Cooling tower blowdown recovery: when silica becomes limiting

Tower blowdown and RO concentrate are different streams

A cooling tower evaporates water while much of the dissolved material remains in circulation. Blowdown controls the resulting accumulation; it cannot simply be removed from the water balance without replacing that function. DOE guidance links flow, conductivity and chemistry when establishing concentration cycles. Silica is one of the species worth tracking separately.

When blowdown feeds reverse osmosis, the system produces reusable permeate and RO concentrate. Membrane recovery is permeate flow divided by feed flow, whereas tower cycles compare circulating-water and makeup-water concentrations. They describe different processes. Returning all RO concentrate to the tower may recycle the very load the blowdown was intended to remove.

What to measure before choosing a recovery

Request silica reported as SiO₂ and record the analytical method: dissolved or reactive silica and total silica may describe different fractions. Include pH, alkalinity, hardness, conductivity, iron, aluminium, suspended solids, organic matter and temperature during relevant changes in loading and chemical dosing. One makeup-water sample does not necessarily represent the actual blowdown.

The FilmTec manual evaluates silica risk under concentrate conditions and considers interactions with metals. A filter can remove particles; general removal of dissolved silica must not be assumed for UF. Conventional sodium-cycle softening that exchanges calcium and magnesium is also insufficient evidence. Each pretreatment step should demonstrate which fraction it changes and what residuals it produces.

A simple balance reveals the concentration effect

Consider an original screening example with no precipitation and negligible silica passage into permeate: 20 m³/h of blowdown contains 80 mg/L as SiO₂. Silica enters at 20 × 80 / 1000 = 1.6 kg/h. At 60 % recovery, permeate flow is 12 m³/h and concentrate flow is 8 m³/h; the latter contains approximately 200 mg/L silica.

Increasing recovery to 75 % raises permeate flow to 15 m³/h and reduces concentrate flow to 5 m³/h, giving approximately 320 mg/L silica. Under those assumptions the general expression is Cf / (1 − Y). These concentrations are neither approved limits nor evidence of operability: they must be assessed against the specific chemistry and membrane design.

If silica passes into permeate, use Qf × Cf = Qp × Cp + Qc × Cc. Add deposited solids or other losses when present. A reduction in dissolved silica may indicate deposition rather than improved treatment.

Select treatment around the required outlet

Compare several recoveries before specifying equipment. Recovering less water, using a demonstrated silica-separation process, or treating a less concentrated stream may be preferable. Antiscalant dosing needs an assessment relevant to both feed chemistry and membrane requirements; it does not make every concentration acceptable.

Check tower oxidants and biocides against RO operating limits as well. Before reusing the permeate, assess every species affecting corrosion, scaling and biological control. A favourable silica result alone does not establish that the water is suitable for the cooling circuit.

Turn the calculation into a project decision

In a pilot, record all three stream flows, paired analyses, differential pressure, normalised production and cleaning events. Test the relevant feed variability and define stop criteria before increasing recovery. The mass balance should explain where the silica ends up.

Calculate freshwater avoided across the combined tower and treatment system, allowing for flushing, cleaning and other losses. Include electricity, chemicals and the management of concentrate or separated solids. The useful comparison is cost per unit of water meeting the intended specification, with a workable outlet for the retained substances.