Cooling tower blowdown ZLD: what sets the size, the energy bill and the salt

Home Cooling tower blowdown ZLD: what sets the size, the energy bill and the salt

Zero liquid discharge (ZLD) for cooling tower blowdown means no blowdown leaves the site as liquid. The water is recovered for reuse and the dissolved salts leave as a solid, or go to an evaporation pond. It is technically proven: in 2016 there were 72 US power facilities using ZLD systems (Plata et al., 2022).

The hard part is not whether it works but how big and how costly it needs to be. Three things decide that, and only one of them is the ZLD plant itself:

  • How many cycles the tower runs. This sets the blowdown volume, and so the size of every step downstream.

  • What is in the make-up water. This sets how many tonnes of salt you will have to dispose of each year, whatever the cycles.

  • How much of the concentration is done by membranes rather than evaporation. This sets most of the energy bill.

This page works through each one with published data, so you can see where your site sits before talking to suppliers.

What is in blowdown, and why it is awkward to concentrate

Blowdown is circulating water bled off to stop dissolved solids building up as the tower evaporates water. The US Department of Energy gives the governing relationship: blowdown equals evaporation divided by (cycles of concentration minus 1) (DOE FEMP, 2011). A tower running at four cycles bleeds off a third of what it evaporates; at eight cycles, a seventh.

By the time it is bled off, the water is already a concentrate. It carries:

  • the make-up water’s hardness, alkalinity, sulfate, chloride and silica, multiplied by the cycles

  • the treatment programme: scale inhibitors, corrosion inhibitors, dispersants and biocides (how these are dosed and controlled is in our guide to the cooling tower water treatment system)

  • suspended solids, iron and biological material scrubbed from the air and the system

That makes it a difficult feed for a concentrator. The water has been held close to its scaling limits on purpose, so any further concentration quickly pushes calcium carbonate, calcium sulfate or silica past saturation. In a long-running forum thread on reverse osmosis (RO) for blowdown, engineers raise the points you would expect: biofouling, silica and magnesium silicate scale that is very hard to clean, and a reject stream that may still breach discharge limits (Eng-Tips discussion).

Start with cycles: a worked mass balance

Before sizing any ZLD equipment, run the tower’s own numbers. The example below is hypothetical, with the assumptions stated so you can swap in your own.

Assumptions: recirculating flow 10,000 m³/h with a 10 °F (5.6 °C) temperature range. Using the US EPA rule of thumb that evaporation is about 1% of recirculating flow for every 10 °F of range (EPA WaterSense at Work, 2012), the tower evaporates about 2,400 m³ a day. Make-up water has 500 mg/L total dissolved solids (TDS) and 25 mg/L silica. Drift and leaks are ignored; with modern drift eliminators they are small next to blowdown.

Cycles of concentration Make-up (m³/day) Blowdown (m³/day) Silica in the tower (mg/L) Dissolved salt in the make-up (tonnes/day)
3 3,600 1,200 75 1.8
4 3,200 800 100 1.6
6 2,880 480 150 1.44
8 2,743 343 200 1.37

Three points follow from the table.

Raising cycles shrinks the water, not the salt. Going from 4 to 8 cycles cuts blowdown by 57%, so every step of a ZLD train gets smaller. But the salt arriving with the make-up water falls by only 14%, because only the make-up volume falls. Whatever ends up in the crystallizer or the pond, about 585 tonnes a year at four cycles and 500 at eight in this example, plus any treatment chemicals, is set by the make-up chemistry. Switching to a saltier make-up source, such as treated municipal effluent, raises that tonnage even if the blowdown volume stays the same.

Silica often decides how far cycles can go. California’s Title 24 compliance guidance for cooling towers lists keeping silica at or below 150 ppm as good practice (2019 Title 24 compliance reference, section 4.6.1.1). With 25 mg/L in the make-up, that ceiling arrives at six cycles. With 50 mg/L it arrives at three. Running further needs silica control, such as side-stream softening, or a treatment programme designed for high silica.

The concentrator inherits the silica problem. If a membrane step then recovers 90% of the six-cycle blowdown, its reject carries ten times the tower’s silica. That stream is where scaling, cleaning frequency and recovery limits are decided. Our Industrial Wastewater Scaling tool runs your blowdown analysis through increasing water removal and shows the point at which gypsum, calcite, barite or amorphous silica crosses saturation. In our view, that crossing point is a better starting number for a recovery target than a supplier’s standard figure.

Cycles can go a long way when the chemistry is managed. A paper by Walters and Duke, based on 2007 data from an automotive headquarters’ central plant, reports towers on softened recycled water reaching tower silica of 350 mg/L and cutting discharge by 95% or more (paper). It was co-written by the company that developed the patented chemistry used, so read it as a demonstration of what is possible rather than a typical result. In an independent federal test, a blowdown recovery system cut blowdown by 53% and total water use by 16% (GSA and NREL evaluation, August 2023).

What does blowdown ZLD cost in energy and water? Modelled data from two US plants

In the most detailed public study, ZLD at one gas-fired plant raised the cost of its cooling water to 2.1 times the non-ZLD baseline with a membrane-led design and 6.3 times with a brine concentrator, and cut water withdrawal by only 18%. The details, and a second plant where the result was different, are below.

The numbers come from a 2022 peer-reviewed study by NREL and partners that modelled two natural gas combined cycle plants in detail (Plata et al., ACS ES&T Engineering, 2022).

Cherokee Generating Station, Denver (being retrofitted for ZLD) Gila River Power Station, Gila Bend, Arizona (built for ZLD, operating since 2003)
Membrane route ZLD with closed-circuit RO: 2.0 kWh/m³ for the whole ZLD system, water cost 0.97 $/m³ (2.1 times the 0.47 $/m³ non-ZLD baseline) RO scenario: 1.4 kWh/m³, water cost 3.5 $/m³
Thermal route ZLD with brine concentrator: 17 kWh/m³, water cost 2.9 $/m³ (6.3 times baseline) Two-brine-concentrator scenario: 23 kWh/m³; existing brine concentrator and RO system: 4.8 $/m³
Effect on water withdrawal 18% lower than baseline Shifting duty to RO cut cost, but RO recovered only about 71%, so withdrawals rose to 1.2 times baseline

What this means for a project:

  • Membranes do the cheap concentration. At Cherokee the brine concentrator scenario used more than eight times the electricity per cubic metre of the closed-circuit RO scenario. In our view, the economic design pushes as much recovery as the chemistry allows onto membranes and sends only the last, smallest stream to evaporation.

  • ZLD is mainly a discharge decision, not a water-saving one. The 18% cut in withdrawals at Cherokee is about the same as the roughly 20% the paper cites for raising cycles from three to six, and less than the 35% it estimates for dry cooling at Cherokee. The study found reuse did not produce large savings where source water was cheap; discharge rules were the main driver.

  • Ponds are not a free pass. Making evaporation ponds bigger lets the treatment plant recover less, but the study found this raised the cost of water, because more water is lost to evaporation instead of reused.

ZLD energy stayed below 0.8% of each plant’s annual generation in every scenario, so on a power plant the energy penalty is usually a cost question rather than a capacity one. On coal plants we would expect the scrubber purge, not the blowdown, to be the harder stream; see FGD wastewater ZLD.

A typical treatment train, and where each step fails

A blowdown ZLD train usually has five steps. Each one has a typical weak point, and those weak points are the questions to put to any supplier.

Step What it does Common failure point
Pretreatment: clarification, filtration, microfiltration or ultrafiltration Removes suspended solids, iron, metals and biological material Biofouling and iron carried over into membranes after shutdowns
Softening and silica removal (lime, soda ash, caustic or ion exchange) Takes out the ions that limit recovery Adds chemical cost and its own sludge
Membrane concentration (RO, closed-circuit RO, high-recovery RO, electrodialysis reversal) Recovers most of the water at low energy Scaling once the reject passes silica or sulfate saturation
Thermal brine concentrator Concentrates the membrane reject further High electricity use; corrosion if chloride and temperature are underestimated
Crystallizer, spray dryer or evaporation pond Turns the final brine into solids, or holds it Mixed, hygroscopic salts that are hard to dewater and costly to landfill

A membrane retrofit at a 360 MW coal-fired plant in Indiantown, Florida, shows what operating this train involves. It put microfiltration and two-stage RO on blowdown from towers running at 4 to 7 cycles, with blowdown designed at 650 gpm. Microfiltration kept RO feed silt density index well below 2.0. Biofouling on the RO was eliminated by dosing the biocide DBNPA, and aluminium fouling was solved by lowering feed pH from 6.5 to 5.5. Iron in the blowdown and fouling after shutdowns were the main concerns found in piloting. The article, co-written by staff of the microfiltration supplier, reports that these early problems were overcome and that the system met its performance goals for more than a year (Water Technology, December 2012).

For how the thermal end compares, see our guide to the ZLD evaporator and crystallizer and our page on multiple-effect evaporators. Where a full ZLD plant is more than the site needs, minimal liquid discharge, which stops before the thermal end and manages a small reject stream, may be enough.

Blowdown ZLD in Texas

Texas sites face the same engineering, but Texas offers outlets that can make full ZLD the last option rather than the first. Any blowdown that is discharged needs an industrial wastewater permit from the Texas Commission on Environmental Quality. Treating blowdown and returning it to the towers as make-up counts as use within the facility, which TCEQ says needs no reclaimed-water authorization under 30 TAC Chapter 210, Subchapter E (TCEQ presentation). On supply, the Texas Water Development Board’s 2027 State Water Plan, adopted in a first phase in July 2026, bases steam-electric demand on the highest recent historical use plus proposed plants, holds it flat to 2080, and relies on conservation and reuse together for 43% of the water its strategies would supply by 2080 (TWDB, 2027 State Water Plan). Our Texas page works through a blowdown example in that order: reuse first, then the outfall, then a Class I injection well, and ZLD only if those close.

Questions to settle before you ask for ZLD quotes

  1. What cycles can the tower actually sustain with its current make-up, and what limits them: silica, hardness, sulfate or the treatment programme?

  2. Is the make-up source going to change? A switch to reclaimed water or groundwater changes the salt load, even if the blowdown volume stays the same.

  3. What does the permit or the receiving sewer allow? If a lower-salinity discharge is acceptable, minimal liquid discharge may beat ZLD on cost.

  4. Where does the recovered water go? Back to the tower, to boiler make-up, or to process use. Each sets a different quality target.

  5. What will the solids be, and who takes them? Mixed salts from blowdown can be hygroscopic. Landfill acceptance and transport cost belong in the comparison from the start.

  6. Is the analysis good enough to design on? Check it with our Water Analysis Validator, then estimate the flows with the Brine Concentration Factor Calculator and the MLD vs ZLD Screener.

Other questions about blowdown ZLD

Can blowdown go straight to an evaporator without membranes? Yes. Gila River’s original ZLD system used brine concentrators, with an ultrafiltration and RO system added in 2021. The study’s figures show why membranes now tend to go first: per cubic metre, the thermal scenarios used roughly eight to sixteen times the electricity of the membrane scenarios.

Does ZLD make sense for a data centre’s cooling towers? In our view, only where discharge is restricted or water is expensive enough. For most data centres we would look first at recovering blowdown for reuse at high recovery and managing a small reject stream. Our article on water reuse for data centres covers those options.

Is ZLD the same as zero blowdown? No. Zero blowdown means running the tower at very high cycles so that drift and other small losses carry the salts out. ZLD treats the blowdown so that no liquid leaves the site. Zero blowdown needs make-up chemistry and a treatment programme that tolerate very high concentration.

How do I compare ZLD offers that use different trains? Put them on one basis: the same design analysis, flow range, recovery and solids specification, then compare cost over the plant’s life. Our ZLD vendor proposal checklist lists the checks.

Talk to us about your blowdown

We are Brine Consulting, an independent industrial water advisory firm in Amsterdam. We sell no treatment equipment, so we have no stake in which train you end up buying. On blowdown, the decisions we help with come before any quote: how many cycles the tower can hold, how much recovery the reject chemistry allows, and where the salt will go. We work remotely from your make-up and blowdown analyses, tower data and permit conditions, or through a defined scoping engagement, and then help you compare supplier offers on one basis. That work sits under our water reuse and ZLD systems service.

Start by running your blowdown analysis through the Industrial Wastewater Scaling tool. Then book a scoping call, send us your project details or email christos@brineconsulting.com with the analysis attached.

Written by the Brine Consulting advisory team. The Brine Consulting advisory practice includes Edwin Muller, a senior industrial water expert with more than 25 years of experience across refineries, petrochemicals, power generation, hydrogen, ammonia and critical industrial infrastructure, and Dr. Christopher Fellows, a senior water technology and physical chemistry expert with more than 30 years of combined academic and industrial research experience. Meet our experts.

Questions about your site's water or brine?

Tell us briefly what you are working on and one of our experts will get back to you.