Search

This paper describes an optimal radiocarbon counting window for liquid scintillation (LS) 14C dating that secures for unquenched as well as for heavily quenched dating samples maximal stability of 14C counting efficiency and theoretically minimal quench correction. In high-precision dating, a balanced counting window with fixed channel limits is frequently used, where about 3% of the highest part of the 14C spectrum is sacrificed for high 14C counting stability. The stability is, however, diminished for quenched samples. Therefore, this window is here replaced by a balanced fixed-energy 14C counting window where the channel limits depend on the quench level. The LS system used must have a linear amplifier and a multichannel analyzer. All samples are measured at a fixed high voltage. For energy calibration and determination of the quench level, the channel number of the middle of the 59.5-keV peak from an external 241Am gamma source is determined before and after measuring each sample. This counting mode is valuable in high-precision dating. It could be widely applied if adapted to systems with a logarithmic amplifier, generally used in LS dating.

An earthquake sequence at the transform plate boundary in South Iceland, that included two magnitude 6.5 earthquakes in June 2000, was anticipated on the basis of a centuries-long seismicity pattern in the area. A program of radon monitoring in geothermal water from drill holes, initiated in 1999, rendered distinct and consistent variations in radon in association with these events. All seven sampling stations in a 50 × 30 km zone covering the epicentral area showed a consistent pattern. Four types of change could be identified: 1) Preseismic decrease of radon. Anomalously low values were measured 101–167 days before the earthquakes. 2) Preseismic increase. Spikes appear in the time series at six stations 40–144 days prior to the earthquakes. These anomalies were large and unusual if compared to a 17-years history of radon monitoring in this area. 3) Coseismic step, most likely related to the coseismic change in groundwater pressure observed over the entire area. 4) Postseismic return of the radon values to the preseismic level about three months later, also concurrent with groundwater pressure changes.

This paper describes a simple and compact liquid scintillation radiocarbon dating system, ICELS, and demonstrates its long-term stability and reproducibility to a precision level rarely presented before, better than 0.04% (3 14C yr). Inexpensive systems of this kind may, in the future, help to meet increasing demand for high precision (±16 to ±20 14C yr) and strict quality control. ICELS comprises a compact detector unit, where a 3-mL dome-shaped vial, with an optimal light reflector, sits on the top of a vertical 30-mm photomultiplier tube. Sample changing is manual. The high voltage is set at the balance point for each sample, securing maximal counting stability. The quench correction method used (spectrum restoration) corrects with 0.04% precision for all parameters that can normally shift the 14C spectrum. For 3 mL of benzene at 71% 14C counting efficiency (recent carbon 23 cpm), the background is 1.72 cpm behind a 5-cm-thick shield of lead (27 kg) and 1.53 cpm behind 10 cm of lead. The background count rate corrected for atmospheric pressure variations was completely stable over 47- and 57-d testing periods for the 2 systems.

After benzene synthesis, radiocarbon dating samples are usually stored for 3–4 weeks before counting to allow an eventual radon contamination to decay to a negligible level. This paper presents a technique that can minimize, and often eliminate, this delay by using a simple single-phototube liquid scintillation counting system, specifically designed for 14C dating. Radon contamination is assessed by pulses of 214Po (a 222Rn decay product, half-life 0.16 μs), identified through pulse-time analysis. For each 214Po pulse, 0.49 beta particle pulses of 214Pb and 214Bi fall in the 14C counting window, and the 214Po pulses are used to correct the 14C count rate. A 14C sample (count rate 11.6 cpm) was measured continuously for 16 days. It was then doped with radon, which increased the first 24-hr count rate in the 14C channel by 3.8 cpm, and the sample was measured for 27 more days. Radon did not measurably affect the 14C-corrected count rate. Counting a sample for 2 min reveals whether it needs storing. If the radon concentration is low, the sample can be measured immediately without degrading accuracy.

The study presented here is the first step in a program aimed at reducing significantly the background count rate of NaI scintillation units. We have investigated: (1) the residual background of a large well type NaI detector, i.e., when shielded with 10 cm of lead and operated deep underground, (2) low concentrations of primordial radioactivity in glass used for photomultiplier tubes (PMTs) and (3) the activity in whole tubes. The residual background of the NaI units is dominated by gamma radiation from potassium, thorium and uranium in the PMT, which severely limits their sensitivity. Activity in tubes made of new high purity glass was close to the detection level. The prospects of a new generation of low-level NaI detectors with these tubes are discussed.

The relative merits of low-level liquid scintillation counters with one and two phototubes, respectively, are discussed and an analysis of their background components is presented. A simple and compact system with manual sample changing is described. It counts four samples simultaneously, and it is based on the single-phototube arrangement where conical vials sit on the top of 28 mm diameter phototubes. A PC computer collects and processes the pulses from a specially designed electronic unit. The system is operated in a small underground counting room, a closet in the wall of a road tunnel under a fjord. Its performance is described.

A 23-yr record of the measuring accuracy of the Copenhagen radiocarbon dating laboratory has retrospectively been provided through a true blind test. A total of 92 samples of oak from old tree trunks were dated in the period 1971 to 1993 and their dendrochronological age determined independently. The 14C activity of the dendrochronological samples measured in the Copenhagen radiocarbon laboratory was compared to the activity of the tree rings of the same age measured by Stuiver and Pearson (1993) for calibration purposes. The average difference was found to be 54 ± 72 14C yr. The results further indicate that the actual standard deviation is only 7% higher than that quoted by the laboratory. The investigation has shown a long-term stability of laboratory accuracy with no systematic laboratory variations either with respect to sample age or to the time of measurement from 1971 to 1993.

Radiocarbon age determination in the 1970s of some 40 archaeological samples strongly indicates a Norse settlement (landnam) in both Reykjavik and the Westman Islands, Iceland, at about AD 700. This early date has generally been rejected by Icelandic scientists because it contradicts the firmly established dating of the settlement (landnam) at about AD 870. In the present article a critical search for an error in the radiocarbon dating work resulted in no weakness being found in the dating material. The history of the landnam of Iceland is thus caught in a dilemma, one which also affects the history of neighbouring countries. A necessary programme for extensive and more precise radiocarbon dating and a critical re‐evaluation of arguments for the conventional date of settlement are outlined.

A recent broad study of the background of all types of low-level beta and gamma detectors has now made it possible to analyze its components more reliably and with greater detail. This general analysis is developed further here for gas proportional and liquid scintillation counters used in radiocarbon dating. The background of gas counters, which is dominated by secondary cosmic gamma radiation, is now well understood and can be described quantitatively. The background of liquid counters is less well understood and can only be described semiquantitatively, its analysis resting partly on estimates. Methods to reduce the background of both types of systems are described and their effectiveness discussed. This analysis may help in evaluating the quality of existing systems as well as in designing better ones.

Low-level counting deals today with a large variety of radioisotopes in different fields, measured with different types of detectors. A short overview is given with the aim of describing its present state, that is frequently limited by insufficient studies in the past, and discussing briefly future possibilities. The main emphasis is on low-level beta and gamma systems and the background per sensitive mass (cmp g−1) of different types of detectors is compared.

A new single phototube liquid scintillation counting system for automatic determination of 222Rn in ground water for earthquake prediction is described. About 40% of radon is transferred from 1.0 L of water with a stream of air into 25 ml of toluene in a sample vial. Delayed coincidence counting of 214Bi and its decay product, 214Po reduces the background to 0.02 cpm. The 214Po detection efficiency is above 90%. Water containing 20 mBq/L of radon can be measured with 10% accuracy in 3 h.

Low-level detector systems for 14C dating are frequently located in underground or basement laboratories to reduce the influence of cosmic radiation. However, if careful precautions are not taken, the presence of radioactive radon gas may severely disturb the analytical results. Measurements of the influence of radon on a proportional counter system, show that a radon level of 100 Bq m−3, a level not uncommon in basement rooms, is sufficient to produce an unacceptable uncertainty in the 14C results. Radon levels of up to 1500 Bq m−3 can be demonstrably reduced to about 30 Bq m−3, using a separate ventilation system that generates a slight overpressure in the laboratory.

Kvartett is a new liquid scintillation counting (LSC) system for radiocarbon dating that takes a radical departure from conventional systems to obtain a compact, low-level counting system measuring four samples simultaneously. Each sample vial, inside the well of a large NaI(Tl) guard-counter crystal (facing down), sits on top of a vertical PMT. The fourfold counting capacity can be used to increase the number of samples being dated or to get higher precision. The increased throughput helps to keep a rigid quality-control standard. We monitored the background count rate almost continuously for 7 months, and measured the count rate of a standard repeatedly for 2 months. The results show the background and system reproducibility to be stable.

I discuss background components of low-level gas proportional counters and show how each component can be estimated based on available data. For more reliable background analysis, further studies are suggested. Based on new information, a generation of low-level gas proportional counting systems for radiocarbon dating may emerge with lower and more predictable background.

A comparison, made eight years ago, of a large number of low-level gas proportional counting systems used for the determination of 14C in radiocarbon dating showed that there can be a large difference in the background of apparently similar systems. This reflects the empirical state of the technique and shows the need for a more detailed and quantitative knowledge of background components. A quantitative evaluation of important background components is presented and the possibility of quantifying others is discussed as well as the advantage of an external guard counting system. Better knowledge of the background components should make it possible to reduce the background below the value of the best counters today.

The high background reduction that has been obtained with a low-level Ge-crystal counting system with external, rather than internal, guard counters is analyzed It is shown that this arrangement eliminates not only the muon component of the cosmic rays, but also a large part of the secondary gamma radiation it produces in the passive shield and therefore gives a much higher background reduction than the conventional arrangement with the guard system inside the main passive shield. The implication of this result for low-level counting is discussed with special reference to gas proportional counting systems used for radiocarbon dating.

I discuss here the basis of a comparison of methods for radiocarbon dating and introduce a new index for the relative merit of a system,factor of counting capacity,that is generally more appropriate than the commonly usedfactor of merit.The merit of a dating system cannot be based on a single figure but other factors must also be considered. A comparison of the gas proportional, liquid scintillation and accelerator mass spectrometry technique is presented that for gas proportional counters is based on a multi-detector system rather than a single detector.

The exceptional sensitivity of gamma-ray spectrometry in underground laboratories has increasing application because of the important science and technology that it allows to be studied. Early work focused on rare fundamental phenomena, e.g. double beta decay, but a growing number of underground measurements is being performed in fields such as environmental monitoring, surveillance of nuclear activities, benchmarking of other physical techniques and materials selection for equipment which require materials with extremely low levels of radioactivity. This report describes the state of the art in underground gamma-ray spectrometry. Backgrounds of HPGe-detectors at various underground laboratories are presented and compared. Improved techniques and detectors are described and needs of deep underground facilities for higher sensitivity measurements are discussed.

An informal collaborative group of radiocarbon dating laboratories, the Low-Level Club, has been established to measure the gamma radiation flux and to test the efficiency of the anticoincidence counting system in laboratories with a NaI detector unit. The detector will record gamma radiation from cosmogenic nuclides, muons and secondary γ radiation formed in the passive shield by charged cosmic-ray particles. We present here the first phase of this work.

A novel compact liquid scintillation counting system with automatic changing of seven samples, dedicated to the assay of low levels of 222 Rn in water, is described. Its has a single phototube, and its operating mode is 222 Rn series selective, as only the close pulse pairs given by 214 Bi and its short lived decay product 214 Po (half life 0.16 ms), are counted. Its 214 Bi / 214 Po counting efficiency is 90.4% and the background, with an unshielded detector, is 3 pulse pairs per hour. 75% of the radon in 200 ml water samples is transferred to the scintillator (15 ml toluene) by circulating air, which is bubbled through the two liquids for 4 min. With a counting time of 3 h, the minimum detection level is 8 mBq l−1. The method is also suitable for various environmental and geophysical studies of 222 Rn and 226 Ra .

Data from three international intercomparison projects are used to study the state of the tritium low-level technique and to set a realistic sensitivity level that should be demanded in ultralow-level tritium work, mainly in hydrology and oceanography. Only two laboratories reach the required goal, a standard deviation ≤0.03 TU for weak samples. The methods used are described and important parameters illustrated graphically. Possible improvements in the measuring techniques are discussed, including the design of better counting systems.

Regional hydrological investigations of the subsurface drainage in the neovolcanic area around lake Thorisvatn on the central Icelandic plateau have been carried out as well as geological exploration. Environmental isotopes, deuterium and tritium, proved decisive in finding the groundwater flow pattern and in separating the different groundwater systems and explaining local deviations as barriers and perched aquifers. The regional groundwater flow is only slightly dependent on the. topography but highly on the geological conditions, as it virtually flows under mountain ranges as well as under the river Tungnaa.

Important advances have been made in reducing the background counting rate of gas proportional counters for14C dating through detailed and systematic study of the background components. Until recently, limited work has been reported on the study of the background of liquid scintillation counters (LSC). During the last few years, commercial systems with greatly reduced background have been introduced. It is shown that the best gas proportional counters and LSC have similar backgrounds for the same amount of sample material. Similar results with less effort may be expected with more detailed and fundamental knowledge of the components of the background of LSC. We report the results of a study of one photomultiplier LSC system where we research all parameters of importance: light collection efficiency, absorbed energy per photo-electron, pulse height spectrum and background counting rate.

The sensitivity of the measurements of tritium in natural water has been increased by using volume reduction factor as high as 300 in a single-stage electrolytic enrichment, instead of a factor of 10–20 as normally used. By continuously filling the electrolysis cell from a reservoir a reduction factor of 1500 can be attained in a single stage. Using this technique the simple Geiger counter gives all the desired sensitivity for hydrological studies. A comparison of the present measuring technique with the more important alternative methods is given.

Liquid scintillation counting (LSC) for radiocarbon dating is a less expensive method than accelerator mass spectrometry (AMS), provides a high degree of accuracy, and is less prone to contamination due to the larger sample sizes. However, to obtain high precision, a long counting time is needed. The Gliwice Radiocarbon Laboratory is seeking to obtain an increased counting capacity with 2–3 mL benzene samples than we presently can achieve with our 2 Quantulus systems. We are therefore investigating the possibility of using a simple, single-phototube LS system (ICELS) for dating samples younger than 5000 yr. We present the first results of this investigation, including the measurement of 3 VIRI and 3 FIRI intercomparison samples.

Acid-bubbling has been proposed as an alternative method for separating the 14C-carbonate from organic matter in determinations of primary productivity, but adhesion of plankton to the wall of the glass vessel used for bubbling can lead to low primary productivity values. A modified procedure is quite efficient compared with the conventional filtration method.