In order to measure radiocarbon ages it is necessary to find the amount of radiocarbon in a sample. This measurement can be made either by measuring the radioactivity of the sample (the conventional beta-counting method) or by directly counting the radiocarbon atoms using a method called Accelerator Mass Spectrometry (AMS).
Measurement of the radioactivity of the sample works very well if the sample is large, but in 9 months less than 0.01% of the radiocarbon ions will decay, so in a reasonable measurement time (typically a few weeks) only a very small proportion of the radiocarbon atoms are detected by this method. AMS, on the other hand, can in principle detect a much higher proportion (typically about 1% of the total) allowing sample sizes to be smaller by a factor of about a thousand.
The method is relatively new because it needs very complicated instruments first developed for Nuclear Physics research in the late 20th century.
In common with other kinds of mass spectrometry, AMS is performed by converting the atoms in the sample into a beam of fast moving ions (charged atoms). The mass of these ions is then measured by the application of magnetic and electric fields.
The measurement of radiocarbon by mass spectrometry is very difficult because its concentration is less than one atom in 1,000,000,000,000. The accelerator is used to help remove ions that might be confused with radiocarbon before the final detection.
The sample is put into the ion source either as graphite or as carbon dioxide. It is ionised by bombarding it with caesium ions and then focused into fast-moving beam (energy typically 25keV). The ions produced are negative which prevents the confusion of 14C with 14N since nitrogen does not form a negative ion. The first magnet is used in the same way as the magnet in an ordinary mass spectrometer to select ions of mass 14 (this will include large number of 12CH2- and 13CH- ions and a very few 14C- ions).
The ions then enter the accelerator. As they travel to the terminal (which is at about 2MV), they are accelerated so much that when they collide with the gas molecules in the central `stripper canal'. All of the molecular ions (such as 12CH2 and 13CH) are broken up and most of the carbon ions have four electrons removed making them into C3+ ions. These are then accelerated down the second half of the tandem accelerator reaching energies of about 8MeV. The second magnet selects ions with the momentum expected of 14C ions and a Wien filter checks that their velocity is also correct.
Finally the filtered 14C ions enter the detector where their velocity and energy are checked so that the number of 14C ions in the sample can be counted.
Not all of the radiocarbon atoms put into the ion source reach the detector and so the stable isotopes, 12C and 13C are measured as well in order to monitor the detection efficiency. For each sample a ratio of 14C/13C is calculated and compared to measurements made on standards with known ratios.
Careful sampling and pre-treatment are very important stages in the dating process, particularly for archaeological samples where there is frequently contamination from the soil.
Before sampling, the surface layers are usually removed because these are most susceptible to contamination. Only very small quantities are required for the AMS measurement (30ug-3mg of carbon) and so the damage to objects can be minimised.
The chemical pre-treatment depends on the type of sample. As an example bones are treated as follows:
Several of these procedures are done in an automated continuous flow system.
After chemical pre-treatment, the samples are burnt to produce carbon dioxide and nitrogen. A small amount of this gas is bled into a mass spectrometer where the stable isotope ratios of carbon and nitrogen are measured. These ratios provide useful information on the purity of the sample and clues about the diet and climatic conditions of the living organism. The carbon isotope ratio can also be used to correct for isotopic fractionation in the radiocarbon measurement.
The carbon dioxide is collected in a glass ampoule or converted to graphite for radiocarbon measurement on the AMS system.
The main advantages of AMS over the conventional beta-counting method are the much greater sensitivity of the measurement. In AMS the radiocarbon atoms are directly detected instead of waiting for them to decay. Sample sizes are thus typically 1000 times smaller allowing a much greater choice of samples and enabling very selective chemical pre-treatment. See also specific advantages for Archaeology, Art History, Environmental Science and Biological Tracer Studies
Small sample sizes do have their disadvantages too: greater mobility within deposits and more difficulty in controlling contaminants. The best conventional counters can still achieve higher precision and lower backgrounds than an AMS system assuming a suitably large pure sample can be found. For this reason, the calibration curves for radiocarbon have usually been measured using counters.
There are a large number of AMS labs worldwide. Many of these perform radiocarbon measurements and some of them will undertake sample pre-treatment. A list of sites is held on the WWW site of the journal Radiocarbon: