SAGD produced water treatment: silica, steam generators and Alberta’s disposal limit

Home SAGD produced water treatment: silica, steam generators and Alberta’s disposal limit

SAGD produced water treatment follows one of two established routes: warm lime softening with weak acid cation exchange feeding once-through steam generators, or mechanical vapour compression evaporators feeding drum boilers. Neither route removes silica; each moves it into a residue that has to go somewhere, and in Alberta the volume a steam-assisted gravity drainage (SAGD) scheme may send to disposal is capped. The cap comes from the Alberta Energy Regulator’s (AER) Directive 081, whose November 2019 edition the AER still lists as current (checked 4 November 2026). It sets each thermal scheme’s disposal limit every calendar year according to the types of water that come in.

So the choice of water plant also fixes which steam generator you can run, how much make-up water you draw and how close you sit to the disposal limit. Brine Consulting, an independent industrial water advisory firm in Amsterdam, reviews those choices for operators of Alberta thermal in situ schemes, remotely or through a defined engagement. We sell no treatment equipment, so we have no stake in whether you pick lime softening, evaporators or a membrane polish.

How the SAGD water loop works

Steam goes down the injector, heats the bitumen and comes back up the producer as an oil and water emulsion. After separation, the hot produced water is de-oiled and treated until it is fit to make steam again. Make-up water replaces what stays in the reservoir, and a small stream leaves the loop as disposal.

The loop is already tight. In 2024, 90% of the water used by Alberta’s in situ oil sands operations was recycled water and 10% was make-up, up from 82% recycled in 2013 (AER, in situ oil sands water use, October 2025). New projects need far more: the AER reports an average of 2.47 barrels of nonsaline make-up water per barrel of oil equivalent in a project’s first year, falling to 0.11 by year fifteen.

What makes the water hard to recycle is not salinity alone. It is the combination of heat, dissolved organics, silica, hardness and oil carry-over:

  • Oil. A report written by an instrument supplier for an Alberta Innovates project recorded 150 to 3,000 ppm of oil at the skim tank and 4 to 30 ppm after the oil removal filter, from samples at Devon’s Jackfish facility (Agar Canada final report, February 2020).

  • Organics and silica. University of Alberta researchers measured SAGD blowdown water at pH 10.3 to 11.2, with 12,000 to 17,000 mg/L dissolved solids, 500 to 2,500 mg/L total organic carbon and 94 to 256 mg/L silica as Si (Guha Thakurta, Maiti and Bhattacharjee, WATERtech 2012). About 40% of the dissolved organic matter in that study was smaller than 500 Da, which means a 500 Da nanofiltration membrane could hold back only about 60% of it.

The two established treatment routes, and a newer one

Route Steam generator Silica in boiler feed Main residues Strengths Watch-outs
Warm (or hot) lime softening, filtration, weak acid cation ion exchange Once-through steam generator (OTSG) Typically 30 to 50 mg/L after lime softening Lime sludge, ion exchange regeneration waste, OTSG blowdown No vapour compressor, so lower power demand than evaporation OTSG steam quality of 80 to 85% without rifled tubes; large footprint and sludge volumes
Mechanical vapour compression evaporator Drum boiler 0.1 to 3.0 mg/L reactive silica in samples from one pilot centre’s distillate and blowdown supply header Concentrated alkaline evaporator blowdown, high in silica and organics Much cleaner feed; fewer unit operations Compressor power and operating cost; blowdown handling decides the disposal outcome
High-temperature reverse osmosis polishing (piloted) Higher-quality OTSG, hybrid or drum boiler Project target 10 to 20 ppm, against a commercial target below 10 ppm; the report says target removals were achieved but gives no measured value Membrane concentrate Modelled energy and water savings Field pilot only; problems above 105 °C

Sources for the table: the lime softening silica range, the 80 to 85% OTSG steam quality, the pilot centre’s header samples and the high-temperature reverse osmosis results all come from Suncor’s final report to Emissions Reduction Alberta on its pilot at the Firebag Water Technology Development Centre, operated by Suncor and partner CNRL (ERA final outcomes summary, May 2025). The same report states that water from the lime softening and ion exchange train cannot meet the feedwater specifications of higher-efficiency drum or hybrid boilers. The footprint, sludge and compressor-power trade-offs follow the comparison published by engineering firm Ausenco (May 2020).

For the evaporator route, supplier literature from the evaporator maker that became part of GE states that 95% to 99% of the feed is typically recovered as distillate with less than 10 mg/L of non-volatile dissolved solids (Heins and Schooley, supplier-written technical paper, 2011). Treat those as supplier figures to confirm on your own water.

In our view the high-temperature membrane pilot is worth watching but not yet worth designing around. Suncor’s report models a 5.3% cut in total greenhouse gas per tonne of oil against a warm lime softening baseline, and up to 40% less make-up water and disposal under select conditions. Those are modelled results. In the field the skid ran at 85 to 105 °C, and the report notes that issues were encountered above 105 °C.

We found no published, verifiable cost figures for SAGD water treatment when we checked in November 2026. Compare routes on what drives their cost instead: compressor power for evaporators; lime, other chemicals and sludge handling for softening; steam generator cleaning; and the volume that ends up in a disposal well.

Neither route removes silica: it moves it

Boiler feed quality is only half of a route comparison; the residue is the other half. Lime softening takes part of the silica out as sludge and sends the rest, about 30 to 50 mg/L, to the steam generator, where it concentrates in the OTSG blowdown. An evaporator keeps silica in solution at high pH and concentrates nearly all of it in a small, very alkaline blowdown stream. Either way, the silica ends up in a residue that you must dispose of, recycle or crystallize.

With an evaporator, the risk shifts to the disposal well. A 2024 University of Calgary doctoral thesis on SAGD evaporator blowdown found that silica tends to scale equipment and to form gels when the blowdown is neutralised, which complicates filtration and deep-well disposal (Rao, University of Calgary, 2024). In her tests, a sodium chloride to silica ratio above 4.5 prevented gel or sol formation during pH reduction, and silica removal was highest at pH 10 and 8. Her geochemical modelling suggests that neutral to slightly acidic blowdown is less likely to clog injection wells than the untreated alkaline stream.

The practical consequences:

  • Specify blowdown conditioning together with the evaporator, not after injectivity starts to fall.

  • Test acidification on your own blowdown before choosing a target pH. Results depend on the salt to silica ratio and on the organics.

  • Check the well class. AER Directive 051, in its July 30, 2026 edition, lists boiler blowdown water among the fluids for class Ib disposal wells and places produced water and brine under class II.

Our free Industrial Wastewater Scaling and Concentration Tool gives a quick first look at how scaling ions build up as a stream is concentrated, before a vendor model is run.

How much can a SAGD scheme dispose of under Directive 081?

Directive 081 sets a disposal limit for each thermal in situ scheme, calculated per calendar year on a scheme basis rather than per facility. Disposal means injection into designated disposal wells plus any water dispositioned out of the scheme.

The limit is a weighted average of four inlet water types. Because the limit is expressed as a percentage of the same total inlet volume, it reduces to a simple rule: each cubic metre that comes in earns a fixed disposal allowance.

Inlet water type (Directive 081, Table 2) What it includes Disposal allowance per m³ in
High-quality nonsaline Surface water, and nonsaline groundwater up to 4,000 mg/L TDS that is not an alternative type 0.03 m³
Alternative type 1 Deep nonsaline groundwater, treated wastewater, industrial runoff, and recyclable produced water (up to the volume of steam injected) 0.10 m³
Alternative type 2 Saline groundwater, oil sands process-affected water, nonsaline groundwater in contact with bitumen 0.55 m³
Alternative type 3 Wastewater otherwise sent to disposal, excess produced water (above steam injected), saline water balance credit 1.0 m³

New or reactivated schemes are exempt for the first twelve months after steam injection starts. Schemes that do not recycle produced water are exempt if they use less than 500,000 m³ a year of high-quality nonsaline make-up and less than 2,000,000 m³ a year of total make-up. Expansions and new phases do not get the start-up exemption.

Worked example (hypothetical)

Assumptions: a scheme injects 3,650,000 m³ of steam a year as cold water equivalent (10,000 m³ a day). It produces back 3,285,000 m³ of water, all of it recyclable because it is less than the steam injected. It brings in 500,000 m³ of make-up water. The two cases differ only in the make-up source, and no saline water balance credit is assumed.

Case A: fresh make-up Case B: saline make-up
Allowance from recyclable produced water (3,285,000 × 0.10) 328,500 m³ 328,500 m³
Allowance from make-up (500,000 × 0.03 or × 0.55) 15,000 m³ 275,000 m³
Disposal limit, volume 343,500 m³ a year (about 941 m³ a day) 603,500 m³ a year (about 1,653 m³ a day)
Disposal limit, as Directive 081 states it (÷ 3,785,000 m³ total inlet) 9.1% 15.9%

Three things follow:

  1. Most of a mature scheme’s allowance comes from its own recycled produced water, not from make-up. Losing produced water returns shrinks the allowance.

  2. Switching make-up from fresh to saline groundwater raises the allowance by 260,000 m³ a year in this example, but it also raises the dissolved solids the treatment train has to handle. Supplier literature puts OTSG feedwater at typically below 8,000 mg/L TDS (Heins and Schooley, above), so a lime softening plant may not absorb saline make-up without changes.

  3. Every stream that leaves the scheme counts. Water sent to a facility outside the scheme is a disposition out, and it uses allowance just as a disposal well does.

To run your own numbers on recovery, blowdown and concentration, try our brine mass balance calculator.

Other Alberta rules that touch the water plant

  • Water licences and source choice. Alberta’s Water Conservation Policy for Upstream Oil and Gas Operations (December 2020) makes water conservation a requirement for major upstream operations. The AER’s Bulletin 2025-17 (May 14, 2025) reminded licensees to prefer alternative nonsaline water over high-quality nonsaline water and to check their licence conditions and water shortage advisories before diverting.

  • Disposal and injection wells. Directive 051 (July 30, 2026 edition, which replaced the April 2023 edition) sets well classes, completion, logging and testing requirements.

  • In situ performance reporting. Directive 054, Performance Reporting and Surveillance of In Situ Oil Sands Schemes (April 3, 2020 edition), covers scheme performance reporting.

Rules change. Check each directive’s current edition on the AER site before you rely on a requirement.

When an independent review pays off

The treatment route is usually fixed at sanction, and changing it later is expensive. Our part is the water balance, the route comparison and the Directive 081 position, worked from your own operating data; detailed design stays with your engineering contractor. That review earns its fee at the points where the choice is still open or the old basis no longer holds:

  • Expansion or debottlenecking. Adding a phase is the moment to decide whether to copy the existing train or change route, for example evaporators on a new phase beside lime softening on the old one.

  • A change of make-up source. Moving to saline or other alternative water changes both the Directive 081 allowance and the feed chemistry.

  • Falling disposal well injectivity. Before drilling another well, check whether blowdown chemistry, silica gels or organics are the cause.

  • New technology offers. Membrane polishing, new evaporator designs and chemical programmes arrive with vendor data. Ask what water they were proven on, at what temperature, and over how many months.

Bring these data to the first conversation: monthly steam injected and water produced, make-up by source category, produced water and blowdown analyses over time (silica, hardness, TOC, oil, TDS, pH), steam generator cleaning records, and disposal well injection pressures. Before sharing analyses with anyone, our free Water Analysis Validator checks charge balance and ion sum against TDS.

For the wider produced water picture, including de-oiling and desalination options outside thermal recovery, see our produced water treatment technologies guide. For other Canadian work, see our Canada page.

Questions operators ask

Why don’t SAGD plants use ordinary reverse osmosis? The water is hot, rich in dissolved organics and high in silica. Cooling it costs energy that the steam generators then have to put back, organics foul membranes, and silica limits recovery. The Suncor high-temperature pilot shows progress, but it reported problems above 105 °C.

Does an evaporator make the plant zero liquid discharge? Not on its own. A blowdown stream remains, and it goes to a disposal well unless a further concentrator and crystallizer or dryer turns it into solids. Our ZLD evaporator and crystallizer guide covers that last step, and our deep well injection post covers the disposal alternative.

Can a scheme use excess produced water to raise its allowance? Excess produced water, meaning water produced above the volume of steam injected, is alternative type 3 and earns a full 1.0 m³ of allowance per m³. It only arises where the reservoir returns more water than was injected, so it depends on the reservoir rather than on plant design.

Talk to us before the next train is sanctioned

Once a new phase is sanctioned, its water train is hard to change. If an expansion, a switch to saline make-up or falling injectivity is on the table, bring a year of steam, production, make-up and disposal volumes to a scoping call: we will tell you where your Directive 081 headroom appears to stand and whether the route question is still worth reopening. You can also send us your project details or email christos@brineconsulting.com. Our water treatment and ZLD consulting service sets out how a review is scoped.

Written by the Brine Consulting advisory team. The Brine Consulting advisory practice includes Steve Coffee, a senior produced water expert and global oil and gas advisor with more than 30 years of experience in produced water treatment, hydrocyclones, sand management, water reuse, and upstream system optimization, and 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. 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.