Systematic errors

Systematic errors

The consistent deviation of analytical results from the "true" value

Sources

Systematic errors can be classified into:

(a) Instrument errors -caused by faults in measuring instruments and instabilities in power supply

(b) Method errors –   arise from non ideal chemical or physical condition of reagents and reactions on which the analytical method is based upon

(c) Personal errors –  errors originating from the individual analyst and are not related to the method or procedure; result from carelessness, personal limitations, ignorance, etc.

Instrumental errors

 Most measuring instruments suffer from systematic errors. Glassware measuring devices like pipettes may deliver or hold different volume from that indicated. These differences may arise if the glassware is used at a temperature significantly different from the calibration temperature, distortions in the container walls due to heating, contaminants on the inner surface of the container. Such systematic errors can be controlled by calibration.

Electronic instruments many sources of uncertainties e.g. decrease in voltage of a battery operated power supply, increased resistance in circuits due to dirty electrical contacts, variations in resistors and standard potential sources. Errors from electronic instruments can be detected and controlled

Method error;Sources of nonideality of reaction conditions include: slowness, and incompleteness of some reactions, possibility of side reactions that interfere with the measurement process, instability of some species, nonspecifity of some reagents. Example - a common method error is the excessive reagent      added in a titration where endpoint is determined by color change.Method errors are the most difficult to detect.   

Personal errors;Apart from the correctness of the method and instruments, all measurements require personal judgment e.g. the color of solution at the endpoint of titration, eye position in reading the volume of a reagent in a burette etc. Such judgments lead to consistent, unidirectional errors i.e. systematic errors. For example an analyst who is color blind or less sensitive to certain color may record either excessive or insufficient volume of a reagent in a titration.preconceived notion of a true value for the measurement or a natural tendency to estimate scale reading in a direction that improves in a set of results.Constant errorsMagnitude is independent of the size of quantity measuredBecome more serious as the size of the quantity measured decreases.

Example: Excess reagent required to show color change at the endpoint constitutes a constant error. Since it is constant irrespective of the volume of the reagent used to reach the endpoint, the effect is more pronounced where small volumes are involved.correct method errors

(a) Analysis of the standard samples,Method errors can be detected by analyzing standard reference materials - materials containing exactly known concentration levels of one or more analytes.

Limitation- Results obtained from the method being evaluated will always differ from those obtained using the standard reference material. Therefore there will be a need to determine whether the difference is due to random errors or not (Application of significance test to data treatment).Standard reference materials are obtained by purchase of standard reference materials from government and industrial authorities. Such authorities market reference materials such as rocks and polymers, rainwater, river sediments, minerals, glasses, gas mixtures, etc.

(b) Independent analysis,In absence of reference standard materials, another independent and reliable analytical method can be carried out in parallel with the method being evaluated.

The independent method should differ significantly from the method being evaluated in order to minimize or preclude common factors from affecting the methods in a similar manner.Statistical test will be required to determine whether difference in results is due to random errors.

(c) Blank Determinations,Useful in detecting certain types of constant errors and even proportional errors.

In blank determination all steps in an analysis are performed in absence of a sample. The result of the blank are then used to correct sample measurement. Most blank determination reveal errors due to contaminants from the reagent or vessel being used in an analysis. In some titrimetric analysis, blank determinations allow for correction of titration data for volume required to change colour at the end point.

(d) Variation of Sample size,Varying the sample size facilitates the detection of constant errors: increasing the size of the sample decreases constant error.