Introduction
The development of the AMS technique as well as the construction of the facility at WHOI have greatly increased the number of individuals and laboratories in the field of oceanography interested in obtaining ¹⁴C analyses, and have greatly increased the importance of collecting the sample in a clean, ¹⁴C-free environment.

Here, we discuss the special precautions necessary for collecting and handling samples for the measurement of natural levels of ¹⁴C. Analysis by AMS requires a sample that is orders of magnitude smaller than that for a conventional radiocarbon analysis (1 mg vs. 5 g). While this greatly increases the scope of radiocarbon studies, it also means that a much smaller amount of contaminant is required to ruin a sample.

For most users, the sample that is submitted will be prepared for AMS analysis in the NOSAMS sample-preparation laboratory. Virtually every AMS laboratory has experienced down time because of the clean-up required after preparing and analyzing a "hot" sample.

We have prepared this document to help users of the NOSAMS facility ensure that they are providing us with a clean sample whose preparation will not compromise the operation of our laboratory.

Many oceanographic research projects use radiocarbon as a spike in experiments at sea and in the laboratory, e.g. the measurement of oceanic productivity, and inadvertent spills can leave isolated spots that are severely contaminated. The levels typically used in tracer experiments can be several million times modern levels and very small residual amounts can ruin the measurement of natural levels of ¹⁴C. Contamination of the sample container and, thus, the sample, can arise from collecting and handling the sample on a contaminated surface. Therefore, we recommend the following procedures to ensure collection and preparation of a "good" ¹⁴C sample.

Laboratory Operations
Collection of natural samples for radiocarbon analysis usually requires the preparation of sample containers and sampling apparatus in a laboratory. All apparatus that is to be used for sample collection should be cleaned and prepared in a laboratory that is known to be ¹⁴C-free. Because it is difficult in many laboratories to be certain that ¹⁴C has never been used as a spike, we recommend conducting low-level swab tests whenever there is the slightest question about the history of a laboratory. The Tritium Laboratory at the University of Miami is one laboratory which has developed a well-established routine and protocol for testing ships and laboratories and which performs swab analysis for a fee. We encourage investigators new to the natural radiocarbon field to consider swab tests specifically designed to measure low levels of contamination in the laboratory as the first step in establishing a radiocarbon research program. In laboratories where a problem is found, corrective action must be taken. When an area is determined to be ¹⁴C-free, we recommend that this area be isolated from general laboratory operations and access be restricted to those familiar with the precautions necessary for handling natural-level radiocarbon samples.

Shipboard Operations
Swab tests of oceanographic research vessels have shown that the use of radiocarbon to measure productivity leaves areas that are severely contaminated. These cannot be used for collection of natural-level radiocarbon samples. We recommend that surfaces where samples are collected or handled should be covered with fresh, disposable sheets of plastic or garbage bags and that disposable gloves (changed often) should be worn during sampling. Sample containers should be handled as little as possible and removed from their packing crates only when necessary. When sample bottles or containers are removed from shipping crates, they should not be placed in direct contact with any surface on the ship either on deck or in the laboratory. A data sheet(s) should be kept for each crate of sample bottles. Information regarding the history of each crate should be recorded on this sheet. This information should include identification of the laboratory in which the bottles were prepared, the shipping and storage history (dates and location) of each crate, information regarding the condition of laboratories and storage facilities (e.g., refrigerated or not) and should identify other sampling programs in progress on the ship. In general, it is not advisable to consider collecting radiocarbon samples on a cruise on which ¹⁴C-spiking experiments will be performed. If laboratory analyses are required at sea, precautions similar to those discussed in the previous section must be observed.

More detailed information can be obtained by contacting nosams@whoi.edu or (508) 289-2513. We have published a detailed protocol for the collection of seawater inorganic carbon samples* and will be happy to discuss the collection of other types of samples with interested researchers.

* McNichol, A.P. and Jones, G.A., 1991. Measuring ¹⁴C in seawater by Accelerator Mass Spectrometry. WOCE Hydrographic Operations and Methods manual, WOCE Hydrographic Programme Office, WOCE Report No. 68/91, Woods Hole Oceanographic Institution, Woods Hole, MA.