Unconventional Resources

Huge heavy oil recovery potential being realized in Arctic fields

Permafrost and other unique factors give Alaskan North Slope fields fewer options for heavy oil recovery than other, warmer areas, but substantial progress is being made using WAG and other advanced technologies.

Perry A. Fischer, Editor

Developing heavy oil is always a challenge, but on Alaska’s North Slope, it’s been a 30-year journey that is only now beginning to get answers. Several methods of heavy oil production have been tried, but permafrost, geology, politics and economics make recovery of this 20-to-30 billion bbl resource more difficult than heavy oil recovery in Canada and Venezuela. What follows is a brief description of the region, the politics and unique considerations that have slowed progress, and what’s been done technically to recover these resources, including more recent investigations into Water Alternating Gas (WAG) methods.

BACKGROUND

Much of the data about North Slope heavy oil is either held tight, or just not agreed on. Nevertheless, numerous press reports, technical articles and DOE projects have produced the range of values shown in Table 1.

TABLE 1. Alaskan North Slope heavy oil fields.1,2,3

Estimates of original oil in place across all North Slope heavy oil fields range between 20 and 36 billion bbl, with 23 billion being a recent, oil company number.⁴ This is as much as the OOIP in the deeper, conventional Prudhoe and Kuparuk fields. Recoverable volume is all that really counts, and this is speculative at present, but 4 to 9 billion bbl have been offered recently in oil-company presentations; only half a billion bbl³ of this is in the proved economically recoverable category, and thus, part of the US’ 21 billion bbl of proved reserves. Viscous oil now accounts for more than 5% of all North Slope production.³

Five fields sit above the deeper producing reservoirs of light oil and comprise most of the heavy oil resource: West Sak, Schrader Bluff, Orion, Polaris and the Ugnu sands, the latter of which are very shallow formations lying mostly above Kuparuk and Milne, Fig. 1. The Schrader Bluff is actually an extension of the West Sak formation, but lies in a different producing unit, Fig. 2. Not shown in the table is a small amount of deep, tarry bitumen at Endicott (~0.5 Bbbl),² and some heavy oil in a few small offshore fields. Permafrost exists at the top of some of these heavy oil reservoirs, at 1,200- to 2,000-ft depths. This permafrost creates special considerations, discussed below.

Fig. 1. Five North Slope fields comprise the resource. The shallow Ugnu sands lie above the producing zones at Kuparuk and Polaris and are not shown in map view.

Fig. 2. Regional Schrader Bluff/ West Sak cross section.

A TAXING ENVIRONMENT

Predicting tax changes is at least as uncertain as predicting the success of experimental IOR (Improved Oil Recovery).⁵ The three stakeholder groups involved – the oil companies, US citizens, and the people of Alaska – all have a common interest in maximizing oil production, but they differ on how to split the revenues. Alaska and the oil companies have been arguing for years about whether these fields should be taxed at 12% or 20%, what kind of development incentives, if any, should apply, and whether there should be any “sunset” (ELF) provisions. All of these issues have recently become the subject of a sweeping energy bill in Alaska, while Federal incentives for heavy oil production remain unsettled.

Alaskan politicians are debating a change in tax structure from a gross production tax to a net profit tax, the key difference being that investment is expensed against profits if the new law passes. Also, there’s a proposal to repeal the ELF provision, which limits the length of time that taxes can be collected on some fields. Alaska’s Governor is pushing the changes, while many in the state legislature are skeptical, fearing, in part, accounting shenanigans, or “gaming” of the new tax scheme. The only predictions that can be made are that comprise will occur, and it will effect the North Slope. Whatever happens, any tax benefit for heavy oil investment will help spur development. And a settled tax structure will bring more certainty to future plans.

CHALLENGES

Besides politics, there are challenges unique to the area, one of which is that arctic temperatures and permafrost conditions makes otherwise moderately viscous oil extremely viscous. API gravity isn’t as important as is reservoir temperature. Hence, these deposits are sometimes referred to as viscous oil deposits, rather than heavy oil. Permafrost affects production, as it will cool oil traveling through the permafrost zone. For thermal recovery, the permafrost can melt as a result of heated fluids injected from the surface, or heated oil that is produced.

In addition to increasing viscosity and friction, the part of the well that penetrates the permafrost should be prevented from melting it, due to potential subsidence and the resulting threat to well integrity. Insulated tubing, now in use, helps mitigate the problem, but recent advances in insulation, such as so-called aerogels, as well as air-entrained cements, could achieve the near-adiabatic insulation needed for high-temperature thermal recovery methods.

Even if well integrity can be maintained, steam injection has caused problems with fines production, sometimes plugging the formation and destroying permeability. Whether or not high-temperature steam can be used at all is not certain. Fortunately, there’s no shortage of gas to burn; but even if the steam can be insulated from the wellbore, it may migrate upward to the bottom of the permafrost, or, in any case, the formation may be too thin for either cyclic steam or SAG-D methods. Despite these potential problems, steam applications are an area of active, ongoing research, as is in situ combustion.

Fines production is a seemingly insurmountable problem, so companies have decided to “go with the flow” for now. The West Sak, Schrader Bluff and Ugnu reservoirs are mostly unconsolidated sand. Production efforts in the late 1990s used conventional downhole sand screens. Unfortunately, the sand is not consistent, with some of it being flour-sized. Attempting to filter it means restricted flowrates, which makes the viscosity problem even worse. Forcing the oil through, damages the expensive screens.

Heterogeneity in these North Slope fields is considerable – much more problematic than that of Venezuela and Canadian heavy oil reservoirs. Compartmentalization and variable oil and formation properties add to the challenge of finding an optimum production solution. It appears that it will take more than one or two technologies to allow heavy oil recovery across all the North Slope fields. Rather, it will take a collection of solutions, each tailor-made for a particular reservoir.

Finally, if you do get the heavy oil to surface, it will sell at a considerable discount to light oil. One bright spot: These oils do not have the sulfur and metals content that plague other heavy oil deposits.

Success is measured in incremental benefits resulting from laboratory work, field trials and thorough reservoir knowledge. A 20% improvement in recovery may sound like a lot – and it is – but it could mean the difference between recovering 10% of OOIP, and 12%, which means that there’s still 88% remaining. To find the best solutions, The North Slope Viscous Team was formed, which includes technical staff from ConocoPhillips and BP, with some ExxonMobil participation.

Over several decades, there’s been a huge effort expended to develop technology that can overcome these challenges. ConocoPhillips says it has spent over $500 million in technology development at West Sak alone.⁴ BP says the total investment across all the heavy oil fields has been closer to $1 billion.³

SOULTIONS

Drilling. The drilling side has provided two major improvements in production and economic performance. Using oil base mud aids both well construction and production (see World Oil, Oct. 2004). Drilling extended reach, multilateral and tri-lateral wells (Fig. 3) has made a big difference in production rates, up 10 times that of vertical well rates, up to 2,000 – 3,000 bopd.

Fig. 3. Multilateral (trilateral and even quad-lateral) wells have had a dramatic effect on production. Sometimes, the bottommost lateral is undulated such that it passes in and out of the formation three or more times, allowing selective isolation of parts of the producing zone.

Let the sand flow. Two years ago, an in-depth study concluded that the best economic route, despite the erosion risk and clogging of production facilities, was to build “sand tolerant” facilities and produce the sand along with the oil. Large slot (40/ft, 2.25 in. X 0.125 in.) liners are now deployed, without any screens. A sand management plan is put into action, including retraining of field personnel. Initial sand production rates tend to eventually taper off and stabilize, but velocities must be closely monitored so that erosional limits are not exceeded. Sand production at West Sak was 3 lbs/bbl oil on average; that is 0.38 % by volume.⁶

By choosing this practice, additional costs are incurred. These include drilling a sand disposal well and associated grinding facility, as well as the cost to recover, haul, process and inject the sand into the disposal well. Such costs are running $150/yd³. Even so, the capital outlay of downhole screens alone are $1 million per lateral, given the length of typical West Sak laterals (6,000 to 7,500 ft), which multiply to $3 million per well for tri-laterals. Further study indicated that oriented perforating did not improve production, nor did fracturing for sand control. Savings from capital outlays, together with improved flowrates, overwhelmed the increase in sand management and disposal costs by more than $50 million.⁶

Lifting costs and methods are an important component. ESPs are generally the preferred way to produce the heavy oil to surface, at least in terms of efficiency, but this also poses a logistical and financial challenge for workover rig scheduling in a remote, harsh climate. Failures can mean a lengthy shut-in time. A partial solution is to use gas-lift, built in as a backup. Although not as good as ESPs, it keeps production online.

Another solution is to use jet pumps. Again, not as good as ESP’s pumping efficiency, but not having to pull the downhole tubing means low maintenance costs and quicker repairs, which, in the long run, may be more important.

Low salinity waterflooding. It’s been known for many years that low-salinity water (<5,000 ppm) improves oil recovery over normal-saline-level waterfloods.⁷ Although the exact mechanism by which this occurs is an unsettled issue, laboratory investigations were confirmed in the field, when BP made four sets of chemical tracer tests at Milne Point that indicated 15% increased recovery. This, together with other experimental work in the North Sea, shows that recovery can be improved by 10 to 40%. BP is understandably very encouraged by these results, and believes that low-salinity waterflooding may have wide-ranging application. More trials are planned for Endicott field in Alaska in 2007.

Gas and WAG flooding. With waterflood, the lightest of these heavy oils can achieve a recovery rate of up to 18%, but even then, over 80% of the oil remains in the formation. Also, the heavier oils cannot benefit as much from waterflood alone. In general, miscible gas flooding is a proven, cost-effective IOR technique in conventional reservoirs. Further research is being directed at improving the recovery efficiency of miscible processes for viscous oils. Such efficiency depends on pore-fluid displacement and sweep efficiency. The application of miscible and immiscible gas flooding needs to be extended to medium-viscosity reservoirs.

Miscible gas injection is used for IOR in conventional reservoirs, but these viscous oils are usually too thick to be miscible. However, non-miscible gas flooding can also enhance production, as can Water Alternating Gas (WAG) with non-miscible gas, and even partial miscibility can lower viscosity considerably.

A common problem in gas flooding is the inability to get the gas front far out into the formation in a uniform manner. The solvent “fingers” through the formation, or otherwise short-circuits a process that needs maximum contact with the oil in order to work well. Improving this flooding process is recognized as the key area for research that will allow recovery of the world’s difficult-to-produce oil, including heavy oil.

The WAG process is being scrutinized for improvements. Since most gases, even when supercritical, have very low viscosities, adding thickeners, such as LNGs, can improve sweep. Dynamically controlling the profile downhole in wells can also help. Variations of WAG include using lean natural gas, raw natural gas and natural gas-CO₂ and LNG-CO₂ mixtures. For example, experiments found that 85% CO₂ mixed with 15% NGL, as well as 60% Prudhoe Bay gas mixed with 40% NGL, developed miscibility with Schrader Bluff oil.⁸ Most of this research is laboratory based, but some of it is now moving into the field.

Lean gas flooding is in pilot testing, with some reports showing a viscosity drop to 10 cp from 60 cp using the method.⁹ In some areas, this type of IOR can increase recovery by 20% over waterflood, improving total recovery to about 22% from 18%.⁹

The latest, DOE-sponsored research has been focusing on this problem, including WAG to improve sweep efficiency for oils in the medium-viscosity (30 – 300 cp) range. For the past few years, the University of Houston, under a DOE grant, has been conducting research into improving the WAG process. A high-pressure quarter five-spot model was built, where multi-contact WAG floods could be conducted and evaluated at approximated reservoir conditions. Dead, 78-cp reservoir oil was used with a pure ethane solvent. A 20-ft-long slimtube was also built to conduct experiments.

Details of this work will be presented at the upcoming SPE ATCE conference in San Antonio.¹⁰ Among the findings were that gasflood, followed by waterflood, improves oil recovery (~0.55 PV) over just waterflood (~0.48 PV) in the quarter 5-spot apparatus. WAG injection slows down gas breakthrough. Compared to continuous gas injection followed by waterflood, which has a pore volume of about 0.55 PV, WAG improves oil recovery to about 0.68 PV in the quarter 5-spot.

As pressure decreases, gasflood oil recovery increases in the pressure range of 660 to 1,380 psi for the dead, viscous oil used. The minimum miscible pressure for ethane with the reservoir oil is between 630 and 1,000 psi.

Decreasing the solvent amount lowers the oil recovery in WAG floods, but, compared to just waterflood (~0.48 PV), a significantly higher amount of oil (~55 PV) can still be recovered with just 0.1 PV solvent injection.

Using a horizontal production well slightly lowers oil recovery compared to a vertical production well during WAG injection. Gasflood at 1,380 psi, followed by waterflood sweep, has an efficiency similar to that of the waterflood in a 5-spot model. Alternate injection of solvent and water at a WAG ratio of 1 increases sweep efficiency slightly. Of course, it’s understood that laboratory work and computer simulations are no substitute for field tests, which often produce lower-than-expected recovery, mostly due to reservoir heterogeneity.

On the North Slope, an immiscible WAG-flooding process, called Viscosity Reduction WAG (VR-WAG), has been proposed for medium viscosity oils.¹¹ Many of these oils are depleted in the light hydrocarbons C7 – C13. When lean gas is mixed with heavier components such as NGL, and the resulting Viscosity Reducing Injectant (VRI) mixture is pumped into the formation, the C2 and higher components condense into the oil, decrease its viscosity up to 90%, making it easier for water to displace oil. Reservoir simulations find that this process should enhance oil recovery by 16% over that of just waterflood (22% vs. 19% of OOIP).

This technology seems suitable for widespread application in Alaska’s heavy oils. The first VR-WAG was applied in the Kuparuk reservoir of Milne Point Unit to replace a miscible gas IOR project, which used purchased NGL.¹¹ In that case, the reservoir was severely undersaturated, and the process was called US-WAG. In the case of nearly saturated reservoirs, such as West Sak 1J, enriched gas can be used in the WAG process, or the injectant can be made by blending rich and lean gases.

The first enriched gas application will be the West Sak 1J IOR Project at Kuparuk, sanctioned in mid-2004, while ongoing studies eye Orion, Polaris and the S-pad IOR Project at Milne Point for feasibility of the technique. Project construction is underway, and will comprise 14 vertical injection and 17 production wells at Site 1J. These producers will be the tri-lateral wells with 6,000-ft-lateral slotted liners shown in Fig. 3. Thirteen more West Sak wells will be drilled at the existing Site 1E. The $500 million project aims to increase West Sak heavy oil production, currently at 10,000 bopd, to 45,000 bopd by 2007.^3,12

CONCLUSIONS

Alaskan heavy oil is being produced in limited volumes. The harsh climate, permafrost and reservoir properties make thermal recovery of these oils more difficult than elsewhere. The need to keep the Trans Alaska Pipeline viable is an added impetus for all parties, but tax schemes need to be stabilized and allow for, if not promote, considerable investment in heavy oil recovery.

A number of IOR technologies for North Slope heavy oil production are reported in the literature, including multilateral drilling, various gases, reduced salinity waterflood and WAG. Thermal production of heavy oil is uniquely difficult, but investigations into its use continue.

The size of the resource and its recoverability, especially economic recoverability, are not agreed-upon numbers. Economic recoverability of this oil is highly sensitive to markets and technology costs, but a minimum of 4 Bbbl should be technically recoverable. Given the level of financial and rhetorical commitment that the North Slope operators have expressed, production should be capable of ramping up above the 100,000 bopd mark in the next few years by employing the newest technologies available for the moderately heavy oils. However, the heaviest of these oils, such as those in the shallow Ugnu, will have to wait until better technologies are developed before they can be recovered. 

LITERATURE CITED AND ANNOTATIONS

¹ Alaska Oil and Statistics Gas Conservation Commission, Annual Reports, Pool Statistics, Updated July 25, 2005.
² Olsen, D. K. and E. C. Taylor, S. M. Mahmood, “Feasibility Study of Heavy Oil Recovery-Production, Marketing, Transportation, and Refining Constraints To Increasing Heavy oil Production In Alaska,” paper NIPER-610, Projects SGP37 and BE11B, IIT Research Institute, National Institute For Petroleum and Energy Research, Bartlesville, Oklahoma, December 1992.
³ BP website: Field Facts: http://alaska.bp.com/docs/1213%20BP%20AKPG.pdf
⁴ Fox, M., PowerPoint presentation by ConocoPhillips to Alaska State Chamber of Commerce http://www.alaskachamber.com/artman/publish/pdfs/OilGas.pdf
⁵  The term IOR is used here, because it is more appropriate then EOR, since, in many cases, recovery is coming from Primary, Secondary and Tertiary recovery simultaneously.
⁶ Burton, R. C., et al., “North Slope heavy-oil sand-control strategy: Detailed case study of sand production predictions and field measurements for Alaskan heavy-oil multilateral field developments,” paper SPE 97279, presented at the SPE Annual Technical Conference and Exhibition, Dallas, Texas, October 9 – 12, 2005.
⁷ McGuire, P. L., et al., “Low salinity oil recovery: An exciting new EOR opportunity for Alaska’s North Slope,” Paper 93903, presented at the SPE Western regional Regional Meeting, March 30 – April 1, 2005.
⁸ Khataniar, S. and V. A. Kamath, S.L. Patil, S. Chandra, M. S. Inaganti, “CO2 and miscible gas injection for enhanced recovery of Schrader Bluff heavy oil,” Paper 54085-MS, presented at the International Thermal Operations/ Heavy Oil Symposium, March 17 – 19, Bakersfield, California, 1999.
⁹ Petroleum News, “Technology driving West Sak development,” Interview with Matt Fox, Jan. 9, 2005.
¹⁰ Lewis, E. and E. K. Dao, K. K. Mohanty, “Sweep Efficiency of Miscible Floods in a High-Pressure Quarter Five-Spot Model,” Paper SPE 102764, presented at the SPE Annual Technical Conference and Exhibition, San Antonio, Texas, September 24 – 27, 2006.
¹¹ McGuire, P. L. and R. S. Redman, B. S. Jhaveri, K. E. Yancey, S. X. Ning,, “Viscosity reduction WAG: An effective IOR process for North Slope viscous oils,” SPE 93914, SPE Western Regional meeting, March 30 – April 1, 2005.
¹² Findings and Decision of the Director of the Division of Oil and Gas, “Approval of the application to revise the West Sak participating area of the Kuparuk River Unit,” Department of Natural Resources, December 15, 2004.