RADIOCARBON DATING

For a description of the principle of the technique one can visit the following sites:

http://www.c14dating.com
http://www.radiocarbon.org
http://info.ox.ac.uk/departments/rlaha/orau/index.htm

NCSR “DEMOKRITOS” - SYSTEM DESCRIPTION

Figure 1. Sample preparation laboratory.

The Laboratory of Archaeometry uses the gas (CO₂) counting technique. Each sample is combusted and the CO₂ produced is measured inside cylindrical gas proportional counters. The laboratory possesses 8 such counters, four with a capacity of 4 litres and four with 3 litres. These counters are identical to those used in the Radiocarbon Laboratory of the Institute of Environmental Physics, University of Heidelberg and were built and tested there. In 1998 an integrated counting system consisting of six counters inside a common guard counter was installed together with a data logger which monitors the meteorological parameters affecting the intensity of the cosmic radiation flux and consequently the counters background. This new counting system combined with the pre-existing counters has improved significantly the precision of the radiocarbon dates obtained as well as the speed with which results are produced and delivered.

Α Β

Figure 2: The multi-counter system consisting of 6 gas proportional counters used for the C-14 measurement of archaeological and environmental samples. Α) Front side, input - output of the gas samples. Β) Interior, 2 big and 4 smaller counters together with the surrounding cylindrical guard counter are shown.

PRECISION

The measurement precision depends not only on the initial amount of sample available and the age of the sample but also on the shape of the calibration curve at the time period to which the sample belongs. The same accuracy of measurement of two samples can result into a different final accuracy after calibration as can be seen in the examples of Table 1.

Table 1. Examples of radiocarbon dates of some routine samples:

LAB CODE	SAMPLE TYPE	C-14 AGE (yr BP)	CALIBRATED AGE
DEM-1680 (Mavropygi Ptolemaidas)	Charcoal	7660 ± 25	6560 – 6455 BC (68.2%) 6590 – 6450 BC (95.4%)
DEM-1798 (Holy Monastery of Pandokrator, Mt. Athos)	Bones	771 ± 25	1225 – 1275 AD (69.2%) 1215 – 1280 AD (95.4%)

Note: The Laboratory of Archaeometry is assigned with the international code DEM (Demokritos).

SAMPLING INSTRUCTIONS AND QUANTITY OF SAMPLE REQUIRED

1. Sampling strategy

The age determination of a sample using the radiocarbon dating method requires a substantial cost and effort from the laboratory involved. For that reason before each sampling a deeper speculation should precede concerning the specific purpose of the research, as well as the best way to accomplish it using absolute dating.

The most important question, which should be put each time a sample is collected for dating, is whether this particular sample is directly related with the historical event or occupation layer whose age is under question. For example, the radiocarbon age of a wood sample from a post, no matter how precise the dating will be in the laboratory, will not be directly related to the specific layer if the post was re-used. Also, as a big tree may live for a long time (100-200 years), if one determines the age of a wooden or charcoal sample originating from the inner rings of such a tree, the calculated age may be 100-200 years older than the date the timber was cut even if it was used for the first time. The event dated in the laboratory is the date of creation and growth of the plant, tree or animal and not the date that it was used as an object. 100-200 years older than the date the timber was cut even if it was used for the first time. The event dated in the laboratory is the date of creation and growth of the plant, tree or animal and not the date that it was used as an object.

It is of course very well known that it is difficult to realise problems of this nature and finally to be able to answer the questions of re-use, the direct relation etc. However, it is necessary to pay special attention to this subject and to check all the possible interpretations before one could select the proper sample. A typical procedure is to separate the short-lived from the long-lived plants (e.g. short-lived nutritional plants, branches from a hearth instead of relatively large trunks of long-lived trees).

More information concerning the correlation probabilities between various samples and events, which may help the sampling strategy, are to be found in the next section.

Finally, we should keep in mind that in order to check the samples for an eventual contamination or material mixing from different layers during the excavation, it is advisable always to collect all the appropriate types of samples even if there is no need for them to be dated immediately.

A point that should be stressed here is that for the best performance of the research a close collaboration between the archaeologist and the scientific staff of the laboratory is needed.

2. Archaeological association

Concerning the sampling, in order to associate the age of a sample with an archaeological question we should bear in mind the following:

a. Absolute certainty: in which case the measured sample originate from the archaeological object itself, e.g. human bones from a grave, canoe from a tree trunk, coach wheel, organic matter embedded in a sherd.

b. High probability: in which case there is a direct functional relation between the analysed organic matter and the archaeological co-finds, e.g. wood remains in a grave with offerings, carbonised seeds in vases, charcoal in an amphorae or from a hearth on a house floor.

c. Good probability: in which case there is no obvious functional relation between the analysed sample and the archaeological material, but the amount of organic matter is in favour of some relation, e.g. charcoal pieces gathered in a dump pit or in an inhabitation layer.

d. Logical probability: the same as it is in case (c), but the pieces of the organic matter are small and scattered around, e.g. "black coloured soil" in an inhabitation layer, charcoal pieces in a grave.

In terms of the distance in time between the natural formation of the sample and its archaeological use, the samples may be grouped as follows:

a) Branches, seeds, leather, bones, outer tree-rings from short-lived tree species, etc: the age difference is so small (< (< 20 years) that may be considered negligible.

b) Charcoal from short-lived trees, outer tree-rings from long-lived tree species, when there is no reason to expect a long duration of use, etc: the age difference may be some decades (between 20 and 100 years).

c) Charcoal from long-lived tree species, which have been probably re-used, etc: the age difference may be some centuries (>(> 100 years).

d) Samples consisted of black coloured soil, ash, or sediment, etc, whose nature and origin of the organic matter to be dated is not precisely known.

3. Sampling instructions

The collection of the samples for radiocarbon dating should be done together with the archaeologists and the researchers of the laboratory in order to obtain the best results. If this is not feasible, the person who will perform the sampling should have in mind the following rules:

1) The sample should never be collected with bare hands; on the contrary clean stainless steel tools (scalpel, tweezers, scoop, etc) should be used.

2) As large an amount of sample as is possible should be collected.

3) The sample should be collected in a region where the stratigraphy is undisturbed.

4) Attention should be drawn to the avoidance of pieces of samples mixing from upper layers during sampling, e.g. falling down of charcoal pieces due to vibrations, from stepping or bore hole drilling, on a close baulk in the sampled region.

5) All visible extraneous materials should be removed from the sample, e.g. stones, plant roots, leaves, soil and sand.

6) Detailed notes should be kept during sampling according to the form attached.

7) Samples with high moisture content should be left to dry in the shade for 2-3 days in order to avoid microbiological contamination during the storage time. Additional attention should be paid during this time period for the samples not to be contaminated or mixed with other ones.

8) Different types of samples should be packed separately, e.g. charcoal, bones, wood, sea shells, land shells, etc.

9) The samples should be properly wrapped in an aluminium foil sheet and then placed inside polyethylenium bags or inside glass bottles. Textiles, cotton, wool, or paper sheets should never be used, as they are carbon rich materials. Plastic bags from PVC and PVA should be avoided as they may contain plasticizers, which may be absorbed by the material of the sample.

10) All samples should be properly labelled and the labels should not be in direct contact with the sample itself.

11) Any known contamination should be pointed out to the laboratory staff in order to prepare the proper chemical pre-treatment for each sample, e.g. the presence of animal excreta and bat droppings in the excavated area. The ash from the cigarettes, fat, oil, human hair or hair from brushes, as well as food remains are common contaminating factors.

4. Required quantity of sample

The optimum required quantity of sample for the determination of reliable dates with the smaller possible error fluctuates in each laboratory according to the radiocarbon technique used. TABLE 1 shows the required quantities for different types of sample in the case of the LABORATORY OF ARCHAEOMETRY of N.C.S.R. "DEMOKRITOS".

In practise though, it is possible to date smaller quantities of sample provided that the age is younger than ca. 18.000 years BP. TABLE 2 shows the minimum required quantities of different types of sample. It is worth noting that the reduction of the quantity of the sample results in the increase of the preparation and measurement time of the sample, as well as to the increase in the age error.

5. Documentation - Submission form

Every sample submitted for dating is one among hundreds, which come yearly to the laboratory. So, it is important that every sample is fully documented with all related archaeological information, which is needed for the laboratory staff to draw the conclusions. For that reason, for every sample, which will be sent to the laboratory a downloadable Submission Form should be filled in and attached to it. download.

The archaeological age estimation is useful as it helps the laboratory staff to decide about the chemical pre-treatment procedures and the measurement time required. Also it helps in the estimation of the expected age error according to the available amount of sample. It should be noted that the chemical pre-treatment procedures are continuously improved and a radiocarbon age calculated a decade or more ago, usually does not have the same reliability with those obtained nowadays.

6. Calibration

It is possible to convert reliably the conventional radiocarbon ages (which are expressed as yr ΒΡ= years Before present) younger than 20,265 yr BP to calendar ages (BC/AD) using the new international calibration curve INCAL98. Though, it is not yet possible to calibrate ages older than 20,265 yr BP.

TABLE 1

Required quantity from different sample types in order to produce an optimum gas sample (8.47 g CO₂).

SAMPLE TYPE	SAMPLE WEIGHT (g)
Charcoal	8
Anthracite	6
Xylite	12,4
Turf peat	62 - 109
Tar	6,65
Sediment	150 - 4000
Wood	13,6
Paper	11,7
Linen Textile	17
Seaweed	23,8
Bones	150 - 300
Shells	30

TABLE 2

Minimum required amount of different type and age of sample in order to obtain a reliable age associated with a minor error.

Age range (yr ΒΡ)	55.000-18.500	18.500-13.000	13.000-9.700	9.700-7.400	7.400-5.600	5.600-2.900	2.900 - 0
Sample type				Weight (g)
Charcoal	8	5,05	3,25	2,63	2,00	1,23	1,00
Anthracite	6	3,78	2,45	1,95	1,50	0,93	0,75
Xylite	12,4	7,83	5,05	4,05	3,13	1,90	1,55
Turf peat	62-109	39,3-68,8	25,3-44,5	20,3-35,8	15,6-27,5	9,6-16,8	7,7-13,5
Tar	6,65	4,20	2,70	2,18	1,68	1,03	0,83
Sediment	150-4000	94,8-2525	61,3-1630	49-1308	37,8-1005	23,1-618	18,7-498
Wood	13,6	8,58	5,55	4,45	3,43	2,10	1,70
Paper	11,7	7,38	4,78	3,83	2,95	1,80	1,45
Linen Textile	17	10,73	6,93	5,55	4,28	2,63	2,13
Seaweed	23,8	15,03	9,70	7,78	5,98	3,68	2,98
Bones	150-300	94,8-189,3	61,3-122,3	49-98	37,8-75,5	23,1-46,3	18,7-37,3
Shells	30	18,93	12,23	9,80	7,55	4,63	3,73

---