 I have this title that's kind of a mashup of things that I'm interested in, and it'll be a little bit easier to explain the title if we go here. So this is a little tongue-in-cheek, but really, I think, to deal with our climate energy problem, we really have to embrace CO2. And I'm going to kind of have a two-part, so maybe three parts to my talk. This is not pointing. The first thing is there's actually some really interesting and, I think, good news that people really don't know about. And I'm going to show you that. And then to make a little bit of a transition, talk a little bit about perhaps the potential availability of CO2. And then I want to get a little bit geeky and talk a little bit about my own research. So hydraulic fracturing, a word probably everybody's heard. And maybe you think you understand it, but it's interesting and complicated. And the aspect ratio here is about right. You would typically, to access these deep shale resources, you might drill down 6,000, 7,000 feet. And then there's a horizontal wellbore. And fluids are injected at high pressure. And the thing that you see in the common media about breaking the rocks up into little pieces is not entirely true. What sort of happens in most cases is that there are preexisting defects in the rock. And you get a little bit of slip on the rock. And that creates a pathway for fluids to flow. What's really important here is this thing that says cemented well casing. Because you see that the aquifer is very far away from what people refer to as the target. And having a good seal against the aquifer is what prevents that aquifer from being polluted. It's virtually impossible, although it's physically possible, but it's virtually impossible for one of these fractures to grow all the way up. So the main contamination path is up the hole that was drilled. So this, in fact, is what a hydraulically fractured system sort of looks like. And the different colors are so-called different stages. And so in this long horizontal wellbore, they tried to hydraulically fracture here and here and here. And what you see in terms of the little dots are where they located, one of these positions where the rock slipped a little bit. So what's pretty interesting about the shale boom is that there's something in this for everybody in the economy. So on the top bullet, that's sort of the top 1%. But there's been a measurable impact on what people pay for energy as well as in GDP growth. So this sector of the economy has outperformed relative to its size of the economy. So pretty interesting in terms of the economics. Climate impacts are a little more debated. And I want to talk about that a little bit. And to do that, I'm going to show you first this figure on the right. So this is the production rate of coal in the US versus time. And so these are annual averages. And you can see there's maybe three peaks. This peak, that's the latest one. That's about 2008 actually corresponds with the onset of a lot of availability of natural gas. So we can try to actually model this and see the effect. So we'll use a logistic model. So you probably all at least heard of the idea of peak oil. So this guy, King Hubbard, tried to predict future oil. So n is the cumulative production. So if we summed up all of that production rate, that would be the cumulative rate. The cumulative, the rate is dn dt. And then r is the rate of change. So it would determine how big your bell curve is. So there's multiple peaks in that. So we have to use a multi-cycle model. So it looks sort of like this. So again, so this is rate. And then each little peak is sort of modeled. So we would sum them up over how many that we need. And we actually do this as an exercise in one of the undergraduate classes that I teach, because it's a nice way of talking about modeling. All models are wrong. Some are useful. So again, this is the data rate versus year. And here in this solid line, the solid gray line is my fit to the data. And this red dashed line is the trend that we were on prior to the onset of availability of natural gas from hydraulic fracturing. So the coal that's not going to be mined in the US is pretty substantial. And you can see we're on this path. And it looks pretty good. So that's all that coal in between there. So we can try and understand how much CO2 that might be. But that coal isn't, well, that electricity from that coal isn't just going away. We're replacing it with other things. And so the US has mostly been on a path where the replacement is about half renewable and half natural gas. So on the right is cumulative productions in gigatons of CO2 versus year. Blue is the path that the US is on. So we're on this blue path. If we look out about 50 years, this is an excess of 100 gigatons of CO2. Has anybody heard of the idea of like wedges, carbon wedges? No one's raised so few people are shaking your heads. Yeah, but 100 gigatons of CO2 is roughly a carbon wedge. It's a little bit less. And so it's an idea from two guys at Princeton, Rob Socklow and Stephen Pakala. But the idea is that you look around for things that you can do in energy. And a wedge is roughly 100 gigatons of CO2. It's actually 92. And if you can do about 10 of these things, basically you've addressed the climate energy problem. So this is just in the US. We're on a pathway to take care of about 10% of the problem. This thing dashed in green is the same 50-50. But if we captured the carbon, well, if we captured 90% of the carbon from the combustion of the natural gas. So you can see that it's substantial, but most of the, it's a substantial difference, right? But most of the benefit is really this switch to natural gas and renewables. And again, this is a pretty amazing thing and something that we don't really talk about very much. A couple other thoughts, just sort of quickly. One of the really interesting things about natural gas and renewables is that they work together very well for electricity. So this is a plot, basically hour in the day for some typical spring day in California. And this is megawatts of electricity. And the yellow and the orange in the top, you see, those are contributions from wind and solar. Clearly the sun goes down even in California every day, right? And so what happens when that solar goes away? California has very little storage of electricity at the moment. And you can see what happens is that natural gas, which is this navy blue and that sort of pale blue, they ramp up, okay? And it offsets for the, you know, offsets for basically the setting of the sun. And estimates are that over time that renewable portion that you see is gonna get sort of even deeper and deeper during the day and then something's gonna have to make up for it at night. So, so far, natural gas renewables are working together quite well. And here is perhaps the surprising thing that you may not be aware of. So this is not carbon emissions from the entire US. This is carbon emissions from electricity generation in the US and it's all emissions with everything converted to CO2 equivalents, okay? So we are here on this side, 2018. I was last number on this chart and we are down to sort of late 80s level for emissions from electricity in the US, right? So the gross amount is down as well as the per capita, right? Because over this time, US population's gone up by 40 or 50 million people, something like that. So pretty amazing. This, if you think about the idea of wedges, there is not a single solution. There's a bunch of activities that we need to do. This looks like a pretty interesting trend that hopefully we can continue to reduce emissions sort of year over year. I won't talk about it, but you can think about other countries in the world that emit a lot of CO2, such as China. China in fact has natural gas resources in shale that are larger than those in the US. So there's a potential pathway to sort of export this idea to other places. And just one other, I won't go into this a lot of detail, but there's a really interesting power plant that's running in Texas and it runs off of natural gas, but it doesn't take in air. It takes in pure oxygen and it actually burns that natural gas in an excess of CO2. And so what's interesting about this is that it basically, you have to bleed some CO2 out of this system because it's gonna accumulate because you're burning natural gas. But the CO2 that comes out is actually ready for industrial use of some sort. And so it sort of solves the problem of carbon capture. So again, this is a pilot plant, promised zero air emissions, same cost or less compared to conventional natural gas fired electricity. And the other kind of thing that's interesting about this as well is that it makes a fair amount of clean water because you think about methane, right? It's carbon, it's hydrogen. So water is also a product that's gotta come out. Okay, so what might we do with CO2? Well, we might go and think about if we can use this beneficially in another problem that we have. I'll just skip this in the interest of time and we'll just focus on water. So these are different places where natural gas is produced in North America by hydraulic fracturing. That's how much water it takes to complete a well. So millions of gallons. So four million gallons of water is a lot of water. So it's about the average daily consumption of 11,000 families. And water reuse and water management is a problem faced by the industry. And so why do they use water to hydraulically fracture? Well, it's because it's available, it's easy to use there. So what we might do is actually instead of using water, we could use carbon dioxide for hydraulic fracturing. There are a lot of other questions about that, like does it flow back? What does it do to the rock? How does it flow? All of those questions and many others really can be answered in a much more holistic way if we have sort of fundamental knowledge of how fluids move through shale. And so in fact, some of us on campus are part of what the Department of Energy calls the Energy Frontier Research Center. That's the name of ours there. But it's really about trying to decrease the impacts of hydraulic fracturing on the environment. So there's a lot of upside in terms of carbon emissions and there's, well, there's a lot of downsides due to water use as well as material and energy demands and a more fundamental understanding could benefit that. So our center is kind of organized in three ways. That's sort of the core. And then most of my work here is characterization. So what does shale actually look like? How do fluids flow? There are, as a multi-physics perspective, so transport, mechanics and reactivity because these rocks actually react. And then over the top is what's called scale translation. So shale pours on the order of nanometers, right? And well spacings are on the orders of kilometers. So there's a big difference in scale there. So I wanna, with my maybe last 10 minutes since trying to save a few minutes for questions, wanna show you a few things that we do in my laboratory. And we do a lot of imaging with computed tomography. And so this cartoon here gives you an idea. So there's an x-ray source and the source is moving or the material you're looking at could be moving. And so there's multiple passes of a beam of x-rays through the same volumetric part of the sample. And it's possible to do a tomographic reconstruction and what that would mean would be for say that yellow square. It's in fact, it's a cube or a cuboid. You could go in that cuboid and understand how much x-rays were attenuated. So one way of actually processing this data with the so-called CT number and there's a bunch of equations here. What really I want you to look at is here on the bottom. So a conventional way that people look at CT images they just look at the plain image and then they play with the gray levels and you try and differentiate things that way. So we do that in my group but what we really like to do is our fluid substitution experiments. So you can have a sample with say no fluid in it, you can put a fluid in it and then you can do a calculation like this where you subtract those images of fluid filled not fluid filled and then you normalize it and then you actually have an understanding of where fluids actually move in your sample which is important when you have something like shale. So this is just some data. So a good penetrant for a sample like this shale is a gas because gases flow very easily. They're much less viscous. The two that we use a lot of are krypton and CO2. So what's relevant here is that air or vacuum would have a CT number about minus a thousand so we actually have a difference with CO2 as well as krypton. So we can actually image where fluids flow and that's what the next one shows you. So this is that sample after we've done our experiment and done all of our image analysis and what is happening is that time is going on as we go from left to right and red is more CO2 and blue is no CO2 and so we can actually see this sample filling up with carbon dioxide. So an interesting fact is that the density of CO2 the density of CO2 at any pressure and temperature is always greater than the density of methane. So practically you can take the methane out of the resource, combust it or extract the energy and put the CO2 back in and have a little bit of room to spare. So this is a nice sample and accepts CO2. One of the things that we tried to do is actually go down in scale and understand what it is about say this sample that makes it accept CO2 versus other samples that don't. So on the top we have that 3D image and it's at a certain resolution, we can spin it around, we can slice and advise it on the computer. We can come to a cross section like this which is interesting because it has blue where no CO2 entered and then it has red where a lot of CO2 entered. We can go take that sample and sample it destructively, so cut it, polish it and then look at it under scanning electron microscope and these are truly multi-scale images because I can look at this region here. Here it is in a larger, more magnified in this square. This region actually looks particularly interesting so we can zoom on that. And in this image there's a bunch of stuff. What do we see? Well we see what looks like spaces to us in the fabric of the shale. This thing called OM is organic matter so it's something carbonaceous. We can actually zoom that in and we get this image here for example. And what the black is are the actually the pores of size about five to 10 nanometers inside this organic matter. And so we can understand again why gases can move around, what makes them good candidates for CO2. We have a whole sort of library of these images that I could bore you with for hours talking about the textural details but we can sort of sort those things out. This is destructive though, right? I have to cut the sample. So what we would like to do is actually not destroy our sample. So we have not a new area but an interesting area. And so what this is is a piece of a shale sample that's been milled and we're actually gonna look at that little thing on the top that says 30 microns. And we're gonna do that up at Slack. They have something called Transmission X-ray Microscope otherwise known as a nano CT. So it has a sort of resolution of 30 nanometers. So those images that we were looking at previously, this image here has a resolution of a couple hundred microns, right? And these have images on the order, these have resolution on the order of nanometers. But so we have a 3D volumetric image this will spin around and do a bunch of really cool stuff but I won't do it. But we can have an image like this which is an image of the top and we can see different minerals and different things in it. But again, the issue here is an issue of resolution. So this is one of those little pillars on the top of our sample and then we'd actually like to image this and have high resolution information but we don't wanna destroy it because we'd like to do an experiment on it. In the middle here is an image, is a nano CT image, okay? And this is an SEM image at almost the same location. So to get this image on the right, we actually did destroy our sample, okay? And you can just see visually, they look sort of similar but this one's a lot less, has a lot less kind of information in it. You can look at this region right here and so yeah, there's a region here but you can look in this SEM image and there's a lot more detail, right? And these little details or something that we're interested in. So what we're trying to do is actually get super resolution on these TXM images by using information from the SEM, okay? So here's an example. Here's a part of a TXM image and again, that's the nano CT. Again, not destructive, okay? Here's the actual SEM from the same location, a lot more information. So we use a lot of the same AI tools that people use for other things but we use them for image analysis and these images on the right that look like this SEM image or actually this image only they've been sort of corrected and so now we have super resolution, okay? So we had to do this, we had to destroy the sample but we only destroyed this part of the sample leaving the rest of this, right? So we have TXM image data on the rest of this part of the sample and so we have super resolution in 3D, okay? So we do convolutional neural networks and generative adversarial networks. Again, just like a lot of people are using for a lot of different applications. So that's the end, I hope I left five minutes for questions. So kind of the take on points are that this switch from coal to renewables and gas so far has had a pretty positive impact on long-term CO2 emissions. Again, this is something that you won't hear much about. The storage capacity in shales for CO2 is pretty good and this sort of this multimodal imaging so using the TXM and using SEM combined with these AI techniques seems pretty promising for getting super resolution so we don't have to destroy our samples to understand everything. So just a couple of slides to finish up. This is the work of these people, some of whom are still in my research group, some of who have graduated and left. If you're looking for energy courses, I encourage you to look at energy resources engineering. We have a full set of classes that go from renewables to petroleum engineering and that's my acknowledgement and I will stop there and see if you have any questions or comments. Yes. My name is Boris, I am with ICME but I have a question about the so the shale gas development, they reduce CO2 long-term and that's great but from what I read in the economist so not scientific publication, when the company started to reduce their water use for fracking and they started using chemicals which are first, they don't use water and second, they are more effective in cracking the shale formations but at the same time they introduce like cancerogenic stuff into the soil, into the groundwater, that kind of stuff so the impact on the environment is reduced via CO2 reductions but increased via health to humans. What is, when you all study this, like do you have to at the same time do the research of the other side or is that left for somebody else to measure? What's the interaction between those? Okay, I'm not sure what the other side is, right? But there are people looking at all aspects. The impact to health. Yeah, all aspects, yeah. So a couple of things. Yeah, so first of all, like in terms of the chemicals that they put in, yeah, they do, they are as a sort of cocktail of chemicals. That's regulated state by states but most states now have a law that people have to document what they're putting in so at least we know what's going in. But generally, right, you're injecting something into what's already a toxic waste dump anyway. It's effectively nature-mated toxic waste dump, right? Because the hydrocarbons that are in there are cancerous themselves. However, one of the things that's very misleading in the media is this idea of groundwater contamination. So as far as I know, there are very few cases of hydraulic fracturing contributing to groundwater contamination because they're so far apart, okay? So when there is a problem, what it is is it goes back to people haven't made sure that there's a good seal between their wellbore when they pass anything that's drinking water, okay? And a lot of what people say is contamination is coming from old wells where someone else may have abandoned it or something. So this is a slide from this reference. And the top are the deepest aquifers across the Marcellus basin. And I guess we're going, I don't remember which way it's going but it's going geographically in some area. And then this is a bunch of information about the wells at those locations but the blue is the top of whatever they did, right? So you still have thousands of feet between where your operation is and the aquifer. So again, if there's going to be contamination it's really from this operation. It's going to be that somehow when you passed the aquifer part you didn't have a good seal between the well casing and the formation. And what's ironic is that most, so this is one of the easier things for people to measure but ironically it's something that's often not measured. So a lot of states are sort of asking that people make sure that they have a better seal. So it's a long answer to a complicated question but we do think about everything. And definitely on campus, so like in my center we're really focused on this problem of water substitution and understanding in the reservoir, the effects of water. But there are other people that are worried about methane leakage and other people worried about more shallow groundwater effects, yeah. Thank you. I just also wanted to point out that I think that the defense that not much has been discovered yet is not a good defense because maybe there were not enough measurements, maybe, I mean, lobbying is a thing in the US and they tend to bend the facts. Yeah, I don't know if I said discovered. I mean, as far, I was saying as far as I know there are very few cases where this operation affected that, right? But there are definitely problems of aquifer contamination. Those are mostly due to older wells that leak, yeah. And things that people didn't know or they just didn't know about, yeah. So definitely, yeah, there are issues but they don't seem to be coming from this problem. So that's the end of my time. I'm easy to find on campus, so I will turn it over. Right? Yep.