sources whereas the dose the sample has received in the past is the sum of multi-energetic a, b, g and cosmic rays (see below). Thus, the experimentally determined dose value is the b or g equivalent of the naturally received dose which is frequently called palaeodose, P.

There are several basic techniques for the determination of the DE value. The additive dose method is most widely used (see Figure 3) in ESR and TL applications. I0 is demonstrably zero or assumed to be zero. In TL studies of sediments, I0 is often a significant percentage of IN and has to be determined experimentally. This is either done with a sunlight simulator (Aitken 1985) or a surface sample is collected and its natural intensity is used as I0 for the samples collected further down in a sedimentary profile (e.g., Readhead 1988).

The regeneration technique is an alternative for DE estimation. First IN is measured and the subsequent aliquots are reset to I0. These aliquots are then irradiated with defined laboratory doses and the projection of IN onto the regenerated dose response curve yields DE. The regeneration method has the advantage of involving smaller errors than the additive dose method and, furthermore, the DE result is little dependent on the mathematical model used for the fitting of the data points (in contrast to the additive and partial bleach techniques). However, some samples show sensitivity changes after resetting and the regeneration DE value may differ from the additive DE value. This problem is addressed in the slide technique (Prescott et al. 1993): the natural and reset aliquots are irradiated so that the additive dose response curve overlaps significantly with the regenerated dose response curve. The horizontal distance between the two curves yields the DE value. The overlapping parts of the two dose response curves can be used for the recognition of sensitivity changes, and re-scaling of the regenerated dose response curve can be carried out, if necessary.

The single aliquot regeneration (SAR) method for OSL dating is an alternative to the multi-aliquot techniques described above (Murray and Roberts 1998, Murray and Wintle 2000). Rather than using several aliquots for the establishment of the regeneration dose response curve, the SAR protocol carries all measurements out on the same sample. First, the natural intensity is measured; after this, the sample is irradiated with defined doses. Sensitivity changes are monitored and corrected for by giving the sample a test dose and measuring its response to it. For dating, the SAR protocol is carried out on a range of sub-samples which allows recognition of sample inhomogeneity as well as partial bleaching (Olley et al. 1999). The precision of the SAR protocol (²2%) for OSL is significantly better than any of the multi-aliquot techniques described above.

Determination of the dose rate

The dose rate is calculated from the concentrations of radioactive elements in the sample and its surroundings (only the U and Th decay chains and the 40K decay are of relevance; a minor contribution comes from 87Rb in the sediment), plus a component of cosmic rays. There are three different ionizing rays which are emitted from the radioactive elements:
(i) gamma rays have a range of about 30 cm,
(ii) beta rays (electrons) have a range of about 2 mm,
(iii) alpha rays have only a very short range of about 20 µm because of the large size of the particles. Alpha particles are not as efficient in producing ESR/TL/OSL intensity as beta and gamma rays, therefore the alpha efficiency, which is usually in the range of 0.05 to 0.2, has to be determined (Aitken 1985).

The concentrations of radioactive elements in the sample are usually very different from those of its surroundings. Thus, internal dose rates and external dose rates have to be assessed independently. Furthermore, it is necessary to estimate the cosmic dose rate, which is about 300 µGy/a at sea-level and decreases with depth below ground, and is also dependent on altitude as well as geographic latitude (Prescott and Hutton 1988).

The conversion of the elemental analysis into dose rates are shown in Table 1. For example, if a large sample, such as a burnt flint, contains 2 ppm U, 5 ppm Th and 1% K, and an a-efficiency (k) of 0.1 is assumed or measured, the total internal dose rate is generated by alpha and beta particles:

=(2 x 0.1 x 2780 + 5 x 0.1 x 732) + (2 x 146 + 5 x 27 + 1 x 782) µGy/a
= 2133 µGy/a.

If an external g-dose rate of 1500 µGy/a was measured and a cosmic dose rate of 150 µGy/a was calculated from the depth of the sample below surface, the total dose rate is:

= 2133 + 1500 + 150 µGy/a
= 3783 µGy/a

Dose rate calculations become more complicated when disequilibrium in the U-decay chains or attenuation factors have to be considered (Aitken 1985, 1998, Grün 1989; see also Chapter 1.6). Typical errors in the estimation of the total dose rate are in the range of 4-7%.

External Dose Rate

The calculation of the external dose rate is dependent on the size of the samples (Figure 9). The removal of the outer 50 µm of the sample eliminates all the externally alpha irradiated volume. If the outer 2 mm can be removed, the external beta dose is eliminated. In this case the external dose rate consists of gamma and cosmic rays only. If the samples are smaller, it receives external beta radiation which decreases with depth and therefore attenuation factors have to be considered (Aitken 1985, Grün 1989). Cosmic rays are high energy particles and are attenuated once they penetrate the sedimentary layers. For practical purposes, the cosmic dose rate becomes negligible at a depth of about 20 m.

The external beta dose rate has to be calculated separately from the external gamma dose rate because the beta dose rate is generated from the sediment immediately attached to the sample, whereas the gamma dose rate originates from all sediment that is within a radius of about 30 cm around the sample (see Figure 9). The beta dose rate from the sediment is derived from a chemical analysis of U, Th, and K. In homogeneous sediments, the gamma dose rate can be calculated from a chemical analysis of the bulk sediment. However, if the sediment contains boulders or intercalation of layers with different radioactivity, the gamma dose rate should be measured in situ with a portable, calibrated gamma spectrometer or TL dosimeters. In situ measurements have the advantage that they include the present-day water contents. Water absorbs some ( and ( rays and its presence in the surrounding sediment has to be considered in the calculation of the beta and gamma dose rate (Aitken and Xie 1990).

Figure 9. The different components of natural radiation relevant for dose rate calculations (based on S. Stokes as shown in Aitken 1998: Fig. 2.2.).

Internal Dose Rate

This parameter is mainly generated by alpha and beta rays emitted from elements in the sample. In luminescence studies, it is often assumed that quartz is free of radioactive elements. This assumption will cause only small errors if the external dose rate is relatively high. However, in low dose rate environments, e.g., quartz sand dunes (Prescott and Hutton 1995), the internal dose rate originated by small amounts of radioactive elements may constitute a significant part of the total dose rate. Feldspars are usually free of U and Th, but K has to be measured. Similar considerations as for quartz apply. Quaternary calcitic samples, including speleothems, shells and corals, but also enamel and dentine of teeth, display disequilibrium in the U-decay chains which is actually the basis for U-series dating. U-series disequilibrium affects the average dose rates and has to be taken into account mathematically (Grün 1989).

Alpha efficiencies range from 0.13 (ESR of tooth enamel) to 0.03 (OSL of quartz; Rees-Jones and Tite 1997). Note that alpha efficiencies can not only change from sample to sample but also from technique to technique on the same samples (e.g., Lang and Wagner 1996). Alpha efficiencies can be routinely measured in luminescence studies whereas the size requirements for ESR measurements make its assessment difficult (Grün and Katzenberger-Apel 1994). If the samples are small, the internal alpha and beta dose rates are not 100% absorbed and self-absorption factors have to be calculated (1 minus the external attenuation factors).

The dose rate can be measured with a precision of about 5%. The overall uncertainty of a trapped charge dating result may be as low as 6 to 7%.

Applications of trapped charge dating techniques

When reviewing the literature, one finds many cases where trapped charge dating results have challenged conventional ideas, as have, for example, the TL and ESR results from early modern hominid sites in the Levant (Valladas et al. 1988, Mercier et al. 1995, Grün and Stringer 1991). One immediate reaction was the dismissal of the results. This partly relates to the fact that there are very few other dating techniques available in archaeology for the pre-radiocarbon time range of older than 40,000 years. Thus, conventional chronological wisdom is sometimes based on preconception. On the other hand, in some cases it has turned out that TCD results have been erroneous because all TCD methods are still in a rapid phase of development and some of the pitfalls have not yet been recognized. However, before disregarding any evidence one has to keep in mind that reliable chronologies are the result of the synthesis of all evidence available, in an ideal case based on detailed archaeological, palaeoanthropological, palaeontological, palynological, sedimentological, geochemical and geochronological analysis.

ESR dating of tooth enamel

ESR spectroscopy was first applied by Ikeya in 1975 when he dated speleothems from Akiyoshi Cave (Figure 10). The main application of ESR in archaeological studies is dating of tooth enamel. ESR dating of speleothems, spring deposited travertines and shells in archaeological contexts has been recently reviewed (Grün 1997) and there is little to add in this respect.

ESR dating of tooth enamel has been applied to a variety of archaeological and palaeoanthropological sites (see Grün and Stringer 1991, Grün 1997, Rink 1997). The method is applicable between a few thousand years and, in exceptional circumstances, up to a few million years (Schwarcz et al. 1994). For details on sample preparation, spectrum evaluation and dose rate determination see Grün (1989, 1997, 2000a-c) and Rink (1997).

ESR dating of teeth encompasses a particular problem: U-uptake. Teeth, bones and some shells show post-depositional U-uptake and this effect further complicates the dose rate determination. The process of uranium uptake cannot normally be determined exactly. Dentine usually accumulates much more U than enamel (by a factor of 10 to 100; Grün and Taylor 1996). For teeth, two models have been suggested (Ikeya 1982);
(i) U-accumulation shortly after burial of the tooth (early U-uptake, EU),
(ii) continuous U-accumulation (linear U-uptake, LU).

Figure 10. Nature 255 with the first paper on ESR dating: Ikeya, M. (1975) Dating a stalactite by electron paramagnetic resonance. Nature 255: 48-50.

As long as the U-concentrations in the components of a tooth are low (<2 ppm in dentine) the discrepancy between EU and LU ages is less than 10%. With increasing U-concentrations this intrinsic uncertainty increases very rapidly. In the extreme, the LU model age is twice the EU age estimate. New developments in ESR and U-series dating have shown a strategy to overcome this problem (see below). However, if there are no specific indications to justify a particular U-uptake model, age estimates for both models ought to be given. The most probable age of most samples is somewhere between the two estimates (for a compilation of ESR results on teeth see Grün and Stringer 1991). Most teeth older than the last interglaciation, which is widely accompanied by a more pluvial climate, have accumulated considerable amounts of uranium. Therefore most ESR age estimates of such teeth are associated with large uncertainties (>25%). It is important to note that the EU age estimate is the minimum possible age. An overall precision of better than 7% can be obtained for teeth with low U-concentrations (e.g., Grün et al. 1990).

The application of ESR dating to the site of Hexian (Grün et al. 1998) is a prime example which illustrates the limited value of stand-alone ESR and U-series analysis on faunal materials.

However, the combination of both methods can lead to very precise results. Portions of several human crania were found at the site of Hexian, Anhui Province, China (Huang et al. 1982). The human remains were attributed to Homo erectus and, based on some progressive features, correlated with the specimen HIII found in Layer 3 of Zhoukoudian (Wu and Dong 1985). The site had previously been dated by Huang et al. (1995a, 1995b) who obtained ESR ages in the range of 160 to 220 ka (EU) and 250 to 350 ka (LU), and Chen and Yuan (1988) who reported U-series results of 130 to 220 ka (combined 231Pa/230Th). The ESR results of Grün et al. (1998) are shown in Figure 11A. These cover a large time span of 344±48 ka (EU) to 465±94 ka (LU). Even though these ages span the time range of about 300 to 550 ka, they were considerably older than the previously-obtained results. Grün et al. (1998) also carried out U-series analyses on different constituents of two teeth obtaining U-series ages of between 150 to 550 ka (Figure 11B). If these results were reported without further consideration they would clearly demonstrate the very limited value of ESR and U-series analysis of faunal materials (mainly due to the unknown U-uptake history).

Figure 11. (A) ESR age estimates on teeth from Hexian (after Grün et al. 1998). (B) Age estimates for samples 1117 and 1118 (EN = Enamel; DE = dentine). Both the ESR and U-series estimates imply that the two teeth have distinctively different ages. The combined U-series/ESR age estimates are virtually indistinguishable and give the best age estimate for the site (412±25 ka). (C) Modelled U-uptake histories. Sample 1117 shows U-uptake which is near to a closed system model whereas the constituents of sample 1118 are closer to a linear uptake model.

Grün et al. (1988) developed a model that can combine ESR and U-series data and solve for U-uptake history (see also McDermott et al. 1993, Grün and McDermott 1994). Both techniques are dependent on uranium uptake, but to a different extent. With the independent measurement of ESR and U-series dating results it becomes possible to establish two equation systems that can solve (i) for age and (ii) for a single parameter U-uptake equation. Two teeth from Hexian were analysed by ESR and U-series and the combination of both methods can explain the very different U-uptake history that both teeth have experienced (Figure 11C). The combined U-series/ESR data result in a very tight age estimation of 412±25 ka (Figure 11B).

This result was first greeted with scepticism, because it implied that the Hexian calvaria were contemporaneous with the morphologically less advanced Homo erectus specimen from Zhoukoudian (LI-LIII) rather than more similar HIII (Wu et al. 1985). However, new precise TIMS U-series results on speleothem layers from layer 2 and 3 at Zhoukoudian (Shen et al. 1996) show that these layers are also considerably older than originally proposed by Wu et al. (1985). As a result, a morphological correlation between Hexian and Zhoukoudian HIII is now supported by chronological evidence, but both specimens are considerably older than previously thought (Grün in press).

TL dating of burnt flint

The most comprehensive review of luminescence applications in archaeology was recently published by Roberts (1997). Thermoluminescence dating was developed in the late 1960s and early 1970s for the dating of pottery (Aitken 1985). TL readers are very sensitive to light emission and samples with ages of a few tens of years can be reliably dated. One of the main application of TL pottery analysis lies in provenance authentication (Stoneham 1983). Wintle and Huntley (1982) pioneered TL dating of sediments in the late 1970s. For sediment dating, TL has been replaced by OSL in recent years, mainly because OSL selectively measures the electrons caught in light sensitive traps whereas the TL signal results from a series of traps with different light sensitivities (see below).

Today, dating of burnt flint is the main application of TL in archaeological contexts. The method is applicable between a few hundred years and a few hundred thousand years, covering the time span of the controlled use of fire by humans. For details on TL sample collection and preparation as well as dose rate evaluation see Aitken (1985), for spectrum evaluation and other laboratory procedures see Wintle (1997) and for specific details on TL dating of burnt flint see Mercier et al. (1995).

TL dating of burnt flint from the site of Qafzeh, Israel, revolutionized our thinking on modern human evolution (Valladas et al. 1988). It has been known for many years that modern humans evolved from or replaced Neanderthals in Europe at sometime in the range of 35,000 to 40,000 years. This Eurocentric picture was then transposed to other areas and it was therefore thought that Neanderthal sites in the Levant, such as Kebara and Tabun were somewhat older than 45,000 years and that the sites where early modern humans were found, Qafzeh and Skhul, were somewhat younger.

The site of Qafzeh contained remains of at least 20 hominids which are generally regarded as early modern humans. The publication of TL age estimates of 92±5 ka carried out on 20 burnt flint samples from layers XVII to XXIII (Figure 12; Valladas et al. 1988) caused a serious disturbance in palaeoanthropological circles. Although Bar-Yosef and Vandermeersch (1981) had already suggested that the site may be as old as 100,000 years, others rejected the TL dates because it was felt that the dates could not be verified independently and may be inaccurate, were still experimental, or would not stand the test of time (Grün and Stringer 1991). Subsequent ESR (Schwarcz et al. 1988) and U-series (McDermott et al. 1993) dating studies on animal teeth as well as non-destructive U-series dating of the human remains (Yokoyama et al. 1997) confirmed the antiquity of the site. Today, the early arrival of modern humans in the Levant is widely accepted. For a summary of the dating results of palaeoanthropological sites in the Levant see Grün and Stringer (1991), Mercier et al. (1995) and Grün (1997).

Figure 12: TL dating of burnt flint from Qafzeh (after Valladas et al. 1988). 20 samples from Layers XVII to XXIII into which more than 20 skeletons of early modern humans were buried. The dates show no trend with depth indicating rapid deposition of the layers. The weighted mean age is 92±5 ka.

OSL dating of sediment

OSL dating was introduced by Huntley et al. (1985). In recent years, OSL has replaced TL for the dating of sediments, because;
(i) OSL uses the same methods as nature to reset the sample (light activation rather than thermal activation),
(ii) OSL measures the light sensitive traps only and the signal is not overlapped by light insensitive signals, thus in OSL I0 is very close to zero whilst in TL I0 is usually a high percentage of IN,
(iii) the OSL signals is far more light sensitive than TL (e.g., Godfrey Smith et al. 1988).

For details on sample collection and preparation, as well as dose rate evaluation, see Aitken (1998), for spectrum evaluation and other laboratory procedures see Wintle (1997). Luminescence dating of sediment can be applied between a few years and about one million years (Huntley et al. 1994).

Luminescence dating of sediments has found wide applications in archaeology (Roberts 1997). Luminescence dating was decisive in demonstrating that the early colonization of Australia took place well before 40,000 years ago, i.e., prior to the invasion of Homo sapiens into Europe (see above). Roberts et al. (1990) used TL dating of sediments at the Malakunanja II shelter in Arnhem Land, Northern Territory, Australia, demonstrating that the earliest artifacts at that site were deposited around 55 ka. Subsequently, a luminescence study was carried out at Deaf Adder Gorge (Roberts et al. 1994), a site about 70 km south of Malakunanja II. As shown in Figure 13, both luminescence methods show very good agreement with calibrated radiocarbon results. In contrast to radiocarbon, the lower layers can be dated with OSL (and TL) and the earliest artifacts are bracketed between 53.4±5.4 and 60.3±6.7 ka. The rubble found in the lower levels at the site could make TL dating questionable, but the short exposure times required for OSL make age over-estimations unlikely. The early arrival of anatomically fully modern humans was confirmed in multi-dating study (U-series, ESR, OSL) of the Lake Mungo 3 skeleton which involved direct dating of the human material (Thorne et al. 1999).

Figure 13: Luminescence dating results from Deaf Adder Gorge, NT, Australia (after Roberts et al. 1994). The luminescence data show near complete resetting at the top of the sedimentary column. Within the radiocarbon dating range, both luminescence methods yield very good agreement with the calibrated radiocarbon age estimates. The lowest artifacts are bracketed between 53.4±5.4 and 60.3±6.7 ka, supporting the earlier claim of Roberts et al. (1990) that the Australian continent was colonized well before 40,000 years ago.

Recently, it was claimed by Fullager et al. (1996) that the site of Jinmium, NT, may be as old as 116 ka. The TL results of this study were indeed obstructed by rubble that decayed in the sedimentary layers without being sufficiently exposed to sunlight. Analysis of the TL glow curves (Spooner 1998) predicted that the maximum age of the site was between 10-30 ka. Roberts et al. (1998) confirmed with AMS radiocarbon and OSL dating that the site was younger than 10,000 years.

Outlook

Trapped charge dating will play an increasing role for the establishment of archaeological chronologies in the pre-radiocarbon time scale. Considering the nearly universal applicability of OSL dating, the routine application of this method should lead to a revolution in our understanding of archaeological processes similar to the introduction of radiocarbon dating in the 1950s and 1960s.

The main advance in ESR dating of tooth enamel lies in the direct dating of human material using enamel fragments (Grün et al. 1996) and, possibly, whole teeth. Uranium of enamel fragments can be measured using laser ablation ICP-MS which allows the precise analysis of very small samples (Eggins et al. 1998) and can be used to establish uranium profiles along very thin holes (50 µm diameter) which are nearly invisible. New generation multi-collector ICP sector mass spectrometers (Halliday et al. 1998) will significantly reduce laboratory preparation for U-series dating. Thus, combined ESR/U-series dating could develop as a routine method within the next few years. ESR measurements can also provide important restraints when hominid fossils are dated non-destructively by gamma spectrometric U-series dating (Simpson and Grün 1998). The key for the establishment of reliable chronologies for a given site lies in the application of as many dating techniques as possible (e.g., Abeyratne et al. 1997, Thorne et al. 1999).

References

Abeyratne M, Spooner N, Grün R and Head J (1997). Multidating studies of Batadomba Cave, Sri Lanka. Quaternary Science Reviews 16:243-55.

Adamiec G and Aitken MJ (1998). Dose-rate conversion factors: update. Ancient TL 16:37-50.

Aitken MJ (1985). Thermoluminescence Dating. Academic Press: New York.

Aitken MJ (1997). Luminescence dating. In Taylor RE and Aitken MJ (eds) Chronometric and Allied Dating in Archaeology. Plenum: New York, 183-216.

Aitken MJ (1998). Optical Dating. Oxford University Press: Oxford.

Aitken MJ and Xie J (1990). Moisture correction for annual gamma dose. Ancient TL 8:6-9.

Bailey RM (1998). The Form of the Optically Stimulated Luminescence Signal of Quartz: Implications for Dating. Unpublished PhD Dissertation, Royal Holloway, University of London.

Bar-Yosef O and Vandermeersch B (1981). Notes concerning the possible age of the Mousterian layers in Qafzeh Cave. InCauvin J and Sanlaville P (eds) Préhistoire du Levant. CNRS: Paris, 281-6.

Berger GW (1990). Regression and error analysis for a saturating-exponential-plus-linear model. Ancient TL 8:23-5.

Bøtter-Jensen L (1997). Luminescence techniques: instrumentation and methods. Radiation Measurements 27:749-8.

Chen TM and Yuan S (1988). Uranium-series dating of bones and teeth from Chinese Palaeolithic sites. Archaeometry 30:59-76.

Eggins SM, Kinsley LPJ and Shelley M (1998). Deposition and elemental fractionation processes during atmospheric pressure laser sampling for analysis by ICP-MS. Applied Surface Science 127-129:278-86.

Fullagar RLK, Price DM and Head LM (1996). Early human occupation of northern Australia: archaeology and thermoluminescence dating of Jinmium rock-shelter, Northern Territory. Antiquity 70:751-73.

Godfrey-Smith DI, Huntley DJ and Chen WH (1988). Optical dating studies of quartz and feldspar sediment extracts. Quaternary Science Reviews 7:373-80.

Grün R (1989). Electron spin resonance (ESR) dating. Quaternary International 1:65-109.

Grün R (1997). Electron spin resonance dating. In Taylor RE and Aitken MJ (eds) Chronometric and Allied Dating in Archaeology. Plenum: New York, 217-61.

Grün R (in press). Electron spin resonance dating. In Ciliberto E and Spoto G (eds) Modern Analytical Methods in Art and Archaeology. Wiley: New York.

Grün R and Brumby S (1994). The assessment of errors in the past radiation doses extrapolated from ESR/TL dose response data. Radiation Measurements 23:307-15.

Grün R and Katzenberger-Apel O (1994). An alpha irradiator for ESR dating. Ancient TL 12:35-8.

Grün R and McDermott F (1994). Open system modelling for U-series and ESR dating of teeth. Quaternary Science Reviews 13:121-5.

Grün R and Stringer CB (1991). ESR dating and the evolution of modern humans. Archaeometry 33:153-99.

Grün R and Taylor L (1996). Uranium and thorium in the constituents of fossil teeth. Ancient TL 14:21-5.

Grün R, Schwarcz HP and Chadam JM (1988). ESR dating of tooth enamel: coupled correction for U-uptake and U-series disequilibrium. Nuclear Tracks and Radiation Measurements 14:237-41.

Grün R, Beaumont P and Stringer CB (1990). ESR dating evidence for early modern humans at Border Cave in South Africa. Nature 344:537-9.

Grün R, Brink JS, Spooner NA, Taylor L, Stringer CB, Franciscus RG and Murray AS (1996). Direct dating of Florisbad hominid. Nature 382:500-1.

Grün R, Huang PH, Huang W, McDermott F, Stringer CB, Thorne A and Yan G (1998). ESR and U-series analyses of teeth from the palaeoanthropological site of Hexian, Anhui Province, China. Journal of Human Evolution 34:555-64.

Halliday AN, Lee DC, Christensen JN, Rehkämper M, Yi W, Luo X, Hall CM, Ballentine CJ, Pettke T and Stirling C (1998). Applications of multiple collector-ICPMS to cosmochemistry, geochemistry, and paleoceanography. Geochimica et Cosmochimica Acta 62:919-40.

Huang HP, Fang DS and Ye YX (1982). Preliminary study of the fossil hominid skull and fauna of Hexian, Anhui. Vertebrata Palasiatica 20:248-56.

Huang PH, Zheng LZ, Quan YC, Liang RY, Xu YH, Fang YS and Fang DS (1995a). Preliminary study on ESR dating of Hexian Man and its fauna. Nuclear Techniques 18:491-4 (in Chinese).

Huang PH, Zheng LZ, Quan YC, Liang RY, Xu YH, Fang YS and Fang DS (1995b). Study on age of Hexian Man of Homo erectus. Acta Anthropologia Sinica 14:262-5 (in Chinese).

Huntley DJ, Godfrey-Smith DI and Thewalt MLW (1985). Optical dating of sediments. Nature 313:105-7.

Huntley DJ, Hutton JT and Prescott JR (1994). Further thermoluminescence dates from the dune sequence in the south-east of South Australia. Quaternary Science Reviews 13:201-7.

Ikeya M (1975). Dating a stalactite by electron paramagnetic resonance. Nature 255: 48-50.

Ikeya M (1982). A model of linear uranium accumulation for ESR age of Heidelberg, Mauer, and Tautavel bones. Japanese Journal of Applied Physics 21:L690-2.

Ikeya M (1993). New Applications of Electron Spin Resonance - Dating, Dosimetry and Microscopy. World Scientific: Singapore.

Jonas M (1997). Concepts and methods of ESR dating. Radiation Measurements 27:943-3.

Katzenberger O and Willems N (1988). Interferences encountered in the determination of AD of mollusc samples. Quaternary Science Reviews 7:485-9.

Krbetschek MR, Götze J, Dietrich A and Trautmann T (1997), Spectral information from minerals relevant for luminescence dating. Radiation Measurements 27:695-748.

Lang A and Wagner GA (1996). Infrared stimulated luminescence dating of archaeosediments. Archaeometry 38:129-41.

McDermott F, Grün R, Stringer CB and Hawkesworth CJ (1993). Mass-spectrometric U-series dates for Israeli Neanderthal/early modern hominid sites. Nature 363:252-5.

McKeever SWS and Chen R (1997). Luminescence models. Radiation Measurements 27:625-61.

Mercier N, Valladas H and Valladas G (1995). Flint thermoluminescence dates from the CFR laboratory at Gif: contributions to the study of the chronology of the Middle Palaeolithic. Quaternary Science Reviews 14:351-64.

Murray AS and Roberts RG (1998). Measurement of the equivalent dose in quartz using a regenerative-dose single-aliquot protocol. Radiation Measurements 29:503-15.

Murray AS and Wintle AG (2000). Luminescence dating of quartz using an improved single-aliquot regenerative-dose protocol. Radiation Measurements 32: 57-73.

Olley J, Caitcheon G and Murray A (1998). The distribution of apparent dose as determined by optically stimulated luminescence in small aliquots of fluvial quartz: implications for dating young sediments. Quaternary Science Reviews 17:1033-40.

Prescott JR and Hutton JT (1988). Cosmic ray and gamma ray dosimetry for TL and ESR. Nuclear Tracks and Radiation Measurements 14: 223-7.

Prescott JR and Hutton JT (1995). Environmental dose rates and radioactive disequilibrium from some Australian luminescence dating sites. Quaternary Science Reviews 14:439-48.

Prescott JR, Huntley DJ and Hutton JT (1993). Estimation of equivalent dose in thermoluminescence dating - the Australian slide method. Ancient TL 11:1-5.

Readhead ML (1988). Thermoluminescence dating study of quartz in aeolian sediments from southeastern Australia. Quaternary Science Reviews 7:257-64.

Rees-Jones J and Tite M (1997). Optical dating results for British archaeological sediments. Archaeometry 39:177-7.

Rink WJ (1997). Electron spin resonance (ESR) dating and ESR applications in Quaternary science and archaeometry. Radiation Measurements 27:975-1025.

Roberts RG (1997). Luminescence dating in archaeology: from origins to optical. Radiation Measurements 27:819-892.

Roberts RG, Jones R and Smith MA (1990). Thermoluminescence dating of a 50,000 year-old human occupation site in northern Australia. Nature 345:125-9.

Roberts RG, Jones R, Spooner NA, Head MJ, Murray AS and Smith MA (1994). The human colonization of Australia: optical dates of 53,000 and 60,000 years bracket human arrival at Deaf Adder Gorge, Northern Territory. Quaternary Science Reviews 13:575-83.

Roberts R, Bird M, Olley J, Galbraith R, Lawson E, Laslett G, Yoshida H, Jones R, Fullagar R, Jacobson G and Hua Q (1998). Optical and radiocarbon dating at Jinmium rock shelter in northern Australia. Nature 393:358-62.

Schwarcz HP, Grün R, Vandermeersch B, Bar-Yosef O, Valladas H and Tchernov E (1988). ESR dates for the hominid burial site of Qafzeh in Israel. Journal of Human Evolution 17:733-7.

Schwarcz HP, Grün R and Tobias PV (1994). ESR dating studies of the australopithecine site of Sterkfontein, South Africa. Journal of Human Evolution 26:175-81.

Shen G, Ku TL, Gahleb B and Yuan Z (1996). Preliminary results on U-series dating of Peking Man site with high precision TIMS. Acta Anthropologica Sinica 15:210-17.

Simpson JJ and Grün R (1998). Non-destructive gamma spectrometric U-series dating. Quaternary Science Reviews 18:1009-22.

Spooner NA (1998). Human occupation at Jinmium, northern Australia: 116,000 years ago or much less? Antiquity 72:173-8.

Stoneham D (1983). Porcelain dating. PACT 9:227-39.

Thorne A, Grün R, Mortimer G, Spooner NA, Simpson JJ, McCulloch MT, Taylor L and Curnoe D (1999). Australia's oldest human remains: age of the Lake Mungo 3 skeleton. Journal of Human Evolution 36: 591-612.

Valladas H, Reys JL, Joron JL, Valladas G, Bar-Yosef O and Vandermeersch B (1988). Thermoluminescence dating of Mousterian 'Proto-Cro-Magnon' remains from Israel and the origin of modern man. Nature 331:614-6.

Wintle AG (1997). Luminescence dating: laboratory procedures and protocols. Radiation Measurements 27:769-817.

Wintle AG and Huntley DJ (1982). Thermoluminescence dating of sediments. Quaternary Science Reviews 1:31-53.

Wu RK and Dong XR (1985). Homo erectus in China. In Wu RK and Olson JW (eds) Palaeoanthropology and Palaeolithic Archaeology in the People.s Republic of China, Orlando:Academic Press: Orlando, 79-89.

Wu RK, Ren ME, Zhu XM, Yang ZG, Hu CK, Kong ZC, Xie YY and Zhao SS (1985). Multi-disciplinary Study of the Peking Man Site at Zhoukoudian. Science Press: Beijing (in Chinese).

Yokoyama Y, Falgueres C and de Lumley MA (1997). Datation directe d'un crâne Proto-Cro-Magnon de Qafzeh par la spectrométrie gamma non destructive. Comptes Rendus de l'Académie des Sciences Serie IIa 324:773-9.