DAF / IGF Flotation — Engineering Fundamentals

1. Why flotation?

Stokes settling becomes impractical for oil droplets below ~30 µm because settling velocity collapses with d². Flotation flips the problem: instead of waiting for droplets to rise, attach a microbubble that gives the droplet-bubble aggregate a 1,000× larger effective rise velocity. Cells process 5–10 µm droplets in 30 minutes that would take 10+ hours by gravity.

2. Bubble-oil capture mechanism

Three sequential physics:

1. Bubble nucleation — DAF: gas dissolved at high P comes out of solution as ~30–70 µm bubbles when released to atmospheric. IGF: shear from an impeller or eductor entrains ~200–1,000 µm air bubbles.
2. Bubble-droplet collision — both rise toward the surface; size-disparate populations have different rise speeds and collide stochastically. Higher A/O ratio = more bubble surface area = more collisions.
3. Attachment + flotation — bubble adheres to droplet (van der Waals + hydrophobic surface chemistry); aggregate buoyancy is positive and rises to the surface skim layer.

3. DAF — pressure saturator

A side stream (typically 20–50% of throughput) is pumped through an air-saturator at 50–90 psig where compressed air dissolves into the water at the Henry's-law solubility. The saturated stream is then released through a needle valve into the cell at atmospheric — the dissolved air comes out of solution as fine bubbles (30–70 µm).

The pressurized recycle stream contacts the inlet at the cell entrance, so the bubbles nucleate near the oil droplets, maximizing collision rate. Saturation efficiency (f = 0.5–0.8) accounts for incomplete dissolution at the saturator residence time.

4. IGF — induced gas

An impeller or eductor pulls atmospheric gas (air or sometimes inert N₂ for sour service) into the cell, creating ~200–1,000 µm bubbles in situ. No saturator, no high-pressure pump — simpler, lower OPEX, smaller footprint. Trade-off: larger bubbles = lower surface-area-to-volume = lower capture for sub-20 µm droplets.

IGF is the standard in offshore deck-space-constrained applications; DAF in onshore SWD facilities where saturator OPEX is justified by tighter outlet spec.

5. A/O ratio derivation

The mass of air actually nucleated as bubbles (per L of treated water) is:

where s_a = solubility at 1 atm (~24 mg/L at 90°F), P_abs = saturator pressure in atm absolute, R = recycle fraction. The 1.3 prefactor converts from mass of air dissolved at P down to volume of bubbles at atmospheric (gas-law expansion). Divide by inlet oil concentration S_o to get A/O. Target A/O ≥ 0.02 mg air / mg oil for reliable 90%+ capture.

6. Cell sizing

Two independent constraints:

• Surface area (hydraulic loading): A = Q / loading, 1–5 gpm/ft² typical. Lower loading = bubbles have more time to rise before water exits.
• Volume (retention time): V = Q · t, 20–40 min typical. Drives cell depth (= V/A).

Depth between 5 ft (too shallow — bubbles don't form fully) and 12 ft (too deep — float blanket collapses) is the sweet spot.

7. Polymer / coagulant

Adding 1–10 ppm of cationic polyacrylamide or polyaluminum chloride coagulates fine oil droplets into larger aggregates before the bubbles arrive. Capture jumps from ~80% to 95%+ in inlet streams with mostly sub-10 µm dispersed oil. The trade-off: polymer cost ($0.01–0.05/bbl) and float-handling load (more recovered solids).

8. References

• API Publication 421 — Design and Operation of Oil-Water Separators.
• Rykaart, E.M. & Haarhoff, J. (1995). "Behavior of air injection nozzles in dissolved air flotation." Water Sci. Tech. 31(3).
• Edzwald, J.K. (1995). "Principles and applications of dissolved air flotation." Water Sci. Tech. 31(3).
• WPCF Manual of Practice 8 — Wastewater Treatment Plant Design.
• NACE SP0192 — Monitoring Corrosion in Oil/Gas Production.
• Vendor lit: Siemens / Wabag, Veolia, Suez, Spec-Drainex.