1. Begin with the decision
Before choosing an instrument, define the measurand—the quantity intended to be measured—and the decision the result must support. Are you establishing presence, concentration, identity, purity, size, charge heterogeneity, molecular structure, or change over time? The answer shapes the selectivity, resolution, calibration, and detection evidence required.
2. Follow the sample
Read Sampling and sample preparation. Ask what target population was sampled, how the laboratory sample and test portion relate to it, which components could be lost or introduced, how stability was preserved, and whether blanks and controls followed the relevant path.
3. Compare the dominant mechanism
Use the CE vs LC vs GC comparison as an orientation tool:
- CE strictly separates dissolved charged species through differences in electrophoretic mobility. Neighboring capillary electromigration techniques may instead add focusing or chromatographic retention, while a sieving medium changes effective mobility;
- LC separates through differential retention between a liquid mobile phase and a stationary phase; HPLC is an instrumental implementation, not a separate retention mechanism;
- GC separates compounds that enter the gas phase through volatility and stationary-phase interactions under controlled temperature.
None is universally best. Sample state, analyte chemistry, required resolution, detector, throughput, infrastructure, and validation needs change the answer.
4. Inspect the detector
Read Detection and coupled methods. A detector controls which separated components become visible and what kind of evidence they provide. Response sensitivity, selectivity, detection and quantification limits, acquisition rate, interface loss, and matrix effects belong in the same conversation.
5. Read one current paper slowly
Open A compact detector gives portable CE a spectrum, not just a signal. Identify the sample and preparation, controls and comparators, calibration and uncertainty, matrix, number of independent observations, and the exact result shown. Then separate four layers:
- bibliographic fact;
- what the authors report;
- SCI analysis of why it may matter;
- what remains uncertain.
Then follow the DOI to the original paper. SCI is a route into the evidence, not a replacement for it.
6. Choose a frontier
Finish with one context where constraints become visible: life science, environment and ocean, space and planetary analysis, or portable analysis.
The goal is not to memorize a technique tree. It is to recognize how sampling, mechanism, detector, context, and evidence boundary fit together.
7. Stop at the evidence boundary
Do not let a familiar instrument name carry more certainty than the experiment provides. A separated peak is not automatically an identified compound; an identified compound is not automatically a biological mechanism, environmental source, or biosignature. State what the data support, what depends on the authors’ interpretation, and which alternative explanations remain open.