Index > 3 Characterisation of radioactively contaminated sites >

3.4.7 Soil and rock samples

Contents
3.4.7.1 Disturbed soil samples
3.4.7.2 Undisturbed soil samples
3.4.7.3 Soil field sample preparation and preservation

Soil and rock samples, both above and below ground level, can provide essential information towards determining the accumulated amounts of contaminants which have been deposited on the ground. It is very important to ensure that the samples taken are seen to provide a realistic representation of both the perceived problem and the area (laterally and/or vertically) over which the contamination is anticipated to exist.

Soil and rock samples are of two main types: mechanically disturbed or undisturbed.

Mechanically disturbed samples are generally adequate for contamination surveys, whereas mechanically undisturbed samples are typically required for geotechnical surveys. There are three main methods of selecting soil and rock samples in the field:

  • Sampling from predefined depth intervals.
  • Sampling based on visual features (i.e., from different geological units or different layers of made ground).
  • Sampling based on the results of radiological or chemical monitoring.

When excavating on a potentially contaminated site, radiation monitors may be used to identify the excavated material with the highest levels of radioactive contamination. This information can then be used to focus sampling, ensuring that at least some of the samples containing the highest levels of radioactive contamination are selected. Care should be taken to avoid over-estimating the volume of contaminated material present if only the most radioactive samples are selected for analysis.

In any survey, it is important that:

  • Samples are representative of ground conditions.
  • Sufficient material is collected to enable all required analyses to be undertaken (including sufficient material for repeat analysis, should this be necessary). The sample size can be significant when undertaking radiological measurements and, in general, large volumes of soil are more representative than small volumes of soil. For example, the time taken to analyse for gamma-emitting radionuclides to a specified detection limit by gamma spectrometry is approximately inversely proportional to the weight of sample analysed. In general, it is sufficient to collect approximately from 100 –1000 g of sample in an appropriate container for gamma spectrometry analysis.
  • Samples have a constant volume, because the volume is related to comparability of the results while the sampling surface area is more closely related to the representativeness of the results.
  • Maintaining a constant surface area and depth for samples collected for a particular survey can eliminate problems associated with different depth profiles. The actual surface area included as part of the sample may be important for estimating the probability of locating areas of elevated concentration.

Certain other analyses require additional field sample preparation. For example, analysis for tritium or volatile organic compounds typically require the soil or rock sample to be stored in a sealed septum vial immediately after collection, the aim being to prevent the loss of volatile compounds during transportation to the laboratory. The chemical or radiochemical analyst will provide advice on the volumes of samples required and on any field preparation required (for example, the addition of ultra-pure water).

In some cases, for example when remediation of the site is a probable outcome of the site characterisation and where a large averaging volume has been agreed with the appropriate regulator, it may be appropriate to homogenise samples from a large volume of material. For example, this approach has been used to determine the average contaminant concentrations in ~30 m3-sized disposal pits.

3.4.7.1 Disturbed soil samples

Disturbed soil may be brought to ground surface using any of the intrusive investigation techniques listed in Section 3.7, Site characterisation: Intrusive methods. Disturbed samples are generally collected from the spoil produced by the excavation process, using a tool such as a stainless-steel trowel, and placed into the appropriate sample containers (as supplied or advised by the analytical testing laboratory).

It will often be necessary to characterise areas of made ground or coarse-grained soil (such as glacial till or rock fill). In this case, samples that are representative of the entirety of the ground cannot be collected because of the presence of coarse gravels, cobbles and boulders. It is neither practicable nor appropriate (given that any contamination will be concentrated within the finer-grained fraction) to analyse these coarse-grained components of the soil. In this case, the coarse-grained fraction should be discarded, and only the finer-grained fraction sent for chemical and radiochemical analysis. The approximate proportion of unsampled material should be recorded to enable the measured contaminant concentration in the finer-grained component to be corrected (i.e., diluted), if required, to account for the presence of the coarser fraction.

It is best practice to consider the extent of any bias introduced by analysing only the finer fractions of the soil samples. This can be achieved by grinding and homogenising soil samples (at least the sub-pebble-sized fraction), and analysing the resulting sample.

3.4.7.2 Undisturbed soil samples

Relatively mechanically undisturbed soil samples are generally collected by using one of the standard drilling techniques (such as cable percussive drilling or coring through the centre of a hollow stem auger). The samples are usually collected using an open tube sampler, such as a U100 tube or a plastic core liner. Rotary coring is typically used to obtain mechanically undisturbed rock samples.

3.4.7.3 Soil field sample preparation and preservation

Proper sample preparation and preservation are essential parts of any radioactivity sampling program. The sampling objectives should be specified before sampling activities begin. Precise records of sample collection and handling are necessary to ensure that data obtained from different locations or time frames are correctly compared.

The appropriateness of sample preparation techniques is a function of the analysis to be performed. Field sample preparation procedures are a function of the specified analysis and the objectives of the survey. It is essential that these objectives be clearly established and agreed upon in the early stages of survey planning (see Section 2).

Soil and sediment samples, in most protocols, require no field preparation and are not preserved. In some protocols, cooling of soil samples to 4°C is required during shipping and storage of soil samples. This is not a practice normally followed for the radiochemical analysis of soil samples.

When replicate samples are prepared in the field, it is necessary to homogenize the sample prior to separation into replicates. There are standard procedures for homogenizing soil in the laboratory, but the equipment required for these procedures may not be available in the field. Simple field techniques, such as cone and quarter, or using a riffle splitter to divide the sample may be appropriate if the sample can be dried. If the sample contains significant amounts of residual water (e.g., forms clumps of soil) and there are no facilities for drying the sample, it is recommended that the homogenization and separation into replicates be performed in a laboratory. It is preferable to use non-blind replicates where the same laboratory prepares and analyzes the replicates rather than use poorly homogenized or heterogeneous samples to prepare replicate samples.