Ultrafiltration of bone collagen provides a method of purification that can be very effective in reducing environmental contamination from soil-derived amino acids as well as removing degraded collagen or other short-chain proteins. The Oxford Radiocarbon Accelerator Unit (ORAU) first implemented ultrafiltration in the pretreatment of bone material for accelerator mass spectrometry (AMS) radiocarbon dating in 2000. However, the filters themselves contain carbonaceous material, and thus stringent quality control is required to demonstrate that this does not affect the accuracy of the dating. Here, we present quality assurance data from the bone pretreatment and dating program at ORAU, including dates on known-age and background-age bones over a range of sample sizes, and measurements of residual carbon contamination present in the filters after cleaning.

As part of our bone dating development, we have tested the ultrafiltration of bone gelatin using 2 different filters—Vivaspin 20™ (VS20), a polyethersulfone, and Vivaspin 15R™ (VS15R), a cellulose, both with a 30,000 molecular weight cutoff—and bone collagen from dated samples ranging in age from 1.5 to >50 kyr BP. A direct accelerator mass spectrometry (AMS) measurement yielded radiocarbon concentrations of ∼0.5 pMC (∼42 kyr) for the polyethersulfone, ∼14.4–17.5 pMC (∼15.6–14 kyr) for the cellulose, and ∼107.4 pMC for the glycerin. The filters were cleaned before use similar to the Oxford protocol (Bronk Ramsey et al. 2004), and a series of freeze-dried archaeological bone gelatin samples and a modern pig-skin gelatin were passed through VS20 and VS15R filters (Vivascience™). We recovered both the eluent (<30-kD fraction) and the liquid that stayed above the filter (>30 kD) in order to obtain a carbon mass and isotope balance. While the >30-kD collagen fraction that is usually selected for AMS analysis does not appear to be significantly contaminated, measurements show significant age differences between the eluent <30 kD and the unfiltered bone collagen, indicating that, despite cleaning, both glycerin and filter still give off contaminants in the eluent. Ultrafiltration with young collagen from pig skin generally confirms these results for the <30-kD fraction but also shows the possibility of small contaminations in the >30-kD fraction. Until a contamination with filter carbon of the >30-kD collagen fraction can be excluded, we would recommend caution in the use of ultrafiltration for cleaning bone collagen with VS20 or VS15R ultrafilters.

In the sample preparation laboratory of CEDAD (CEnter for DAting and Diagnostics) of the University of Lecce, a protocol for the quality control of bone samples based on infrared spectroscopy has been set up. The protocol has been recently developed as a check-in test with the aim to identify the presence of collagen in the samples, assess its preservation status, and determine whether the submitted bone samples are suitable for accelerator mass spectrometry (AMS) radiocarbon measurements or not. We discuss in this paper the use of infrared-based techniques to identify the presence of “contaminants” such as restoration and consolidation materials, humic acids, and soil carbonates, which, if not removed by the sample processing chemistry, can be sources of exogenous carbon and can thus influence the accuracy of the 14C determinations.

Bone samples recovered in well-defined and previously dated archaeological contexts were intentionally contaminated, submitted to the standard method for collagen extraction and purification, and then characterized by means of Fourier transform infrared (FTIR) spectroscopy analyses performed in attenuated total reflection (ATR) mode before being combusted, converted to graphite, and measured by AMS. The study shows that the ATR-FTIR technique is an extremely powerful method for the identification of both the collagen and its contaminants and can supply important information during the selection and processing of samples to be submitted for 14C dating.

Animal bones and tortoise shells were used for divination by the Chinese royal family during the Shang Dynasty (∼16th–11th century BC), and the divination results were recorded as inscriptions on oracle bones and shells, which are very valuable cultural remains and record many important events in the Shang Dynasty period. Thus, radiocarbon dating of oracle bones was used to build a precise chronology of the late Shang Dynasty. Due to their original burial conditions and the fact that in subsequent decades the pieces were traded or archived in museums, oracle bones are expected to be contaminated with exogenous materials from the environment and the conservation process. During dating, we found that some samples were contaminated by conservation chemical reagents. The contaminated samples were purified by removing exogenous chemicals with a series of organic solvents, in a method modified from Bruhn et al. (2001). Both whole bone and gelatin samples were processed with this purification method, resulting in satisfactory improvements in dating results.

We present our new graphite target production system, developed in the Institute of Nuclear Research of the Hungarian Academy of Sciences (ATOMKI), for accelerator mass spectrometry (AMS) radiocarbon dating measurements. The system consists of a gas handling line and a graphite target production system. Results of AMS measurements, stable isotope mass spectrometry measurements, and gravimetric/pressure yield determinations have been used to find the proper conditions for the graphitization process. We have also investigated the 14C contamination and the memory effect of the system during the graphitization processes. This paper covers the details of these experiments and a discussion of the results.

A system with several lines for the preparation of graphite targets for radiocarbon analysis has been built at the new accelerator mass spectrometry (AMS) facility in Caserta, Italy. Special attention has been paid in the design to the reduction of background contamination during sample preparation. Here, we describe the main characteristics of these preparation lines. Results of tests performed to measure 14C background levels and isotope fractionation in several blank samples with the Caserta AMS system are presented and discussed.

Accelerator mass spectrometry (AMS) radiocarbon measurements of organic samples require combustion to obtain CO2 for graphitization. Furthermore, determination of δ13C values is required in order to correct the 14C age due to carbon isotope fractionation effects. δ13C analysis is commonly carried out by stable isotope mass spectrometry because most applications demand high-precision δ13C values in addition to the requirements of 14C dating. A simplifying step is therefore to combine the combustion for stable isotope analysis with cryogenic trapping of CO2 for AMS graphite targets. Presented here is a simple CO2 trapping device based on a modified Gilson 220XL sampling (manifold) robot coupled to the inlet manifold system of a GV Instruments IsoPrime stable isotope mass spectrometer. The system is capable of batch combustion and analysis of up to 40 samples and is under full computer control by the mass spectrometer software. All trapping parameters such as flush time prior to trapping and total trap time are adjustable through the standard software user interface. A low 14C activity of background materials and high precision and accuracy of stable isotope analysis of carbon and nitrogen are demonstrated.

The effect of nitrous oxide (N2O) upon the graphitization of small (∼40 μg of carbon) CO2 samples at the ANSTO and University of California, Irvine, radiocarbon laboratories was investigated. Both laboratories produce graphite samples by reduction of CO2 over a heated iron catalyst in the presence of an excess of H2. Although there are significant differences between the methods employed at each laboratory, it was found that N2O has no effect upon the reaction at levels of up to 9.3% by volume of CO2. Further, it was systematically determined that more effective water vapor trapping resulted in faster reaction rates. Using larger amounts of the Fe catalyst generally resulted in higher yields or reaction rates (but not both). The effects of changing the type of Fe catalyst on the final yield and reaction rate were less clear.

We present an overview of accelerator mass spectrometry (AMS) radiocarbon sample preparation and measurements, describing the technical upgrades that now allow us to routinely obtain 0.2–0.3% precision for 1-mg carbon samples. A precision of ∼1% on samples with 100 μg of carbon can also be achieved. We have also developed graphitization techniques and AMS procedures for ultra-small samples (down to 0.002 mg of carbon). Detailed time series are presented for large and small aliquots of standards such as NIST OX-I and OX-II; FIRI-C and -D; IAEA-C6, -C7 and -C8; and 14C-free samples.

Recent progress in preparation/combustion of submilligram organic samples at our laboratories is presented. Routine methods had to be modified/refined to achieve acceptable and consistent procedural blanks for organic samples smaller than 1000 μg C. A description of the process leading to a modified combustion method for smaller organic samples is given in detail. In addition to analyzing different background materials, the influence of different chemical reagents on the overall radiocarbon background level was investigated, such as carbon contamination arising from copper oxide of different purities and from different suppliers. Using the modified combustion method, small amounts of background materials and known-age standard IAEA-C5 were individually combusted to CO2. Below 1000 μg C, organic background levels follow an inverse mass dependency when combusted with the modified method, increasing from 0.13 ± 0.05 pMC up to 1.20 ± 0.04 pMC for 80 μg C. Results for a given carbon mass were lower for combustion of etched Iceland spar calcite mineral, indicating that part of the observed background of bituminous coal was probably introduced by handling the material in atmosphere prior to combustion. Using the modified combustion method, the background-corrected activity of IAEA-C5 agreed to within 2 σ of the consensus value of 23.05 pMC down to a sample mass of 55 μg C.

We investigated a simple, reliable radiocarbon measurement procedure for water samples using the direct absorption method for sample preparation, followed by low-level liquid scintillation spectrometry. This process has involved quantitative evaluation of the conversion steps in order to estimate the appropriate working parameters. The 14C activity of dissolved inorganic carbon (DIC) for several types of water (ranging from seawater to groundwater) has been measured, paying attention to the preparation requirements of each type of water. The main advantage of this method is the simplified sample preparation, allowing measurement of a great number of samples in less time. This method was designed for routine analysis of water samples, and it is proposed particularly for use in 14C monitoring programs of CANDU-type reactors.

The new facility Artemis was installed in 2003 in Saclay, France. This 3MV NEC Pelletron is dedicated to high-precision radiocarbon measurements for French 14C laboratories. We will present information on Artemis along with our sample preparation methods. Results from measurements on some intercalibration samples will be given along with the values of measured blanks. Finally, we report on some problems we have encountered when measuring sputter cathodes with high CH− outputs.

We report on 2 recent developments in an ongoing program of characterizing and improving the National Electrostatics Corp. (NEC) MC-SNICS ion source at University of California (UC) Irvine's Keck AMS laboratory. First, we have investigated the possibility of modifying a large-body (134-sample) MC-SNICS to incorporate the UC Irvine Cs oven and vacuum-insulated Cs feed tube, which provide better confinement of Cs than the standard NEC setup. In our 40-sample source, the feed tube enters the source housing directly below the ionizer assembly. This area cannot be accessed for machining on the 134-sample source, but we have successfully tested a modified geometry where the delivery tube enters the body via the source end flange. Second, we recently installed a second beam profile monitor in the injection line of our spectrometer to allow us to make online emittance measurements. At full output (150 μA of C− at 55 keV), the emittance of our source at 8 kV sputtering voltage is approximately 40π mm mrad.

The novel tabletop miniaturized radiocarbon dating system (MICADAS) at ETH Zurich features a hybrid Cs sputter negative ion source for the measurement of solid graphite and gaseous CO2 samples. The source produces stable currents of up to 6 μA C− out of gaseous samples with an efficiency of 3–6%. A gas feeding system has been set up that enables constant dosing of CO2 into the Cs sputter ion source and ensures stable measuring conditions. The system is based on a syringe in which CO2 gas is mixed with He and then pressed continuously into the ion source at a constant flow rate. Minimized volumes allow feeding samples of 3–30 μg carbon quantitatively into the ion source. In order to test the performance of the system, several standards and blanks have successfully been measured. The ratios of 14C/12C could be repeated within statistical errors to better than 1.0% and the 13C/12C ratios to better than 0.2%. The blank was <1 pMC.

We present a broad and detailed study of the background components of a liquid scintillation (LS) detector, using a simple laboratory-built system, ICELS. It was specifically designed for radiocarbon dating and is compact and easily transportable (total weight 35 kg). Its flexible LS detector unit has a dome-shaped vial with 3 mL benzene to which 45 mg butyl-PBD is added. The vial sits on the top of a vertical 28-mm-diameter phototube. The gamma radiation, to which the benzene is exposed under varying conditions, was measured by replacing the vial with a 38-mm-diameter Nal crystal. The pulse-height spectra of the 14C LS background and the Nal gamma background were measured in a surface laboratory and in a deep underground counting room with: 1) a lead shield of varying thickness; 2) lead of normal and low 210Pb concentration; 3) phototubes of 2 different types; and 4) varying benzene volume. The beta emission from the face of the tubes was measured with a low-level Geiger counter.

A preliminary investigation of the biocarbon fraction in carbon dioxide emissions of power plants using both fossil- and biobased fuels is presented. Calculation of the biocarbon fraction is based on radiocarbon content measured in power plant flue gases. Samples were collected directly from the chimneys into plastic sampling bags. The 14C content in CO2 was measured by accelerator mass spectrometry (AMS). Flue gases from power plants that use natural gas, coal, wood chips, bark, plywood residue, sludge from the pulp factory, peat, and recovered fuel were measured. Among the selected plants, there was one that used only fossil fuel and one that used only biofuel; the other investigated plants burned mixtures of fuels. The results show that 14C measurement provides the possibility to determine the ratio of bio and fossil fuel burned in power plants.

Progress in radiocarbon accelerator mass spectrometry (AMS) techniques enables much more access to wiggle-matching techniques for high-precision 14C dating with relatively low costs than before. Recently, we have applied wiggle-matching for a number of wood samples where dendrochronology is difficult because of various limitations imposed for dendro-dating. In most cases, wiggle-matching gave rather unambiguous calendar ages, but we found that in some cases the calibrated date was very sensitive to a systematic error of the 14C date. Here, we present a wooden artifact from the Ujishigai archaeological site as a case where the highest wiggle-matched date did not agree with the date given by dendrochronology. An age with lower probability agreed with the tree-ring age of AD 389, which marked the beginning of the production of Sue ware (unglazed stoneware) in Japan. We show that systematic errors must be carefully taken into account while interpreting 14C wiggle-matching results, whether they are due to instrumental errors (statistical) or due to a regional offset from the IntCal04 (Reimer et al. 2004) calibration curve.