Previous Lecture Complete and continue  

  1. Introduction

Lecture content locked

Enroll in Course to Unlock
If you're already enrolled, you'll need to login.


- [Ravinath] So, today's talk is gonna mainly focus on relaxation measurements in unconventionals and, by that, the most common and well-known relaxation measurement is that of T2. So, T2 is a simple relaxation distribution that we have been very accustomed to seeing in conventional plays but, when it comes to unconventionals, it brings with it additional complexity. In other words, a simple T2 measurement and interpretation in unconventionals is much more complex and difficult to do in comparison to what we're used to in conventionals. So, there's no simple answer product, as in there is one single cutoff number and you talk about free fluid or bound fluid. Things get a little bit more messy. And why is that? The reason for that is unconventionals are, by nature, made up of much more complex constituents. You have these new constituents which you do not normally come across in conventionals, by which I mean the kerogen and the bitumen. And other than these two, of course, you have the light hydrocarbon, which you want to look at, which can be bound or which can be mobile or free. And this light hydrocarbon, let's say, is about 30 API or more is exactly the kind of hydrocarbon that you want to produce. But the new complexity now that you have, is that's not the only hydrocarbon in your system. The kerogen is a hydrocarbon. It is a solid hydrocarbon. It forms the matrix of the rock so it has porosity in it and it is solid so it doesn't flow. So, that's also hydrocarbon but definitely not something that's producible. Then you have the bitumen. The bitumen is defined. Both kerogen and bitumen are defined operational. Bitumen is the fraction of the total organic carbon which is soluble in organic solvents. It's a very, very heavy oil, maybe half a million centipoise or so. Maybe a little less, little more. And bitumen is one fraction of this hydrocarbon which is also not producible. So the kerogen and bitumen, one a solid, the other is a very viscous heavy oil which is dissolvable in organic solvent and neither of which are easily, or in any way, producible. And then, of course, you have two fractions of the hydrocarbon which you are interested in, which is the light oil, some of which might be bound to the kerogen and not be part of the ineffective porosity, or in other words, not formed part of channels, which lead to production so it might still stay behind. And the fraction of it which is mobile. And, other than this, of course you have the free water and the bound water which is associated with the clays and the free water which can flow out. How to these look on a T2 distribution? On the T2 distribution, and I take the example of a simple Eagle Ford Shale, this is kind of true universally for the different shales that you find, not just in U.S. land but maybe even worldwide. You have the kerogen which you mainly do not see, because it's a solid. And then, of course, you have the bitumen and the heavy oil which overlaps with the bound water so it's not two different T2's, they overlap with each other. Similar to this, you can have the bound light oil and the mobile light oil, overlapping, and then of course, you have the free water. So the main message over here is that the T2 distribution is not simply about drawing some kind of cutoff and identifying fluids because things are more complex. Things overlap with each other so, in other words, they are on top of each other on the T2 distribution. So, you need additional information. You need more complex or richer experiments with other dimensions to identify these different fractions. And only if you can identify these different fractions can you really talk about production or sweet spot or whatnot, because you're not just interested in total organic carbon anymore, you're interested in the fraction of the organic carbon which is the mobile light oil, which is what your production in play's all about. So, how do you go about doing that? That's what we gonna focus in this talk. Now, firstly before embarking on this which are a little bit more complex, I want to place some kind of basic ground rules. This is a plot. This is the NMR echo amplitude, or nothing but the NMR signal vs. time. And the first thing I want to state over here is that the kerogen, which is a solid, has almost no signal. Or, in other words, an NMR measurement made at two megahertz, which is similar to what you have in the lab, will not be able to see the kerogen signal because it's a solid. The first point over here corresponds to a time of 200 microseconds, which is one of the smallest echo times you can measure downhole. And it still will not catch the kerogen signal because kerogen, being a solid, dies very fast, so it doesn't catch it. On the other hand, the bitumen signal and these two correspond to two different kinds of bitumen, both produced from tectonic tide organic shale, at two different maturities, so they actually have two different viscosity. And based on their viscosities, you can actually measure the bitumen signal using NMR. So you can see finite signal. And I've just plotted a signal from a water sample just for comparison. So, the message over here is the viscous fluid bitumen you can see partially or sometimes, if you're lucky and the viscosity is low enough, fully. But the kerogen is a solid that you really do not catch with NMR, because it is a solid. Now, going ahead, the problems that you encounter in drawing cutoffs and identifying fluids, is not something that you have in T2 alone but also in T1. In this plot what I have is a distribution of T1 and a distribution of T2 at about eight or 10 different depths in the Eagle Ford. The bold line is the T2 and the dotted line is the T1. And you will see that drawing a cutoff is a problematic endeavor in both the T2 dimension or the distribution and the T1 distribution. You cannot go into the T1 distribution and say that I can draw a line over here, anything below that is bound hydrocarbon and then you have free hydrocarbon. A simplistic analysis like that is neither successful on the T2 nor on the T1 because things overlap in both the T2 and the T1 distributions. So this means that you need something beyond simple 1D distributions to enable you to do fluid typing using an NMR in shale. So, this is gonna be the topic for today. How do we go beyond simple 1D? And the solution we're gonna put forth is 2D NMR T1 T2 maps. And the key message that I want to convey today is these 2D maps are sensitive to molecular mobility. Or, in other words, instead of doing a one dimensional distribution, if you're doing a 2D NMR T1 T2 correlation experiment, the maps that you get out of it are sensitive to how things move. Or, in other words, molecular mobility. And this is the key message today and what we're gonna do is, we're gonna take advantage of the sensitivity to mobility, molecular mobility, and we're gonna apply it to shales and then show you how fluid typing can be done, how interesting information can be got out from shales using this measurement. So, I'm gonna show examples of T1 T2 experiments on oil saturated samples, shale samples saturated with light oil. I'm gonna show them on some model porous glass. I'm gonna show them in shale samples, real eagle ford shale samples from U.S. land. And based on what we understand, I'm gonna draw some kind of a universal picture of 2D NMR T1 T2 response in tidal and gas shales. Like, how do different compounds behave? Something very general, very universal. Based on that, I'm gonna go ahead and discuss ways in which we can do this in the laboratory. And bring these two together for a routine core-log integration. To be able to do that in a routine way.