Modernizing Vitrinite Reflectance Models for Paleothermal History Calibration

Easy%R₀ has been perhaps the most widely used model over the past 30 years to help calibrate paleothermal histories. It was published in the AAPG Bulletin in 1990 by Jerry Sweeney and me, and it is one of the most highly cited articles in the Bulletin in the last 40 years. As the person who constructed the Easy%R₀ algorithm, I would like to share some of the drivers for how the parameters were chosen and why it is appropriate now to move on to more recent models that appear to be more universally correct.

The Evolution of Vitrinite Reflectance Models

My story begins with Lopatin’s Time-Temperature Index, which was popularized by Doug Waples in a paper published the AAPG Bulletin in 1980. It posits that changes in vitrinite are driven by chemical reactions that double in rate every 10 degrees Celsius. While that rule of thumb may be good for estimating the time to bake a pizza, it is not very good for chemical reactions involving strong chemical bonds.

Image Caption

Figure 1. A comparison of vitrinite reflectance calculated as a function of depth for Easy%R₀ (kinetics) compared to TTI, as first published by Burnham and Sweeney in the 1991 AAPG Special Volume, “Source and Migration Processes and Evaluation Techniques,” which was edited by current AAPG Bulletin Editor Robert Merrill.

Please log in to read the full article

Easy%R₀ has been perhaps the most widely used model over the past 30 years to help calibrate paleothermal histories. It was published in the AAPG Bulletin in 1990 by Jerry Sweeney and me, and it is one of the most highly cited articles in the Bulletin in the last 40 years. As the person who constructed the Easy%R₀ algorithm, I would like to share some of the drivers for how the parameters were chosen and why it is appropriate now to move on to more recent models that appear to be more universally correct.

The Evolution of Vitrinite Reflectance Models

My story begins with Lopatin’s Time-Temperature Index, which was popularized by Doug Waples in a paper published the AAPG Bulletin in 1980. It posits that changes in vitrinite are driven by chemical reactions that double in rate every 10 degrees Celsius. While that rule of thumb may be good for estimating the time to bake a pizza, it is not very good for chemical reactions involving strong chemical bonds.

At the other extreme, Charles Barker and Mark Pawlewicz from the U.S. Geological Survey proposed that geological time is essentially irrelevant to vitrinite reflectance because it equilibrates at the maximum temperature within a few years. A figure from Easy%R₀’s original publication shows that it has less dependence on time than other contemporary models, but not to the extreme of Barker and Pawlewicz.

The relationship between reflectance and time depends on a parameter called the frequency factor, A, which in simple terms is approximately the number of times a molecular bond vibrates per unit time. At the time I derived Vitrimat, which is the more complete model that Easy%R₀ mimics, I estimated A to be 1´1013s-1 based on the best kerogen pyrolysis kinetics then available. The activation energies were next adjusted to give the right answer for geological conditions, and they also gave approximately the right answer for high-pressure laboratory pyrolysis. However, the laboratory results were different for coals and oil-prone shales for reasons not understood at the time, and Easy%R₀ was closer to the oil-prone shale values. Also, the optimization process did not place sufficient emphasis on reflectance from high-rank terrigenous organic matter.

We now fast forward to 2015, where Nielsen and workers renewed a criticism that Easy%R₀ did not follow reflectance trends at high maturities, and more specifically, that the reflectance trend should have a sharper break (dogleg, DL) near the end of oil generation. Catalyzed by a discussion with Doug Waples, we concluded that the median A for kerogen pyrolysis was actually closer to 2´1014s-1, so I chose that value to create a new version, Easy%R₀DL. Activation energies for this version were optimized in collaboration with Ken Peters and Oliver Schenk using vitrinite reflectance data from the Alaska North Slope, where paleo heat flows were thought to be well known. Easy%R₀DL also had a sharper dogleg versus depth than Easy%R₀ but not as sharp as proposed by Nielsen and coworkers.

Part of the reason for the weaker calculated dogleg versus depth was that we thought Nielsen’s model had been influenced too much by vitrinite suppression in oil-prone shales. Vitrinite suppression had been known for a very long time, but arguments about its cause had not been resolved. In 2018, Ken Peters published a paper in Organic Geochemistry with Paul Hackley of the USGS and J.J. Thomas and Drew Pomerantz of Schlumberger that showed definitively that oily products from oil-prone kerogen can react with and suppress the reflectance of true coaly vitrinite, which indicates that vitrinite from oil-prone shales should not be used with vitrinite reflectance models.

During that time frame, I reexamined the much-larger literature on vitrinite reflectance from high-pressure pyrolysis experiments. Not only did it confirm the fundamental difference in reflectance between coals and oil-prone shales, but it also indicated that using a frequency factor of 1´1015s-1 enabled a new version, Easy%R₀V to match both geological maturation and laboratory pyrolysis results. This work was also published in Organic Geochemistry in 2018. It is compared to other models in figure 2.

Time to Move On

The full story is more complex, in that Easy%R₀V is based on an updated version of Vitrimat, which was also the basis for the original Easy %R₀. That work was motivated in large part to have a better model for bitumen reflectance, and the corresponding model for that material is Easy%R₀B. All these models are based on the original concept dating back to the 1950s coal literature, which related reflectance to the H/C ratio of the vitrinite. That relationship is embedded in Vitrimat 2018, which in principle, can be tailored to model the reflectance of any sedimentary organic matter.

As understanding progresses, the best models change. Just as the internal combustion engine replaced horses and electric batteries and motors are replacing the internal combustion engine, it is time to move on from Easy%R₀ to either Easy%R₀DL or Easy%R₀V, which have only minor differences in the sharpness of the dogleg near the end of oil generation.

You may also be interested in ...

Popular articles