Industries / Energy, water and waste / Oil and gas extraction / Oilfield services / Saltwater disposal
Treatment
Treatment is the work of taking oil, grit, gas and dissolved salts out of produced water so that it can be injected, released or used again.
6
Milestones achieved, from 1987 to 2026.
2
To watch for: announced, not yet confirmed outside.
2
Open deliverables. Nobody has shipped them.
The road
AchievedWatch forOpen
Achieved
1987
Hydrocyclones spin out oil
Offshore platforms cleaned produced water in large settling tanks and flotation cells, which are heavy and work badly when a floating platform moves. A deoiling hydrocyclone spins the water inside a narrow cone, so lighter oil droplets gather at the center and leave through their own outlet. In 1987 Noel Meldrum of Conoco reported what he called the first full-scale commercial use of the "4 in 1" hydrocyclone design, on Conoco's Murchison platform, along with field tests of larger 60 mm cyclones on the Hutton tension-leg platform, where platform motion had upset conventional equipment. A 2025 review names hydrocyclones and induced gas flotation (bubbling gas through the water to float oil off) as the two main offshore deoiling methods.
Credit
Conoco; Noel Meldrum
2002
Oil-sands water distilled for steam
In steam-assisted gravity drainage (SAGD), steam is pumped into oil-sands rock to melt the tar-like bitumen, and the hot water that comes back with it is cleaned and turned into steam again. An evaporator boils the water and collects the distilled vapor. A 2013 GE brochure lists one at Petro-Canada's MacKay River plant near Fort McMurray, Alberta, taking about 165 US gallons a minute of feed in 2002, the earlier of the two plants it names, and says GE had supplied six oil-sands evaporators that make distilled water for high-pressure drum boilers and four that feed once-through steam generators. The Alberta government says oil sands projects recycle 80% to 95% of the water they use.
Credit
Petro-Canada; GE Water & Process Technologies
2011
Sewage plants stop taking brine
Pennsylvania let shale gas wastewater go to municipal sewage plants, which could take out some metals but not dissolved salts such as bromide, so the salt passed into rivers. Bromide reacts with the chlorine used at drinking-water plants to form trihalomethanes, chemicals linked to cancer; 33 western Pennsylvania water systems exceeded the federal trihalomethane limit after 2008, by state records reported by WESA. In April 2011 the state Department of Environmental Protection (DEP), under Secretary Michael Krancer, asked drillers to stop sending Marcellus Shale wastewater by 19 May to the 15 plants still taking it, down from 27 a year earlier; most complied. In June 2016 the US EPA barred discharges from onshore unconventional (shale and similar) oil and gas wells to municipal sewage plants nationwide.
Lessona plant built to break down sewage dilutes salt; it does not remove it.
Credit
Pennsylvania DEP; Jeanne VanBriesen (Carnegie Mellon University) and Stanley States (Pittsburgh Water and Sewer Authority), who tracked bromide in the rivers
2011
Reed beds clean oilfield water
At the Nimr oil fields in southern Oman, Petroleum Development Oman (PDO) had pumped its oily, brackish produced water more than 1.5 km down into deep disposal wells. From the start of 2011 a passive oil-water separator, 234 hectares of reed wetlands and 300 hectares of evaporation ponds took more than 45,000 m³ a day, flowing by gravity. By the operators' figures it used 0.06 kWh per m³ in 2011, against 3.6 to 5.5 kWh per m³ for deep-well disposal. In 2017 a Shell engineering team (Shell is a PDO shareholder) wrote that the wetland covered more than 360 hectares, treated more than 95,000 m³ a day and had kept oil and grease in its outflow below detection, under 0.5 mg/L, since 2011.
Credit
Petroleum Development Oman; BAUER Nimr LLC
2013
Radium settles below a plant
Nathaniel Warner, Avner Vengosh and colleagues at Duke University sampled a western Pennsylvania plant that treated oil and gas wastewater and discharged it to a stream. The plant cut barium and radium by more than 90%, yet radium-226 in stream sediment at the outfall measured 544 to 8,759 becquerels per kilogram (a measure of radioactivity), about 200 times the 22 to 44 Bq/kg upstream and above thresholds for radioactive waste disposal.
Lessonremoving 90% of a contaminant is not enough when the rest collects in one place.
Credit
Nathaniel Warner, Avner Vengosh and colleagues (Duke University)
2026
Five pilots, one laboratory
The Texas Produced Water Consortium, based at Texas Tech University, had Eurofins, an outside certified laboratory, analyze 60 samples taken from August 2024 to June 2025 at five Permian Basin pilots that treated real produced water, at feed rates of 20 to 800 barrels a day. Raw and pretreated water averaged more than 120,000 mg/L of dissolved solids, and desalting brought it to an average of 317 mg/L; 582 of the 765 substances tested for were not detected. After a final polishing step, 75 of 76 comparable measures met Texas drinking-water limits and 85 of 95 toxicity endpoints (tests on algae, water fleas and fish) showed no toxicity in undiluted water, while desalted water that had not been polished passed only 3 of 16. The report grouped the five pilots together and did not compare their methods.
Lessontaking out the salt is not enough; trace chemicals left behind still made the water toxic until it was polished.
Credit
Texas Produced Water Consortium
Watch for
Desalting for the Pecos River
NGL Water Solutions Permian has accepted the terms of a draft permit from the Texas Commission on Environmental Quality (TCEQ) for a Reeves County plant that would desalinate produced water and release it to Salt Creek and the Upper Pecos River below Red Bluff Reservoir. Discharge would start at an average of 650,000 gallons a day and rise in phases to 16.9 million gallons, about 400,000 barrels, a day. The draft now goes to public notice and comment.
Achieved whenTCEQ issues the final permit and a year of the plant's public discharge monitoring reports show it meeting the limits.
Claimed by
NGL Energy Partners, November 2026
Freezing salt out of brine
Texas Pacific Water Resources says its 10,000-barrel-a-day pilot, which freezes produced water and then uses membranes and reverse osmosis, has three times passed a five-day test that organisms can reproduce in the cleaned water and a seven-day test that aquatic life can grow in it, and that it could receive a permit to release water to rivers as soon as early 2026. A paper by New Mexico State University (NMSU) researchers, with two Texas Pacific co-authors, reports about 99% removal of dissolved ions over a ten-month field run of freeze desalination followed by seawater reverse osmosis, from six sampling rounds covering more than 440 substances. As of April 2026 the state environmental agency (TCEQ) had released no draft permits for surface discharge or land application, by the Texas Produced Water Consortium's account.
Achieved whenTCEQ issues a discharge permit and a laboratory with no company co-authors repeats the toxicity results.
Claimed by
Texas Pacific Water Resources, December 2025
Open
01
One set of test brines
Open recipes for synthetic produced waters that match the median chemistry of five real sources (for example the Permian's Delaware Basin, the Marcellus, the Bakken, a Wyoming coalbed-methane water and a low-salt California heavy-oil water), a short bench protocol, and a round-robin in which several laboratories run thermal and membrane desalination on the same brines and publish water recovered, energy used and scale formed.
The briefCredit · unclaimed02
Every desalting pilot, one basis
A public table of every reported field pilot that desalinated oil and gas produced water, one row per pilot: salt in and out, share of water recovered, days run, electricity and heat per barrel of feed on one stated boundary, how much concentrated brine was left and where it went, and whether anyone outside the operator measured the results.
The briefCredit · unclaimed
Open 01 · the brief
One set of test brines
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
Open recipes for synthetic produced waters that match the median chemistry of five real sources (for example the Permian's Delaware Basin, the Marcellus, the Bakken, a Wyoming coalbed-methane water and a low-salt California heavy-oil water), a short bench protocol, and a round-robin in which several laboratories run thermal and membrane desalination on the same brines and publish water recovered, energy used and scale formed.
Why it is still open
Dardor and colleagues (Qatar University and ConocoPhillips, 2021) wrote that "there is no clear scientific guideline" for making such recipes and published two, but theirs are oily brines for testing oil-water membranes, not a set matched to basins by salt content. The New Mexico Produced Water Research Consortium's 2022 testing guide sets out how to run pilots on real produced water, not shared laboratory brines. A GitHub tool from July 2026 (kalx-cyber/Synthetic_Brine_Tool) calculates a recipe from any water analysis but publishes no reference set or results.
Built with what exists
The USGS National Produced Waters Geochemical Database; the Texas Produced Water Consortium's spreadsheet of 2024–2025 raw-water analyses from five Permian pilots; USGS PHREEQC, free software that checks whether a recipe stays dissolved; laboratory-grade salts; bench reverse osmosis, membrane distillation and small evaporator rigs that university water laboratories already run.
First step
Pull every Delaware Basin record from the USGS database, drop those whose positive and negative ions do not balance, and write the median as a first PHREEQC recipe.
Done looks like
Five recipes published under an open license with the analyses they came from; at least three laboratories reporting at least two methods on all five brines; results for the same method in different laboratories agreeing within a stated margin, such as 10% on water recovered.
Open 02 · the brief
Every desalting pilot, one basis
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A public table of every reported field pilot that desalinated oil and gas produced water, one row per pilot: salt in and out, share of water recovered, days run, electricity and heat per barrel of feed on one stated boundary, how much concentrated brine was left and where it went, and whether anyone outside the operator measured the results.
Why it is still open
Pilots report energy in ways that cannot be compared. A membrane distillation pilot on pretreated Permian water, run by University of Pittsburgh researchers at the University of North Dakota's Energy and Environmental Research Center and shown at a National Energy Technology Laboratory (NETL) review in April 2024, gave electricity per cubic meter of feed. Crystal Clearwater's 2022 New Mexico pilot gave heat, 0.18 to 0.20 MMBtu (million British thermal units) per barrel of distilled water. The Texas Produced Water Consortium's April 2026 report pooled water quality from five Permian pilots but grouped them on purpose, gave no energy figures and left comparing their methods to "future research". A search for this brief found no table that puts pilots on one basis. Open planning tools such as NETL's PARETO and the WaterTAP treatment-modeling library take such figures as inputs but do not collect pilot results.
Built with what exists
Pilot reports posted by the New Mexico consortium, NETL meeting slides and journal papers found through the Crossref and Europe PMC search services; an AI model that reads PDFs to draft each row, with every figure checked by a person against its page; a CSV file kept on GitHub.
First step
List every pilot report on the New Mexico consortium's research page and fill in the first ten rows.
Done looks like
At least 25 pilots, each figure linked to its page in the source, energy converted to kWh per barrel of feed with the conversion shown, and a column saying whether the result was measured outside the operator.