1 The Sample

In the text the word sample is used in two ways. The sample referred to in Part 1 is the portion of material submitted for chemical analysis, e.g. a 10 mL sample of olive oil removed from a one litre bottle of oil for the purpose of chemical analysis. When statistical procedures are applied to the results of analytical measurements (Part 6) the term sample refers to the segment of all the possible results, i.e. the population that is being used to calculate the statistic. Ten measurements are used to calculate the sample standard deviation (see random errors, Section 6.2); the ten results are only a portion of the infinite number of possible measurements.

Terms used in describing samples and sampling procedures are many and varied. The text below aims to describe those in common usage. It is clear that there is potential for confusion. It should therefore be a golden rule for any analytical chemist to define clearly the terms used in the context of the particular problem being addressed so that any procedure can be understood and repeated.

1.1 Sampling

Sampling Sampling schedule
Sampling plan Sampling scheme
Sampling programme Sample variability

One of the first considerations in any analytical work is what information is required from the analysis. Fundamental to this is the sample and the procedure by which it is obtained, i.e.sampling. Consideration needs to be given to whether the sample is part of a larger whole, how it should be removed, and how it should be stored and processed prior to the analytical measurement.

Samples are taken from an infinite variety of sources and the sampling procedure has to be designed accordingly. There should be a requirement to hold discussions between the customer, the analyst, the sampling officer etc. to reach agreement on what is a representative sample.

Example: 1.1

Sampling

Sampling a day's production of shampoo, already divided into several thousand individual containers, is very different from sampling water from a reservoir. Various means of removing samples are available (sampling method).

The method of sampling selected should be appropriate for the analysis. Sampling a tank of stratified liquid by dipping an empty sample bottle in from the top may be inappropriate, as this would discriminate with respect to the top layer. This method of sampling may, however, be suitable if the top layer was the object of the analysis.

With the exception of portable on-site testing kits, analytical equipment is rarely sited adjacent to the sampling point. Consideration should be given to the need for sample storage and transportation. For unstable samples it may be appropriate to stabilise the sample in some way before analysis can be carried out, e.g. this may involve chilling the sample.

Example: 1.2

Sampling programme, sampling plan, sampling schedule, sampling scheme

The use of a sampling programme is common practice in the operation of a major chemical manufacturing plant. Analysis is carried out to ensure the plant is running efficiently and that feedstock and products are within the specified limits. Samples of the feeds and production streams are, therefore, taken regularly. The sampling plan is formulated for each analyte or group of analytes and combined into a sampling schedule or sampling scheme where the sampling point, sampling times, and frequency of sampling are defined.

The following require different sampling plans:

If feedstock is taken from a well-mixed supply tank for periods of days without the tank being refilled it is clearly unnecessary to analyse the feedstock each hour.

If the composition of production streams is known to vary slowly on changing the operating parameters of the plant then the sampling should reflect the fact that, even if operational changes are being made, analyses will not be required at a higher frequency than the time constant of the process.

Samples are taken from designated and carefully chosen sample points to meet the needs of the process. Therefore a sample taken to ensure product quality may be taken prior to the product storage tank so, if problems are identified, the material can be diverted to an off-specification tank for reprocessing.

Sampling of materials which are at non-ambient conditions also require special considerations - this is clearly the case with compressed gases and liquids where removing the sample under different conditions can result in a change to the sample.

Generally, the sampling process used must reflect a detailed consideration of the requirements of the analysis and the material being sampled.

A sampling plan needs to be established which involves decisions as to when, where, and how the sample should be taken, and whether the process of sampling should be a one-off or whether it should be repeated and how often. Safety implications of sampling procedures, sample preservation and transportation should be considered in any sampling plan. When there is a regular requirement for analysis the sampling plan is referred to as a sampling scheme or a sampling schedule.

The term sampling programme is commonly used to describe a combination of procedures where several related sampling schemes are combined.

Example: 1.3

Sample variability

Samples taken from a water supply reservoir may vary with the time of day and the position from which they are taken. These factors would have to be considered in order to determine an overall analytical result for the reservoir.

Samples can vary with time, or with location, and the range of samples conceivably attainable is referred to as the sample variability.

1.2 Types of Sample

Aggregate sample Judgmental sample
Batch Lot
Bulk sample Random sample
Composite sample Representative sample
Gross sample Retained sample
Heterogeneous sample Selective sample
Homogeneous sample

The term bulk sample, sometimes used interchangeably with gross sample, is an amount of material from which further sub-divisions are to be made. The lot is the amount of bulk material from which a particular sample is taken and can in some cases be identical to a batch, which is material produced or taken at the same time.

Figure 1 shows the relationship between the various operations in a sampling scheme up to sample taken for analysis. The terms used provide an alternative to those used here.

A homogeneous sample is one for which there is no evidence of variation throughout its extent.

Example: 1.4

Homogeneous sample

A drinking water sample would be homogenous for sodium ions.

Conversely, a heterogeneous sample is one that varies throughout its extent; any sub-samples taken from it may be expected to vary in composition.

Example: 1.5

Heterogeneous sample

A soil sample made up of stones, clay, organic matter, etc. would be heterogeneous for iron(II) ions.

A composite sample, sometimes referred to as an aggregate sample, is one that is made up of several individual samples. Thus, a heterogeneous material may be sampled several times and these samples combined to produce a composite sample that is aimed to have the overall properties of the material.

A representative sample is one that is regarded as having identical composition to the larger bulk or batch from which it has been taken. Results obtained from a representative sample can thus be regarded as also applying to the bulk of material from which the representative sample was taken (see also Integrity, Section 1.3).

A selective sample is one which has been chosen to demonstrate a particular aspect of the material from which it is taken, e.g. a sample of water taken from immediately downstream of an industrial outfall to assess the worst possible situation with regard to potential pollution. Some analysts refer to judgmental samples as those chosen to illustrate a particular property of the material under consideration.

A random sample is chosen from a bulk material at random, i.e. so that any part of the bulk has an equal chance of selection, and without any particular reference to the properties of the bulk.

A retained sample is a sample taken for future reference purposes, e.g. for use in analysis where the result is the subject of a dispute.

1.3 Sample Handling

Aliquot Sample accountability
Integrity Sub-sample
Laboratory sample Test portion
Primary sample Test sample

The integrity of a sample should be maintained during the process of sample handling. The integrity of a material is described as the closeness of the state of the material to its original form. In the case of a certified reference material (see Section 3.1) it would be the closeness of the state of the material to that described on the certificate. During the sample handling and preparation stages of an analysis, the procedures should not change the information contained within the sample.

Example: 1.6

Integrity

Along with the identity of a sample, its integrity is also important in forensic cases. An athlete has supplied a urine sample for drug testing. If the sample is contaminated at any stage then the integrity of the original sample (as described on the sample label) has been destroyed and the analytical results are worthless and cannot be used in evidence.

A term related to integrity is sample accountability, which is the procedure by which the appropriate documentation regarding the sampling process and subsequent actions prior to analysis is established and maintained. The sample will have a unique identity and each stage of handling the sample will be documented so that its whole 'life cycle' is traceable. Sample accountability is thus part of the audit trail (see Section 5.1) of a sample.

A primary sample is the sample taken from the lot or batch; in most cases a portion of this sample will be taken and known as a sub-sample. The sub-sample or part of the sub-sample is used to provide the laboratory sample.

A laboratory sample is the material sent to or received by the laboratory. The test sample, which is sometimes the same as the laboratory sample, provides the source of material used to carry out the analysis. In many cases the laboratory sample has to undergo some type of sample preparation before it is ready for analysis.

Example: 1.7

Laboratory sample [right arrow] Test sample

100 g of wet sediment was submitted to the laboratory for analysis. The laboratory sample was dried at 3O °C for 24 hours and then homogenised to provide the test sample, a portion (test portion) of which was used for analysis.

A test portion is the amount of the test sample that is used for a particular determination.

An aliquot in this context is a part of a laboratory sample used in the analytical determination: the term is used commonly for the individual portions taken to repeat a single analysis.

Example: 1.8

Aliquot

To perform the analysis, the analyst takes two 1O mL aliquots from the 100 mL test (sample) solution.

1.4 Sample Preparation

Clean-up Extraction
Digestion Pre-concentration
Dissolution Pre-treatment

Samples, as taken, are often unsuitable for direct analytical measurement. These samples will require some pre-treatment. Sample pre-treatment is a term used to encompass a variety of sample preparation procedures, including pre-concentration, clean up, extraction, dissolution, digestion and homogenisation.

Samples are pre-concentrated in order to allow detection to be achieved. This is required when the concentration of analyte in the sample is too low or when as a result of pre-treatment the concentration in the resulting solution is too low for detection. Care is needed in any pre-concentration procedure to ensure that it does not lead to contamination from, for example, the equipment or chemicals used.

Example: 1.9

Pre-concentration

A water sample requiring heavy metal determination may be evaporated to a smaller volume prior to using atomic absorption spectrometry.

Extraction procedures involve the removal of the analyte from a solid or liquid sample so that the analytical requirements can be achieved more satisfactorily than before, e.g. liquid-liquid extraction or solid-phase extraction.

Dissolution is the conversion of a solid or gaseous material into a solution. Digestion is the process of bringing a material into solution using, e.g. a solvent, often with the application of heat. Both are processes of preparing a sample for analysis. In doing so a heterogeneous sample is made homogeneous. Dissolution is commonly used in the pharmaceutical industry for preparing tablets for analysis.

Clean-up procedures involve the separation of analyte(s) from other sample components that might otherwise make the analysis more difficult. Filtration, sublimation and centrifugation are techniques that can be used for sample clean-up.

1.5 Sample Blanks

Blanks as part of sampling are described in this section whereas blanks used in the analytical procedure are covered in Section 5.4.

Field blank
Sample blank
Transport blank

Sample blanks are samples that are used to establish the effect of the sample matrix or to ensure that the matrix is not affecting the analysis. They approximate to the analytical samples except that they do not contain any measurable concentration of the analyte under investigation.

Sample blanks used for specific purposes may have more specific titles, such as transport blanks and field blanks. Transport blanks are taken through exactly the same journey as the analysis samples apart from the inclusion of the analyte. Field blanks are taken through the sampling process and are subject to all the possible sources of contamination without having the analyte included.

Example: 1.10

Transport blank, field blank

Organic vapours in air can be sampled by pulling air, by means of a pump, through a tube containing charcoal. A transport blank would be a charcoal tube taken to the sampling site but never removed from the sampling case. A field blank would be a charcoal tube taken to the site, the ends of the charcoal tube broken and then connected to an air sampling pump without the pump being switched on.

2 The Analytical Method

Analytical methods are used to qualitatively and quantitatively establish the composition of a material. An analyst is often faced with answering questions of the type:

• Can this unknown substance x be identified?

• How much is there of x in y (e.g. lead in water, pesticides in fish, oil in a shale) or is there a suitable way of estimating the amount of substance x present? Is it necessary to take account of other chemicals present in y when performing the analysis?

• If the analysis is repeated using identical samples will the results be identical?

• If two different people use the same analytical method to analyse identical samples, will they both find the same amount of x in y?

• If an alternative analytical method is used, will it give the same result for identical samples?

This Part deals with the language associated with analytical methods.

2.1 Analysis

Analysis is the measurement of the amount of substance x present in the sample using an analytical method; everything else in the sample is the matrix. Some general names that one can give to x and the process of analysis are:

Analyte
Matrix
Measurand
Test

When an analysis is performed the amount of a chemical component present in a sample is identified and measured. The chemical entity being investigated (qualitatively or quantitatively) is called the analyte. There may be components, other than x, present in the sample. All the 'other' components are called the matrix. The process of analysing the sample to determine the amount of the analyte in the matrix is called a test.