From Bitumen to Fuel: Upgrading and Refining in Alberta

Why bitumen needs work before it becomes fuel

Refineries are configured to handle specific blends of crude oil ^[11]. Alberta's bitumen is a difficult feedstock for most of them. As the oil sands guide explains, bitumen is the extra-heavy oil recovered from the Athabasca, Cold Lake and Peace River deposits. Before anyone can make gasoline or diesel from it, it has to be made lighter, thinner and cleaner, or at least thin enough to pump.

Four properties cause the trouble:

• Density. Athabasca bitumen has a density of about 1,015 kg/m³, or roughly 7.9° on the American Petroleum Institute (API) gravity scale. That is heavier than water ^[12].
• Viscosity. At 25°C its viscosity (resistance to flow) ranges from about 5,000 to 300,000 centistokes. Pipelines leaving Alberta accept a maximum of 350 centistokes and a maximum density of 940 kg/m³, which is about 19° API ^[12].
• Sulphur. A typical bitumen contains around 5 per cent sulphur by weight ^[12]. For comparison, the benchmark West Texas Intermediate (WTI) crude carries less than 0.5 per cent sulphur, and the Western Canadian Select (WCS) heavy blend carries 3.0 to 3.5 per cent ^[2].
• Too much carbon, too little hydrogen. Heavy, sour crudes are made of large, complex molecules with more sulphur and a higher proportion of carbon to hydrogen. Light, sweet crudes have smaller molecules and a higher hydrogen-to-carbon ratio ^[13]. Upgrading therefore aims to raise the hydrogen-to-carbon ratio, by removing carbon or adding hydrogen ^[13].

Producers have two basic choices. The first is dilution: blending bitumen with a light liquid called diluent so it meets pipeline limits. Condensate diluent typically makes up about 30 per cent or more of a diluted bitumen (dilbit) blend. When synthetic crude is used as the diluent the product is called synbit, and it can be as much as 50 per cent synthetic crude ^[12]. The second choice is upgrading: chemically converting bitumen into a lighter oil in a dedicated plant called an upgrader.

Upgrading began in the late 1960s because there was no other way to move bitumen to market ^[14]; mined bitumen also held too much sediment and water to ship by pipeline ^[12]. Today dilution dominates growth. In 2024 Alberta shipped out 2.03 million barrels per day of non-upgraded bitumen and 828,600 bbl/d of upgraded bitumen. To move the diluted barrels, 751,100 bbl/d of pentanes plus left the province as diluent ^[15].

Two ways to lighten bitumen: remove carbon or add hydrogen

The AER describes upgraders as plants that chemically alter bitumen "either by adding hydrogen or by removing carbon," and sometimes both, making a lighter, sweeter oil that can fetch a premium price ^[2]. Every upgrader follows roughly the same five steps: distillation, residue conversion, hydrotreating, sulphur recovery, and utilities such as steam, power and hydrogen ^[13].

It starts with distillation: the diluent is boiled off and returned to the producer, and a vacuum unit separates gas oils from the heaviest material, called vacuum residue, which goes on to conversion ^[13].

Carbon rejection: coking

Coking is the most common carbon-rejection technology in the oil sands. Cokers heat the residue to about 450 to 500°C at relatively low pressure, about 350 kPa. The heat cracks large molecules into smaller ones and leaves behind solid petroleum coke, a coal-like, carbon-rich by-product ^[13].

• Delayed coking is a batch process. Residue is fed into a pair of tall coke drums. While one drum fills with coke over a cycle of about 10 to 16 hours, the other is cooled with water and cleaned out with high-pressure water jets ^[13]. Suncor's base plant and Canadian Natural's Horizon upgrader use delayed coking ^[16]. In 2025 Suncor finished replacing eight coke drums on its Upgrader 1, which it expects to extend that unit's life by about 30 years ^[17].
• Fluid coking is continuous. Hot residue is sprayed onto a bed of hot, fluidized coke particles, and the coke is reheated in a burner and recirculated. Fluid cokers make less coke and slightly higher liquid yields, but are more complex to operate ^[13]. Syncrude uses fluid coking ^[17].

The trade-off is yield. Up to about 20 per cent of the bitumen ends up as coke, so coking recovers about 80 to 85 per cent of the feed as liquids ^[13]. The Canada Energy Regulator (CER) puts coking yields at 80 to 90 per cent, with a lower capital cost than the hydrogen route ^[14].

A related method, solvent deasphalting, uses a light solvent to precipitate the heaviest molecules (asphaltenes), usually paired with a gasifier that turns them into gas for making hydrogen and power ^[13]. The idled Long Lake upgrader used this approach ^[16].

Hydrogen addition: hydrocracking and the LC-Finer

Hydrocracking breaks large molecules in the presence of hydrogen and a catalyst (a substance that speeds a reaction without being used up). The most common hydroconversion unit in the oil sands is the LC-Finer, an ebullated-bed residue hydrocracker ^[13].

Residue and hydrogen flow up through a reactor at more than 425°C and 18,000 kPa, and recirculation keeps the catalyst suspended in a churning bed. Because metals in the residue foul the catalyst, it is continuously removed and replaced ^[13].

Because no carbon is rejected, hydroconversion loses almost no yield and avoids coke handling, but it costs considerably more to build and run ^[13]. The CER notes that hydroconversion can yield 100 per cent or more of the bitumen volume ^[14]. The Scotford Upgrader uses LC-Fining technology ^[18]. Syncrude runs an LC-Finer alongside its fluid cokers, and the Sturgeon Refinery pairs LC-Fining with a gasifier ^[16].

Hydrotreating

Hydrotreating is the clean-up step, often called secondary upgrading. The gas oil and distillate streams from distillation and conversion are mixed with hydrogen and passed over a catalyst at high temperature. Sulphur and nitrogen react to form hydrogen sulphide (H₂S) and ammonia, and metals and olefins are removed. The cleaned streams are then blended into synthetic crude ^[13].

Sulphur recovery and hydrogen supply

Bitumen's sulphur ends up as hydrogen sulphide in the plant's gas streams. Amine units absorb the H₂S and send it to a sulphur recovery unit, usually the modified Claus process, which combusts the gas and passes it over a catalyst to produce elemental sulphur and water. With a tail gas treatment unit added, the circuit typically captures at least 99.8 per cent of the sulphur in the feed ^[13].

All this hydrotreating needs hydrogen, which upgraders mostly make from natural gas by steam methane reforming. Gasifiers can also make hydrogen from residue or asphaltenes ^[13]. Making hydrogen releases carbon dioxide, and Alberta's two large carbon capture projects capture CO₂ from hydrogen production at the Scotford Upgrader and the Sturgeon Refinery ^[2]. Upgraders also make an off-gas rich in natural gas liquids and olefins, which companies increasingly separate for use as petrochemical feedstock ^[2].

Synthetic crude oil: what upgraders make

Synthetic crude oil (SCO) is the blended product of an upgrader. The classic grade, Syncrude Sweet Premium, is a light, sweet "bottomless" blend of hydrotreated naphtha, distillate and gas oil fractions made in a coker and hydrocracker based upgrader ^[19]. "Bottomless" means it contains no heavy residue. Over five years to August 2026, Syncrude Sweet Premium averaged a density of 859.5 kg/m³, a gravity of 33.0° API and 0.19 per cent sulphur by weight ^[19]. That makes it lighter and far sweeter than WCS.

SCO is not a single product. Other grades and streams include:

• Synthetic Sweet Blend, a pipeline blend of Suncor Synthetic A, Syncrude Sweet Premium and CNRL Light Sweet Synthetic ^[20].
• Premium Albian Synthetic and Albian Heavy Synthetic from the Scotford Upgrader, which also makes vacuum gas oil sold to Shell under a long-term agreement ^[18].
• Suncor sweet and sour SCO, plus ultra-low sulphur diesel made directly at the base plant upgraders ^[17].

The AER's 2024 figures by upgrader were ^[4]:

Output rose 4.5 per cent in 2024 as all four plants ran at high utilization. Suncor's upgraders set a production record, Horizon reached its highest output after a maintenance optimization project, and Scotford grew after a debottlenecking project ^[1]. For 2025, Statistics Canada reported synthetic crude production of 76.0 million m³, up 2.7 per cent ^[3].

The AER's base-case forecast has upgraded output reaching 204.5 thousand m³/d (1.29 million bbl/d) by 2034. Because raw bitumen output is expected to grow faster, the upgraded share of Alberta's bitumen supply falls from 41 per cent in 2024 to 38 per cent in 2034 on that measure. The forecast assumes no new upgrading facilities ^[1]. These are projections, not commitments.

Where does SCO go? In 2024 Alberta's own refineries ran 357,900 bbl/d of it ^[4], and 828,600 bbl/d of upgraded bitumen left the province ^[15]. Horizon's SCO travels by pipeline to the Edmonton area for distribution ^[18]. Suncor also sends SCO to its refineries in Sarnia and Montreal ^[17]. Some SCO is also blended into raw bitumen as diluent ^[21].

Alberta's upgraders

The AER counted four active bitumen upgraders in 2024 ^[4]. The CER says they can process about 1.4 million barrels per day of bitumen ^[21]. Three sit beside mines north of Fort McMurray. The fourth, Scotford, sits near Edmonton and receives diluted bitumen by pipeline.

Sources for the table: ^{[17][16][18][22][23][14][24]}. The Syncrude gross figure is calculated from Suncor's net share.

Suncor base plant. The original Great Canadian Oil Sands plant started in 1967, and a second upgrader followed in 2001 ^[16]. It is the hub of Suncor's regional network, linked by pipeline to Fort Hills, Firebag and Syncrude ^[17].

Syncrude. Syncrude began producing in 1978 and has been operated by Suncor since September 30, 2021. Its upgrader is rated at 206,000 bbl/d of SCO net to Suncor's 58.74 per cent share, and it mainly makes a sweet SCO that each owner markets separately ^[17].

Horizon. Canadian Natural made its first SCO in 2009 from a 110,000 bbl/d first phase. Expansions through 2017 took capacity to about 250,000 bbl/d, and a 2024 reliability project lifted average capacity to about 264,000 bbl/d ^[18].

Scotford. Diluted bitumen from the Muskeg River and Jackpine mines travels south on the Corridor Pipeline, and a 2024 debottleneck raised gross capacity to about 328,000 bbl/d ^[18]. In a 2025 asset swap, CNRL took full ownership of the mines and Shell raised its share of the upgrader and the Quest carbon capture facility to 20 per cent ^[18][23].

Long Lake. Nexen's Long Lake project combined steam-assisted gravity drainage with an on-site upgrader. After a July 2015 pipeline rupture and a January 2016 explosion in the hydrocracker unit that killed two workers, Nexen put the upgrader into "cold-stack" mode in July 2016 and cut about 350 jobs ^[24]. The CER no longer counts it in active capacity ^[14].

Two more upgraders, in Regina and Lloydminster, operate in Saskatchewan and are not counted in Alberta's figures ^[14].

Alberta's refineries

Alberta has five refineries, the largest refining capacity of any province. The CER notes that they process only western Canadian crude, including a large proportion of blended bitumen and SCO ^[21]. Nationally, Canada has 17 operating refineries with roughly 2 million barrels per day of capacity, not counting upgraders ^[11].

Capacities are from the Canadian Association of Petroleum Producers (CAPP); throughput is from the AER ^[11][4]. Capacity totals differ slightly between sources: the CER put combined capacity at 569,000 bbl/d in 2023 ^[21].

Imperial Strathcona. Imperial calls it "the world's most travelled refinery." After the 1947 Leduc discovery, the company moved a refinery from Whitehorse to the outskirts of Edmonton. In 2025 it processed about 186,000 barrels of crude a day. It makes gasoline, diesel, aviation fuel, lubricating oils and asphalt for western Canada and the northern United States, and employs about 400 people ^[25]. In July 2025 it began producing renewable diesel from hydrogen and Canadian bio-feedstocks such as canola oil, with capacity of up to 20,000 barrels a day ^[25].

Suncor Edmonton. This refinery processes a wide range of feedstocks from Suncor's oil sands operations and other producers in the Wood Buffalo and Cold Lake regions ^[17]. In 2025 it processed 142,700 bbl/d: 55,200 of sweet SCO, 43,600 of sour SCO, 37,800 of diluted bitumen and 6,100 of conventional crude. Utilization was 98 per cent (105 per cent in 2024). Its output was 43 per cent gasoline, 52 per cent distillates such as diesel and jet fuel, and 5 per cent other products ^[17].

Shell Scotford. Operating since 1984, the refinery can process up to 100,000 bbl/d of SCO from the neighbouring upgrader. It makes gasoline, diesel, jet fuel, benzene, propane and butane. An adjoining Shell chemicals plant can make about 475,000 tonnes of styrene monomer and 548,000 tonnes of ethylene glycol a year ^[22]. Shell owns the refinery and chemicals plant outright ^[23].

Cenovus Lloydminster. A small asphalt refinery in the border city of Lloydminster, rated at about 29,000 bbl/d and counted by the AER among Alberta's five refineries ^[11][4].

The Sturgeon Refinery, the first built in Canada in more than 30 years, more than offset the closure of the small Bowden refinery (6,700 bbl/d) ^[11].

What a refinery makes and where Alberta's fuels go

A refinery works in three broad stages. Distillation heats crude and separates it into fractions by boiling point. Conversion units such as catalytic crackers and hydrocrackers break heavier fractions into more valuable lighter products. Hydrotreating then cleans the products to meet fuel standards ^[11]. Light crude yields more gasoline and jet fuel with less work. Heavy crude needs more processing and yields more asphalt and heavy fuel oil ^[11].

Across Canada in 2025, refineries turned their crude into 38 per cent motor gasoline, 36 per cent distillate fuel oil (mostly diesel), 6 per cent jet fuel and 5 per cent asphalt. Smaller shares went to still gas used as refinery fuel, petroleum coke, heavy fuel oil and petrochemical feedstocks ^[11]. Individual plants differ. Suncor's Edmonton refinery tilts toward distillates ^[17], and more than 80 per cent of the Sturgeon Refinery's production is ultra-low sulphur diesel and diluent ^[6].

Utilization. Alberta's refineries ran at 103 per cent of rated capacity in 2024 ^[4] and nearly 99 per cent in 2023 ^[21]. A rate above 100 per cent means the plants processed more crude than their rated capacity.

Where the fuel goes. Alberta makes more refined products than it uses, and nearly all the gasoline consumed in the province is made there ^[21]. Demand is high: in 2022 Alberta used 1,219 litres of gasoline and 1,393 litres of diesel per person, 18 and 80 per cent above the national averages ^[21]. Products move:

• within Alberta by truck, rail and the Alberta Products Pipeline, which carries an average of 48,400 bbl/d from Edmonton refineries to southern Alberta ^[21];
• to British Columbia largely through the Trans Mountain pipeline ^[21];
• to Saskatchewan and Manitoba mainly through the Enbridge Mainline ^[21];
• to the northern United States, which Strathcona also serves ^[25].

Canada as a whole exported about $19 billion of refined petroleum products in 2025 ^[11].

The Sturgeon Refinery: Alberta's royalty-in-kind experiment

The Sturgeon Refinery, about 45 km northeast of Edmonton ^[7], began as a policy to add value to Alberta's own royalty barrels and became a multi-billion-dollar public commitment.

The idea: royalties taken as barrels

Alberta can collect some oil royalties "in kind," as barrels rather than cash. Under its Bitumen Royalty-in-Kind (BRIK) policy, the Alberta Petroleum Marketing Commission (APMC) agreed to supply 37,500 barrels per day of the Crown's bitumen, 75 per cent of the refinery's intake, and to pay a processing fee, or toll, to have it refined ^[5][21]. Canadian Natural supplies the other 25 per cent and pays 25 per cent of the tolls ^[5].

The refinery is designed to process about 79,000 bbl/d of diluted bitumen blend and produce about 40,300 bbl/d of ultra-low sulphur diesel, along with diluent, naphtha, low-sulphur vacuum gas oil and light ends ^[26]. It uses LC-Fining hydroconversion with a gasifier ^[16]. CO₂ from its hydrogen production is captured and shipped on the Alberta Carbon Trunk Line for enhanced oil recovery at the Clive field ^[2]. The CER noted in 2018 that capacity could triple if further phases were built ^[27].

It made its first diesel in late 2017, initially from SCO, and began processing bitumen under the tolling agreement on June 1, 2020 ^[27][21].

How the deal works

The APMC pays a monthly toll made up of debt principal, debt service and operating costs. The processing agreement runs 40 years from June 1, 2018 to May 31, 2058. The APMC has very restricted rights to end it and is unconditionally obligated to pay its 75 per cent share of the debt portion of the toll ^[6]. The province then sells the diesel and other products. It makes money only when the products are worth more than the bitumen plus the toll ^[5].

The arrangement also acts as a partial hedge. Sturgeon's feedstock is priced at a discount to WCS, so when the heavy oil discount widens and royalty income falls, the refinery's feedstock gets cheaper ^[6][5].

Rising costs and the Auditor General

The capital cost climbed repeatedly ^[5]:

• February 2011: agreement signed, facility cost estimated at $5 billion.
• November 2012: agreement amended, estimate $5.7 billion.
• December 2013: estimate raised to $8.5 billion.
• April 2014: agreement amended again to remove the cost cap, with the APMC providing half the subordinated (lower-priority) debt.
• May 2017: estimate raised to $9.4 billion.

In February 2018 the Auditor General of Alberta reported that the arrangement committed the province to about $26 billion in payments over 30 years. The estimated net present value to the government had fallen from a range of $200 million to $700 million in 2011 to under $200 million by early 2017. The APMC's loan to the partnership had grown from $324 million to $419 million ^[5].

The Auditor General found that the APMC did not have well-designed risk management systems. It had not audited the refinery's capital costs since those incurred in 2012, had done no lessons-learned analysis, and gave Albertans limited information about the deal's risks. The report also observed that there was "no plausible scenario where it would make financial sense to pull out" ^[5].

A 2018 University of Calgary School of Public Policy study projected a first-full-year loss of about $24 million for the two toll payers, roughly $1 per barrel of feedstock ^[28].

Losses and the 2021 restructuring

The final cost reached $10.1 billion. The APMC lost $1.88 billion in 2019, and as of March 31, 2020 the processing agreement carried an estimated negative net present value of $2.52 billion ^[29].

On June 30, 2021 the province, through the APMC, took over North West Refining's 50 per cent stake in the partnership ^[26]. The partnership paid $425 million to North West Refining and $400 million to Canadian Natural, and the processing agreement was extended by a decade. The government said the change would save $2 billion over the project's life, while estimating the refinery would still lose $2.5 billion over that period ^[7]. The APMC says the renegotiation improved the government's net present value by $2 billion and freed up $1 billion of cash flow through 2026 ^[30].

Views differed. Energy Minister Sonya Savage said the deal gave the province "a seat at the table." Former APMC chief executive Richard Masson said ownership offered more control than risking a partner's failure. University of Alberta economist Andrew Leach called the project a "bitumen boondoggle" ^[7].

Recent results and status

• Fiscal 2023–24: the APMC's refinery segment lost $1.6 billion, mostly because of a $1.263 billion non-cash charge to its "onerous contract" provision, driven by lower forecast crack spreads (the margin between product and crude prices). The APMC also reported that the refinery generated $273.4 million of EBITDA (earnings before interest, taxes, depreciation and amortization) for it that year ^[6].
• January 2024: the province said it was examining options for its investment, including a sale ^[31].
• Fiscal 2024–25: the APMC recorded a net loss of $0.6 billion from the processing agreement, mainly due to higher feedstock costs ^[32].
• Operations in 2025: after its second major turnaround, the refinery processed more than 52,900 barrels of bitumen a day (up from 45,800 in 2024) and produced more than 41,800 barrels of diesel a day (up from 37,800). It captured about 1.12 million tonnes of CO₂ and had 339 full-time employees ^[33].
• Budget 2026: the APMC was projected to lose $604 million in 2025–26 and is forecast to lose $313 million in 2026–27, reflecting higher production after the turnaround. Losses are forecast at $343 million in 2028–29 ^[8]. These are forecasts.

Supporters point to a steady supply of low-sulphur diesel, several hundred skilled jobs, and more than a million tonnes a year of captured carbon ^[33]. Critics point to a cost that nearly doubled and long-term losses borne by taxpayers ^[7][5].

Alberta's Industrial Heartland

Alberta's Industrial Heartland is a 582 km² industrial cluster northeast of Edmonton. Its planning association, founded in 1999, brings together Fort Saskatchewan, Lamont County, Strathcona County, Sturgeon County and the City of Edmonton. The association counts more than 40 companies and about $50 billion of existing capital, and describes the region as Canada's largest hydrocarbon processing region ^[34].

The area was once nicknamed "upgrader alley." Seven major projects were announced there in 2007, but by 2010 local officials expected only one or two to proceed within a decade ^[35]. Today it hosts a mix of fuels, chemicals, hydrogen and carbon capture:

• Refining and upgrading: the Scotford complex (upgrader, refinery, chemicals plant and the Quest carbon capture project) near Fort Saskatchewan ^[22][18], the Sturgeon Refinery near Redwater, and Edmonton's Strathcona and Suncor refineries ^[21].
• Polypropylene: Inter Pipeline's Heartland Petrochemical Complex combines a propane dehydrogenation (PDH) plant, which turns propane into propylene, with a polypropylene plant. It can make 525,000 tonnes of polypropylene a year ^[36]. The plastics plant was commissioned in July 2022 ^[37], and commercial production began in early 2023 ^[36]. It received a $408 million grant from the Alberta Petrochemicals Incentive Program ^[38], after earlier qualifying for up to $200 million in royalty credits under the first round of the Petrochemicals Diversification Program ^[39].
• Ethylene and polyethylene: Dow's Path2Zero expansion at Fort Saskatchewan is billed as a net-zero-emissions ethylene cracker, costing $10.1 billion and designed for 3.2 million tonnes a year of polyethylene and ethylene derivatives. Dow reached a final investment decision on November 28, 2023, delayed construction in April 2025, and in January 2026 moved completion to 2029 for phase one and 2030 for phase two ^[40][41][42].
• Hydrogen: Air Products is building a $4.6 billion hydrogen plant in Edmonton designed to make 1,500 tonnes of hydrogen a day, capture 3 million tonnes of CO₂ a year and supply hydrogen by pipeline to Imperial's Strathcona refinery. The province lists it as under construction, with a 2022–2027 timeline ^[43].
• Upgrader off-gas: plants in Fort McMurray and the Heartland recover natural gas liquids and olefins from upgrader off-gas as petrochemical feedstock ^[2][10].

Not all of Alberta's petrochemical industry is in the Heartland. NOVA Chemicals' large Joffre site, with three ethylene crackers and two polyethylene plants, is about 21 km east of Red Deer ^[44].

Provincial incentives shaped the cluster. The Alberta Petrochemicals Incentive Program pays grants worth 12 per cent of eligible capital costs ^[38]. Budget 2026 says the province has committed $647 million to five projects representing more than $9.1 billion of private investment ^[8].

Carbon capture links. The Heartland hosts Alberta's two large operating capture systems. Quest, at the Scotford Upgrader, began operating in 2015, can capture up to 1.08 million tonnes of CO₂ a year and had stored more than 10 million tonnes. The Alberta Carbon Trunk Line, operating since 2020, carries CO₂ from the Sturgeon Refinery and Nutrien's Redwater fertilizer plant, has capacity for up to 14.6 million tonnes a year and had moved more than 8 million tonnes. Governments contributed $745 million to Quest and $495 million to the trunk line ^[45].

Partial upgrading: the middle road

Partial upgrading tries to do just enough processing to meet pipeline specifications with little or no diluent. Alberta Innovates defines it as technologies that process bitumen to enhance its value while reducing or eliminating the need for diluent ^[46].

The prize is real. Diluent takes up about 30 per cent or more of a dilbit pipeline barrel, and in the second quarter of 2017 the diluent needed to ship bitumen to the U.S. Gulf Coast cost producers nearly US$14 per barrel of bitumen ^[12]. Cutting diluent frees pipeline space and could raise the value of each barrel. The most promising technologies rely on thermal cracking, some combined with removing part of the asphaltenes ^[12].

A 2018 white paper for Alberta Innovates found no partial upgrading plants in operation and no fully commercial technology ^[12]. Alberta Innovates funded several projects, including CNOOC's BituMax technology from 2016 to 2023, but its partial upgrading program is now listed as inactive ^[46].

The 2018–2019 policy push

• 2016–2017: the province's Energy Diversification Advisory Committee studied value-added options and recommended support for partial upgrading in its report Diversification, Not Decline ^[47].
• February 2018: the government announced up to $1 billion in grants and loan guarantees to help build up to five partial upgraders ^[39]. Bill 1, the Energy Diversification Act, was introduced on March 8, 2018 ^[47].
• January 2019: the province signed a letter of intent for a $440 million loan guarantee for Value Creation Inc.'s planned $2 billion partial upgrader in Strathcona County, which would process 77,500 bbl/d of diluted bitumen into medium synthetic crude and ultra-low sulphur diesel ^[48]. The company targeted operation by 2022 ^[49].
• October 2019: the province discontinued the Partial Upgrading Program and the Petrochemical Feedstock Infrastructure Program. Petrochemical support continued through the Alberta Petrochemicals Incentive Program, which opened on October 30, 2020 ^[38].

Company efforts

MEG Energy developed a process called HI-Q to make pipeline-ready crude without diluent. A field pilot, backed by $5.2 million from Emissions Reduction Alberta within an $88.4 million project, passed stability and desalting tests and increased yield by 4 per cent. MEG sold the technology to Suncor in 2019. A follow-up Suncor demonstration approved in 2021 was cancelled because of high capital costs ^[50].

A simpler form of partial processing already operates at the mine. Suncor's Fort Hills uses paraffinic froth treatment to make a partially decarbonized bitumen that needs less diluent and no on-site upgrader ^[17]. No stand-alone commercial partial upgrader was reported operating in Alberta in the sources reviewed for this guide.

Why so few new upgraders were built after 2008

In late 2008 Petroleum News reported that about C$90 billion of upgrader investment near Edmonton was in limbo as financial markets collapsed ^[51]. Canadian Natural's president Steve Laut summed up the mood: "We will not build in a high-price environment for a moderate-price world" ^[51]. Since then, new capacity has come mainly from Horizon's phased expansions, debottlenecking at existing plants and the Sturgeon Refinery ^[18][14].

Several forces explain the pattern.

Capital cost. Upgraders are among the most expensive plants in the industry. When Suncor cancelled Voyageur in 2013 its estimated cost was $11.5 billion, about $60,000 per barrel a day of capacity ^[12]. Suncor said market conditions had changed significantly since 2010 and took a first-quarter charge of about $140 million ^[53]. Sturgeon's rise from $5.7 billion to $10.1 billion showed the same risk ^[7].

Narrowing differentials. An upgrader earns the gap between cheap heavy bitumen and valuable light crude. The CER says soaring capital costs and shrinking light-heavy price differentials, largely a result of the U.S. shale oil boom, made new upgraders hard to justify, so only expansions were completed ^[14]. The WTI-WCS differential averaged US$14.73 per barrel in 2024, down from US$18.65 in 2023 after the Trans Mountain expansion started ^[2].

Upgrading capacity already exists elsewhere. Many U.S. refineries, especially in the Midwest and Gulf Coast, have coking units designed for heavy sour crude, originally to process Mexican and Venezuelan oil ^[11]. The AER notes that a significant share of refining capacity in those regions is designed to process bitumen ^[15]. Expanding existing refineries in distant markets is generally cheaper than building new greenfield capacity in Canada ^[12].

The case for more upgrading. Supporters argue that processing more bitumen at home adds jobs, investment and tax revenue, reduces diluent use and frees pipeline space. That reasoning drove the Energy Diversification Advisory Committee's recommendations ^[47], and a 2017 School of Public Policy study cited by Jacobs saw significant potential benefits for Alberta and Saskatchewan ^[12]. Sturgeon, however, shows the risk: the Auditor General found the province had taken on many of the risks as if it were building the refinery itself ^[5].

Today the industry favours squeezing more from existing plants. Recent examples include Scotford's 2024 debottleneck, Horizon's reliability project and Suncor's coke drum replacement ^[18][17]. The AER's forecast assumes no new upgrading facilities through 2034 ^[1].

Environmental basics: sulphur, coke and air

Upgrading leaves two large solid by-products, sulphur and coke, and both upgrading and refining emit air pollutants. Greenhouse gases are covered in the oil sands guide.

Sulphur. In 2025 Alberta produced about 4.18 million tonnes of sulphur. About 2.90 million tonnes came from oil sands facilities, 1.13 million tonnes from gas processing plants and 0.15 million tonnes from refineries. Reported sulphur inventory fell from about 11.99 million tonnes at the start of the year to 11.41 million tonnes at the end ^[9]. Upgrader sulphur is either stockpiled or shipped to plants that turn it into sulphuric acid, mainly for fertilizer ^[2]. Sulphur recovery units with tail gas treatment typically capture at least 99.8 per cent of the sulphur in the feed ^[13], and the Sturgeon Refinery reported 99.0 per cent recovery in 2025 ^[33].

Petroleum coke. At the end of 2025 the AER's plant statistics showed about 80.5 million tonnes of coke in inventory at Syncrude's Mildred Lake, 64.7 million tonnes at Suncor's base plant and 38.9 million tonnes at Horizon, roughly 184 million tonnes in total. The three plants produced about 13.3 million tonnes of coke in 2025 (totals calculated) ^[10]. Coke can be burned as fuel or used in reclamation and water treatment ^[13]. Suncor stopped relying on it for steam in late 2024, when an 800 MW cogeneration facility replaced the coke-fired boilers at its base plant ^[17].

Air monitoring. Independent airshed groups run monitoring networks around both processing hubs. The Wood Buffalo Environmental Association operates 29 ambient air stations in the Fort McMurray region, measuring pollutants such as sulphur dioxide, hydrogen sulphide, fine particulate matter and nitrogen dioxide, and publishes its data online ^[54]. In the Heartland, the Heartland Air Monitoring Partnership (formerly Fort Air Partnership), formed in 1997, covers about 4,500 km² north and east of Edmonton ^[55]. It is one of 10 airsheds in Alberta and reports real-time air quality readings ^[56].

FAQ

What is the difference between an upgrader and a refinery? An upgrader turns bitumen into a lighter synthetic crude oil by removing carbon or adding hydrogen and stripping out sulphur ^[2]. A refinery turns crude oil, including SCO, into finished products such as gasoline, diesel, jet fuel and asphalt ^[11]. The Sturgeon Refinery blurs the line by taking diluted bitumen straight to diesel ^[26].

Is synthetic crude oil the same as dilbit? No. Synthetic crude has been chemically upgraded into a light, low-sulphur oil with no heavy residue ^[19]. Dilbit is raw bitumen mixed with about 30 per cent or more of a light diluent so it can flow in a pipeline; the bitumen itself is unchanged ^[12].

Why doesn't Alberta build more upgraders? New upgraders cost billions, and the price gap between heavy and light crude has often been too narrow to pay for them ^[14]. Many U.S. refineries already have coking units to process heavy oil ^[11]. Several projects were cancelled after 2008, including Suncor's Voyageur upgrader in 2013 ^[53].

How much has the Sturgeon Refinery cost Albertans? The refinery cost $10.1 billion to build ^[7]. The province's marketing commission lost $1.88 billion in 2019 ^[29], $0.6 billion in 2024–25 ^[32], and is forecast to lose $313 million in 2026–27 ^[8]. Final returns depend on diesel prices, bitumen prices and interest rates over a contract that runs to 2058 ^[6][5].

What happens to the sulphur and coke taken out of bitumen? Sulphur is recovered as elemental sulphur and either stockpiled or sold, mainly for fertilizer production ^[2]. Coke is mostly stockpiled at the mine sites; about 184 million tonnes were held at three upgraders at the end of 2025 (calculated) ^[10].

Where do Alberta's gasoline and diesel come from? Almost all gasoline sold in Alberta is refined in the province, mostly in the Edmonton area ^[21]. Surplus fuel moves to British Columbia on Trans Mountain, to Saskatchewan and Manitoba on the Enbridge Mainline, and to southern Alberta on the Alberta Products Pipeline ^[21].

Is anyone building a partial upgrader in Alberta? Not at commercial scale, based on the sources reviewed for this guide. The province's Partial Upgrading Program was discontinued in October 2019 ^[38], and a Suncor demonstration of MEG's HI-Q technology was cancelled because of high capital costs ^[50].