Chemical Analysis Fundamentals and Quality Assurance
Chemical Analysis Fundamentals and Quality Assurance
This domain tests whether you can judge the metrological quality of a result, not just perform a measurement: accuracy, precision, selectivity, calibration, quality control, traceability, and the quality-system records behind a reported value.
Core rules
- Accuracy is closeness to a reference value; precision is dispersion among replicates. A method can be precise and still biased.
- Selectivity is the ability to measure the analyte in the presence of matrix components; specificity means no interference at all.
- Calibration curves should span the working range, use at least five levels, and be checked with independent standards. A high r² is weak evidence by itself; inspect residuals and lack of fit.
- For a calibration line with slope b and residual standard deviation s of the response, LOD = 3.3 s/b and LOQ = 10 s/b. State the basis of s.
- QA includes blanks, duplicates, spikes, certified reference materials, control charts and proficiency tests. QC must be independent of calibration.
- ISO/IEC 17025 requires traceability, validated methods, documented uncertainty and data integrity.
Worked example
A calibration gives b = 0.0250 A per mg/L and residual SD s = 0.00080 A. LOD = 3.3 × 0.00080 / 0.0250 = 0.106 mg/L; LOQ = 10 × 0.00080 / 0.0250 = 0.32 mg/L. Keep s in absorbance units before dividing by slope; reporting 0.0008 mg/L would be a unit error.
Traps
- Equating LOD with LOQ, or estimating LOD from a single blank.
- Using a calibration standard as the QC check; it cannot reveal calibration bias.
- Quoting precision as accuracy; a low RSD does not prove trueness.
- Ignoring matrix effects and blank correction.
- Treating r² = 0.999 as sufficient without residual checks.