It was well thank you for the introduction. And I just wanted to take a brief moment at the beginning here to thank Dr Sun ho Troy and Dr Chris Jones for giving me the opportunity to come here and talk about my research to such a broad audience it's not very frequent that you know graduate students get this opportunity so I was very excited when this opportunity arose. So the title of my talk today is molecular design of liquid sorbent for C O two capture shown here is a coal fired power plant located in Wyoming son who so that's where I'm from and so C O two capture literally hits home for me it's a big problem where I come from or it will be soon. And when I start I want to define the C O two problem. For everyone because I'm going to use that term frequently throughout the presentation and for the context of this talk today the C O two problem as I define it is the problem industry is going to face when Congress passes regulations that are going to enforce separation that are going to force regulations to limit C O two and that's happening as we speak. Congress is literally burning the midnight oil to see that you know. They're talking about twenty twenty. We're going to have regulations in place that are going to cause industry to you know either go to innovative technologies where lower amounts of C O two are emitted or simply capture that C O two and they're talking about doing that by establishing a cap and trade system and what is that cap and trade system going to do. It's basically going to put a dollar sign on C O two and it's not an easy problem for or an easy number for Congress to come up with if the numbers too low. Well then companies. It's just cheaper for them to you know pay the penalty and continue to continue with business as usual and then just pass it off to the consumer which is us. And if the numbers too high that could be devastating. So they're trying to find that happy medium until they do find that happy medium really any technology is on the table because we don't know what C O two is going to cost so anything is economically viable at this point. But I do want to emphasize that they are looking at targeting you know not just the industrial sector but also agriculture in pulp and paper as well. So this is going to be you know a broad problem that a lot of different sectors are going to have to deal with. And soon the first number that they're given is to Earth the first day is two thousand and twenty which means we have to have a viable solution within the next you know five to ten years and that's not too far away and in my mind the sea. To problem as I defined it is like a snowball rolling down a hill. It's just getting bigger. It's a fact that there will be more C O two tomorrow than there is today. And so you know it's just going to continue to grow until we do something about it and this graph shows metric tons of C O two for the entire world over time and you know. It's increasing sharply after the turn of the millennium it's shooting up like a rocket and this has to do with several factors. The biggest being countries like China and India are developing at a very fast rate China just passed the United States for the leading C O two emitter in the entire world and although they're C O two emissions per capita is about twenty five percent of what the United States is their population makes up for that. So as you know populations continue to grow and the standard of living continues to grow across the world this problem is really only going to become pounded so we've got to deal with it today because tomorrow it's just going to be a big bigger headache and how can we deal with the C O two problem well there are some strategies to reduce the amount of C O two one obviously is consume less which you know is easier said than done because nobody wants to turn off their air conditioner in the middle of the summer and not only that like I had mentioned population is exploding and those countries. You know the people in those countries they want the same standard of living as everyone else. So we can consume less and we should consume less but that alone is not going to you know solve the problem and then I'm going to jump down here to innovation. We need innovative technologies we need technologies that will emit lower amounts of C O two. But that takes time. That's completely dependent on markets and cost right now there is no market for C O two. So it's cheap. You know it's free for companies just to dump it into the atmosphere and that puts us here at C O two capture. I define soon to capture as a bridging technology it's going to allow us to get to the innovative solutions for the C O two problem because it's not going to happen overnight. We need to have some way to capture the C O two until we hit the point where you know emitting C O two is so expensive the these innovative technologies which you know could be gasification or you know co-gen and those types of things when they become economically feasible. So this is where I focus my talk on today is C O two capture this bridging technology and when we talk about C O two capture you really have to start with an application a lot of different industries release C O two and their feed stocks their flu gas could all be different and the conditions of the flue gas the compositions really dictate what type of technology is going to be applied. And I want to start talking about power plant flue gas because this is really the heart of the problem we can see here that you know in two thousand and three a third of all our our C O two emissions in the United States came from the combustion of coal and that's primarily for you know power generation and this is a picture of a power plant. I took it about two miles down the road from where my parents live. There's six hundred of these in place in the United States. So if Congress says tomorrow we need to start capturing C O two. Well this is the industry that's going to be hit the hardest and not only that we need to have the capability to retrofit some sort of capture technology on to this power plant so that it can continue to produce electricity the way it has in the past in the way it will tomorrow. And that's where we're really at. But you know power plant flue gas is not an easy problem it's quite challenging actually and that's due to several reasons one. I showed the. Characteristics for just the flu gas from a typical power plant. You know they come out of the high temperatures. You would expect there's a lot of it. That's a big problem seventy eight million cubic feet an hour. And it's really really dilute in C O two. So the techniques separation techniques that use you know partial pressure as a driving force are going to be tough to apply to this feedstock simply based on the fact that carbon dioxide is in such low concentrations couple that with you know the amount of C O two. That's being released. We're talking about six thousand tons a day from just one power plant. Multiply that by six hundred. So whatever we do to capture C O two. We have to be able to reuse it. If we're going to react C O two is something and then just dispose of it. Well you know we're going to run out a place to put that. So whatever we come up with that has to be regenerated oil and it has to be effective for this type of composition and the kicker. It's got to produce a high purity product stream without that you can't really do anything with it if it's not high purity then transportation costs are going to be enormous. And you know whatever we decide to do the C O two has to be pure so this is really the problem as I see it. How are we going to separate that C O two. Well here in Georgia Tech. We're looking at a bunch of different technologies and like I said because there's no dollar sign on C O two yet. Everything's on the table. Dr chorus and company they're looking at membranes for the separation. Dr Jones and company they're looking adsorb and and well as my advisor always says start with what you know and we do solvents So that's where we're going to focus our C O two capture process is on utilizing liquid solvents. Well how do liquid solvents capture C O two. There's two main mechanisms the first being chemical absorption which is the most popular especially for flu gas from a power plant. Chemical absorption works as the name suggests through a chemical reaction. So the C O two. You react with some sort of absorbent. And it's because of that it's highly selective nitrogen doesn't so you get a really selective separation. It's really efficient. It's very fast and it's capable of reacting you know a lot of C O two. The problem with that is it's thermally driven and as will show later. It costs a lot of money to get yourself back because Right. We have to get our solvent back otherwise it's just not going to work. However you know it has been thoroughly research and it's proven so you know this is where we'd like to start on the other side we have physical absorption Now this is where things like ionic liquids generally fall into it's basically Vander walls forces that are giving you a separation and you know people have reported high capacities and they have reported high selective eighty's and it's generally the separation is generally pressure driven. So for a flu gas stream that's really dilute in C O two. You can't really you know it's not going to be economic to pressurize to the partial pressures you need to get these separations they do have the benefit of the low heat for a generation but again they're just simply not economic based on the proof on the pressure. And I like to say in here not effective alone as you'll see later on so on. How does a typical absorption system work. Well it's quite simple really. The flu gas comes into an absorber a big tall tower where the solvent comes in through the top it reacts with the flue gas in the scrub scrub gas which is mostly nitrogen at this point goes out the top then we send the C O two rich solve it through a heat exchanger where we try to recover some of the sea to a stripper where we heated up that releases the C O two and then we can recycle the solvent again. So you see flue gas coming in the two products being C O two in the scrub gas and typically the absorbers run it like. Forty to sixty degrees centigrade stripper's run round one hundred to one hundred thirty degrees centigrade somewhere in there. And this as I'll show later on is where the real energy sink is for this process so what if we had to separate C O two tomorrow. You know one of Congress said OK. Tomorrow's the day. C O two. It's just getting out of control. Let's just start doing it tomorrow. Well this is what we're going to do. It's a chemical absorption process using mono ethanol mean as the solve it and why are we going to do that. Well because it's been thoroughly researched it's been proven. It's been applied. So for a power plant standpoint it's a low risk technology. They know it's going to work it. They know it's going to cost a lot of money but it's not going to disrupt their power generation cycle. So this is where we want to to make you know an impact. We want to take the process that would be used tomorrow and make a significant improvement on it. And like I said this is well researched and proven so if we can develop a unique solvent system utilizing the same processing technology the same pilot plants. It could be implemented in a short amount of time which would help us meet the goal of twenty twenty as a start date for the C O two regulations. Additionally as engineers of course we have to do simulations and come up with dollar figures for our processes and so this is a good start point for us because it's going to be similar to how we operate our process so it can help us validate our simulation and it also helps us evaluate efficiency and economic targets for our compound so this is really what we're trying to improve upon. I ran a simulation myself on an Emmy a process for C O two capture using the flu gas from that typical three hundred fifty megawatt power plant. I assume ninety percent recovery and ninety five percent C O two products STREAM and these are numbers that that deal we has put out there for targets for the. Type process and the bottom line is yeah cost a lot of money. Two thirds of the total operating costs come from regenerating the solve it. So I'm a big picture kind of guy. This is where I want to make an improvement. If I can get this number down right here. Well then suddenly if I can cut it in half will now suddenly the process cost thirty three percent more to operate. So this is really where our approach comes from identifying the biggest problem for this process and trying to make an improvement and I don't do this alone. I'm a member of the Ecker Liotta group and we are chemists and chemical engineers working together to solve real world practical problems and like I said you know them a pretty a process it works but you know that the problem comes from engineers are really limited to what's available. You know when we design a C O two capture process we kind of have to work with the materials that are available and then chemists are on the other side. Usually you know developing the separate or the solvents themselves so we need to combine forces here so that we can come up with an efficient solution in a short period of time and that's really what our group is set up to do and like I had said you know energy is the real problem and the M.E. a process and I forgot to mention this. The real problem behind that is that it's dilute they run thirty thirty percent M E A and the rest is water. So in that stripper when you have to heat it up to one hundred to one hundred twenty degrees C. seventy percent of that energy roughly is going into basically boiling water. It does nothing to regenerate yourselves. It's just added energy and that a lot of that has to do with the corrosion of Amy. So if we can eliminate that need for that added solvent will we just made a huge impact on the operating costs of this you know already proven process and that takes us to our approach reversible ionic liquids before I start talking about reversible and click woods. I felt it appropriate to. Discussed briefly traditional ionic liquids traditional ironically quids have been getting a lot of interest in academia as of late comes from a couple main reasons one is you can come up with almost a limitless number of combinations that will give you an ionic liquid and each one has their own unique properties so people are looking at using them for catalysts or solvents and and things of that nature. So it's no surprise to see that people are actually looking at using them for C O two capture. People also tout them as green and this mainly has to do with the fact that they have zero vapor pressure and that means you have no solvent losses you know you don't have any VO Cs now that are contributing to pollution. However they can be very toxic. This is a very common on a click which shone down here it's based on and the image is only a Marine. And it's beam in P.F. six. It's found at high temperatures it will actually release H F which as you can imagine is problematic for a process using this as a solvent in operating at high temperatures so you can come up with a lot of different ironically quids but they certainly have their drawbacks. As I said a lot a lot of researchers are looking at using them for C O two capture and why not. You can come up with a bunch of different ones. And this is work done by JOHN BRANNOCK at Notre Dame and on the Y. axis we have C O two more fraction and on the X. axis is pressure of C O two partial pressure of C O two and we do see that we get very high absorption. But if you look at the pressures associated with that. That's extremely high. So for a flue gas stream that's you know fifteen percent C O two to get that up to eighty bar partial pressure of C O two is out of the out of the picture. We can't do that but you know they do show promise. So we want. To take the idea that you know ironically quids can give you high absorption capacity but really you need chemical absorption to give you high selective ity and that's where our reversible and a click would come in. So this is the reaction that gives us our reversible and quick woods and we start with a cycle amine and we can alter the substitutions on the group and basically when you just bubble C O two through a mixture it can be atmospheric C O two you get full reaction to this ionic liquid. And if you add heat or you sparge with an inner gas you go back to the molecular liquid again. So it's completely regenerate will which is what we need and not only that we have the freedom to tune our own I click was you know just like traditional an eclipse can be tuned by changing the cat I honor and pairs. So I'm going to focus on the four main reversible one a click woods where our equals and I thought See I thought see Ethel and pro ball and as I had mentioned I work in a group with chemists and chemical engineers we started with meth oxy and I thought the silo propylene because it was commercially available. And you know we needed somewhere to start and sure enough you know we form the on a click when they should good properties however the S.I. Obama. Is known to cleave in the presence of water which is problematic for flu gas because it's got a lot of water so we knew we needed something different and then we went to the chemists and said You know I need to get rid of that S.-I Obama and what can you do for me and that's where the Ethel in the purple came from and sure enough I ran stability studies for the Ethel and pro ball doesn't matter how much water is there over two months there was no degradation at all. So they have a long shelf life which is also good. And sure enough the mythos enough doxie degraded. So these are the reversible and click woods and what makes some really cool is that there are tuneable if we change those are groups on the SRI we actually chain. The properties of not just the starting molecular liquid shown here in gray but also the ionic liquid in black. So this graph shows you lambda Max here on the Y. axis and just think of it as plenty lambda Max is an indication of the player D. of the solvents and it's determined by. Looking at the spectra of the Nile red probe in these solutions and it shifts according to the plenty of the solvent. There's a database of like five hundred solvents were Nile red has been run in and the lambda Max has been reported so it allows for easy comparison it's a quick way to evaluate the plenty of solvents and I showed just three reference lines here but it's really quite interesting. Take for example C.M.'s say where we have trouble trying to thaw propylene. If we start out with the molecular liquid it has a player he about like acetone and then we just bubble through C O two and form the Scion liquid we shoot up to something that's about like methanol and that's a pretty big property change just by you know switching from this molecular time cliquish But what I really want to point out is that it changes both the magnitude of the switch. And then the magnitude of the starting and molecular or the molecular and ironically quids also changes as you change that our group. So the point I want to get across is that we can tune these compounds we can have we can have them perform how we want by changing the substitution on that S I. Group. Well another important factor is you know the sustainability I said it's got to be reusable if we want to have a liquid solvent for C O two capture has to behave the exact same every single time you use it and it has to do that over and over and over again and on this graph we have conduct to video on the Y. axis versus time and we can see that when we bubble C O two. Sure enough we get an increase in conductivity. And then when we add he it goes back down again and this was just repeated for three cycles but it can be done over and over and over again. So so far so good. I mean these reversible look like they have favorable properties for use as chemical absorbance and C O two capture. And one more slide here is a D.S.C. differential scanning calorimetry so Y. axis we have heat flow and on the X. axis is temperature what happens is we had heat to a an ionic liquid and then we see right here around one hundred ten degrees the C O two comes off. So now we're back to the molecular liquid and then at two hundred forty degrees. We see that the molecular liquid boils. Which is great. That's one hundred forty degree C. difference. What does that mean that means we get a very clean separation. That also means that we're going to minimize the solvent loss in the stripper and avoid problematic processing by trying to recover this solvent. So so far so good it looks like they they could work as chemical torments But how are we going to decide which ones the best one. Well you know we can do the shot going to approach we can just stick a whole bunch of our groups on here and look and see how well they capture C O two. I prefer to take a more calculated approach rather than trying a thousand different compounds. So the idea is that ultimately the thermodynamics are going to dominate the cost. It's not really the capture of C O two. That's the problem it's the release for chemical absorbance and that is given by the heat of reaction which is the same for the forward as the reverse. And if we want to determine the heat of reaction for this equilibrium based reaction. We need to understand the equilibrium constant you know the equilibrium constant is a function of activities of the products over the reactance. You can think of these activities basically like concentration so now for the you know the experimental part I want to know you know what is this equilibrium constant at a bunch of different operating conditions temperatures pressures and ideally for a bunch of compounds so we need a technique that can do that. Can give us a quick equilibrium time allow us to change the temperature from you know low which would be forty degrees seats a very high temperatures one fifty two hundred high pressure and it's got to be high throughput we need to do these quickly because I want to graduate some time. So the technique we're using is spectroscopy actually it's A.T.R. F.D.R. spectroscopy and we have a custom made reactor and I don't want to spend a lot of time into the experimental set up but if anyone's really interested into it in it. I'd love to talk to you about it. So just send me an e-mail. Basically this is an F.B.I. our spectra of tri Ethel silo propylene the red curve here is just the molecular liquid. No C O two present at all. And I've always been intrigued by F.D.R. I think it's a very powerful tool we can see right here. These are C.H. bonds and when we bubble through C O two through the mixture. We see a couple interesting things happen. One is you know we see the appearance of this carbon you know Carbondale comes from the formation of the I click wood so if we can understand if we can quantify this carbon you know peak relative to conversion will then we know what the conversion is additionally in this was kind of a added bonus to this measurement technique this big giant peak this shows up here is C O two. That's physically absorbed into our ironically quids So ironically quids were you know we're marketing them as utilizing this dual capture mechanism chemical reaction to give you nine a liquid physical absorption in the liquid to give you further capacity so. I. I can collect both those sets of data with this one measurement technique I say here we do see some interesting behavior. Typically it's assumed that for car bomb information you have a one to two more ratio of C O two to mean in those situations you see a very sharp Carbondale peak which is typical of carbon eels and I are we do see some interesting behavior in here which may suggest that we do increase the capacity beyond that one to two ratio. So her you know currently were investigating that to see if we can increase that capacity. So what are the results. These are just initial results on the four ironically quids reversible liquids that I had mentioned. And I still hear the grey is C O two captured through reaction. And that was calculated based on the what were the one to two more ratio. Like I said if we can go beyond that that ratio will see even increased capacities and I should mention that i Report capacity in terms of more of C O two particular gram of solvent which is very useful for the next slide that I'm getting to and then we look at the black lines here which is the C O two. That's physically absorbed into the eye a click would show here beam in P.F. six right. It's a very common ionic liquid will how well does it do sixty bar and thirty five degrees C. It's got a capacity just over five hours as well be of that above that and then premie a this line here was calculated for a thirty way percent solution of M E A. And we can see that you know our solvent do better than that as well when they're run neat. So when you combine the two you see in even a substantial increase in the C O two capacity so so far these ionic liquids do show very promising capacities in terms of C O two loading. But what does that really mean and I kind of hinted to earlier on in the presentation but let's revisit. The flow diagram. This is the energy sink rate two thirds of our operating costs are going here into the stripper. And how much heat is that how do we calculate that he well it's basically mass times the heat capacity the configuration times the the change in temperature. Let's just make for comparison sake the assumptions that the heater reaction is the same for a means where we have a means M.E.'s And I mean it's about the same. Let's also say that. OK that C C P is the same he capacity of our configurations about the same as their swell if we can double the capacity of our C. R. capture a molecules compared to N.B.A. that results in a fifty percent reduction in cost and that's huge. The bottom line is less solvent equals less energy. If you get a five percent increase in capacity that's huge. If we can tune the molecules so that they deserve or about a lower temperature or abs or before actively at a higher temperature that also has huge ramifications so we're really trying to attack the problem at the you know at the heart and I think we're there we're getting there. We still need some more research that needs to be done but for an immediate solution. I think we're really on to something and I have to mention you know ironically quids reversible they're not perfect. If anyone's ever worked with ironically quids they know that they're extremely biscuits and that can be problematic for processing. Shown here is again light gray from molecular liquid dark gray for Ironically quid. We get a three order of magnitude jump in viscosity when we change from molecular liquid to ionic liquid. That's a problem but. Observations in the laboratory have suggested that it's not a linear change in viscosity it increases sharply at higher conversions. We have some theories on on why. But we're working in collaboration with Dr breed veld to try to understand our. What conversion does the viscosity start to increase drastically and then it becomes more of an optimization problem. Well we may only run to eighty percent conversion if we can keep the viscosity down and so there's a lot of different parameters that go into this and you know we're still doing research every day I'll probably go back to research after this is done. So we're not there yet but we have some very interesting results and I'd just like to conclude that say you know real world problems are interdisciplinary and Georgia Tech you know is one of the I think it's a it's a great university for encouraging interdisciplinary research collaborative research because you know real world problems require that you know real world problems need real world solutions. We all need to work together. So I encourage everybody to go out work with other people see what they're doing and see if you can't work together to come up with solutions that have a bigger impact than you could ever do alone. We need to retrofit solution now that's a fact we need some sort of solution to capture C O two from power plants within Ideally the next five years because it's going to take a couple years to build that. Test it out and then get it up and running. So I want to emphasize again that C O two capture is really a bridging technology. I want to also encourage everyone to go out and pursue those innovative technologies that are going to get us away from emitting C O two and having to worry about how we're going to capture it. But those are going to take time. And then you know I have to sum up by saying that the reverse of one o'clock Woods have shown phenomenal results so far in terms of C O two capacity and I think they could be the next generation of liquid solvents for C O two capture and I just have to acknowledge now my my funding sources which is deal we N.S.F. Professor surrogate because they're in from Imperial College London who help me design a high pressure high temperature reactor for the A.T.R. after our experiments. The entire accurately OTA group without any of them I could've done this you know we work as a team and so we share the credit as a team and I think you know the support staff here at Georgia Tech doesn't get enough credit from the machine shop to the stock room to the administrators without them we all wouldn't be here right now doing what we're doing so a big shout out to the support staff at Georgia Tech and question. Thanks very. Now we're looking at water absorption now. We're going to get by carbonate formation. It's a little more stable so we have to determine what effect that has on the thermodynamics of reversal. So that's what we're currently in right now everything's been done with just pure C O two up to this point Jeff I don't know. So we prefer formed the Ironically quids they're actually you can preform the on a click would stick them on a shelf and they'll stay like that for a week or so before they. Well it's a mass transfer problem at that point they're so viscous that. Yeah they start to reverse back a little bit but that C O two stays in train in the solution. So the reversal at ambient conditions takes a really really long time. So no we just preform and run the Imation and so on and then we saw a nice clean straight lines versus share stress and we reversed it so we're confident in those numbers. You. Joe your yard there by see is the real ride made by Lee Yeah we have to do the measurements at a as a function of temperature because equilibrium constants as a function temperature so we're working our way up right now we're at seventy five degrees C. right now. Yeah of course Yeah we're looking at. I'm actually coupling the experiment so we have low pressure experiments which is about zero to fifty P.S. I just to get a handle on you know the conversion and then we're looking at the high pressure experiments to look at physical absorption once the on a liquid fully formed and we're doing that as a function of temperature and right now like I said we're at seventy five degrees C. So the end would probably be around one fifty depending on what we see around one twenty one twenty five and we need to look at mix gas as well. I mean I did. Yeah but you need pure C O two for that fact and that's that any separation technique you use has to be capable of producing a high purity product stream and I base that solely on the fact that transportation of the C O two. Whatever you do with it and it's probably made it to do that and that's what I was trying to me. I think he live. Well I'm looking at here. And that's a very good point I'm trying to look at the therms for physical absorption in our in a quick woods not for the application of you know post combustion flue gas treatment and it came because I can I can get that measurement using the same sample the same set up that I aready I'm using speed. But that's using the same problem that I would like to see. So that's kind of like you know looking for something to go right. Really. In general and conventional in ours so. You know we just don't have that that's going to probably stop yours. So if you look at the big picture and. There's a problem. Three. Already. That you're talking about say. Well you know yeah. Actually that doesn't. Because that's really what the problems of them just that's why. This property is free. Are you talking about missions that go to. Three. Each.