UV-Vis Spectrophotometer: How It Works

A UV-Vis spectrophotometer measures how much light a sample absorbs at specific wavelengths in the ultraviolet (roughly 190–400 nm) and visible (400–700 nm) range, and it is one of the most heavily used instruments in a research lab — not for a single dedicated assay, but as the underlying measurement principle behind dozens of everyday lab tasks: checking how much DNA or RNA is in a prep, tracking bacterial growth, quantifying protein, or reading a colorimetric assay endpoint. Most searches for the instrument itself land on vendor product pages comparing models; this guide instead covers the actual physics and workflow — what the instrument is doing, why the math behind it works, and where people get the reading wrong.

What a UV-Vis spectrophotometer actually measures

The instrument shines light of a chosen wavelength through a sample and compares how much light comes out the other side to how much went in. Molecules absorb light at wavelengths that correspond to their electronic structure — conjugated systems, aromatic rings, and certain metal complexes all have characteristic absorbance peaks. The instrument reports this as absorbance (A), a unitless log-scale value, or as %transmittance (%T), the fraction of light that made it through.

The core hardware path is the same across nearly every bench UV-Vis instrument:

Light source — typically a deuterium lamp for the UV range and a tungsten-halogen lamp for the visible range (many instruments switch between the two, or use a xenon flash lamp that covers both).
Monochromator — a diffraction grating that isolates a narrow band of wavelengths from the source and directs it toward the sample, so the instrument can scan across or fix on a specific wavelength.
Sample compartment — holds the cuvette containing the sample; a matched reference (blank) is measured separately or simultaneously depending on the instrument design (single-beam vs. double-beam).
Detector — a photodiode or photomultiplier tube that measures the intensity of light after it passes through the sample and converts it to an electrical signal.

Cuvette material matters more than it looks: standard disposable plastic cuvettes absorb strongly below roughly 300–340 nm, so they work fine for visible-range colorimetric assays but will distort or block a true UV reading (e.g., nucleic acid quantification at 260 nm). Quartz cuvettes are transparent across the full UV-Vis range and are required whenever the measurement wavelength is below the plastic cutoff.

The Beer-Lambert law: why absorbance is proportional to concentration

The reason a spectrophotometer can report a concentration, not just a light reading, is the Beer-Lambert law:

A = ε × c × l

A — absorbance (unitless)
ε (epsilon) — molar absorptivity (or molar extinction coefficient), a constant specific to the substance and wavelength, in L·mol¹·cm¹
c — concentration of the absorbing species, typically in mol/L
l — path length the light travels through the sample, typically 1 cm for a standard cuvette

Because path length is usually held constant (a standard 1 cm cuvette) and molar absorptivity is a fixed property of the molecule at a given wavelength, absorbance and concentration are directly proportional — double the concentration, and (within the law’s valid range) absorbance doubles too. That linear relationship is what makes a calibration curve possible: measure absorbance for several samples of known concentration, plot absorbance against concentration, fit a line, and use that line’s equation to convert an unknown sample’s absorbance back into a concentration.

The law breaks down at high absorbance (commonly above roughly A = 1.0–2.0, depending on the instrument), where stray light and detector non-linearity cause the relationship to curve away from a straight line. A reading in that range should be diluted and re-measured rather than trusted directly — see the common errors section below.

Blanking, and why it’s not optional

Before measuring a sample, the instrument is “blanked” (or “zeroed”) using a reference that contains everything the sample contains except the analyte of interest — the same buffer, solvent, or diluent. This subtracts out absorbance from the cuvette itself, the solvent, and any background contribution so that the reported absorbance reflects only the compound being measured. Using the wrong blank (plain water instead of the actual assay buffer, for example) is one of the most common sources of a systematically off reading, because buffer components, detergents, or even cuvette scratches can absorb at the wavelength of interest.

Building and reading a standard (calibration) curve

To convert absorbance into an actual concentration for a compound without a known, reliable molar absorptivity value, most labs build an empirical standard curve rather than relying on the Beer-Lambert equation directly:

Prepare a dilution series of known concentrations of the analyte (a serial or simple dilution from a stock — see the molarity and solution-preparation guide for the underlying math).
Measure absorbance for each standard at the chosen wavelength, blanked against the same buffer.
Plot absorbance (y-axis) against concentration (x-axis) and fit a linear regression; a good standard curve typically has an R² close to 1.0 within the linear range.
Measure the unknown sample’s absorbance and solve the fitted equation for concentration — diluting first if the raw reading falls outside the linear range established by the standards.

Common research-lab applications

UV-Vis absorbance underlies several assays researchers run routinely, each using a different characteristic wavelength:

Nucleic acid quantification (A260) — DNA and RNA absorb maximally around 260 nm due to their nitrogenous bases. The A260/A280 ratio is used as a rough purity check: a ratio near 1.8 is generally taken as an indicator of relatively pure DNA, and near 2.0 for RNA; a lower ratio can indicate protein or phenol contamination carried over from extraction.
Protein quantification (A280) — aromatic amino acids (mainly tryptophan and tyrosine) absorb around 280 nm, allowing a rough direct protein estimate, though colorimetric assays (Bradford, BCA) that read absorbance in the visible range are often used instead for better accuracy at low concentrations.
Bacterial growth (OD600) — optical density at 600 nm is a standard, wavelength-specific way to track culture turbidity as a proxy for cell density over the course of a growth curve.
Colorimetric and enzymatic assays — many kit-based assays (protein quantification kits, enzyme activity assays, plate-based colorimetric readouts) generate a colored product whose absorbance in the visible range is read out as the assay’s endpoint.

Common sources of error

Wrong cuvette material — using plastic instead of quartz for a UV-range measurement (see above).
Air bubbles or particulates — scatter light and produce artificially high, noisy readings; spin down or filter samples first.
Fingerprints or scratches on the cuvette’s optical faces — handle cuvettes by the ribbed/frosted sides only.
Reading outside the linear range — absorbance readings above roughly 1.0–2.0 (instrument-dependent) should be diluted and re-measured rather than accepted at face value.
Wrong or stale blank — re-blank whenever the buffer, solvent, or cuvette changes.
Wavelength drift — instruments should be periodically checked against a certified reference material (such as a holmium oxide or didymium glass filter) as part of a documented calibration/verification program, particularly in a regulated or quality-managed lab.

Fixed wavelength vs. full spectral scan

Routine quantification (OD600, A260, A280, a colorimetric endpoint) reads absorbance at a single fixed wavelength. A full spectral scan — sweeping across a wavelength range and plotting absorbance at every point — is used instead when characterizing an unknown compound, confirming a peak wavelength before setting up a fixed-wavelength assay, or checking a sample for contamination signatures that would show up as an unexpected secondary peak.

Frequently asked questions

What is a UV-Vis spectrophotometer?

It is an instrument that measures how much light a liquid sample absorbs across the ultraviolet and visible wavelength range, and uses that absorbance — via the Beer-Lambert law and/or a standard curve — to determine the concentration of a light-absorbing compound in the sample.

How does a UV-Vis spectrophotometer work?

A light source generates light across the UV and visible range; a monochromator isolates a specific wavelength; that light passes through the sample in a cuvette; a detector measures how much light made it through compared to a blank reference; the instrument converts that ratio into an absorbance value, which is proportional to concentration under the Beer-Lambert law.

What is the difference between absorbance and transmittance?

Transmittance (%T) is the direct fraction of light that passes through the sample. Absorbance (A) is the negative base-10 logarithm of transmittance (A = −log₁₀(%T/100)); absorbance is used for quantification because it is linearly proportional to concentration, while transmittance is not.

Why is my absorbance reading negative or unstable?

A negative or drifting reading is most often caused by an incorrect or stale blank, a dirty or scratched cuvette, an air bubble in the light path, or a lamp/detector that hasn’t warmed up; re-blank with fresh reference solution and check the cuvette before troubleshooting further.

Do I need a quartz cuvette?

Only if the measurement wavelength falls in the UV range below roughly 300–340 nm (this includes 260 nm nucleic acid and 280 nm protein readings) — standard plastic cuvettes absorb too strongly in that range to give a reliable reading. For visible-range colorimetric assays, plastic cuvettes are generally fine.

This guide covers the underlying measurement principle and bench workflow, not a specific instrument model or vendor. For dilution and solution-preparation math used to build calibration standards, see the molarity and solution calculations guide.