OIL Shale BASICS
Many so-called shales are really silts, shaly silts, or laminated shaly sands or silts, such as the Green River Shale of Western USA (shown at the right). In this image, the light coloured laminations are volcanic tuff, the dark colours are kerogen bearing silt.
Some oil shales contain significant amounts of calcite, dolomite, and siderite, making log analysis difficult due to the varying matrix density of the rock. Many are laminated, adding to the log analysis problem (see Laminated Reservoirs for details).
Some clean silts and sands are called oil "shales" because they are radioactive and look like shale on logs, such as the Bakken "Shale" in Saskatchewan and North Dakota. The Upper and Lower Bakken are real organic-rich shales, but the oil producing Middle Bakken is a dolomitic quartz sand with little clay. These are analyzed with a standard shale corrected complex lithology porosity model coupled with a shale corrected Archie-type water saturation model.
There are two basic classes of "oil shale". One contains immature kerogen with no liquid oil. Oil is obtained by heating the rock in-situ or at the surface to "mature" the kerogen. The second type is more like a shale gas play; the kerogen is mature enough that there is oil in free porosity, but not so mature that it becomes a gas play. There are hybrid combinations, particularly popular are liquids rich shale gas plays. Petrophysical analysis of mature oil shale plays are done exactly the same way as gas shales and tight gas or tight oil plays, depending on the level of organic matter and the hydrocarbon phase.
The distinguishing characteristic of an immature "oil shale" is that it contains significant organic carbon but no free oil or gas. This hydrocarbon is immature, not yet transformed into oil by natural processes, and are usually termed "source rocks". Some adsorbed and some free gas may also exist. Immature oil shales require a specialized log analysis model because the Archie saturation model is often inappropriate.
Immature oil shale can be mined on the surface or
at depth and the rock heated in a retort to convert the organic
content to oil. Some valuable by products such as vanadium may
also be extracted, but dry clay, ash, and other minerals are a
serious waste disposal issue. In-situ extraction using
super-heated steam, air, carbon dioxide, or some other heat
transfer system is used to convert the organic carbon to oil.
Collector wells then extract the oil.
CLASSIFYING OIL Shale
A useful classification of oil shales was developed by A.C. Hutton. He divided oil shale into three groups based on their deposition environment: terrestrial, lacustrine, and marine, and further by the origin of their organic matter.
Terrestrial oil shales include those composed of lipid-rich organic matter such as resin spores, waxy cuticles, and corky tissue of roots and stems of vascular terrestrial plants commonly found in coal-forming swamps and bogs. Lacustrine oil shales include organic matter derived from algae that lived in fresh, brackish, or saline lakes. Marine oil shales are composed of organic matter derived from marine algae unicellular organisms, and marine dinoflagellates.
1. Cannel coal is brown to black oil shale composed of resins, spores, waxes, and cutinaceous and corky materials derived from terrestrial vascular plants together with varied amounts of vitrinite and inertinite. Cannel coals originate in oxygen-deficient ponds or shallow lakes in peat-forming swamps and bogs.
Resistivity image log in lacustrine oil shale. White is high resistivity, black is low resistivity.
4. Torbanite, named after Torbane Hill in Scotland, is a black oil shale whose organic matter is composed mainly of telalginite found in fresh- to brackish-water lakes. The deposits are commonly small, but can be extremely high grade.
5. Tasmanite, named from oil-shale deposits in Tasmania, is a brown to black oil shale. The organic matter consists of telalginite derived chiefly from unicellular algae of marine origin and lesser amounts of vitrinite, lamalginite, and inertinite.
6. Kukersite, which takes its name from Kukruse Manor near the town of Kohtla-Järve, Estonia, is a light brown marine oil shale. Its principal organic component is telalginite derived from green algae. Kukersdite is the main type of oil shale in Estonia and westtern Russiaa, and is burned instead of coal to generate electricity in power plants.
OIL Shale IN CANADA
Canada produced some shale oil from deposits in New Brunswick in the mid-1800's. The mineral was called Albertite and was originally believed to be a form of coal.
Albert Mines, New Brunswick, in 1850's
Later, the nature of the mineral and its relation to the surrounding oil shale was described correctly. Abraham Gesner used Albertite in his early experiments to distill liquid fuel from coal and solid bitumen. He is credited with the invention of kerosene in 1846, and built a significant commercial distillery to provide lighting oil to replace whale oil in eastern Canada and USA. In the 1880's, shale oil was abandoned as a source of kerosene in favour of distillation from liquid petroleum.
Canada's oil-shale deposits range from
Ordovician to Cretaceous age and include deposits of lacustrine
and marine origin in at least 20 locations across the country.
During the 1980s, a number of the deposits were explored by core
drilling. The oil shales of the New Brunswick Albert Formation,
lamosites of Mississippian age, have the greatest potential for
development. The Albert oil shale averages 100 l/t of shale oil
and has potential for recovery of oil and may also be used for
co-combustion with coal for electric power generation.
Oil shales contain predominantly Type I kerogen, as opposed to coal and coal bed methane reservoirs, which contain mostly Type III. Gas shales contain mainly Type II kerogen.
Various methods for quantifying organic content from well logs have been published. The most useful approaches are based on density vs resistivity and sonic vs resistivity crossplots. Other approaches using core measured TOC versus log data, for example density or sonic readings are also common. See TOC Calculation for details.
The heating value is useful for determining the quality of an
oil shale that is burned directly in a power plant to produce
electricity. Although the heating value of a given oil shale is
a useful and fundamental property of the rock, it does not
provide information on the amounts of shale oil or combustible
gas that would be yielded by retorting (destructive
Determining OIL YIELD (Grade) of
Oil Shale FROM WELL LOGS
By crossplotting Fischer assay oil yields with corresponding log data, regression lines are generated that provide a decent average oil yield from logs. Problems related to matrix density or matrix travel time variations due to mineral variations with depth are masked by this method. Separate transforms are usually taken when mineralogy is known to change. Logs average about 3 feet (1 meter) of rock compared to much finer detail available from the core assay, so crossplots tend to show considerable scatter in laminated intervals, as shown in the examples below..
Sonic log data versus oil yield from Utah example. Reduced major axis best fit is the most appropriate regression method (red line). Y-on-X and X-on-Y regression lines are also shown. Sonic is in usec/foot, oil yield is in US gallons/ton (gpt or g/t)of rock.
Equation of the line is:
Density versus oil yield for same data set. Density is in grams/cc
Equation of the line is:
Data is from "Basin-Wide Evaluation of Uppermost Green River Oil Shale Resources, Uinta Basin, Utah and Colorado" by M. D. Vanden Berg, Utah Geol Survey, 2008.
Equations for each individual well were also presented, showing considerable variation from well to well and zone to zone.
literature search quoted by R. M. Habiger and R. H. Robinson in
1985 gives the following equations for estimating oil yield:
I have reduced all equations to 3 significant digits, which is all that log analysis can support. The reader should refer to the appropriate technical papers to see the data spread and regional environment before using any of the above equations.
NUMERICAL EXAMPLE DENS 2.2 1.8 g/cc DTC 100 130 usec/ft Smith 3: 26.7 58.6 US gal/ton 4: 23.4 53.6
Bardsley and Algermissen 5: 24.9 51.5 6: 24.3 52.6
Tixier and Alger 7: 24.3 48.1 Cleveland-Cliffs 8: 24.0 63.6 9: - 23.3 61.0
They were also faced with very poor quality density log data
from poorly calibrated slim hole, non-contact tools. As a
result, they had to normalize the density logs using histograms
and correlated density "variation" (DV) to oil yield instead of
raw density. DV was calculated from:
This also had the effect of handling some of the matrix density variations between wells, but not from layer to layer within each interval in a single well.
A clay index was generated by regression:
This equation set is inverted by Cramer's Rule or with spreadsheet functions to obtain the unknown volumes. Parameters must be adjusted to suit local conditions. Minerals chosen must be guided by local knowledge, based on petrography or XRD results. If a log curve is unavailable or faulty due to bad hole conditions, the data can be synthesized or the equation set reduced to eliminate that curve, with the loss of one of the minerals in the answer set.
The volumetric results must then be converted
to mass fraction, as is done for tar sands, potash, and coal
Density parameters must match those used in the original simultaneous equation set.
Kerogen mass fraction should be close to Oil Yield mass fraction from Fischer
analysis, or a simple linear conversion to account for "gas plus
loss". If Fischer analysis is given in US gal / ton or liters /
ton, suitable conversion factors must be used to obtain mass
fraction (ton / ton) for comparison to the log analysis results.
I have had no chance to test simultaneous or PCA approach on oil shale, but have used it successfully in potash and conventional multi-mineral oil reservoirs.
SPREADSHEET -- OIL SHALE ASSAY FROM LOG ANALYSIS
Page Views ---- Since 01 Jan 2015
Copyright 1978 - 2018 E. R. Crain, P.Eng. All Rights Reserved