A topic in the Open Knowledge Graph — a free, open map of 15,290 topics and the order to learn them in.
Selectivity (the ability to distinguish an analyte from interferences) and sensitivity (the ability to detect small amounts) are often inversely related in analytical methods. High selectivity may require longer analysis times or more complex sample preparation, while maximizing sensitivity can increase background noise and reduce the ability to differentiate signals. Method development requires understanding these trade-offs and optimizing for the specific application requirements.
Compare selectivity and sensitivity parameters across different LC and GC methods for the same analyte. Use detector types (UV, mass spectrometry, electrochemistry) as case studies showing how detector choice affects both properties. Design experiments where improving one parameter degrades the other.
From your introduction to analytical chemistry, you know that a good analytical method must detect your target analyte reliably (sensitivity) and distinguish it from other substances in the sample (selectivity). What becomes clear at the method development stage is that these two qualities pull against each other in most instrumental techniques, and optimizing one often degrades the other. Understanding this tradeoff is essential for choosing and tuning methods appropriately for each analytical problem.
Consider a concrete example with UV detection in HPLC. Measuring at 254 nm (a common default wavelength) gives you broad sensitivity — many organic compounds absorb there — but poor selectivity because your analyte peak might overlap with dozens of other UV-absorbing compounds. Switching to a wavelength where only your analyte absorbs strongly (say, 340 nm for a compound with an extended conjugated system) improves selectivity dramatically but reduces sensitivity for compounds that absorb weakly at that wavelength. A mass spectrometer as a detector can monitor a specific mass-to-charge ratio (selected ion monitoring), giving exceptional selectivity for your target compound's molecular ion, but in doing so it ignores all other ions — if your analyte fragments or ionizes poorly, you lose sensitivity. Tandem mass spectrometry (MS/MS) in selected reaction monitoring mode pushes selectivity even further by requiring a specific precursor ion to fragment into a specific product ion, virtually eliminating chemical noise — but the signal intensity drops with each stage of mass filtering.
The tradeoff extends beyond detector choice into sample preparation and chromatographic conditions. A highly selective extraction procedure — say, immunoaffinity cleanup that binds only your target mycotoxin — produces a very clean extract with minimal background, but the antibody binding step may not capture 100% of the analyte, reducing recovery and effective sensitivity. Running a longer HPLC gradient improves selectivity by spreading peaks further apart in time, but the peaks broaden, reducing peak height and thus detection sensitivity for the same injected mass. Adding ion-pairing reagents to the mobile phase can dramatically improve selectivity for charged analytes on reversed-phase columns, but they may suppress ionization in a mass spectrometer, hurting sensitivity.
The practical resolution of this tradeoff depends on what your application requires. Screening methods for unknown contaminants prioritize broad sensitivity — you want to detect anything that might be present, even at the cost of occasional false positives from co-eluting interferences. Confirmatory methods for regulated analytes prioritize selectivity — you need to prove beyond doubt that the signal is from your target compound, not an interferent, even if that means a higher detection limit. The best method development approaches evaluate both parameters explicitly, often plotting figures of merit like signal-to-noise ratio and resolution as functions of adjustable parameters (wavelength, mobile phase composition, extraction conditions) to find the operating point that best serves the specific analytical question. Recognizing that no single method maximizes both selectivity and sensitivity simultaneously prevents the common mistake of chasing ever-lower detection limits without considering whether the measured signal is actually coming from the right compound.