Biosensors explained

Reference guide written and published by biochip.com. Draft, not yet independently reviewed; last substantive update September 7, 2026.

A biosensor is a measuring device that uses a biological component to recognise a target and a physical component to turn that recognition into a readable signal. The glucose meter used by hundreds of millions of people with diabetes is a biosensor. So is the pregnancy test, the continuous glucose monitor worn on the arm, and many of the rapid antigen tests that became familiar during the COVID-19 pandemic. Biosensor chips are one of the main branches of biochip technology, and much of the research covered on this site is about making them more sensitive, more selective, cheaper or wearable.

The parts every biosensor has

The IUPAC definition, set out in a 2001 technical report on electrochemical biosensors, describes an integrated device in which a biological recognition element is in direct contact with a transduction element (Thévenot et al., 2001). In practice:

• Bioreceptor: the molecule or structure that binds or reacts with the target. Enzymes (glucose oxidase), antibodies, nucleic acid strands, aptamers, receptors, whole cells and, more recently, CRISPR enzymes are all used. Selectivity comes almost entirely from this component.
• Transducer: converts the binding or reaction into a signal. Electrodes measure current, voltage or impedance; optical structures measure light intensity, colour, fluorescence or refractive index; mechanical resonators measure added mass.
• Interface chemistry: the layer that attaches the bioreceptor to the transducer and resists everything else in the sample. Much of the practical difficulty of biosensors lives here.
• Electronics and readout: amplification, signal processing and display, increasingly in a phone or wearable.

The main ways of transducing a signal

Electrochemical

The largest category by far. Leland Clark's 1962 enzyme electrode, which trapped glucose oxidase against an oxygen electrode, is the ancestor of every glucose strip (Clark and Lyons, 1962). Modern strips use screen-printed electrodes and a mediator molecule that shuttles electrons from the enzyme to the electrode; the current is proportional to glucose concentration. Variants measure potential (potentiometric), impedance, or the current at field-effect transistors whose gate is functionalised with a bioreceptor. Electrochemical sensors are cheap, small and easy to integrate with electronics, which is why they dominate consumer and point-of-care products.

Optical

Fluorescence and colour changes are the workhorses of laboratory assays and of the lateral-flow test, where captured target concentrates coloured nanoparticles into a visible line. Label-free optical methods, such as surface plasmon resonance and silicon photonic ring resonators, detect the tiny change in refractive index when molecules bind to a surface, and can follow binding in real time. Optical sensors are often more sensitive than electrochemical ones but need light sources and detectors that are harder to miniaturise.

Mass and mechanical

Quartz crystal microbalances and microcantilevers shift their resonant frequency when mass binds to them. They are label-free and physically simple but sensitive to temperature, viscosity and vibration, so they remain mostly research tools.

Performance terms you will see in every paper

• Limit of detection (LOD): the smallest concentration reliably distinguished from zero. Often reported in buffer, where it is far better than in blood or saliva.
• Selectivity: response to the target compared with similar molecules and with the messy background of a real sample.
• Dynamic range: the span of concentrations over which the sensor responds usefully.
• Response time and reversibility: how fast the reading settles, and whether the sensor can be reused or measures continuously.
• Stability and shelf life: proteins denature, surfaces foul, and dried reagents degrade; commercial sensors must survive months of storage.

Where biosensors work well, and where they struggle

They excel when the target is abundant, the sample is simple or well understood, and the bioreceptor is robust. Glucose in blood fits all three. Lateral-flow tests work because they trade sensitivity for simplicity and speed; Martinez, Whitesides and colleagues extended the same logic to patterned paper for very low-cost assays (Martinez et al., 2007), and Yager and colleagues surveyed how such devices could serve settings with no laboratory at all (Yager et al., 2006).

They struggle when the target is scarce (early-stage cancer markers, single viral particles), when the sample is complex (whole blood, wastewater), or when continuous operation is needed. Biofouling, the accumulation of proteins and cells on the sensor surface, degrades any implanted or wearable sensor over days. Non-specific binding produces false signals. Calibration drift is why continuous glucose monitors still need periodic checks. And a very low detection limit in a research paper often depends on long incubation times, washing steps and laboratory instruments that disappear in a real product.

How to read biosensor claims

Ask what sample matrix was tested, how many real samples, whether the comparison was against a validated reference method, and whether the reported detection limit is clinically relevant rather than merely impressive. A sensor that detects a protein at femtomolar concentration in buffer has shown something about chemistry; it has not yet shown it can diagnose anyone.

Related terms

• Aptamer: a short synthetic DNA or RNA strand selected to bind a target; a more stable alternative to antibodies.
• Lateral-flow assay: a paper-strip immunoassay read by a coloured line; the pregnancy and rapid antigen test format.
• Label-free: detection that needs no dye or tag on the target molecule.
• Wearable biosensor: a sensor in contact with skin, sweat, tears or interstitial fluid that reports continuously.