HPLC Column Volume Explained
Geometric column volume = πr²L. A 150 × 4.6 mm column is 2.49 mL. But only about 68% of that is liquid — the rest is silica — so the void volume is 1.70 mL. Both get called "column volume", and which one a protocol means changes the answer by a third.
The geometric volume
A column is a cylinder, so it uses the cylinder formula. The only trap is units: internal diameter is quoted in millimetres, length in millimetres, but you want an answer in millilitres (= cm³).
| Step | Working | Result |
|---|---|---|
| Radius | 4.6 ÷ 2 = 2.3 mm | 0.23 cm |
| Length | 150 mm | 15 cm |
| Volume | π × 0.23² × 15 | 2.49 mL |
That is the empty tube. Nothing about packing has entered yet.
| Column | Geometric | Void (ε = 0.68) | t₀ at 1 mL/min |
|---|---|---|---|
| 250 × 4.6 mm | 4.15 mL | 2.83 mL | 2.83 min |
| 150 × 4.6 mm | 2.49 mL | 1.70 mL | 1.70 min |
| 100 × 4.6 mm | 1.66 mL | 1.13 mL | 1.13 min |
| 50 × 4.6 mm | 0.83 mL | 0.57 mL | 0.57 min |
| 150 × 2.1 mm | 0.52 mL | 0.35 mL | 0.35 min |
| 50 × 2.1 mm | 0.17 mL | 0.12 mL | 0.12 min |
Why only two-thirds of it is liquid
A packed column is full of silica particles. The mobile phase occupies only the space around and inside them:
- Interstitial volume — between the particles, roughly 0.40 of the tube.
- Intraparticle volume — inside the particles' pores, roughly 0.28.
- Silica skeleton — the remaining ~0.32, which holds no liquid at all.
Total porosity ε is the first two added, about 0.68 for conventional fully porous particles.
| Particle | ε | Why |
|---|---|---|
| Fully porous | ~0.68 | Liquid fills the pores throughout |
| Superficially porous (core-shell) | 0.55–0.60 | The solid core holds no liquid |
| Monolith | 0.80–0.90 | Very high through-pore fraction |
| Non-porous | ~0.40 | Interstitial space only |
Core-shell columns are a common trap. Fit one and assume ε = 0.68 and you overstate the void volume by roughly 15%, which shifts every retention factor and every gradient calculation. Their solid cores genuinely hold no mobile phase. When precision matters, measure with an unretained marker rather than assuming any of these figures.
The two conventions — and why protocols clash
This causes more confusion than the arithmetic ever does. "One column volume" means different things in different fields:
| Field | "Column volume" means | For 150 × 4.6 mm |
|---|---|---|
| Analytical HPLC / UHPLC | Void volume V₀ | 1.70 mL |
| Preparative / FPLC (e.g. Cytiva protocols) | Packed bed volume (geometric) | 2.49 mL |
The gap is ~32%. A protocol saying "wash with 5 column volumes" means 8.5 mL under one reading and 12.5 mL under the other.
Analytical chromatographers use void volume because that is what governs retention and gradient behaviour. Preparative work uses bed volume because it maps to how much resin was packed. Neither is wrong — but check which is intended before scaling anything, and state which you mean when you write a method.
Dead time, and why it anchors everything
Dead time t₀ is how long an unretained compound takes to cross the column:
t₀ = V₀ ÷ F
Every retention factor is measured relative to it: k = (tR − t₀) ÷ t₀. So an error in t₀ propagates into every k value in the method — which is why it is worth measuring rather than estimating.
To measure: inject an unretained marker (uracil on reversed phase; toluene on normal phase) and take the time of the first baseline disturbance. Multiply by flow rate for V₀. This captures the column's actual packing rather than a textbook porosity.
Scaling a method between columns
Two rules, and they work together:
- Flow scales with the square of the internal diameter — to keep linear velocity constant.
- Gradient volume in column volumes stays constant — to keep selectivity.
| From → To | Factor | 1 mL/min becomes |
|---|---|---|
| 4.6 → 2.1 mm | 0.208 | 0.21 mL/min |
| 4.6 → 3.0 mm | 0.425 | 0.43 mL/min |
| 4.6 → 4.6 mm | 1.000 | 1.00 mL/min |
| 2.1 → 4.6 mm | 4.798 | 4.80 mL/min |
Note how severe the 4.6 → 2.1 mm step is: a fifth of the flow. Injection volume scales the same way as the column volume ratio, so overloading a narrow column is easy if you forget.
Expressing the gradient in column volumes is what makes the transfer dimension-independent. Ten minutes on a 150 × 4.6 mm column and ten minutes on a 50 × 2.1 mm column are entirely different separations; eight column volumes is the same separation on both.
Does column volume change as a column ages?
The tube does not, but the void volume can. Bed settling or a void forming at the inlet changes the packing and therefore the liquid inside.
If your measured dead time drifts noticeably over a column's life, that is usually bed collapse, not a calculation error — and it is a reliable sign the column is nearing replacement. Tracking t₀ across a column's life is a cheap and underused diagnostic.
Common mistakes
| Mistake | Effect | Fix |
|---|---|---|
| Using geometric volume as void volume | ~32% overstated | Multiply by porosity |
| Using ε = 0.68 for core-shell | ~15% overstated | Use 0.55–0.60, or measure |
| Not checking which convention a protocol uses | Wash volumes out by a third | Confirm before scaling |
| Using diameter as radius | 4× too big | Halve the ID first |
| Scaling flow linearly with diameter | Wrong linear velocity | Scale with diameter squared |
| Expressing gradients in minutes across sizes | Different separation | Use column volumes |
| Estimating t₀ instead of measuring | Every k value shifted | Inject an unretained marker |
Frequently asked questions
How do I calculate HPLC column volume?
Treat the column as a cylinder: πr²L. For a 150 × 4.6 mm column the radius is 0.23 cm and length 15 cm, giving 2.49 mL. That is the empty tube volume, before any allowance for the packing inside it.
What is the difference between column volume and void volume?
Column volume is the empty tube. Void volume is the liquid actually inside the packed column — between and within the particles — and it is only about 60–70% of the tube, because the silica occupies the rest. A 150 × 4.6 mm column is 2.49 mL geometric but about 1.70 mL void.
What porosity should I use?
About 0.68 for conventional fully porous particles, 0.55–0.60 for superficially porous (core-shell) because their solid cores hold no liquid, and ~0.40 for non-porous. These are typical values, not certainties — measuring with an unretained marker always beats assuming.
Why do people disagree about what a column volume means?
Two conventions are in use. In analytical HPLC, a column volume usually means the void volume, since that governs retention and gradient behaviour. In preparative and FPLC work — Cytiva protocols especially — it means the whole packed bed volume. The gap is about 32%, so check before scaling.
How do I scale a method to a different column size?
Keep gradient volume in column volumes constant, and scale flow with the square of the internal diameter. Moving 4.6 → 2.1 mm means multiplying flow by 0.208, so 1 mL/min becomes about 0.21. Injection volume scales the same way as the column volume ratio.
What is dead time and how is it related?
Dead time t₀ is how long an unretained compound takes to pass through: t₀ = V₀ ÷ F. For 150 × 4.6 mm at 1 mL/min that is about 1.70 min. Every retention factor is calculated relative to t₀, so an incorrect value shifts every k in the method.
Does column volume change as a column ages?
The tube geometry does not, but the void volume can. Bed settling or a void forming at the inlet alters the packing and therefore the liquid inside. Noticeable drift in measured dead time usually signals bed collapse rather than a calculation error — and that the column is nearing replacement.
Why express gradients in column volumes rather than minutes?
Because minutes are meaningless across different column dimensions. Ten minutes on a 150 × 4.6 mm column is a completely different separation from ten minutes on a 50 × 2.1 mm one. Expressed as a number of column volumes, the gradient becomes dimension-independent and selectivity survives the scale.