ELEMENTS, VOL. 10, PP. 277-284 AUGUST 2014277

1811-5209/14/0010-277$2.50 DOI: 10.2113/gselements.10.4.277

Oil Sands and Heavy Oil: Origin and Exploitation

INTRODUCTIONOutput from Alberta’s oil sands bitumen reserves, the world’s third-largest proven crude oil deposit, is expected to reach 2.3 million barrels (bbl) per day in 2015, an increase of 26% over production in 2012, and then rise to 5.2 million barrels per day by 2030. Hein et al. (2013) estimate that global bitumen and heavy-oil resources are around 5.6 trillion barrels, with most of these located in the Western Hemisphere. Much of the enabling technical developments have occurred in the largest bitumen and heavy-oil fi elds of the Canadian oil sands, the Orinoco heavy oil belt of Venezuela, the heavy oil on the North Slope of Alaska, and the heavy-oil fi elds of California. These areas have served as proving grounds for the commercial development of the in situ recovery technologies (mostly thermal) that will be used to extract most of the remaining unconventional bitumen and heavy-oil resources.

The American Petroleum Institute gravity (or API gravity) is the oil density–based standard by which oil quality is measured. It is reported in degrees (= [141.5/specifi c gravity at 60 °F] − 131.5). Even with such standards in place, industry terminology is inconsistent and confusing. For example, many of the so-called “extra heavy oils” of Venezuela would be considered “oil sands” in Canada or

“tar sands” in the United States. Heavy oil is defi ned as oil with 10° to 20° API (oil with <10° API is denser than water) and a viscosity of more than 100 centipoises (cp; 1 cp = 1 mPa.s). Bitumen is defi ned as oil having less than 10° API and a viscosity of more than 10,000 cp at ambient conditions. The main distinction between the two is that the high viscosity of bitumen prevents it from fl owing to a wellbore under in situ reser-voir conditions, whereas heavy oils will fl ow under the same condi-tions. Heavy oil and bitumen can be regarded as part of a continuum of heavily to severely biodegraded oil and, in oil sands, are found in unconsolidated sandstone reser-voirs (Hein et al. 2013).

Bitumen is distributed across Alberta in Lower Cretaceous sandstones and underlying Devonian carbonate reservoirs of the Western Canada Sedimentary Basin (FIG. 1). These biodegraded oils exhibit wide variations in fl uid properties laterally and vertically, on both local and regional scales, with API gravities as low as 6° and with dead-oil viscosity in the range of thousands to tens of millions of centipoises at reservoir conditions (FIG. 2). Dead oil refers to oil without its dissolved gas. In the Peace River and Lloydminster areas of the Albertan oil sands, the high bitumen viscosity means that primary production is only possible in the few places where oil viscosities of <35,000 cp are found. These opera-tions yield low oil recoveries (much less than 15 to 20%). Thus, much of the oil sand reserves must be extracted using enhanced oil recovery (e.g. steam-assisted gravity drainage, SAGD, or via mining operations; Hein et al. 2013).

THE OIL SANDS MACHINELarge oil sand deposits are found in foreland basins adjacent to orogenic belts, with source-rock “kitchens” (where oils are generated) charging large, shallow, cool reservoirs at the basin fl ank, where conditions are suitable for severe biodegradation of the oils (Creaney et al. 1994; Adams et al. 2013). The world’s largest oil sand deposit, located in western Canada, is hosted in Early Cretaceous sandstones in a basin adjacent to the Canadian Rocky Mountains (foreland basin) (FIG. 1). Petroleum was derived princi-pally from marine-shale source rocks, with the petroleum migrating eastward up to several hundred kilometers to accumulate and become biodegraded on the northeastern margin of the basin. The main phase of accumulation was around 84–55 million years ago (Adams et al. 2013; Tozer et al. 2014). The petroleum accumulated in tide-infl uenced

Oil sands are a mixture of “bitumen” (a very viscous, heavily biode-graded crude oil), unconsolidated sand, and water bound together by the bitumen and confi ning stresses. Economic incentives to produce

reserves from the western Canada oil sands have driven geological and geochemical mapping to assess fl uid quality controls and improve our under-standing of the fundamental principles of the biodegradation of oils. While much of this activity has been for practical application, researchers have also had the opportunity to make fundamental advances in our understanding of subsurface biogeochemical processes and the boundaries of life in Earth’s crust. Indeed, the huge size and shallow location of oil sands, coupled with the many thousands of wells drilled, mean that on a per cell basis, oil sands represent a most accessible portion of the deep biosphere. Perhaps the most exciting future for the oil sand resource is on the biological front rather than as an energy resource.

KEYWORDS: deep biosphere, biodegradation, heavy oil, bitumen

Stephen R. Larter1 and Ian M. Head2

1 Department of GeoscienceUniversity of Calgary2500 University Drive NW, Calgary, Alberta T2K 3J8, Canada E-mail: slarter@ucalgary.ca

2 School of Civil Engineering and GeosciencesNewcastle University3rd Floor, Devonshire Building, Newcastle upon Tyne NE1 7RU, United Kingdom E-mail: ian.head@ncl.ac.uk

The surface and deep biospheres meet:

Oil sands reservoir and boreal forest in

northern Alberta, Canada

ELEMENTS AUGUST 2014278

river and estuarine sediments. Oil similarly accumulated in foreland basin settings in the Ofi cina Formation in Venezuela, another major heavy-oil resource.

Bitumens are crude oils depleted in saturated hydrocarbons and enriched in aromatic hydrocarbons (such as alkylnaph-thalenes or alkylphenanthrenes) and non-hydrocarbon compounds containing sulfur, nitrogen, and oxygen (as well as hydrogen and carbon). Non-hydrocarbon compounds make up around 25 wt% of the western Peace River oil sands and nearly 60 wt% of parts of the Athabasca oil sands. Canadian bitumen is signifi cantly non-hydro-carbon in nature and is commonly denser than water. Geochemical studies suggest that the bitumen deposits of Alberta share common source rocks and similar thermal maturities with the oil sourced predominantly from the Mississippian/Devonian Exshaw Formation and with oil in the western Peace River oil sands sourced from the Jurassic Gordondale member (Creaney et al. 1994; Adams et al. 2013).

The primary control on oil composition and viscosity is in-reservoir biodegradation (Larter et al. 2008). Oil API gravity in the Alberta Lower Cretaceous reservoirs ranges from 38° API (light oil) in the barely biodegraded oil pools west of the Peace River oil sands to 6° API (bitumen) in severely biodegraded eastern Athabasca oil sands bitumens, and to even lower API values in the most extremely degraded

bitumens found in karsted Grosmont carbonate reservoirs underlying the oil sands. Oil sulfur contents range from 1 to >10 wt%, with the western Peace River oil sands having the highest sulfur contents, due to an additional sulfur-rich oil charge from Jurassic source rocks. This overall variability in fl uid properties correlates roughly with the level of oil biodegradation, which broadly increases from west to east and from south to north (FIG. 1B). Field observa-tions typically record a coincidence of the lowest oil quality (highest viscosity and lowest API gravity) and strongest biological and molecular evidence for hydrocarbon biodeg-radation at or near oil–water transition zones (OWTZs) in the deepest oil-fi lled parts of individual sandstone reser-voirs, suggesting most petroleum biodegradation occurs near the interface (Bennett et al. 2013) where the biosphere meets the geosphere (FIG. 2).

HEAVY OIL: THE BASE OF THE DEEP BIOSPHERE AND MADCORSeveral lines of biogeochemical and microbiological evidence suggest that the maximum temperature for biolog-ical activity in petroleum reservoirs is around 80 oC (Head et al. 2003; Adams et al. 2013). Connan (1984) suggested that in-reservoir oil biodegradation ceases at a reservoir temperature of about 80 oC. However, reservoirs containing non-degraded oils are found at low temperatures, and the

Dept

h (m

KB)

Temperature (°C)

0

0

15

45

80

1000

100 50

2000

Time (Ma)

onse

t of c

harg

e

3

2

1

Athabasca

Steriliza!on

Peac

e Rive

r

Wester

n Reserv

oirs

Mannville Reservoirs

Regionalseal

Main reservoironlapping ontobasement

Extensive severelydegraded tar sand

Basement

Foreland basinOrogenic belt

Non-degraded oilsin deeper reservoirs

Unit containing source rock 1Unit containing source rock 2

Oil migrationdirections

Foreland basement fill

Eroded section

c

a b

d

1 km

50 km

SedimentsWater

Thrusting

Mature zone for oil

M

ature

zone for gas

FIGURE 1 (A) Foreland basins produce the largest oil sand deposits (after Head et al. 2003). In Canada, the

Lower Cretaceous sandstone reservoirs (McMurray Formation of the Manneville Group) host the largest resource (Athabasca). It is a complex sandstone reservoir with vertical (interbedded thin silts and shales) and lateral barriers (compartment-forming mud plugs) (Fustic et al. 2013). (B) Variation of oil quality in oil sand and heavy-oil reservoirs across Alberta. Whole-oil chromatograms are shown for oils typically found in each region. To the west of the oil sands, the oils have n-alkanes, while to the east the oils are heavily biodegraded with no n-alkanes. The API gravity and viscosity (centipoises, cp) ranges are reported in black and blue texts,

respectively, and the Peters and Moldowan (PM) biodegradation level of the oils is reported in red (scale ranges from 0 [no biodegradation] to 10 [extreme biodegradation]). The inset plot shows the variation of the API gravity and dead-oil viscosity for the oils in Lower Cretaceous reservoirs along an east–west cross-section (the red line on the fi gure), through the Athabasca and Peace River oil sands and beyond into the pasteurized reservoirs to the west of the oil sands (after Adams et al. 2013). (C) Generalized burial- and temperature-history curves showing low maximum reservoir temperatures for Athabasca, intermediate for Peace River, and high (pasteurized) maxima for reservoirs to the west of Peace River (after Larter et al. 2006; Adams et al. 2013).

A

B

C

ELEMENTS AUGUST 2014279

explanation for their occurrence invokes the paleopasteur-ization hypothesis (Wilhelms et al. 2001). According to this hypothesis, the upper thermal limit for hydrocarbon-degrading microbial life in petroleum reservoirs is typically around 80–90 oC. Once a reservoir has been heated to temperatures in this range, it is typically not recolonized by hydrocarbon-degrading microorganisms, even if the reser-voir is subsequently uplifted to shallower depths where in situ temperatures are below 80 oC. Subsequently a relation-ship between the extent of biodegradation, maximum reservoir burial depth, and maximum burial temperature was observed in the Western Canada Sedimentary Basin (Adams et al. 2006). Reservoirs containing light oils west of the Peace River oil sands experienced pasteurization above 80 oC during deeper burial. The western boundary of the oil sands in the Lower Cretaceous reservoirs is the boundary of the petroleum-degrading deep biosphere in Alberta (FIG. 1B, C).

Why does life in oil fi elds cease at temperatures lower than the 121 oC observed for organisms in high-tempera-ture hydrothermal systems? Thermophilic, hydrocarbon-degrading microorganisms are rare. One moderately thermophilic, alkane-degrading, sulfate-reducing bacte-rium, with optimum activity at 60 oC, has been isolated (Rueter et al. 1994). Only recently has hydrocarbon degra-dation at relatively high temperatures been reported from

petroleum reservoir microbial cultures under both sulfate-reducing and methanogenic conditions (Gieg et al. 2010). Observations from laboratory and fi eld studies suggest that hydrocarbon-fermenting bacteria—which provide methanogens with H2, CO2, and acetate—are more heat sensitive than the methanogens themselves (pasteuriza-tion at 95 oC for 2 hours stops methanogenesis associated with oil degradation, but does not kill all methanogens). There are also a few reports of subsurface hyperthermo-philes isolated at temperatures above 90 oC (Grassia et al. 1996). Methanogenesis and sulfate reduction have been measured only at 70–83 oC in waters produced from Californian petroleum reservoirs, which can be as hot as 120 oC (Orphan et al. 2003). The base of the heavy-oil zone appears to correspond to the base of the hydrocarbon-degrading biosphere in Earth’s crust!

Where do the organisms responsible for oil biodegradation in petroleum reservoirs come from? To the west of the Peace River oil sands, reservoirs have remained pasteurized for over 50 million years, and we know of no good examples of reinoculation of pasteurized heavy-oil fi elds. Work from McIntosh’s group (e.g. Schlegel et al. 2011) demonstrated that organic-rich shales, and some coals located in interior United States basins, locally matured to temperatures well over 100 oC, became methanogenic after uplift, cooling, and exposure, and underwent microbial reintroduction

BacteriaMethanogenic Archaea

MetabolismFermenta!onSmithella/SyntrophusFirmicutesThermotogalesSynergistalesFlexis!pes

Smithella/SyntrophusFirmicutes

Acetoclas!cmethanogenesis

Syntrophicacetate oxida!on

Methanoculleus MethanocorpusculumMethanogenium

MethanosaetaMethanosarcina

MethanogenicCO2 reduc!on

MethanogenicCO2 reduc!on

4 C16H34 + 64 H2O

68 H2 + 32 CH3COOH

64 CO2 + 128 H2 + 68 H2

32 CH3COOH + 68 H2

68 H2 17 CO2 + 15 CO2 + 32 CH4

17 CH4 + 34 H2O

49 CH4 + 15 CO2 + 34 H2O49 CH4 + 15 CO2 + 34 H2O

64 H2O

CB Oil zone-sandstone30% porosity. 85% oil satura!on≥ 1 Darcy permeability104 cells/gm

Biodegrade︢ 10

-4kg.m 2.yr -1

degrade> charge

charge> degrade

Oil Water Transi!on Zone Sandstone. 0-50% water satura!onCa 106 cells/gm MADCOR

OWC

~10-4 kg.m2 .yr-1

Oil Charge

OWTZ

OilLeg

Shale/silts

Conglomerate

Sands

Sands

Shales/silts

0.01 1.0 100

Oil ViscosityMcp (at 20°C) Wt % Bitumen

0.0 5.0 10.0

Resis!vity log

0 50 100Ohm meters

Lithology

Water

OWTZ

OilLeg

0 500 1000

n-C30 Pr

μg/g oil0 100 200

2-MP 26,27-DMN

μg/g oilA

-2

3

8

13

18

23

28

Dept

h (m

)

FIGURE 2 Processes and bioreactors. (A) Geophysical and geochemical logs through an oil column and oil–water

transition zone (OWTZ) near 600 m depth, in a Canadian oil sand reservoir, show gradients of hydrocarbon destruction and increasing oil viscosity with depth (n-C30 = alkane; pr = pristane; 2-MP = 2 methylphenanthrene; 26,27-DMN = dimethylnapthalenes) (MODIFIED AFTER BENNETT ET AL. 2013). The OWTZ has lower oil contents, as indicated by the resistivity log and the bitumen content log. (B) During the fi lling of the reservoir with oil over geological timescales, oil charge and biodegradation rates had similar magnitudes and the oil–water contact (OWC) could have

migrated back and forth competitively until the reservoirs fi lled, with gas leaking through the shale caprock above the sandstone reservoir. The multimeter-thick OWTZ contains the main biological resource of the reservoir; the OWTZ is the site of the MADCOR process (methanogenic alkane degradation dominated by CO2 reduction), in which H2O and hydrocarbons react to make CH4 and CO2, reducing the bitumen content in the process. (C) Major processes involved in MADCOR. Points where H2 is a key interme-diate are highlighted in red. In oil reservoirs, methane production from acetate (acetoclastic methanogenesis) is subordinate and most methane is generated via (syntrophic) acetate oxidation coupled to methanogenic CO2 reduction. MODIFIED AFTER JONES ET AL. (2008)

A

B C

ELEMENTS AUGUST 2014280

with groundwater fl ow. For the most part, in the oil sands as in other oil reservoirs, the petroleum deep biosphere is most likely derived from the near-surface biosphere through subsidence: the organisms present at depth are descendants of those living in the sediments when they were deposited.

While much progress has been made, there are still confl icting observations about microbial processing of heavy oils. Recent work suggests that while the bulk processes forming the oil sands must have been anaerobic, “aerobic-organism” genomic signatures (e.g. Pseudomonas species) have been identifi ed in oil sands petroleum columns (An et al. 2013). Our group recently suggested that these organisms might actually be capable of functioning anaerobically, and anaerobic activity by Pseudomonas spp. has been reported by Budwill (pers. comm. 2011) in anaerobic, methanogenic, coal-biodegradation micro-cosms. Certainly, the abundant associated methane, low levels of CO2, and heavy carbon isotope signatures of the CO2 observed in heavy-oil and oil sand reservoirs (Jones et al. 2008) are consistent with anaerobic, and specifi -cally methanogenic, processes driving overall oil degra-dation in these systems. However, it is likely that there are processes of which we have little appreciation at present. For example, at the end of the last glacial period during melting of the thick ice sheet that covered the oil sands (Grasby and Chen 2005), shallow oil sand reservoirs could have been fl ooded with surface waters and thus potentially infected with new organisms.

The extent of petroleum degradation in reservoirs over geological timescales is controlled by the reservoir tempera-ture, the chemical compounds being degraded, recharge and mixing of fresher oils with degraded oils, and the relationships between bioreactor elements (oil and water volumes; oil–water contact area) (Larter et al. 2006). Reservoir-water salinity appears to act as a second-order control on the process, and variations in these factors across oil fi elds control the lateral and vertical oil compo-sition and viscosity gradients observed (Larter et al. 2006, 2008). Biodegradation fl uxes for fresh petroleum in clastic oil reservoirs, assessed using advection/diffusion models and geochemical gradients (Larter et al. 2003), are in the range of 10−4 kg petroleum m−2 OWC y−1 (OWC = oil–water contact). The biodegradation fl ux increases with decreasing reservoir temperature, from a fl ux of zero near 80 oC to a maximum fl ux at the OWC of much less than 10−3 kg petroleum m−2 OWC y−1 at temperatures less than 40 oC. The biodegradation fl ux subsequently drops to much lower levels at very low reservoir temperatures and with severely degraded oils, such as seen in oil sands today where reser-voir temperatures can be as low as 10 oC (Adams et al. 2013).

Quantifi cation of bacteria in the OWTZ of an oil sand reservoir, using polymerase chain reaction approaches, suggests that about 105 to 106 cells per gram of sediment are present, with approximately 103 to 104 cells/g outside the transition zone (Bennett et al. 2013). This is in agreement with the notion that microbial activity and abundance in the deep subsurface is elevated at geochemical interfaces. Bacterial abundances in the OWTZ are consistent with the trend of decreasing bacterial abundance with depth that has emerged from extensive analysis of microbial cells in deep subsurface sediments (Parkes et al. 1994). Thus, oil reservoirs are nothing special in the context of the deep biosphere except for the much greater access provided by the oil industry! The Canadian oil sands biosphere likely contains well in excess of 1023 bacterial cells (FIG. 3).

Variation in biodegradation levels in heavy-oil columns is ubiquitous, refl ecting biodegradation processes at the oil column base and mass transport processes within the oil column itself (Larter et al. 2003). Modeling suggests that where continuous chemical gradients in oil columns are seen, biodegradation must have been occurring near continuously, for many millions of years. It is estimated that ~35 million years of biodegradation, close to the time span of main oil charge, reservoir burial and subsequent reservoir uplift (ca 85–50 Ma), resulted in biodegradation of ~30–50% of the oil mass to produce the current oil sand bitumen and large volumes of gas, up to 7 times the reser-voir volume of methane and CO2 (Adams et al. 2013). The absence of signifi cant gas caps today is indicative of major gas leakage from the deposit during this time. Without this leakage, the gas would have displaced much of the oil from the reservoir. Much of this methane and CO2 would have been released during the Laramide uplift and may have made a small contribution to the warming seen at the Paleocene–Eocene Thermal Maximum (Adams et al. 2013).

In light-oil reservoirs, a consortium of syntrophic bacteria (one species lives off the products of another species) adding functional groups to hydrocarbons and oxidizing acetate to H2 and CO2 are linked with methanogenic archaea that use hydrogen to reduce CO2 to CH4. MADCOR (methano-genic alkane degradation dominated by CO2 reduction) is the likely process by which alkane-rich crude oil is turned into methane, CO2, and residual heavy oil or bitumen (Jones et al. 2008). While the number of methanogenic, crude oil–degrading microbial communities that have been characterized remains small, it is clear that a number of different organisms may have a role in converting crude oil components to methanogenic substrates.

It is intriguing that the bacterial abundance in the most biologically active part of oil sand reservoirs is compa-rable with that seen elsewhere in the deep biosphere. This suggests that the supply of electron donors (organic matter) is not limiting in petroleum biodegradation processes (Head et al. 2003). It remains unclear whether nutrient limitations, microbial inhibition by petroleum compo-nents, water availability, or other factors control net rates of biodegradation in reservoirs. Similar levels of activity and cell abundance in the deep biosphere, within or outside oil fi elds, suggest that that there is a large-scale, common control on biological activity in the deep subsurface. The observation that oil charge rates (known to be largely driven by source-rock heating rates) and oil biodegrada-tion rates are similar in heavy oil fi elds (Larter et al. 2003) may suggest a causal linkage related to common physics of mass and energy transport in the subsurface (or it may be purely coincidental).

H2 is a key intermediate in heavy-oil and bitumen forma-tion (FIG. 2C). In oil reservoirs, acetoclastic (acetate-split-ting) methanogenesis (anaerobic archaea converting acetic acid to CH4) seems subordinate, and most methanogenesis (>80%) results from the reduction of CO2 with hydrogen, with both H2 from water and hydrocarbons being liber-ated during the degradation process. Dolfi ng et al. (2008) demonstrated that a low PH2 (<4 Pa) is required for thermo-dynamic feasibility of the methane-producing MADCOR process, and this is achieved naturally in oil fi elds by methanogens using H2 for CO2 reduction. Although it is theoretically feasible to recover H2 from biodegrading oil fi elds, the low concentrations would render this diffi cult as a viable low-carbon, large-scale energy-recovery process.

ELEMENTS AUGUST 2014281

ENERGY RECOVERY Biodegradation reduces oil hydrocarbon concentration and increases oil molecular weight, resulting in large increases in oil viscosity. Light oils may have viscosities of around a centipoise or less (water viscosity at 20 oC), whereas bitumen may have viscosities from tens of thousands (cold maple syrup) to tens of millions (cold peanut butter) of centipoises at reservoir conditions. Thus severely biode-graded oil (bitumen) will not fl ow under native reservoir conditions. Two sets of processes to liberate heavy oils from sandstone reservoirs have been invented. The fi rst set of processes is a variant of the Clark process (Clark 1931), which takes mined oil sand “ore” and processes it with alkaline warm water to separate an oil phase from the sand; the oil is then concentrated, upgraded, and refi ned. The second set of processes, used for oil sand reservoirs deeper than ~200 m, involves steam injection into the reservoir itself. The most common process used today for in situ recovery of bitumen is called steam-assisted gravity drainage (SAGD; Butler 1997); it uses injected steam, gener-ated in natural gas–fi red boilers, to heat oil in situ to lower its viscosity. Heating the oil to temperatures above 200 oC reduces its viscosity to around 10 cp, enabling production. Given that these deeper reservoirs make up 97% of the total oil sand resource, thermal recovery processes will dominate future production from oil sands.

Current in situ oil sand development focuses on SAGD technology and, to a lesser degree, cyclic-steam stimula-tion (CSS) for deep deposits. Both are clever technologies involving steam injection and oil and condensed-water recovery from horizontal wells that are several hundred meters long (FIG. 3; Butler 1997). Other emerging technolo-gies include in situ combustion, in which air or oxygen is injected and oil is ignited to produce in situ heat and steam. The addition of solvents such as butane to injected steam, the electrical heating of reservoirs, and passive heating-assisted recovery methods (Hein et al. 2013) are also used but are currently minor players.

SAGD (FIG. 3) is the primary in situ process and uses two horizontal wells. The top well constantly injects steam, at near reservoir pressure, into a depletion chamber that forms in the oil sands reservoir. The steam releases its latent heat through condensation at the edge of the chamber, heating the surrounding oil sand and causing the viscosity of the heated oil to decrease. Under the action of gravity, the heated oil drains through the reservoir down the edges of the steam chamber to the production well at the base of the chamber (Butler 1997). The key controls on the produc-tivity, effi ciency, and emissions of the SAGD process are reviewed by Gates and Larter (2014), who discuss fl uid-property variations and intrareservoir shales as the major obstacles to effi cient oil production.

The recovery, extraction, and processing of bitumen is an energy-intensive process, using mostly natural gas and electricity. It also results in signifi cant greenhouse gas emissions (Bergerson and Keith 2010). Open-pit mining projects typically have lower energy use, lower CO2 emissions, and better resource-extraction effi cien-cies (i.e. less energy and resources required per barrel of bitumen produced) than in situ projects. In situ projects, however, have lower water use and a smaller direct land-use footprint, though if habitat fragmentation is counted, their life cycle land-use impacts may be greater. Looking only at extraction emissions, oil sand emissions are up to 5 times higher than those of conventional oil. If, however, one looks at the full well-to-wheel emissions, across the range of emissions from oil sands and conventional crudes, then oil sand projects are about 10–20% higher than conventional oil projects (Bergerson and Keith 2010).

THE AGE OF BIOLOGY AND ALTERNATIVE FUTURES FOR “UNCONVENTIONAL RESOURCES”The discovery of MADCOR raises many possibilities for reduced emission processes as methane is produced from biodegraded hydrocarbons, utilizing water as a coreactant,

FIGURE 3 The present and some futures: The

block diagram (modifi ed after Hein et al. 2013) shows typical, optimal conditions for steam-assisted gravity drainage (SAGD) in situ schemes used in the Athabasca oil sand area. Most of the bitumen produc-tion is from the McMurray Formation. The horizontal injector and producer wells commonly have 5 m of vertical separation and extend for 600 to 800 m from the toe to the heel in the reservoir. The producer well is commonly placed 2–5 m above basal water zones in the reservoir and typically has more than 100 to 500 m of overburden. While the recoverable oil resource is large, the biological resource is immense, yet largely uncharacterized. Thus, in addition to being an energy resource, these formations are a major portal to the unexplored microbial world of oil reservoirs and may be valuable as a source of knowl-edge and novel genes!

From SAGD to beyond Syntrophic Bacteria

Resource Op!ons

ResourceFossil fuel (100 Bbbl+)

Biological resource >1023 cellsLargely uncharacterized

Today

OilMicrobial conversion to methaneKnowledge of the deep biosphere

SAGD

Futurescape: • Discovering new subsurface bacteria, archaea, viruses, and their genomes• Discovering novel biosynthesis routes and biotransforma!ons• Development of subsurface/surface applica!ons of in situ microbial and viral genes to prac!cal ends • Developing in situ processes for recovering feedstock chemicals (e.g. alcohols, acids, biopolymers)• Developing biosensors using low-energy, high-affinity enzymes of deep biosphere microbes• Developing deep biosphere microbial or viral gene pools as sources of an!bio!cs or other products• Developing routes to novel specialty chemicals based on deep biosphere microorganisms • Developing approaches to clean up contaminated areas inspired by deep biosphere organisms• Developing novel approaches for controlling bio-influenced corrosion problems in industrial processes

ELEMENTS AUGUST 2014282

with H2 being the principal intermediate (Jones et al. 2008). This raises the prospect of a truly green energy-recovery process if hydrogen production could be dramatically accelerated by engineering in a reservoir and if H2 could be recovered in situ. Unfortunately, the thermodynamic window in which H2 is generated limits its production to conditions of low PH2 (Dolfi ng et al. 2008). This requires dynamically managed recovery systems for H2 produc-tion that will be technically challenging. Accelerated methane production through nutrient addition to heavy oil fi elds is feasible today, but with low natural gas prices in North America, there is currently little economic incen-tive for such schemes. However, increased oil production together with gas is being considered as a biologically assisted engineering recovery process, following on from conventional primary and secondary oil recovery. An alternative to terminal methane production could be the use of methanotrophs (methane-consuming organisms) to convert any produced methane in situ or ex situ to higher-molecular-weight hydrocarbons, other species, or methanol. Biological approaches that use methanotrophs to produce long-chain hydrocarbons have been identifi ed as a priority by the United States government (Chu and Majumbdar 2012). Although using methanogenic and methanotrophic methods to convert heavy oil to methane to heavier hydrocarbons or other fuels is possible, it seems to be a technological stretch at this time and would not please William of Occam!

The most interesting future for oil sands may lie outside the energy sector, and that, in the end, will be largely driven by politics and economic interests. The lack of technologies beyond SAGD capable of recovering energy from the vast chemical reductive potential of oil sands, without emissions of large amounts of CO2, refl ects the low level of invest-ment in energy-related R&D over the past decades and is a global societal/business failure rather than a technical one. Today, United States (and Canadian) consumers spend more on potato chips than their governments devote to energy R&D (US NAS 2010). Low spending and lack of basic research have resulted in few breakthrough technologies for “the market” to optimize in the fossil fuel area. Marchetti (1977) showed that the time for market-share replacement of one energy source by another (e.g. replacement of wood by coal then by oil) is long, with 1% to 10% market-share replacement taking ~40 years and with 2% to 50% taking a century. Driven solely by current market economic models with their legacy investment effects, few radically different developments in fossil energy seem possible in the timetable to 2050, by which time drastic emissions reductions will necessarily already have been implemented.

Are alternate futures possible? The bitumen represents a vast chemical resource which could drive various chemical or electrochemical processes in situ or ex situ that might be used for low-emission energy recovery. Bitumen could also be a source of hydrogen or electrons via biofuel cells or other systems, and its extraction could encompass a range of in situ microbial processes having biotechno-logical applications (FIG. 3). The resource is of intriguing scale and complexity, but such radical futures would also need new business models. To date, the energy sector has been reluctant to experiment with very different energy-recovery processes involving fossil fuel resources, and current, market-driven technology-replacement cycles are too slow to signifi cantly impact emissions reductions by mid-century. A market-driven model of technology devel-opment has not been the only model tried through history, and, as with much large-scale infrastructure, the develop-ment of SAGD itself was driven by government investments (Alberta Oil Sands Technology and Research Authority).

What may be required is a focused investment and R&D program equivalent to the United States’ Manhattan Project during the 1940s. That project established an inter-national, focused synergy of government, industry, and universities to develop weapons (Kelly 2009). Its postwar, nonmilitary legacy included the U.S. National Laboratories, nuclear power, radioactive isotopes, modern technological industry, and industrial R&D, and also set the pattern for “Big Collaborative Science” (e.g. the Large Hadron Collider, the Square Kilometer Array, and the Human Genome Project). The Manhattan Project demonstrated that it is possible to transform technology and have it deployed at scale within a decade. Surprisingly, the Manhattan Project was cheap, with a cost of US$22 billion (equivalent cost in 2008 dollars) over 5 years and a peak spending of less than 0.5% of the United States’ GDP (Stine 2009), though the environmental legacy and cost is large and complex. In contrast, hundreds of billions of dollars are planned to be invested in oil sands developments in the next 25 years, with over $20 billion (1.3% of Canadian GDP) invested in oil sands development in 2008 alone (Alberta Energy 2014), but with only a tiny fraction of that dedicated to R&D. A coordinated government–industry–academia megaprogram focused on a rapid transition away from traditional fossil fuel use could be most cost effective and would likely cost much less than 2% of GDP (equivalent to the annual retail spending on Christmas in Canada). It could transition technology and economies towards zero emissions, while transforming incumbent industry which, from historical precedent, may otherwise resist change—a techno-economic fi nesse! The oil sands may be the place to start such a transformative program!

CONCLUDING REMARKSThe oil sands and heavy-oil belts of the world represent the most viable access point to the deep biosphere, which from a cell-balance perspective is the largest biome on the planet (Kallmeyer et al. 2012). Although these belts may not be biodiversity hotspots like the Amazon, they do represent a huge, largely untapped (micro)biological resource and a repository of biological novelty (FIG. 3). Mapping the distribution of bacteria, archaea, and viruses in subsurface reservoirs and assessing gene complements and biotech applications are necessary activities. To survive petroleum reservoirs, microorganisms must generally survive high pressures bathed in a complex mix of organic solvents and potentially toxic chemicals, and depending on the partic-ular formation, high temperatures and/or salinity. The organisms present may, therefore, have evolved survival and lifestyle strategies that may prove benefi cial for a variety of applications (pollutant degradation, producing industrial chemicals, etc.) and for biosensor develop-ment with possible nano/bioengineering applications. It is still unclear whether bioactive species with medical applications from deep-biosphere bacterial or viral gene pools are possible (FIG. 3), but our needs for novel antibi-otics have never been greater. Given the great biological novelty and potential for technological, environmental, and economic advances, the oil sands metagenome may be a more valuable and less contentious resource than the heavy oil and bitumen!

Oil sands represent an end-member of a global resource consisting of oil reservoirs containing trillions of barrels of viscous, heavily biodegraded oil which are increasingly being produced as energy needs grow worldwide. While

An expanded, unedited version covering oil sands emissions and alternative futures in more detail is available online at www.ucalgary.ca/prg/publications.

ELEMENTS AUGUST 2014283

REFERENCESAdams JJ, Riediger CL, Fowler MG, Larter

SR (2006) Thermal controls on biodeg-radation around the Peace River tar sands: Paleo-pasteurization to the west. Journal of Geochemical Exploration 89: 1-4

Adams JJ and 5 coauthors (2013) The dynamic interplay of oil mixing, charge timing, and biodegradation in forming the Alberta oil sands: Insights from geologic modeling and biogeochemistry. AAPG Studies in Geology 64: 23-102

Alberta Energy (2014) Oil Sands Facts and Statistics. www.energy.alberta.ca/oilsands/791.asp#Economics, accessed July 2014

An D and 32 coauthors (2013) Metagenomics of hydrocarbon resource environments indicates aerobic taxa and genes to be unexpectedly common. Environmental Science & Technology 47: 10708-10717

Bennett B and 7 coauthors (2013) The controls on the composition of biode-graded oils in the deep subsurface – Part 3. The Impact of microorganism distribution on petroleum geochemical gradients in biodegraded petroleum reservoirs. Organic Geochemistry 56: 94-105

Bergerson JA, Keith DW (2010) The truth about dirty oil: Is CCS the answer? Environmental Science & Technology 44: 6010-6015

Butler RM (1997) Thermal Recovery of Oil and Bitumen. GravDrain Inc., Calgary, Canada, 234 pp

Chu S, Majumbdar A (2012) Opportunities and challenges for a sustainable energy future. Nature 488: 294-303

Clark KA (1931) Process and apparatus for separating and treating bituminous sands. US patent # 1791797

Connan J (1984) Biodegradation of crude oils in reservoirs. In: Brooks J, Dietrich H (eds) Advances in Petroleum Geochemistry, volume 1, pp 298-335

Creaney S and 8 coauthors (1994) Petroleum generation and migration in the Western Canada Sedimentary Basin. In: Mossop GD and Shetsen I (compilers) Geological Atlas of the Western Canada Sedimentary Basin, Calgary. Canadian Society of Petroleum Geologists and Alberta Research Council, pp 455-469, www.ags.gov.ab.ca/publications/wcsb_atlas/atlas.html

Dolfi ng J, Larter SR, Head IM (2008) Thermodynamic constraints on metha-nogenic crude oil biodegradation. ISME Journal 2: 442-452

Fustic M and 5 coauthors (2013) Impact of reservoir heterogeneity and geohis-tory on the variability of bitumen properties in the Athabasca Oil Sands, Canada. AAPG Studies in Geology 64: 163-205

Gates ID, Larter SR (2014) Energy effi ciency and emissions intensity of SAGD. Fuel 115: 706-713

Gieg LM, Davidova IA, Duncan KE, Sufl ita JM (2010) Methanogenesis, sulfate reduction and crude oil biodegradation in hot Alaskan oilfi elds. Environmental Microbiology 12: 3074-3086

Grasby S, Chen Z (2005) Subglacial recharge into the Western Canada Sedimentary Basin—Impact of Pleistocene glaciation on basin hydro-dynamics. Geological Society of America Bulletin 117: 500-514

Grassia GS, McLean KM, Glenat P, Bauld J, Sheehy AJ (1996) A systematic survey for thermophilic fermentative bacteria and archaea in high-temperature petro-leum reservoirs. FEMS Microbiology Ecology 21: 47-58

Head IM, Jones DM, Larter SR (2003) Biological activity in the deep subsur-face and the origin of heavy oil. Nature 426: 344-352

Hein FJ, Leckie D, Larter SR, Suter JR (2013) Heavy oil and bitumen petro-leum systems in Alberta and beyond: The future is nonconventional and the future is now. AAPG Studies in Geology 64: 1-21

Jones DM and 12 coauthors (2008) Crude-oil biodegradation via methano-genesis in subsurface petroleum reser-voirs. Nature 451: 176-180

Kallmeyer J, Pockalny R, Adhikari RR, Smith DC, D’Hondt S (2012) Global distribution of microbial abundance and biomass in subseafl oor sediment. Proceedings of the National Academy of Science 109: 16213-16216

Kelly CC (2009) The Manhattan Project: The Birth of the Atomic Bomb in the Words of Its Creators, Eyewitnesses, and Historians. Black Dog & Leventhal Publishers, 495 pp

Larter SR and 8 coauthors (2003) The controls on the composition of biode-graded oils in the deep subsurface —part 1: Biodegradation rates in petro-leum reservoirs. Organic Geochemistry 34: 601-613

Larter S and 9 coauthors (2006) The controls on the composition of biodegraded oils in the deep subsur-face: Part II—Geological controls on subsurface biodegradation fl uxes and constraints on reservoir-fl uid property prediction. AAPG Bulletin 90: 921-938

Larter S, Adams J, Gates ID, Bennett B, Huang H (2008) The origin, prediction and impact of oil viscosity heteroge-neity on the production characteristics of tar sand and heavy oil reservoirs. Journal of Canadian Petroleum Technology 47: 52-61

Marchetti C (1977) Primary energy substitution models: On the interac-tion between energy and society. Technological Forecasting and Social Change 10: 345-356

Orphan VJ, Boles JR, Goffredi SK, Delong EF (2003) Geochemical infl uence on diversity and microbial processes in high temperature oil reservoirs. Geomicrobiology Journal 20: 295-311

Parkes RJ and 8 coauthors (1994) Deep bacterial biosphere in Pacifi c Ocean sediments. Nature 371: 410-413

Rueter P and 6 coauthors (1994) Anaerobic oxidation of hydrocarbons in crude oil by new types of sulphate-reducing bacteria. Nature 372: 455-458

Schlegel ME, McIntosh JC, Bates BL, Kirk MF, Martini AM (2011) Comparison of fl uid geochemistry and microbiology of multiple organic-rich reservoirs in the Illinois Basin, USA: Evidence for controls on methanogenesis and microbial transport. Geochimica et Cosmochimica Acta 75: 1903-1919

Stine DD (2009) The Manhattan Project, the Apollo Program, and Federal Energy Technology R&D Programs: A Comparative Analysis. Congressional Research Service Report for Congress 7-5700, http://fpc.state.gov/documents/organization/128397.pdf

Tozer R, Choi A, Pietras J, Tanasichuk D (2014) Athabasca oil sands: megatrap restoration and charge timing. AAPG Bulletin 98: 429-447

US NAS (2010) Rising Above the Gathering Storm, Revisited: Rapidly Approaching Category 5. The National Academies Press, Washington, DC, 85 pp

Wilhelms A and 5 coauthors (2001) Biodegradation of oil in uplifted basins prevented by deep-burial sterilization. Nature 411: 1034-1037

emissions related to oil sands use are higher than those of conventional crude oil use, the overall end-use differences are small. Most future oil production from this resource will use energy-ineffi cient steam-based recovery processes, but there are many exciting possibilities for using biological processes for much cleaner energy recovery within and crucially beyond the oil sands. Current politics and market economics mean that it is unlikely new recovery technolo-gies will develop quickly, and the most exciting future for the oil sands may be one in which the resource is a biological one rather than an energy source.

ACKNOWLEDGMENTSWe acknowledge the contributions made by the very many scientists, engineers, and technicians, both academic and industrial, referenced here, who worked in our research groups on petroleum biodegradation during the last 15 years, in both Newcastle and Calgary. We also acknowl-edge material contributions from Canada Research Chairs, Bacchus Consortium, Genome Canada, Genome Alberta, Carbon Management Canada, NSERC (National Science and Engineering Research Council, Canada), the NERC (Natural Environment Research Council, UK), and PRG (Petroleum Reservoir Group). We acknowledge material previously published by AAPG, Nature Publishing Group, and Elsevier.