The process of sample collection is often minimized during the planning phase of remediation and other on-site activities. However, sampling can provide the empirical data needed for reliable decisions to be made. Therefore, greater emphasis on collecting samples will or should provide support to decision-makers when they have to answer “why, how, and where” questions.
The challenges involved in collecting samples are both multifold and multilayered. They usually begin when the decision to collect samples has been made. Proper planning is essential to successful sample collection, though, and largely for that reason a number of questions should be asked ahead of time – specifically including the following: (1) “Why sample?” (2) “What is the purpose of the sampling?” (3) “How will the resulting data be used – and/or what decisions will it be expected to support?” (4) “What methods will be used to collect the samples?”
The effectiveness of the sample collection required to find answers to these questions usually depends primarily on the use of proper statistical techniques – which themselves almost always are determined by the purpose, or intent, of the decision to collect samples. One of the techniques used is to collect “discrete” or “grab” samples. This method is particularly useful when: (1) the areas being searched are and/or have been exposed to an equal or near-equal level of contamination; and (2) it has become obviously necessary to find the proverbial “needle in the haystack.”
In the second of these situations, a large number of samples – or,eally, samples from the entire population/decision area – must be collected. If it has been decided that the “average” level of contamination in the area has to be determined, the more effective way to sample is toentify, as precisely as possible, the full extent of the decision area and carry out a multi-incremental sampling process, usually preceded by developing a sampling grid.
Selection, Use, Documentation & Other Challenges Even after the purpose of the sampling has beenentified and the sampling strategy has been selected, several additional challenges remain. Among the more important of those challenges are: (1) selecting the best tools and equipment; (2) using those tools correctly; (3) properly documenting the entire process; and (4) preserving the collected samples from the time of collection until their receipt at the laboratory.
The error rate associated with sample analyses usually ranges from 2 to 20 percent, and is linked to and/or dependent on: (1) complete extraction of the target analytes (the substances being analyzed); (2) correct calibration of the analytical instrument(s); (3) proper execution of analysis; and (4) verification that the correct samples are being analyzed. In a typical laboratory environment, these issues are carefully and effectively met, reducing the error rate to a very low and therefore scientifically acceptable level.
Here it is important to remember that sample preparation usually is associated with a higher level of error – ranging from 100 to 300 percent, depending on the type and level of preparation required. The problems associated with preparation are primarily linked to the subsampling process, where the goal is to obtain a homogenous sample representative of the larger sample.
The error rate associated with sample collection is significantly greater – up to as much as 1000 percent, in fact – than the rates associated with sample preparation and analysis. The errors found are linked to any of several factors involved, including but not limited to the following: insufficient sample mass; the improper selection and/or use of tools; limited access to the population that should be sampled; sample contamination, loss, and/or reactions; and even the improper or erroneous selection of the sampling location.
Detailed Planning, Beforehand, Is the Key Requirement Most if not all error rates can be reduced to at least some extent by careful planning. The planning processeally should address all of the numerous issues (not all of them scientific in nature) related to sample collection. For example, the purpose of the specific task(s) involved needs to beentified. Also, an adequate budget should be developed and approved, and an optimum strategy selected to mesh the need for the collection of samples with the budget available.
The development of a detailed sampling plan is also an essential prerequisite. That plan should address such generic topics as: sample collection, preservation, and shipment; the sample collection tools needed; the documentation required; and the statistical sampling process.
Because a significant cause of sample collection error is linked to the sample process itself, particularly careful attention must be paid to the choice of sample collection tools and equipment. Use of the “wrong” tools (and/or improper use of the “right” tools) can introduce an unintentional bias into the findings. Unfortunately, there seems to be no practical way to measure such biases, so the level of error also cannot be calculated.
To avoid cross contamination, sample collection tools should therefore, whenever possible, be sterile, single-use in nature, easy to use, and compatible with both the sample being collected and the analytical method used to evaluate the sample.
Cost, Compatibility, and Other Decision-Making Factors QuickSilver Analytics offers several examples of collection systems specifically designed for the collection of biological samples while avoiding cross contamination. The company’s All-in-One Swab, for example, is a self-contained swab system designed for sampling either a very small location or a specific point-source sample. The All-in-One Swab is also compatible with the Critical Reagents Program’s Hand Held Assays, and provides a low cost per sample. In addition, the company’s B2C (Bulk Bio Collection) and SP2C (Swab Powder Sample Collection) Kits are designed for the collection of powders from nonporous surfaces. The B2C is capable of sampling a large area, while the SP2C is designed for a smaller area. Moreover, the B2C and SP2C kits meet the sample collection requirements postulated both by the American Society of Testing and Materials (ASTM) and the Association of Official Analytical Chemists (AOAC – the latter acronym is an “unofficial” name derived from an earlier (1884) acronym for the Association of Agricultural Chemists).
In addition to having the “right” tools available and meeting the other requirements mentioned earlier, each sample must be thoroughly documented. That documentation should include, but not necessarily be limited to: (1) the exact sampling location (determined by GPS and/or photography); (2) the means (e.g., tools, methods) by which the sample was collected; and (3) the person(s) who collected and/or witnessed the collection process.
Equally important is the chain-of-custody documentation, which records exactly how the sample was transferred from one individual to another and from the sample-collection location to the laboratory. Documenting the chain of custody preserves the traceability of the sample from the time of collection through analysis and reporting. However, without proper packaging and preservation during transit, such documentation is almost always useless.
The requirements for sample preservation, which will vary based on the analyte(s) of interest, must also be fully and correctlyentified during the planning phases of the project. Extensive thought and planning are necessary, therefore, to select, collect, and handle the samples needed to provide all of the data required to support decision making.
To assist in this planning, there are several published guidelines such as those available not only from ASTM but also the U.S. government – more specifically, the Environmental Protection Agency. In short, with proper planning, sampling can and should be used as a valuable tool whenever critical decisions must be made.
Patti Riggs is the Principal Chemist and Quality Manager at QuickSilver Analytics, where she has been employed for the past fifteen years. She has a Master of Science degree from the University of Delaware, and a strong interest in improving all aspects of chemical analysis.