Where is the most likely place in any sedimentary basin to find oil and gas? The source rock. Once the rock has been identified, some might assume the process shifts into manufacturing mode, with hydrocarbon extraction the only remaining complexity. But as sports broadcaster Lee Corso might say, “Not so fast, my friend.”

So, how can the feasibility of present-day commercial petroleum development from an unconventional system be determined? First, one must thoroughly understand the geology, assessing the source rock and its depositional history, stratigraphy, lithofacies and more, as well as the overlying and underlying sedimentary column. In terms of the petroleum system, is retained petroleum alone accessible, or is there expelled petroleum nearby in nonshale lithofacies? Is the right geological framework present to tap into the unconventional system, which, by definition, requires stimulation?

The maturation level of the source rock also must be determined: Is it in pregeneration (immature) phases, oil/gas generation phases, or post-mature? This, along with kerogen type, affects whether oil or gas will be found, as well as the quality of the oil. Different kerogen types generate different quality oils at the same level of thermal maturity. Some high-API-gravity oils may not flow readily due to higher wax contents, for example. This is more typical of lacustrine shales than marine shales.

With the industry’s current shift toward the higher economic payback of oil production, having an idea of the source rock’s oil saturation or the associated lithofacies within the source rock is invaluable in understanding the likelihood of oil flow in these typically ultra-tight rocks.

Hybrid system

The starting point is to ascertain the primary product objective of the exploration search—oil or gas—and then to evaluate the petroleum system in the context of an unconventional system. A conventional discovery contains expelled hydrocarbons only. An unconventional discovery has the potential for both retained and expelled petroleum, or a “hybrid system.” Thus, a key difference between the Barnett and the Haynes­ville shales is the former produces only retained gas, while the latter produces both retained and expelled gas.

Also, the porosity of the two systems is quite different (e.g., 5% on average for the Barnett shale, and about 9% on average in the Haynes­ville shale), as is the carbonate content. The Barnett shale’s storage capacity has been described as largely derived from organic porosity, i.e., porosity created by the conversion of organic matter. The Haynesville shale has organic porosity, but in addition it often has a much higher level of matrix porosity largely associated with carbonate.

Considering oil- or condensate-producing systems such as the Eagle Ford, Bakken, Niobrara—and soon, many others—the close association of carbonate or other nonshale lithofacies is critical, because this is primarily recovery of the expelled rather than the retained oil. And carbonates provide something that quartz content does not—a relatively non-sticky matrix relative to oil.

Sticky issues

Quartz content is crucial in stimulating tight mudstones such as the Barnett shale, because it is a component of the rock fabric’s brittleness. But quartz in these mudstones is also closely associated with organic matter. Petroleum in this system will stick to the carbon, causing immediate or eventual petroleum-flow issues, particularly when black oil is present.

Black oil is a less mature oil and contains both polar (relatively non-sticky) and polar (very sticky) molecules. It is really the polarity or stickiness of the molecules that causes occlusion of the tight pore throats in a mudstone and quartz-rich mudstone system.

Certainly, size and drainage rate can also cause issues. Like a funnel filled with sugar or sand, the system will drain thoroughly if infill/outflow rates are optimized. Too much flow will block the funnel, and, if the walls of the funnel are wet, i.e., sticky, the sugar or sand will begin to accumulate and block drainage, virtually regardless of drainage rate. This implies both flow rate and oil composition can affect initial and ultimate flow rates and recoveries. Controlling flow rates may help to optimize recovery, but its role is not as dramatic as that of oil quantity and its composition.

On the other hand, carbonates, although present in small amounts throughout many shales, are often separate beds, or interfingered lenses or beds having storage capacity. This potential storage capacity is non-sticky, because the interfingered lenses are not organic-rich and so avoid adsorption issues. The most obvious example of an oil-producing system with such features is the Bakken formation, or the ongoing development of the Austin Chalk (Eagle Ford source rock).

In this context, during late diagenesis, organic matter releases relatively high amounts of water, carbon dioxide and organic acids. While the amounts of water and carbon dioxide released decrease in the hydrocarbon-generation phase, they continue to be released, albeit in lesser quantities. Thus, saturation of any pore volume quickly occurs until a migration pathway is created by the excess pressure from either the nonhydrocarbon or the hydrocarbon generation.

The generated carbon dioxide and acids in water have the ability to dissolve carbonate, and it becomes a matter of volumetrics to understand if they can create both migration conduits and potentially secondary porosity in associated carbonates. If so, the source rock plays an important role not only in generation, but also, potentially, in the creation of increased pressure associated with compaction and generation, organic porosity, migration pathways, and secondary porosity in carbonates.

Early techniques in geochemistry used measurement of petroleum yields via solvent extraction. This approach is still used today, particularly for characterizing petroleum either for quality or for source characteristics derived from fingerprinting, biomarker, and isotopic data. With the advent of the Rock-Eval instrument, however, an approximation for extract yields could be obtained. Rock-Eval free-oil yields (S1) are similar but not identical to extract yields, and can be used to identify high oil saturations in shales or nonshales.

As a final point, obtaining maximum gas flow from a shale cannot be achieved when there is oil in the system. The reason? Not all the oil has been converted to gas, thereby limiting the volume of gas generated, the pressure in the system, and the presence of polar-black-oil components that are strongly adsorptive.

This does not mean that oil cannot be produced from shales, however. Oil can be produced from shales (and has been for more than a century) and gas will almost always be associated with that oil, because both oil and gas are generated in the oil window. In fact, it is much more difficult to obtain high oil-flow rates, depending on the shale-oil-system type: either mudstone-dominated systems, or the less restrictive non-mudstone-dominated systems, particularly systems containing high amounts of carbonate.

The better economics of oil versus gas during recent decades have given shale oil and any liquids co-produced with gas a premium. In fact, the process becomes a matter of identifying not where the shale oil is (it’s everywhere), but rather where one can locate the highest oil saturations (which change during the oil window due to episodic expulsion), thereby achieving the best flow rates and better estimated ultimate recoveries.

Obviously, any oil and gas developer would prefer to be in a manufacturing mode for producing hydrocarbons from reservoirs, whether conventional or unconventional. But, for the sake of minimizing exploration and development risk, a petroleum systems approach is essential in unconventional reservoirs. And, the ultimate end-game can pay out big dividends.

Daniel M. Jarvie is president of Worldwide Geochemistry LLC, based in Houston. Contact him at danjarvie@wwgeochem.com.