How To Measure Concentration From Absorbance

You stare at the cuvette. Now, the spectrophotometer spits out a number: 0. 642. Here's the thing — absorbance. Unitless. Now what?

If you’ve ever stood in front of a UV-Vis spec with a fresh sample and a vague memory of Beer’s Law from undergrad, you know the feeling. The theory is clean. The practice? Messy. Still, path length isn’t always 1 cm. The blank drifts. This leads to the extinction coefficient you found in a paper might not match your buffer conditions. And nobody tells you that the linear range of your detector is usually narrower than the textbook claims.

Let’s walk through how to actually turn that absorbance reading into a concentration you can trust — without the hand-waving.

What Is Absorbance-Based Concentration Measurement

At its core, this is the Beer-Lambert Law in action. Because of that, light goes in, some gets absorbed, the rest hits the detector. The relationship between absorbance (A) and concentration (c) is linear — if you play by the rules. Easy to understand, harder to ignore.

$A = \varepsilon \cdot b \cdot c$

$\varepsilon$ is the molar absorptivity (extinction coefficient), usually in M⁻¹cm⁻¹. Consider this: $b$ is the path length in centimeters. $c$ is concentration in molarity.

$c = \frac{A}{\varepsilon \cdot b}$

Simple. But the variables hide traps.

The extinction coefficient isn’t a universal constant

You’ll see papers cite $\varepsilon_{280} = 1.So two proteins at 1 mg/mL can give wildly different A₂₈₀ values. The real value depends on sequence, buffer composition, pH, and even temperature. Or $\varepsilon_{260} = 50$ for dsDNA. Those are averages*. This leads to for proteins, the aromatic amino acid content (Trp, Tyr, Phe) drives absorbance at 280 nm. 0$ for 1 mg/mL IgG. If you’re using a literature $\varepsilon$ for a recombinant protein with a non-native buffer, you’re already off by 10–20% before you pipette.

Path length lies

Standard cuvettes are 1 cm. ) use 0.But that correction assumes perfect geometry. Think about it: 1 mm or 1 mm paths — and they adjust* the reading internally to report a “1 cm equivalent” absorbance. Day to day, microvolume pedestals (NanoDrop, DeNovix, etc. But dust, bubbles, or a slightly off-center drop changes the effective path length. Here's the thing — the instrument doesn’t know. It just reports what it calculates.

Why It Matters / Why People Care

You’re not measuring absorbance for fun. You need to know how much protein to load on a gel. Whether your DNA prep is clean enough for sequencing. If your enzyme assay has the right starting concentration. A 20% error in concentration propagates into every downstream calculation — specific activity, molar ratios, stoichiometry, dosing.

I’ve seen a whole week of crystallization trials fail because someone trusted a NanoDrop reading on a buffer with 500 mM imidazole. Imidazole absorbs at 280 nm. The protein concentration was overestimated by 2x. Day to day, the drops never equilibrated. Crystals never grew.

It’s not just about “getting the number right.” It’s about knowing how wrong* the number might be — and deciding if that matters for your next step.

How It Works (or How to Do It)

Step 1: Pick your wavelength and know why

• 260 nm — nucleic acids. Peak absorption for purine/pyrimidine bases.
• 280 nm — proteins. Aromatic side chains.
• 230 nm — peptide bonds, but also carbohydrates, phenols, thiols. Messy.
• 340 nm — NADH/NADPH assays. Clean, specific.
• Visible range (400–700 nm) — colored assays (Bradford, BCA, Lowry) or dyes.

Don’t just default to 280 nm because the button exists. If your sample has nucleic acid contamination, A₂₈₀ is inflated. If it’s a peptide with no Trp/Tyr, A₂₈₀ is near zero — you’ll get noise, not signal.

Step 2: Measure a proper blank

This is where most people rush. The blank must match the sample matrix exactly*. Here's the thing — same buffer, same pH, same additives, same temperature. Still, if your sample is in 20 mM Tris, 150 mM NaCl, 5% glycerol, 2 mM DTT — your blank is that exact buffer. Not water. Not “Tris buffer.” The exact thing.

Why? Because of that, high salt can shift baselines. This leads to dTT absorbs. Because glycerol absorbs at 260/280. A mismatched blank introduces a systematic offset that no math fixes later.

Step 3: Check the linear range

Every detector has a sweet spot. For most UV-Vis specs, absorbance between 0.1 and 1.0 is reliable. That said, above 1. 5, stray light and detector nonlinearity creep in. Below 0.05, signal-to-noise tanks.

Want to learn more? We recommend the smallest unit of a compound and write the electron configuration for a neutral atom of chlorine for further reading.

If your A₂₈₀ is 2.Dilute. On top of that, 3, don’t just plug it in. Back-calculate. And record the dilution factor*. 5 mg/mL instead of 5 mg/mL. Measure the dilution. I’ve seen people forget to multiply by 10 and report 0.It happens.

Step 4: Apply the right extinction coefficient

Three main approaches:

1. Literature/generic values

• dsDNA: 50 µg/mL = 1 A₂₆₀ unit (1 cm)
• ssDNA: 33 µg/mL = 1 A₂₆₀ unit
• RNA: 40 µg/mL = 1 A₂₆₀ unit
• IgG: 1.4 mg/mL = 1 A₂₈₀ unit (approx)
• “Average protein”: 1 mg/mL = 1 A₂₈₀ unit (very rough)

Use these only for quick checks. Not for publication data. Not for stoichiometry.

2. Sequence-based calculation (proteins)
Tools like ProtParam (ExPASy) compute $\varepsilon$ from the amino acid sequence. It counts Trp, Tyr, Cys (disulfide-bonded), and applies known molar absorptivities. This is much* better than generic values — but it assumes the protein is folded, fully oxidized, and in a non-absorbing buffer. If your protein has a heme group, a flavin, or a bound chromophore, the calculated $\varepsilon$ misses it entirely.

3. Experimental determination
Gold standard. Purify your protein. Determine concentration by amino acid analysis (AAA) or quantitative NMR. Then measure A₂₈₀. Calculate $\varepsilon = A / (b \cdot c)$. Now you have your* extinction coefficient for your* prep in your* buffer. Do this once per batch if you need high accuracy.

Step 5: Correct for contaminants (if you can)

Step 5: Correct for contaminants (if you can)

Even with a perfect blank and the right extinction coefficient, contamination can still skew your results. The most common offenders — nucleic acids, free dyes, residual detergents, or particulates — all introduce spectral interference that simple A₂₈₀ measurements cannot resolve.

For nucleic acid contamination, use the A₂₆₀/A₂₈₀ ratio as a diagnostic:

• A ratio of ~1.Consider this: 0 indicates pure protein
• A ratio >1. 2 suggests significant nucleic acid carryover
• A ratio <0.

If you suspect nucleic acids, apply the Pace & Schiebinger correction:
$ \text{[Protein]}{\text{corrected}} = \text{[Protein]}{\text{uncorrected}} - \left( \frac{A_{260} \cdot \varepsilon_{\text{protein}}}{\varepsilon_{\text{NA}}} \right) $
where $\varepsilon_{\text{NA}}$ is the extinction coefficient of the nucleic acid contaminant (typically 50 for dsDNA, 33 for ssDNA). This is an approximation — better than nothing, but not a substitute for actual cleanup.

For dye contamination (e.Here's the thing — g. Plus, , from Bradford or Coomassie staining), you can often dialyze or run a quick desalting column. For detergent carryover, measure A₂₈₀ before and after dilution — detergents cause nonlinear absorbance curves, so if doubling the dilution doesn’t halve the A₂₈₀, you’ve got interference.

And if you’re working with turbid samples — precipitates, aggregates, or membrane fragments — don’t trust absorbance at all. Switch to a colorimetric assay (BCA, Bradford, or Lowry) that measures only soluble protein, or use light scattering corrections if your instrument supports them.

The bottom line: Stop guessing. Start measuring.

UV-Vis spectroscopy is powerful precisely because it’s simple — but simplicity invites complacency. Every shortcut compounds: a wrong blank, an unvalidated extinction coefficient, an uncorrected contaminant — each silently erodes accuracy until your “precise” concentration is off by 20%, 50%, or more.

Before you click “calculate,” ask yourself:

• Did I match my blank to my sample matrix exactly?
• Is my absorbance within the linear range? In real terms, - Is my extinction coefficient appropriate for this protein, this buffer, this state? - Could anything in this sample be absorbing at my chosen wavelength?

If you can’t answer yes to all four, go back and fix it. Think about it: not tomorrow. Now.

Because the difference between a good experiment and a wasted one isn’t always in the pipetting — sometimes, it’s in the blank.

Takeaway:
Concentration determination is not a checkbox. It’s a chain of assumptions, each one capable of breaking the whole result. Treat it like the foundational step it is. Get it right, and everything downstream — binding curves, activity assays, structural models — stands on solid ground. Get it wrong, and even the most elegant downstream analysis becomes a beautifully constructed house on sand.