How it works
- The pore will not wet — The fibre wall is polypropylene, and water will not enter a pore that small against its own surface tension. So the pore stays full of gas, and dissolved CO₂ crosses it by diffusion and leaves — no bubble, no gas–liquid interface, nothing to disengage. Exceed the breakthrough pressure, or put oil or surfactant in the feed, and the pores flood. The contactor then does nothing at all, and rinsing will not bring it back.
- Counter-current, and the floor — Sweep gas enters at the clean end and leaves at the dirty one, so the water always faces gas leaner than itself. Nothing can be stripped below equilibrium with the gas on the far side of the membrane, and that floor is why the sweep is scrubbed: ambient air at 420 ppmv of CO₂ would hold the outlet near 0.1 mg/L however much membrane you bought.
- The pump is mostly steam — The pore is open to water vapour just as much as to CO₂, so the gas side leaves saturated at the water temperature. At 200 mbar and 25 °C that is 31.7 mbar of the 200 — a sixth of the volume. Only two things cross the membrane, and the pale ones are water: at these conditions about fifty molecules of steam leave for every one of carbon dioxide. Pull the vacuum deeper and that ratio climbs with it — which is why chasing a deeper vacuum buys a bigger pump far faster than it buys a better outlet.
- Why oxygen is a different machine — Air is a fifth oxygen and four parts in ten thousand carbon dioxide. The same trace of air leaking into the shell is invisible to a CO₂ duty and decisive for an oxygen one: it sets an oxygen floor of tens of ppb that no amount of pumping gets under. That is the whole reason ppb-oxygen duties run on nitrogen sweep, and why vacuum alone is fine for CO₂.
- What governs it — Membrane area buys outlet quality and nothing else. The pump is set by the vapour and the sweep gas, neither of which has anything to do with how much CO₂ you are removing — the CO₂ itself is under half a per cent of the suction volume. Add a standby train in service and the contactors improve while the pump gets 50 % bigger, because every module in service still wants its sweep.
- 6 × 28 in modules — The workhorse, and the size vendor sizing reports are usually written around. Most industrial degasifiers between 5 and 60 m³/h are banks of these. At 37.4 m² and 10 m³/h rated, each one is worth about 1.6 transfer units at its rating — and since every size is rated at much the same shell velocity, that figure barely changes between them. Module size is a pipework and reliability decision, not a process one.
Design parameters
The panel opposite runs the same correlations as the MD-630 design sheet, so the animation and the calculator cannot disagree. Every parameter below is live — drag one and the picture responds.
| Parameter | Unit | Default | Range |
|---|
| Contactor |
|---|
| Module size | | 6 × 28 in | 10 options |
| Area and rated flow move together, so the duty barely changes — what changes is how many shells, how much pipework, and how much of the plant one failure takes with it. |
| Duty |
|---|
| Feed water flow | m³/h | 17.5 | 2 – 200 |
| Free CO₂ in | mg/L | 10 | 1 – 60 |
| CO₂ out, design max | mg/L | 0.5 | 0.05 – 5 |
| Water temperature | °C | 25 | 5 – 45 |
| Gas side |
|---|
| Vacuum, absolute | mbar | 200 | 40 – 500 |
| Sweep gas, × module rating | – | 0.6 | 0.05 – 2 |
| Arrangement |
|---|
| Contactors in parallel | – | 0 | 0 – 8 |
| Contactors in series | – | 0 | 0 – 6 |
The full MD-630 design calculation — governing equations, accepted envelopes and the worked sizing — is in the design calculator workbook. Everything on this page stays free.