High Resolution NMR Spectroscopy and Molecular Modeling of Confined Fluids
This presentation was given by Professor R. James Kirkpatrick during his investiture as an MSU Foundation Professor on October 29, 2019. Sadly, Professor Kirkpatrick passed away on January 8, 2020.
Transcript for video below was completed via an automated process.[applause] [Professor Kirkpatrick steps up to the podium with presentation slides on the wall behind him.]
Thanks Rob. Thank you all for coming.
Um, when I started to prepare this talk, I, uh, uh, didn’t really quite know what to do.
This is, I’ve been at this now almost 50 years and question is what, what the heck can I talk about 50 years?
Um, so if you’ll indulge me a little bit, I’ll talk about a little history and, uh, and then talk about the work I’m doing supported by the, uh, the foundation professorship, which is I think, very, very exciting. So, as Rob said, early in my career, I was a pretty hardcore geologist and running around the world to Cypress and, and Iceland doing field work and had a pretty big experimental program in crystallization kinetics in the early 1980s. Uh, there were huge advances in nuclear magnetic resonance spectroscopy that allowed, uh, really high resolution studies for solid materials.
And at that point I was fortunate enough to, to hook up with a group that, that was really at the cutting edge of doing this. And, um, the more I talked to him, the more I was convinced that this was really an extraordinary, uh, extraordinary opportunity to, to, to make some critical, critical contributions.
So in effect, I did a second PhD. I actually took a bunch of courses. Um, you know, all kinds of the study I knew about NMR, but I didn’t know all the, all the, the, the details, uh, and spent lots and lots of late nights in the lab and, and that way, you know, really got really got into this. Um, so I’ll talk a little bit about that and then we’ll, we’ll talk about, uh, what’s going on now.
So just for the development officers watching it, I put that in purple so that, that we recognize the color of your, of your operation. It’s all for you guys.
Okay. So, um, most people here probably know what, what NMR is. Uh, it’s the same basic physics is magnetic resonance imaging. Uh, and so how does all this work? Well it’s a a big magnet with a whole bunch of really fancy electronics in that box. Those boxes, fancy stuff.
So that thing’s taller than a person typically. And then you take a sample, put it in these little teeny tiny sample holders, uh, put that in this thing here, which spins it at the experiments. We do, spins it, um, much faster than a dentist drill. And there’s all kinds of fancy radio-frequency electronics in that too. This then fits in the top of that. That thing goes in the magnet and then we collect the data.
So basically, what is it we’re doing?
We’re taking a rock or piece of glass or whatever biological sample, we’re putting it into a magnet. We’re making it a radio transmitter where detecting the signal from that radio transmitter and it’s going to tell us about the arrangement of the atoms, the molecules in the structure and how they’re moving around. And every time I think about it in those terms, I think how amazing is that, that that something like that can be, can be done.
So, so I just, I love this stuff. So, so there were, in the early eighties, there were several groups around the world, uh, working on these class of problems and Oh yeah. So what’s the data look like? So there’s some, some NMR spectrum, uh, and that access there is basically frequency. And this is chemically useful because the frequency of this quantum phenomenon called spin is modified by the local chemical binding environment.
And, and also how the atoms and molecules are moving around. So that’s chemical bonding environment spreads that frequency out over some range. Uh, and that’s what we can, so we get, so how the heck do you take the wiggles like that on a computer screen and make any sense out of it? And it’s, you know, it’s still amazing to me. Well, it turns out that the fundamental theory of NMR is really extraordinarily well-developed and, and there’s, there’s all kinds of good ways one can interpret It doesn’t matter what this is, this, this, this is just an illustration.
So the theory is very highly developed and one can extract lots of information. There’s also a very large empirical database from different kinds of samples that we know do, have contributed to that helped with help with that interpretation. So we get something like that and you’ll see a bunch of them, uh, and we make some interpretations.
So we, and several other groups around the world started, um, just studying these, materials and we made quite a lot of contributions to understanding basic mineral structure. And this is just, again, this is illustrative of solving a longstanding problem about the structure of mixed layered clays. But again, you can take something like that that’s the spectrum and really extract very useful information.
So we did a lot of that, uh, with just fundamental kind of aluminum silicate chemistry. I guess. We also did a lot of work on glasses and we actually sold this to funding agencies on the basis of understanding glass structure early on. Um, initially it was all, uh, alumina silicate glasses related to, to geochemistry crystallization, lava and magma, which is what we work on at the time. But that dragged us off into, uh, studying basic glass science. Lots of issues there. Those are alumina, silica glasses. There’s some aluminum spectra.
We, uh, learned a lot about their structure.
We started working on a whole bunch of technologically important glasses, just illustrating some aluminum phosphate glasses. Those things have very high thermal expansion coefficients used in glass to metal seals. So why do we want to understand the structure stuff like this? Well most of the chemists in this audience, you’ll, you’ll, you already know understanding the fundamental structure of material and its relationship to the properties is really lets us provide a scientific basis for understanding of those properties, but also predicting the behavior of the property, say as we modify, um, composition. And that’s what happened with these uh, aluminum phosphate classes.
We did a lot of work on Pyrex kind of stuff. Uh, um, mostly from nuclear waste applications and, and lots of other things, things like that that uh, uh, I think it took us all different kinds of directions and understanding natural materials and technological materials.
Um, we also did an awful lot of work on cement chemistry. We were part of one of the initial nine NSF science and technology centers, um, that was supporting, um, uh, cement chemistry and did, did I think some, some very interesting work, we learned a lot. Cement science flies below the radar screen at Michigan state, but most of the big engineering colleges around the country will have a civil engineering materials program and cements at the core of it. And of course, cement industry, cement and concrete industry is an immense industry worldwide with a huge, uh, a huge research infrastructure all over the world. And, and, and our contribution based on Silicon NMR data that looked like that was basically the fundamental understanding of the call it molecular scale crystal graph unit cell scale structure offer the variation in the composition of this stuff that, uh, called CSH is the main binding phase for, for cement and concrete.
And, um, I think that that provided a basis for, for understanding a lot about, about the sort of chemical and then hierarchical behavior of, of this stuff. So that was a really quite enjoyable, uh, set of work.
Now this is, you know, getting to what’s this, we’re working on supported by the foundation professorship. One of the themes, uh, that we followed for many years is trying to understand what’s happening at the interface between water and some fluid water and a solid or some of the fluid in the solid. We’ll, we’ll see more later on. It turns out this is extremely hard to study and NMR is a very powerful tool for, for looking at it. So for example, these are Caesium one 33 NMR spectra for uh, for a clay and uh, there’s room temperature.
Hey look at this spectrum and it’s pretty boring. Not much going on isn’t okay.
It turns out that if you go down to low temperature, you resolve two peaks.
This one, Oh yeah. Okay. This was two peaks.
So we’re going to see a lot of figures like this, this, this cross hatch kind of stuff is the solid, it’s a clay, there’s clay there. And here we have a pore in that clay filled in this case with water and Cesium
And this turns out to be a nuclear waste issue. Cesium one 37 is one of the major components of the U S nuclear waste inventory and a really high energy gamma… it’s bad stuff.
So we wanted to understand what the Cesium’s done and it turns out what this is telling us is in fact that there’s the cesium is near the surface in two different kinds of environments. One is surrounded by water molecules, but close to the surface. And the other is actually directly coordinated to the atoms of the surface with some water molecules around it called inner sphere. And outer sphere is you raise the temperature. What happens is those two peaks merge together and that represents Cesium hopping between those two environments. And it turns out at that temperature, at about minus 50, we can tell that the frequency of that hopping is about 10,000 times a second. And you can model it in more, more detail than that, but here’s the point.
You get structure and dynamics from, from this kind of, uh, of, um, spectrum.
Okay, why do we care about this? Well, there’s all kinds of process chemistry that goes on at fluid solid interfaces, but from a geochemical standpoint, uh, surface interactions between the solid surfaces and the fluid have huge, exert huge control over all kinds of things going on in the environment.
So service interactions, control behavior, water, all kinds of natural and synthetic dissolved materials, natural organic matter. We’ll talk a little bit about that. And of course pollutants for all types. So that’s just a soil profile showing that. So we’re interested because of sort of near surface, uh, environmental issues.
It’s also really important deeper in the earth. There’s a oil rig and uh, oil well down there. Um, the, the, the rocks deep in the earth where they’re producing oil or trying to do carbon sequestration are pores, uh, and again, the interaction of the fluid in those pores and, and the rock is really controls a lot of things. So, so, you know, it’s a relevant situation. Oil and gas reservoirs, CO2 floods, hydrofracking whether you like it or not. Uh, geologic sequestration.
So it turns out now our approach is to use combined experiment and calculation. And I want to talk a little bit about how these, the core of the talk here, how calculations work with the NMR to, to resolve some of these issues. And really where we’re taking that with support from the foundation professorship.
Ignore those equations.
The, the computational methods we use, uh, are mostly molecular dynamics type calculations. Very different than I think what, what Andrew is going to talk about in a bit. Uh, but we use it because we want to understand the complex cooperative behavior between the surface, the fluid dissolve species. We want to understand it through time, both structure and, and dynamics. So we use this, Molecular dynamics method. Um, people go to war over the details of all this, but fundamentally pretty easy. You have a computer and in that computer you build a box, you put a bunch of atoms and molecules in the box, you, um, give those, uh, atoms and molecules, some properties, charging that kind of stuff shake it up and let it go. And, uh, provide you all kinds of information about, about structure and dynamics. And, ah, I want to introduce Nuresh Loganathan Raise your hand.
Uh, he’s the guy in our group who, who’s really responsible for most of this.
And one of the things that the foundation professorship does is support Nuresh.
So it’s really critical to, to, uh, this, uh, uh, this, this whole research program.
So we do computational modeling. Um, as Rob alluded to, we do most of our NMR at the Pacific Northwest national lab place called the environmental molecular sciences lab, which has truly amazing NMR facility. It’s a national use facility. You write proposals to get time like you would for a telescope. Uh, we’ve been very successful. The important thing is they have a really unique, uh, set up there for doing the so-called magic angle, spinning NMR at elevated pressures and temperatures. And as far as I know, there’s no place else in the world that does this. And 400 bars, 20 C or two 25 bars in two 50 C and for my chemist colleagues, if you have interest in understanding, uh, reactions involving solids under those kinds of conditions, come talk to me because the P and L people are really pushing this to understand all kinds of chemical processes. And it really is very, very powerful.
So I want to talk about modeling. I’m talking about high temperature.
Yes. NMR. Okay.
So as an example, um, we’re interested in what carbon dioxide is doing if you pump it down into the earth, uh, and how it interacts with the minerals. So, uh, we collected a carbon NMR spectra of a CO2 in the Nanopore galleries between two layers of clay. And there’s, uh, there’s the scale four angstroms here and the, the spectrum of this is a terrible figure, but there’s spectrums in blue and everything else is in theoretical spectrum. Uh, what we concluded is in fact that the only way you can get that spectrum is if the Co2 molecules are sitting in there, uh, kind of parallel to the clay layers and rotating rapidly around an axis perpendicular to their molecular axis it reorients the chemical shift.
We couldn’t tell that from a molecule doing the same thing but also wobbling back and forth, just, you know, don’t have the spectral resolution.
So Nuresh did some MD calculations, uh, for that kind of situation. And, uh, here he is plotting the angle between the clay and the Co2 molecule, uh, from these calculations. And sure enough the average position of that CO2 is, um, is parallel to the clay. But what is showing in fact is that it is doing this kind of wobbling motion and that the wobbling is, you know, plus or minus 20 degrees or so. You’d never be able to learn that from, from just the experimental data. And uh, yeah.
And so this really, really provides a hard basis for understanding what CO2 is doing in these, in these clays.
Okay. Another example, this is a carbon NMR spectra of Co2 and methane in in a compressed clay sample. artificial shale we’re calling it. We’ll focus on the methane region here. Uh, you just putting methane, methane goes into the inner layers to I didn’t, I never believed it. And we see two residences. This one here is methane. That’s natural gas, uh, in those inner layer galleries of the clay. And then this peak here that the breadth represents methane that’s exchanging between those inner layer galleries. And, uh, I said macro pores is really the bulk, bulk space in the, in the rotor.
Okay. So if you’re going to pump CO2 down into the earth, this methane around, what’s the CO2 going to do to the methane? Well, we did that experiment yet at high temperatures and pressures.
And what happens is the methane peak changes its position becomes a lot smaller, Methane is being replaced. And this peak over here that represents a methane exchanging, uh, um, becomes narrower, much less methane exchange going on.
So what did we say is that, uh, CO2 forces methane out of the small pores into the larger pores and decreases the rate that the methane’s exchanging between those, those two environments.
Okay. Another example of that, for instance, now for hardcore NMR people, this is a 2D, so called exchange spectrum for methane in this little bit different sample system. Um, there’s methane and these little teeny tiny pores that is methane in a middle-sized pore. And there it is in macro pore. But with this exchange spectrum, you get these so called cross peaks that demonstrate that the methane is exchanging between those environments. If you get across peak, the methane’s jumping back and forth between those environments. And I didn’t show all the data, but, uh, you do it as a function of, uh, so-called contact time and you can get an idea of the characteristic time characteristic frequency for that exchange. And it turns out that for this sample, it’s of the order of a quarter of a second, which is a pretty long time, uh, for a chemical process. But I don’t know any other method that would give you that kind of information. This is really quite extraordinary.
Okay, so how do we, how do we do these calculations?
So, uh, this is a relatively new method that we use a molecular dynamics method. We call it constant reservoir composition. It was created by a professor at university college London, but he’s also an adjunct professor here working with us. And another thing that foundation support provides is money to fly him here every once in awhile and really work with us on advanced techniques like this. So this particular method, uh, there’s a, uh, uh, clay samples call it, that could be anything. It could be, you know, membrane or, or, or whatever you want. This is very general.
But in this case, we have a clay sample and a pore that we vary the width of. And in this technique, what you can do is you can control the composition there and there, uh, in a way that’s not possible with other mechanisms. with other computational procedures. So you can get, in our case, we looked at the steady state partitioning of CO2 and methane, but, uh, you can also look at the steady state diffusive and pressure gradient driven transport rates. So this is really a very powerful tool. And again, without foundation support, we wouldn’t be working on this.
So, so, uh, Nuresh did these calculations What did he find? Well, absolutely the CO2, just connecting back to what we said about that CO2 is absolutely preferred in small pores relative to the methane. So pumping CO2 into the ground and you know, then say a carbon sequestration situation, we’ll release the methane. Uh, and definitely drive it from small pores, rocks or heterogeneous stuff, uh, small pores into, into the larger pore. And one thing that might happen is that that would make the natural gas more readily recoverable. And I, the PNL guys who are pushing that.
Uh, so, uh, so once again, these calculations really support, um, they really help interpret the NMR results that says, yes, CO2 does drive methane into larger pores. Uh, and uh, we also talked about exchange rates. So the calculations to understand the effect of CO2, methane in these pores on the diffusion coefficients in those pores, we’re getting going on that. So that’s against something supported by, by the foundation.
Okay. So, um, just a couple other set of themes that are groups working on, again, supported by the, by the foundation. We’re very interested in the interaction of a mineral surfaces and natural organic matter and other organic pollutants in the environment. And here’s a, uh, just a snapshot from an MD simulation. There’s a surface there. There’s a whole cluster of NLM, uh, molecules, natural organic molecules and the MD simulations are showing in fact is that, uh, these things stick to the surface, um, through, cation bridges. That’s where the, the calcium especially will coordinate to the surface and also coordinate to the carboxylate groups of the, of the nlm sticking to the surface. And we’re in the middle of writing a really long paper connecting the MD simulations and a bunch of experimental work. Huh? Talking about that. So it’s mineral organic interactions are our main theme.
We’re also very interested in reactivity and these high pressure NMR, uh, um, capabilities. Let us study insight to, uh, reactions and I don’t want to go through all that, but, um, this, this paper on a reaction of Clay’s with, uh, uh, water saturated CO2 solutions, uh, just got published.
So, so those are the themes, mineral surface interactions, including, uh, organic stuff, um, transport, structured dynamics, and in a moly scale, porous materials and reaction processes, uh, in, in, uh, especially in, uh, CO2 rich environments.
Finally, the last thing that, uh, foundations the supports for is it, let me put together a symposium for next spring’s ACS. Uh, 40 years of high resolution NMR spectroscopy of inorganic solids. And we just got all the abstracts in and we’re probably going to have two days worth of worth of talks. Uh, looking at a sort of bit of the history of, of all of this and also looking forward to, to where NMR is going to be useful in, in, uh, uh, and understanding inorganic systems going forward. And again, uh, we wouldn’t have been able to do that without the foundation professorship, so thank you.