Powered by
JSPWiki v2.8.2
g2gp 17-01-2009
View PDF
This is version . It is not the current version, and thus it cannot be edited.
[Back to current version]   [Restore this version]

Thermoluminescence Dating: A Guide to Good Practice#

Nikolaos A. Kazakis & Nestor C. Tsirliganis, ATHENA#

2016#

Section 1. Introduction#

Section 2. Acquiring and Processing Thermoluminescence Data#

Section 3. Archiving Thermoluminescence Data#

Section 4. Case Study: Dating Pottery from Ancient Abdera (Greece)#

Bibliography#

DANS logo
ARIADNE logo

Previous | Next | Contents


Section 1. Introduction#

1.1 Scope of the Guide#

The present document serves as a guide to good practice for the collection and archival of data produced by Thermoluminescence (TL) measurements (analyses) of archaeological materials, such as ceramics, in the frame of the archaeological research. This guide does not elaborate on the methods involved in the Thermoluminescence analysis in general, but aims at informing researchers involved in archaeological studies about the key elements and the important metadata that should be documented in Thermoluminescence analyses during the determination of the age of archaeological materials.

It should be noted that specific metadata can be very important since they are descriptive of the procedure followed for the treatment of the physical samples and the protocols/techniques used during the analysis which are solidly interconnected to the produced data. Special attention should be given documenting such metadata, which would allow not only the easy archival, but the reuse of the produced datasets. This would ensure the re-evaluation of the samples and the inter-comparison of the results among various laboratories.

1.2 What is Thermoluminescence?#

Thermoluminescence is the emission of light during the heating of a solid sample, usually an insulating one, which has been previously excited. The source of the emitted light is the initial excitation, which is typically created by irradiation, while heating acts as a trigger which contributes to the releasing of this accumulated energy.

To elaborate on the above, a solid sample, e.g. ceramic can be excited by ionizing radiation at a certain relatively low temperature. This irradiation can either take place in the laboratory or in a radiative environment. However, another version, more related to the archaeological use of thermoluminescence, is when a material is irradiated by the radiation field in its natural surrounding. At the end of this stage, the sample is placed in an appropriate instrument where its temperature is gradually raised at a constant heating rate. Then the emitted light (photons) is recorded as a function of the temperature using a light sensitive detector, such as a photomultiplier (Chen & McKeever, 1997) and a glow-curve is acquired (Fig. 1).

Glow curve

Figure 1: Typical glow curve of a pottery sample recorded after artificial radiation administered in the laboratory


1.3 Applications of Thermoluminescence in Archaeology#

a) Dating

The most common and important application of Thermoluminescence in Archaeology is dating of archaeological objects, mainly ceramics, such as pottery, bricks or terracotta. The rapid heating of such objects up to 500°C results in the weak, but measurable emission of light, which stems from some of the constituents minerals. The TL from the specimen is mostly due to TL sensitive mineral inclusions (mostly quartz) in the host clay matrix of the pottery. The basis of TL dating in archaeology is a definite event, the kiln firing. This event is regarded as the starting of the "TL clock" for archaeological dating, since any TL previously stored in the object is completely erased during the firing process. After the onset of the TL clock, the pottery starts to build up TL due to its exposure to the weak flux of nuclear radiation emitted by radio-active impurities in the pottery and the surrounding burial soil, the most active of which are the potassium-40 (K-40), thorium (Th) and uranium (U). The above radioisotopes have half-lives of more than 10years, which results in a constant radiation flux.

Based on the above, the amount of the material's Thermoluminescence can provide an estimation of the elapsed time since the pottery was fired. This is better described by Eq. 1 (Aitken, 1985), while the whole process is illustrated in Fig. 2.

Glow curve

Figure 2: Flow chart of the process for pottery dating


b) Authenticity testing of ceramic objects

Thermoluminescence technique can also be used in the battle against art forgery, since it has been widely employed in the field of authenticity testing. Porcelain and/or ceramic objects can be the subject of such a test with Thermoluminescence in order to determine their archaeological value. A modern object which has only been fired recently will carry only a trivial level of stored TL energy, compared to its ancient prototype (Fleming, 1979). Consequently, it is relatively easy to distinguish a fake object whose age is less than a hundred years rather than more than five hundred.

c) Pottery provenance and transactions among ancient communities

Thermoluminescence can also provide information about the provenance of pottery, which can further shed light on the local production and the socio-economical relationships amongst ancient communities. The above can complement the archaeologists' observations of the typology and the recognition of special fabric textures of the excavated pottery and lead to a more concrete decision about their provenance. Several studies have been accomplished towards this direction (e.g. Vaz et al., 1997; Rasmussen, 2001) and they are based on the differences of the TL sensitivity and/or characteristics of pottery of different provenance. These variances are the outcome of the different tempering materials and concentration of the trace elements found in the clay used for the manufacture of pottery of different source (geographical region).

1.4 Current issues or concerns#

a) Efficiency and diffusion of the Thermoluminescence data

In many cases, Thermoluminescence data of various archaeological objects (mainly pottery) are produced to support the work of archaeologists, without being stored with additional information about the object itself (e.g. type), mainly because the archaeologists most often neglect to provide such information. The above limits the reusability of TL data for the benefit of the wide archaeological community, since they are only produced to give a certain answer (e.g. the age of an object). The above situation is further deteriorated by the fact that such data are not published, since only the result of the process (e.g. age) is of interest for the archaeologists. However, TL data could provide further information about the temporal or spatial distribution of certain pottery when compared with similar data of other objects.

b) The need for additional metadata

To ensure the use of TL data in other archaeological studies and to allow the comparison between data acquired in different laboratories or the repetition of the same measurement several metadata should also be stored. These metadata include both details of the under-study object (e.g. sampling procedure, excavation site, picture, etc) and the measuring conditions (method, number of sub-samples, doses, temperatures, etc).

1.5 Data and metadata#

Definition of Data and Metadata is a controversial issue with multifold approaches, since it extensively depends on the usage purpose. In the case of Thermoluminescence, raw data, i.e., results of every measurement (directly produced by the scientific instruments without any intermediate calculations), are considered as data. In fact they are temperature-photon counts pairs, whose population depends on the heating rate and final temperature selected for the measurement. On the other hand, metadata comprise all subsequent information that stems from the post processing of the raw data along with every other sample-related and pre-measurement (sample treatment, operational conditions etc) information which allows the interpretation of the results or the comparison with previous measurements. Examples of the above include the location of the archaeological material, the method/technique, the measurement parameters etc.


Previous | Next | Contents

Previous | Next | Contents


Section 2. Acquiring and Processing Thermoluminescence Data#

2.1 Project Planning and requirements#

The requirements of a project planning are not standard and can vary depending on various factors, such as the size and/or available mass of the sample, desired precision, the sampling procedure, the archaelogist's estimation of the object's age. All the above information are significant and will determine the protocol/technique that will be used for the TL measurement (fine grain, coarse grain etc), the number of the sub-samples (aliquots) that will be prepared and the doses that will be applied for the calibration.

2.2 Sources of data and file types#

Production of TL data is accomplished worldwide either by commercially available or custom made instruments. As a result, both the source (platform or software) and the type of the data are instrument dependent and proprietary. All the same, in all cases, the TL-data can easily be converted to an ASCII text format, which allows their easy processing with spreadsheet programs, such as MS Excel. A difference, however, can be the ability of the various types of software used to store along with the data, the measurement-associated metadata, which could hamper or even block their comparison with other TL-data.


Previous | Next | Contents

Previous | Next | Contents


Section 3. Archiving Thermoluminescence Data#

3.1 Deciding what to archive#

The size of the acquired files with the TL-data are very small (in most cases only few kbytes) and thus there is not practically a dilemma about choosing the appropriate files to archive. In any case, the resulting TL-data file(s) should always be archived along with the file(s) which describe the detailed sequence of the steps followed (complete experimental conditions and procedure) during the TL measurement of the various samples or sub-samples. This would allow the prompt match of the results with the appropriate measurement and sample.

Furthermore, it is a good practice to store and archive additional information describing the objects, such as photos, archaeological observations and information, related research reports in order to enhance the usability of the TL-data by other researchers or future studies.

3.2 Deciding how to archive#

To the Authors' best knowledge there are not any established standards which would dictate the content and the related fields that a database of TL data should be composed of. As a result, TL data are mainly stored in custom-made repositories of "arbitrary" structure. In any case such a database should always be structured on the premise that TL data should be publicly available and freely shared to other researchers in order to enhance the archaeological research and contribute to the cultural heritage management.

Apparently, the minimum requirements regarding the structure of such a database should at least fulfill the need to allow future re-evaluation or verification of certain Thermoluminescence measurements. However, one should always have in mind that it is a good practice to build such a generic repository covering a wide range of metadata, which would allow the inter-comparison of the results among different TL-data-producing instruments and their use in follow-up research.

3.3 File types for archiving#

As previously stated, the format of the TL data is native or proprietary. Consequently, besides their original format, raw data should also be converted and stored in ASCII format in order to enhance their readability by other researchers and in other studies. Furthermore, most metadata have descriptive nature and thus should be stored preferably in rich text format (.rtf), while images/photographs of the objects should be stored in a commonly used image file format (jpg or tiff being the most appropriate).

3.4 Archiving TL-related metadata#

The required metadata could be divided into three levels, Project level, Object level and Measurement level metadata. Each one provides different information. The project level metadata mainly refers to information related to the archaeological site at which the object(s) was found. The object level metadata provide information (mainly archaeological) and description of every object to be measured. Finally, the measurement level metadata is related to all available information about the experimental/measuring procedure including any post-measurement data treatment. Based on the above a project may include multiple measurements of several objects found in the same excavation site.

Although the more metadata the better, in most cases a certain number of metadata is enough to establish their usability for future purposes. Table 1 presents all metadata that should be stored when archiving TL data.

Table 1. Metadata fields (mandatory, desirable and optional) that should be filled when archiving TL data. *M: Mandatory, D: Desirable, O: Optional

ElementDescriptionField importance*
Project level
Project nameName of the projectM
Project dateDate the project startedM
Name of stakeholderName of excavation, museum, collection, archaeological project etc, where the objects are coming fromM
Object locationGeographical area, where the object(s) was found or stored or excavation site or name of MuseumM
Site geographical longitudeLongitude of the excavation siteD
Site geographical latitudeLatitude of the excavation siteD
Object level
Object ID codeThe ID of the object as registered in the excavation or museumM
Object lab IDThe ID of the object registered by the laboratory performing the measurementM
Object lab dateDate the object was delivered to the laboratoryM
Batch protocol numberThe protocol number of the batch in which the object belongsD
Group object IDThe ID of the group at which the object belongs (archaeological excavation)O
Excavation sectionThe section of the excavation site at which the object was foundO
Area descriptionDescription of the area where the object was found (e.g. in a kiln)M
Object depthThe depth from the surface where the object was foundM
Object typeThe type of the object (e.g. soil, clay, sherd, brick, other)M
Object categoryThe category where the object belongs (mainly applicable for ceramics) according to the archaeological classificationO
Object dimensionsThe dimensions of the object in cm (length, width, height). Not applicable for powder objects (e.g. soil)O
Object descriptionFree text describing the object (colors, texture, etc)O
Object picturePicture of the object in tiff or jpg formatD
Related documentsOther related documents (e.g. other object pictures, pictures from the area where the object was found, bibliography etc.)O
Measurement level
Sample preparationFree text describing the preparation/pretreatment of the sample (cleaning, sieving etc)M
Measuring techniqueTechnique used for the TL measurement (e.g. fine grain, coarse grain etc)M
Grain sizeSize of the grains used for the measurementM
Number of aliquotsThe number of sub-samples used for the measurement for statistical purposesM
Measurement protocolFile describing the steps and the operational conditions of the measurement protocolM
FiltersDetails of the optical filters used for the TL measurementM
Instrument(s)Instrument(s) used for the TL measurement (brand name, model and serial number)M
Instrument(s) detailsOther instrument details (especially those that are unique for the specific instrument (s) used)O
LaboratoryName of Laboratory performing the measurementM
Researcher(s)Name of the Researcher(s) who conducted the measurementsD
Start measurement dateThe date the measurement of the sample startedM
End measurement dateThe date the measurement of the sample endedM
Date of resultsThe date the results (age) are producedD
Technical Report fileTechnical report of the measurement (.doc or .pdf file)M
Annual dose estimationMethods and/or assumptions used for the determination of the annual doseD
Method for potassium (if applicable)Method used for the determination of the sample's potassium (K) concentration (ICP, XRF etc)D
Method for uranium/thorium (if applicable)Method used for the determination of the sample's uranium (U) and thorium (Th) concentrationD

Previous | Next | Contents

Previous | Next | Contents


Section 4. Case Study: Dating pottery from Ancient Abdera (Greece)#

4.1 Project Background/Scope#

The site of Abdera, city of Democritos, was settled in the middle of the 7th century B.C. by colonists from Clazomenae and in 545 B.C. by the inhabitants of Teos. The site has undergone various changes during its history, from a prosperous city during the Roman period to a cemetery during the Byzantine period. According to the archaeologists, ceramics found in the site are both local and imported products, while their age may vary. The scope of the project was the dating of several ceramic objects (mainly pottery) found in the archaeological site of ancient Abdera in Greece using Thermoluminescence.

4.2 Project deliverables#

For each object the age and the corresponding error were calculated following the established TL protocol along with all necessary complementary measurements as previously mentioned. The datasets which were selected for archival include:

  • Raw TL-data file
  • Converted TL-data file in ASCII format
  • Raw data from every other measurement conducted for the estimation of the "annual dose"

4.3 Archival metadata example#

In order to present an example of a complete metadata set, one of the analyzed objects was selected. Metadata of the Project level are filled once, since they refer to the archaeological site and are common for all objects of the project. On the other hand, the rest of the metadata (object and measurement level) are different for each object (in most cases the majority of the measurement level metadata may also be common for all objects). All archived metadata are in given Table 2.

Table 2. Metadata fields filled for the ancient Abdera project

ElementField entry
Project level
Project nameAncient Abdera
Project date10 May 2005
Name of stakeholderAbdera museum
Object locationSite of Ancient Abdera, Greece
Site geographical longitude40.934249
Site geographical latitude24.975368
Object level
Object ID codeAB.ST.s(1)10
Object lab IDAA10
Object lab date15 May 2005
Batch protocol number-
Group object IDGroup A
Excavation section-
Area description-
Object depth20 cm
Object typesherd
Object categoryamphora
Object dimensions25 x 28 x 4 mm
Object description-
Object pictureAB.ST.s(1)10.tiff
Related documents-
Measurement level
Sample preparationA piece of the sherd is removed. Then after removing few µm of the external layer, it is gently pulverized and then sieved
Measuring techniqueFine grain
Grain size2-10 µm
Number of aliquots16
Measurement protocolAA10.SEQ
FiltersCorning 7-59 & Pilkington HA-3
Instrument(s)Riso TL/OSL reader (model TL/OSL-DA-15), S/N: RISO 105/00/06/b/OSL-B42-IR/830
Instrument(s) details-
LaboratoryLaboratory of Archaeometry and Physicochemical Measurements, R.C. Athena
ResearcherDr Nikolaos Kazakis
Start measurement date1 June 2005
End measurement date2 June 2005
Date of results20 June 2005
Technical Report fileAncient_Abdera_May_2005.doc
Annual dose estimationMeasured K, U, Th activities, cosmic ray dose equal to 0.015 rads/yr
Method for potassiumMicro-XRF
Method for uranium/thoriumELSEC 7286 Low Level Alpha Counter

Previous | Next | Contents

Bibliography#

Aitken, M.J. (1985) Thermoluminescence Dating. Academic Press, Orlando, Florida.

Chen, R., McKeever, S.W.S. (1997) Theory of Thermoluminescence and Related Phenomena. World Scientific Publishing Co. Pte. Ltd, Singapore.

Fleming S. (1979) Thermoluminescence Techniques in Archaeology. Clarendon Press, Oxford.

Rasmussen, K.L. (2001) 'FOCUS: Provenance of Ceramics Revealed by Magnetic Susceptibility and Thermoluminescence'. Journal of Archaeological Science 28(5), 451-456. DOI: 10.1006/jasc.2001.0610

Vaz, J.E., LaBrecque, J.J., Cruxent, J.M. (1997) 'Determination of the provenance of majolica ceramics from Europe by thermoluminescence employing principal components.' Fresenius' Journal of Analytical Chemistry 358(4), 529-532. DOI: 10.1007/s002160050460