Workshop: Innovative Sorption Technology Solutions for Porous Materials

Hello everyone and thank you for joining today's Surface Measurement Systems Online workshop, organizing collaboration with Imperial College London. Today we are exploring the topic Innovative Absorption Technology Solutions for Porous Materials and we'll have three guest speakers with us today to explore absorption techniques and their applications for the study of porous materials. As mentioned, we're really lucky today to have three guest speakers from us, two from the Imperial College of London and one from the University of Edinburgh. So thank you to all of our guest speakers and their guest institutions at the outset for being with us here today to share their findings and their insights. Just before we begin today, I wanted to remind our audience of some of the housekeeping rules, so how you can get the most out of today's workshop. We do ask that all attendees please keep their microphones muted for the duration of the presentations. This is to ensure top audio quality for all of the listeners so we avoid any feedback or background noise. If you would like to ask a question of any of our speakers, either arising from your queries on this topic or queries that arise from their presentations, there is a live Q and A at the end of each presentation and there are two ways you can submit questions for this. If you want to submit your question in writing, I can ask it on your behalf to the speaker. This can be submitted in the questions panel in the control panel to your right. Alternatively, if you would like to unmute your microphone and ask the question yourself to the speaker, you simply need to raise your hand using the control panel and then I will come to you and let you know when you can unmute your microphone to ask the speaker a question. We just ask that. If you do raise your hand, have your microphone set up so that we have as little downtime between meetings prompting you and you unmuting so that we can keep things moving and on schedule. Without further ado, let's begin with today's presentations. We are very lucky with our first speaker to be visiting us. From the University of Edinburgh, this is Dr. Giulio Santori. Dr. Is a senior Lecturer at the University of Edinburgh where he leads the Emerging Sustainable Technologies Laboratory performing research towards the development of new heat power technologies for zero and negative carbon future, including direct air capture. His work ranges from thermodynamic dynamics to actual prototyping and testing using thermodynamics as a necessary framework for the early design of novel processes. He is an author of 52 articles in international journals on energy technologies and thermodynamics and has presented at more than 40 papers at international conferences and meetings. He is a member of the Scientific board of the Heat Powered Cycles Conference series and of the Editorial board of Energy and Environmental Journal and Low Carbon Energy, Environmental and Management Journal. So a man of real expertise. So I'm sure you'll all join me in welcoming Dr. Santori to today's session and I will now pass over to him to give his presentation.

Thanks John, for introducing me and thanks Surface Measurement System for this nice invitation. Just one second to set up.

Everything.

I hope you can see my screen.

Julio, I think you just need to switch the screens again because we've got the two shot. That's perfect. That's great.

Okay, wonderful.

So.

Today I've chosen this topic ionically with ionogel for watersoption in low grade heat driven applications, which is one of the few things we are doing in the Emerging Sustainable Technologies Laboratory in Edinburgh. The presentation is organized in various number of sections. So we are going through the what ionic liquids and ionogels are first the factors affecting water absorption in ionic liquids. So the focus of this presentation is water water solption in ionic liquids first. And then we will move to two types of ionogels and the comparison between ionogels and pristine ionic liquids. So what differences which differences are introduced Once we fix the ionic liquid on a on a solid surface, then we move to the equilibrium properties of ionic liquids and ionogels and we focus on a specific technology which is the utilization of the ionogels in adsorption desalination, which is a heat driven technology. And then we draw some conclusions and we highlight which are the knowledge gaps which still open and should be filled. So the ionic liquid. An informal definition of ionic liquid is a salt. We are talking about a salt in liquid phase. We used to salt in to absorbs in solid phase. These are salts in liquid phase in a wide range of temperatures proximal to the ambient temperature. We use the range of temperature between 25 and 100. So we have a liquid salt in between ambient temperature and 100C and showing extremely low saturation pressure. Essentially they remain even at a very low pressure in liquid in a liquid state. So their aspect is the one in picture here. So they are liquid. They can come in various colors depending on the type of anionic liquid. And our emphasis, the emphasis on this presentation will be especially on the imidazolium ionic liquids with a special interest because this is the interest of my laboratory on immo acetate. We will also understand why we ended up on immune acetate. But in general you will see many references to other types of imidazolam ionic liquid. So the type of ionic liquid showing anemidazolium cation which can come in different version. Then we need to introduce a second definition which is the ionogel. The ionogel is turning the ionic liquid in a solid state. And we can do that. In this presentation we will see two ways to do that. We can do that in many different ways. The two ways we focus on today are by trapping the ionic liquid in a polymer matrix or by supporting the ionic liquid on a nanoporous material. In particular, in this presentation we will see silica supported ionic liquids. Now, the capability of ionic liquid to retain water of ionic liquids and accordingly ionogens to retain water is very well known. And usually this is an unwanted behavior of the ionic liquid. Because the majority of research on ionic liquids, on sorption in ionic on ionic liquids focuses on CO2 capture utilization of the ionic liquids in supercapacitors and where the water is unwanted because it is going to reduce the performance of the technology. What the interest in my laboratory is the opposite. We want to have even large amount of water in the ionic liquid so we can learn from the research done in the past and extend it to have more water. Now, we start from this graph. First of all, this is a collection of equilibrium data. You see the graph show relative humidity. Sorry for the spelling mistakes sometime in the in the presentation. But in any case. So we show the relative humidity on the horizontal axis against the uptake. That is the amount of water absorbed. The way I'm showing here all the data on the beam im, which is another amid azolium ionic liquid very well measured. Where the water sorption is very well measured on a number of different ionic liquids. So you see, I've used here the uptake. I put the uptake on a logarithmic scale just to emphasize the fact that at very high relative humidities, about 80% usually we can reach amounts of water above 1 this ionic liquid. So 1 gram of water per gram of dry ionic liquid or kilogram of water per kg of dry ionic liquid is something outstanding compared to. If we want to compare the water sorption with the normal nanoporous material. All the isotherms are, as you can see here, type 3isotherms. And usually the best way to fit type 3isotherms is by using the Guggenheim Manderson de Boer isotherm. We have used it often sometime in our laboratory we switch to the Dubin Nastrocope because just because it is more popular and it looks easier to use other research group. But for a precise fitting honestly, we should use Guggenheim, Anderson Debar, which can fit the type 3 isotherm much better than other types of isotherms. This is we said type 3 isotherms. Type 3 isotherms are often called unfavorable. But I mean this is not a nice attribute for this type of isotherm because we should remember that it is favorable or unfavorable depending on the specific application. And there are some applications working in this region, in the region of the high humidity, high relative humidity requiring where this performance these shapes of the isotherms are very welcome. So we see that when we compare all these ionically with on the same cation BMINIM cation, what really makes the difference is the presence of the type of the anion. And in particular acetate anion and chloride anion are dominating over all the other types of anion and give the ionic liquid larger absorption across a wide range of relative humidities. We can already rank the hydrophilicity on the basis of the anion from the large amount of water retained to the small amount of water retained. In any case, even the TFSI anion is going to show some minimal amount of water retained and we will discuss later that there is a benefit even sometime in the utilization of this type of anion instead of the large amount of those showing large amount of water retained. So the message in this slide is the type of anion often controls more than other factors watersorption equilibrium such as the other factors might be for example the type of cation. Now a good number last remark from this slide is that a good number of of these ionic liquids here are toxic flammable so are not easy to use in the lab. And therefore this long list of ionic liquids is in the end massively reduced if we apply a constraint on the the possibility to use it without problems in lab. Just let's move now to the next slide. So here I've just focused in the previous slide we have changed the anion, but what happens if we change the cation? Now these three curves show the BIM cation, the IMIM cation. Let me use the laser pointer. So the BMIM cation, the imine cation, blue line and a third cation which is the choline acetate cation. Choline cation in a choline acetate ionic liquid. Now I just want to emphasize that bimin acetate is acute, has acute toxicity and irritant while moving twimmim. It shows only irritable irritation. The third ionic liquid is choline acetate with choline cation. This is still less known than the other two. So all the tests on safety have not ended yet. But in general people say that this might be used also as a supplement, a nutrient supplement. So it looks it might be even edible, although there is no evidence, strong evidence yet on that. So for sure it is going to be more friendly than the other two.

Now.

The special aspect of this ionic liquid, and in particular the special aspect of this cation, is the presence of an OH group on the cation. We used to have OH groups on the anion instead. And this makes a difference in terms of watersorption. As you can see, it plays the isothermic on top of the other two other two cations. So here the message is the shorter is the cation, the more is the water sorbet. Again, I just want to point to highlight that again. In the high relative humidity region, we can heat values of 10 kilogram of water per kilogram of ionic liquid, which is almost outstanding, almost unbelievable. So let's now see how we can fix the ionic liquid on the surface of solids. Or we can trap the ionic liquid in polymer matrix. In general, we call it immobilization of the ionic liquid. We can think of nano confining ionic liquid on nanoporous materials through impregnation. And this is what we have initially done in our laboratory by using a silica. This is silica siloid. Al1FP is a particular silica with high density of OH groups on its surface. When we impregnate 50% by weight of this silica with imine methane sulfonate in this case. This was at the beginning of our research where we didn't identify the acetate anion yet as the best. But what happens is that we have when we impregnate the anionically with on a nanoporous material, we can have a degree or degree of different isotherms depending on the level of impregnation. The limit we can impregnate the silica is 60% of ionic liquid. So the composite material is going to be up to 60% the ionic liquid in weight. It's not possible to impregnate more than 60%. And from the pure silica isotherm we move to the orange, the curve with orange symbol. Gradually we see we have a reduction when we impregnate just around 2% of ionic liquid, a reduction of the isotherm likely because we are going to fill Already the nanopore, the micropores, sorry. And we start to see an advantage above 70% of impregnation in the higher humidity region. And this advantage increases if we impregnate more and more ionic liquid into the matrix. Now, if we are going to check what is through the scanning electron microscopy microscopy to check how the material has changed, we see that the pure silica is evidently a porous material. But when we impregnate up to 60%, this is a different ionic liquid. That doesn't make any difference, because even with a different ionic liquid, the shape, the morphology will be the same. Essentially you are losing all the porosity and you are working with a thin layer of liquid solvent, the ionic liquid, the EDF shows what you have just said. So the pores are partially filled with ionic liquids, but are not actually contributing to to the absorption mechanism. Because when you impregnate at a high level of impregnation 60%, you are basically having a thin film of ionic liquid spread on the surface. Now, one last remark from this slide is that I'm showing here only 25c isotherms. The reason is that when plotted against relative humidity, the ionic liquid and even the confined ionic liquid show temperature independent, so it becomes temperature independent. Mostly of them are temperature independent, so the only temperature dependence is the temperature dependence of the saturation pressure. But when you plot the isotherma gans against the relative humidity, if you repeat the isotherm measurement on several temperature, you will not see a significant variation of the isotherm itself. The second way to fix the ionic liquid is in a polymeric network. And this is meant to achieve higher loadings. If the amount of water absorbed, as we have seen in the previous slide, depends on the amount of ionic liquid that we can load into the material, the first thing one might think of is to achieve higher loadings. And the immobilization of an ionic liquid in a polymer network can achieve. People have reported even loadings of 90%, 90% ionic liquids, 10% polymer magics. We did our measurement on that, measurements on that.

We.

Basically prepared a material which is still 60% amine acetate, 40% PEGDA, which is a polymer. Now, with this polymer here, we couldn't go more than 60% of the impregnation. There was not really an advantage what we have seen. And when you compare the isotherm of this material with the isotherm of the pure ionic liquid and the isotherm of the pure ionic liquid, it's scaled 60% of the amount of water Retained because you have only 60% of ionic liquid. You see that the data collected in the DBS in the DBS instrument, the data point collected in the DBS instrument directly on this material here are essentially the 60% of the pure ionic liquid with a difference at higher loadings, which comes, which is a small difference, which becomes larger at higher level relative humidity, which becomes from the fact that this material shows some porosity and in particular some macroporosity, not in the range of the. Not a nanometer scale, but a micrometer scale, almost, let's say hundreds of nanometer scale, as you see here from the same measurements. So you have some cavities left in the material. And again the material shows some bumps, is similar to the previous one that we have seen. Still you have a film of ionic liquid trapped in the polymer matrix. And these cavities at high relative humidity promote condensation of border. So that would explain the difference between 60% curve and the actual data. So in the end we didn't have. We continue to work on this concept, but it looks so the achievement of higher loadings is still a challenge in our laboratory. Now we have talked about the materials. Now let's see how these materials, when these materials are usable in technologies. What are the technologies in which these materials can be used and can provide some advantage. In particular, we focus in this presentation on technologies which are powered by extremely low grade heat. And in particular we are targeting heat below 60 because this is really wasted. No one is gonna ever Recover Heat below 60C. Nevertheless, there are various technologies, emerging technologies that can recover this waste heat. These are technologies. I'm listing just some options here for drying, for heat storage, even greenhouse gases, capture heating and cooling and desalination, which is the focus of the next slides. Now we are going to talk about adsorption desalination, how adsorption desalination process works. It is composed by three units. The suction compressor, we call it sorption compressor, is the suction bed. Where we have our adsorption material is connected with a condenser and an evaporator. Each one of these two vessels show an interchanger. We feed the evaporator with sea water. We give some heat input and we have some evaporation. Here the evaporation is both because of the heat input and because of the presence of the absorption material triggers and enhances the evaporation. So this paper is first fixed in the sort of material. Then we close this connection valve, as you see here, and open the other valve, we heat up the material and we regenerate the bed and condense the water vapor generated on a cold surface, which is usually at ambient temperature, producing pure water and producing also brine, the evaporator. So the system works according to four phases. We start this is the plot of the temperature revolution of the adsorption material. We start from one, when the adsorption material is cold and full of water. Then we heat up up to the point three until we have complete regeneration. Then we cool down back to one to half the absorption again in a thermodynamic diagram. This is usually visualized in a temperature pressure diagram in which the we can visualize together pressure and temperature. And this is very useful. And also the amount of water retained in the material, which is the isostere. The isostere moves from the high loading to the low loading are straight lines in this diagram here. So again we move from low pressure to to a higher pressure point three while heating and a high pressure, low pressure while cooling, going from three to one. And the pressure is set by the value of the saturation or the evaporation and condensation pressure of the salted water, which is close to the saturation pressure of water. But because of the presence of salt it not actually the saturation pressure of water. Another concept we should keep in mind is that in adsorption desalination we tend to work with the evaporator temperature and the condenser temperature as close as possible. So the delta t between these two vessels here will be it's ideally 1c 5c 10c but we tend to have them at the same temperature. That means that this curve, this straight line is pretty close, goes up compared to the way I've sketched here, goes up and is pretty close. The transformation 41 becomes pretty close to the transformation 2 3, ideally coincident with the transformation 2 3. We should also remember that the concept of relative humidity is essentially the ratio between the saturation pressure, so sorry, a ratio between the pressure of the operational pressure and the saturation pressure of water up to the same temperature. If this comes closer here, so one, let's say it's on a point here, the relative humidity will be pretty high because the pressure that we can read is pretty close to the pressure of 0.5. The message here, the final message is that when we talk about the source and desalination, we should expect a technology working at high relative humidities. So where the adsorption bed works at high relative humidities. Now here some results from our laboratory. When we plot on a real device on A real adsorption desalination. If we plot the performance of the adsorption desalinator on a reduced temperature, which is the pressure between the condenser minus the evaporator and the regeneration minus the condenser, we see that these curves collapse all the data point collapse in curves. And this is pretty useful because the problem the process is defined by three temperatures in general. But in this way we can define the process performance only against one single temperature, one single reduced temperature and in particular in adsorption thisalination on the vertical axis, the value primarily we see we check is the specific daily water production, how much water is produced per day. And you see that when we work at low T red close to zero, then we have high water production. When we are far from zero at higher T red then we have lower water production. There is a correspondence between T red correlation between T red and relative humidity. So low T red means high relative humidity and high T red means low relative humidity. What is the advantage then to work at high value of tred? So if this is the zone defining desalination, this will be the zone defining cooling. So if we want to use the same process for cooling, what is clear from here is that we need to sacrifice something in water production. So there is a trade off between the process can provide desalination and cooling. But if you want more cooling, you need to sacrifice the amount of water and vice versa. Now let's see now what each point in this graph means on the isotherm. Each point on this graph defines two points in each isotherm which is the operational zone. They define the relative humidity in at the soption. Sorry, this has to be fixed. This is at disruption and the relative humidity in desorption state. As you see, you make the difference between these two points. You see that the difference is pretty or in terms of uptake can be pretty. Pretty significant number of units, let's say of grams of water per gram of ionic liquid. And this is really the zone where the, where the, this is really the zone where the adsorption desalination works. So we can expect from these materials great performance. And in fact if we check the same information for immune macetate considering T re 0.07 you see where we have an extremely low regeneration temperature and very close evaporation and condensation temperature, we see that we should expect from the mean massetate a working capacity of 1.5 gram of water per gram of ionic liquid. Fantastic. So let's see if in the real machine we can achieve this value. So, given this analysis, we have built a device an absorption desalination, a very small one. You see, the scale here is 5 centimeters. This is the evaporator that we have seen before. The two is a two bed system, just to make the process continuous. While one bed is absorbing, the other one is desorbing and a condenser. The adsorption beds are pretty small as well. We have tested initially the system with a benchmark material silica gel. And we moved later to a silica supported ionic liquid. And in particular we used in this device immunos date.

Now.

The second issue is okay, we have the material. Now we need to fill the material into the system. The material should be structured. So the way the simplest way to structure a composite material like a 40% silica siloid 72 FP containing 60% of amino acetate is by compressing the material in between the fins. So we obtain monoliths, small monoliths. The material, as you see as soon as when it is dry is powder, which can be compacted. So it's pretty fragile. But is this type of systems are not mechanically stressed. Is enough to work in an absorption desalination system. And it becomes gummy when we have some. You have some content of water. So the stability of the material even increase. And this is its aspect after 300 cycles in the test rig. Now we have measured the again the equilibrium isotherms of this material at 25. And this is the equilibrium isotherm measured before and after 300 cycles. Nothing was changing. With a dopinine NASDAQ overfit following pretty well in this case the material behavior. As you see, we expect this machine working at dry relative humidities and at high relative humidities. We are working in a region of uptakes which is above 1. If we then plot from the Dubin NASDAQ Advisor term the net differential sorption enthalpy. We will see that the by working at high relative humidity, we are really spending just latent heat of water to regenerate the material. The higher energy is the heat of absorption increases when we work at lower relative humidity. But this is not the case. The case of adsorption for adsorption desalination. And this is when compared to the silica gel net differential sorption enthalpy. This is amazingly lower. So big advantage here. When we plot the isotherm, instead of plotting the isotherm against relative humidity, we plot the isotherm now against the pressure. The pressure we see. We realize that we have a significant working capacity, A significant working capacity even at a low regeneration temperature. You see, the cooling is set at 25. The heating of the material is 30. So very just 5 Celsius degree would be enough to give us a large working capacity. So we went to the system, we did our temperature swings of 5C, as you see from the graph here, with the material I've showed before to you. And we could really observe some amount of water produced. Now then we compared the amount of water produced, the specific daily water production. Remember, this is the parameter of the key parameter for adsorption desalination. We compared the results on the specific daily water production for silica gel and the ionogel that we have seen. So in the same device, the silica gel stopped to work with a delta T regenerate. Delta T is here the regeneration temperature minus the condenser temperature 15. While the ionogel could produce water even with 5 Celsius degree difference. And the other important aspect here is the cycle time. So the cycle time, as you can see from this graph here, each temperature swing lasts just five minutes. So four times faster than silica gel, even with lower temperature. So fantastic results. We were extremely happy. Then we went to check why this was going on. Because we have seen also some unexpected behavior of the material. So remember that now the unexpected behavior is linked to the dynamics of the material, the dynamics of the adsorption. Remember that the material works in between two streams of relative humidity. The relative humidity set by, in other terms, the pressure set by the the evaporator and the pressure set by the condenser. So if we check how the uptake evolves, we will see this trend. Here you immediately see that the two limiting values, upper and lower value, moves from 0.3 to 0.6. So the working capacity is only 0.3, which is dramatically lower than the equilibrium value we have seen before. So this, the red line, the black line is the equilibrium uptake. So what we should expect if the bed works at equilibrium. But the bed, the source of material, shows some dynamics. You never achieve the equilibrium uptake. You work with the actual uptake. When you consider the actual uptake, you have even lower working capacity of 0.15. So which is lower than the 0.3 we are set, which comes from the limit set by the evaporator saturation pressure and the condenser saturation pressure. And we can see also another unexpected behavior. So we spend our effort to heat up and cool down the bed. But the advantage that we get in heating and cooling. So this loop here and the slope here in five minutes is really small compared to the total. So the message here is that the change, which happens mostly here, so the curve is going down and then down. So this step is the majority of the working capacity. And this step here as well happens every time we have a change of relative humidity. So is the change of relative humidity that makes really the difference rather than the heating and cooling of the bed. And this has to do with two factors in our opinion, which are matter of investigation. So the dynamics of the absorption bed is short because we need to heat at low regeneration temperature. As you see, as you seen, the temperature we are talking about here is just a temperature difference from sometimes 30 to 40 or 25 to 30, so 5 or 10 delta T. But we don't need to spend time to heat up up to 60 or 80c the bed. And also this material shows a pretty complex adsorption kinetics which is very fast at the beginning and then it slows down. If you want to achieve the equilibrium values that we have seen before. And this happens because when water comes in contact with the surface of the ionic liquid, you have a first instantaneous condensation on the surface of the ionic liquid and then a very slow diffusion inside the ionic liquid film. So we cannot with this material really exploit the equilibrium properties and we should be happy with the fast kinetics. Now if the ionic liquid has to be used for its fast kinetics, then going back to the initial slides, maybe the ionic liquids with lower water content might be good as well in making the task of absorbing the water vapor. Because it's very well known that non protic ionic liquid usually show a very fast kinetics in water sorption of their small amount of water salt. Now let's move to the conclusion. What we have seen in this presentation is that some protic ionic liquids show equilibrium working capacity larger than 1 greater than 1 gram of water per gram of ionic liquid. He needs driven absorption processes operating in the high relative humidity range. In the high relative humidity range, the differential enthalpy of adsorption is coincident with the latent heat of water. And this is good. The corresponding ionogels inherit similar equilibrium properties in the same order operating range similar because they are often scaled. If we load the material with 60% loading. Unfortunately the equilibrium working capacity is inaccessible in practical devices. However, practical working capacity is good enough to align with the competitor materials and the productivity. So the amount of water produced is much improved thanks to the number of times faster dynamics likely from low regeneration temperature and faster kinetics although the kinetics is something under explored yet. And with that I conclude my presentation.

That was brilliant. Thank you, Julio. We have a. We're running a bit behind, but I think we have a few minutes just for any questions. So if anyone has any questions for the doctor arising from that presentation on the topic he covered, please let us know in the. In the questions panel. So we have our first question. Giulio, I may ask, is this an industrial project?

So it started as an Horizon 2020.

Project.

And yeah, there are some industry interested. There have been some industries interested because especially on drying and recently also in desalination, but in general it was funded by Horizon 2020 and also EPSRC. So it wasn't. There was some industry, some comp. There was some company involved that was research and innovation.

Thank you.

Just as we are, we are pressing our time schedule a bit. I think we'll move ahead. But if anyone does have any questions for Dr. Giulio, please submit them and we can try to either take them at the end of today's session if we can, or we will forward him to address them after the presentation. So please do submit those if you have them. So moving on to our next speaker, if you give me a moment, we will be hearing now from Elwyn Hunter Sellers. Elwin Hunter Sellers has completed his Master's degree in chemical engineering at the University College of London in 2017, studying his final year abroad at Columbia University where he specialized in surface phenomena and material science. He is currently a PhD candidate at I Imperial College London, researching the impact of adsorption competition on the performance of porous materials in air pollution abatement. His work combines equipment development, material characterization and synthesis of structured MOFs for improved filtration efficiency. He has applied these skills to several research partnerships within fields such as VOC filtration from indoor air carbon capture and heritage preservation science. So without further ado, I would like to thank Elwin for being with us to share his presentation with us today and I will hand over to him now.

Hi everyone. Hopefully you can hear me. All right. As I said, thank you very much. My name is Elwyn. I'm going to be speaking to you today about the impact of relative humidity on hydrophobicity and the selectivity of a number of different porous materials for VOC capture.

Hear and see you.

Okay?

Perfect.

Okay, so volatile organic compounds are a. They're a broad class of chemicals which are typically characterized by high vapor pressure and low boiling point. They have a variety of chemistries, so polar, non polar, aromatic, aliphatic, they have chlorinated groups, they can have pretty much anything you can imagine. And if you worked in a chemistry lab, you know, the sort of chemicals that I'm talking about, things like toluene, ethanol, formaldehyde, these are emitted from a variety of sources. So the typical ones, like chemical and petrochemical industries, but also sources you might not expect, such as building materials like woods and paints and varnishes, and also a number of consumer products, so fragrance, candles, cleaning products like bleachers and detergents. And because of this variety of sources, both in outdoor air and in indoor air, you can end up with fairly elevated levels of these VOCs. And this can actually contribute to a wide variety of health problems. So there are, in extreme cases, there are acute conditions like asthma, increased risks for heart disease and cancer, but at sort of elevated but not extreme levels. VOCs are associated with sick building syndrome. So anyone who's worked in an old office building for an extended period of time probably knows about this. You get things like dry eyes, dry throat, headaches, nausea, and this is due to these elevated levels of harmful compounds. So if we want to improve the quality of life and the sort of health prospects for people in these affected areas, we want to try and reduce the concentrations of VOCs. And one of the main ways we do this is through adsorption using porous materials. So porous materials are, again, a fairly broad class, but they're characterized by having a high degree of free space or poor volume, and they typically possess high surface areas. So in the natural world, examples of porous materials are things like wood, dirt, and clay. They're used extensively in industrial applications, Things like zeolites, charcoals, and silica gels. And more recently, there's a lot of work on different synthetic forest materials like polymers and also metal organic frameworks, which I'll be mentioning a little bit later. The great thing about porous matures is they're fairly cheap, and it's fairly easy to understand how they work. They essentially. But because they have a large amount of free space and surface area, there are a large number of sites for VOCs, or carbon dioxide or whatever you're interested in to absorb onto. But one of the problems is that there is a limited number of adsorption sites. You can only have so many of these sites filled. And so in applications in indoor air, for example, you have a significant amount of competition from water vapor. So relative humidity indoors and outdoors will often BE at concentrations 4, 5, maybe 6, 7, or 8 orders of magnitude greater than that of VOCs. So VOCs are normally present at part per billion and part per million concentration levels. And so if we're designing or selecting a material for removing VOCs we need to make sure they have significant enough selectivity to absorb the VOCs while not interacting significantly with water. And so that's what we're going to be investigating today. So in this presentation I'm going to be going over three ways that we assess adsorption hydrophobicity. The first way, and the simplest way, is to derive hydrophobicity indexes using single component gravimetric absorption data. So using adsorption isotherms generated using the DVS Resolution and the DVS Endeavour, we can also use these two machines to carry out wet adsorbent gravimetric experiments, which is where we're going to adsorb a voc, in this case toluene, after we pre expose our samples to a certain level of relative humidity. So that way we're giving the water ample time to sort of compete with the VOC before it's exposed. And finally we're going to look at wet adsorbate. So it's the, it's the gas phase that's wet, in this case wet absorbate. Breakthrough curve experiments where toluene and water vapor are absorbed simultaneously in a fixed bed column of our material. And we tested these materials out on a number of industrial adsorbents that are typical in the industry right now, and also a few promising high density metal organic frameworks that were also synthesized during this project. So the initial stage to this project we worked on screening a number of industrial adsorbents. So we looked at amorphous silica, so silica gel, activated charcoal and three different kinds of zeolites. We have hydrophilic molecular sieve 13x and high silica forms, so hydrophobic forms of zeolite Y and ZSM5. We characterize their physical properties using nitrogen absorption. So we're looking at ETCEPs area, the level of microporosity in the material and their pore size distributions. We use techniques such as x ray fluorescence and contact angle for measuring the silica alumina ratio in the zeolites, which directly affects the surface chemistry of the material. And finally, we use DVS vaporized therms to calculate the capacity and the absorption kinetics these materials have for a variety of different volatile compounds in the vapor phase. So here are the nitrogen isotherms for the five materials. I'll quickly go over what their names mean, just because it's going to be used a fair amount so AC is activated charcoal, as is amorphous silica Ms. 13x is molecular CIV 13x and A88Y and A14Z are zeolite Y and Zsyn5. So all samples were found to have high BET surface areas, typically above 300 meters squared per gram. And the highest observed was at vedicle, which was around 1200, followed by zeolite x or molecular sieve 13x and zeolite y. All samples were found to have sufficiently large pores. So these are the pore size distributions here to absorb our largest target voc, which was toluene. So each material has a pore size of at least around 5.9 angstroms. And we found that there was additional MESA porosity in our zeolite Y and ZSM5 samples compared to the zeolite X sample. And this is due to these materials undergoing something called DE illumination, which is where the aluminium ions within the material and within the framework are removed. And this can lead to some changes in porosity. We also confirmed the DE illumination process in X ray fluorescence experiments. And we see that the if you look down at the bottom, we see that the silica alumina ratio in the zeolite Y and ZSM5 is much higher than that of molecular CIV13X around 30 to 35. And so DE illumination is carried out to remove or reduce the amount of aluminum in the zeolite framework. But this can lead to an increase in non framework amorphous species and also cause the aluminum to migrate and form defect sites. And it can also cause some amount of microporous collapse, which actually can be beneficial in some cases because you're generating mesopores, which often can improve your diffusion processes. But the main reason that we have DE illumination is that it enhances the hydrophobicity because metal ions such as sodium, calcium and aluminium contribute to the overall acidity of the zeolite, which improves its interactivity with water vapor. So that's a bit of background why the illumination is taking place. And we can see that this hydrophobicity, at least surface hydrophobicity, has been increased in the zeolite Y sample since it was the only zeolite found to actually present some sort of water contact angle, in this case around 73 degrees. And the other hydrophobic material, activated charcoal, also had a contact angle. Although these are fairly low and aren't really considered hydrophobic by any means, they are at least higher than the other materials we tested in which water was found to wet the surface immediately. So moving on to Vapor isotherms. You've seen some isotherms before, so I'll assume you know what we're measuring here. But we can see that for toluene and hexane we have fairly similar isotherm shapes and these largely follow the trends we were seeing in nitrogen uptake. So nitrogen, toluene and N linear hexane predominantly interact through non polar or dispersive interactions. So they don't contain any significantly polar groups. And so there's little to no impact of surface chemistry on the absorption performance. And it's largely dictated by the microporosity and the BT surface area of the materials. So at a partial pressure of around half a percent, so this is around 180 to 200 ppm of toluene, we found that ativid charcoal the best, with an absorption capacity of around 260 milligrams per gram. So that's around 25% uptake by mass, followed by the two forgesite zeolite, zeolite x and y and then ZSM5 and amorphous silica. So in the case of ZSM5 and amorphous Silica, these samples were found to be predominantly mesoporous. So although they had a high surface area, a large amount of their free space was located in larger pores. So as your pore size increases, it requires a larger partial pressure or a larger driving force to fill those pores. So if we're working in applications such as VOC capture, where we're at low concentrations, you can imagine that the smaller the pore or the smaller the pore with significant pore volume, the better the capacity. And we do see that with activated charcoal and molecular sieve and zeolite. Yeah. With polar vapors we begin to see some more interesting results. If we look at the orange plot, which is activated charcoal, we can see that as we move to more and more polar species, the essentially the gradients of the isotherm become less and less steep, so it's absorbing less of the compound as polarity increases. This is due to obviously at better charcoal being a hydrophobic adsorbent. And so as you increase polarity, the interactivity there decreases and this becomes more and more significant. As in this case we've shown the dielectric constants, but pretty much any metric of polarity will show that interactivity with activated charcoal decreases with polarity. And you may have also noticed in the blue we actually see a shift in the isotherm shape. So for zeolite Y, this the hydrophobic effect is Much more significant. And we see. So these are plots of toluene, isopropanol, ethanol, methanol and water. And so zeolite Y, due to the framework aluminum being removed, it has this sort of hydrophobic electrostatic adsorption screening effect. So as you increase the polarity of your molecule, larger and larger driving forces are required for these molecules to penetrate this barrier and enter the micropores. So this initial step in the isotherms of the alcohols can be seen as adsorption purely onto the surface of the zeolite. So this is likely on those defect sites of aluminum that I mentioned previously. And once a certain driving force is reached, and we generally find that the higher the polarity of the molecule, the higher the driving force required. Once this driving force is reached, it's able to penetrate the barrier and fill the pores, which is shown by a large increase in capacity. So this is a type 5 isotherm. It generally means that you have initially unfavourable interactions followed by favorable. And so this, this hydrophobic effect can also be seen in the water isotherms. It has very low capacity for water vapor pretty much at all levels. It shows type 3 isotherms, which again shows unfavorable absorption capacity. But this plot here, combined with the hydrophobicity index we calculated, which I'll show in a second, highlight the importance of the adsorption environment on the hydrophobicity of a porous material. So even though activated charcoal, which you can see in orange, has incredibly low capacity for water vapor at low relative humidities or low partial pressures, it does actually increase fairly dramatically past 40% relative humidity. And in fact, at 80% relative humidity, it was found to have the highest capacity for water vapor out of any material, despite supposedly being hydrophobic. So to capture this change in hydrophobicity, we calculate a static hydrophobicity index, which is calculated using a fixed quantity or a fixed parameter. In this case, it's the uptake of toluene at a partial pressure of half a percent. And we divide this by, by the uptake of water vapor by that material at different relative humidities. So in this way we can very easily illustrate the change in hydrophobicity of the material as relative humidity increases. So at low humidities, we see activated charcoal as far above all other materials. But shown by the inset plot, we can see that zeolite Y begins to overtake activated charcoal at higher levels of relative humidity. So this shows us that the mechanistic, that there could be mechanistic or structural changes. So in terms of the adsorption mechanism for water vapor. So changes in this adsorption mechanism can directly influence the overall hydrophobicity of the material. So to test whether these these hypotheses were correct. So basically whether Z Lite Y would outperform activated charcoal at extreme levels of relative humidity, we we calculate or we carry out wet absorbent tests. So in these experiments shown by the plot here, we initially dry the sample. We then expose it to a certain level of relative humidity. Once that sample has reached equilibrium, as you can see on the plot here, we can calculate the uptake of water vapor. And then while maintaining the same level of relative humidity, because we want to ensure that there's no desorption occurring there, we introduce a step change in toluene concentration, in this case half a percent of partial pressure. And in this way we're able to essentially pre expose to relative humidity while also measuring the potential uptake a material has while exposed to these high levels of water. And this is done using the dual solvent mode of the DVS endeavour. So this lets us control partial pressures of two different vapors, but it can also be extended to water and CO2 in carbon capture applications. So following this pre exposure, we see that hydrophilic materials like especially like Molecular Civ 13X have a significant drop off in capacity for toluene after being exposed to water vapor. And because these materials have a high amount of acidity, they have a high amount of hydrophilic interaction chemistry. Water vapor is able to occupy all available adsorption sites and not leave any leftover petulluen. When we look at hydrophobic materials, we do see some decrease and this becomes much more significant at high levels of relative humidity, which is especially noticeable for activated charcoal. So despite having a high capacity at 10% and 30% relative humidity, the capacity quickly drops off. And this is actually quite interesting because it matches what we were seeing in our hydrophobicity indexes, that as water partial pressure increases, the supposed hydrophobicity of the material does also decrease. And again, similar to the hydrophobicity index calculations, Zeolite Y was found to actually have a significant amount of toluene uptake at all partial pressures of water, which is very promising. This means that our single component isotherm data was actually able to give a fairly good prediction of what performance would be like in two component systems. And it also provides an easy to visualize data set. And it's fairly quick to collect these data sets. However, because we're working purely gravimetrically, we're lacking any definitive information regarding the mechanism of competition. So although we can semi quantitatively, semi quantitatively determine how much competition is occurring, we're not sure how it's happening. And so to do that, we wanted to carry out breakthrough curve analysis experiments. So I'll go a little bit over how these techniques work. So a breakthrough curve analyzer consists of three components. The first is generation. So that's where we generate a certain headspace. We generate a certain gas stream for our material to absorb. This could be water plus a certain VOC like toluene. You could do water and CO2 depending on your sensors. You can pretty much imagine any system that you'd like to generate. You then pass this through a column packed with your absorbent material and you measure the outlet concentration of these species afterwards. So initially your column is able to absorb whatever is exposed to. And so you initially see no concentration in the outlet. Eventually these species can break through just where the name comes from. And by looking at the profile and integrating downwards from the inlet concentration, you can get an idea about capacity and also the efficiency of the packed bed. And so by using a number of different probes, we're able to independently measure the concentrations of different vapors and gain quantitative understanding about how they're interacting. And so for this project, we built our own breakthrough analyzer with facilities in Pearl College, London, and also at Surface Measurement Systems. So we used a DVS mass flow controller manifold for generating concentration. We have a low sample volume absorption column. This is very important because a lot of the MOFs and metal organic frameworks that we're working with are either difficult or expensive to synthesize. So we want to use as low a sample volume possible. And we can measure our outlet concentrations using a retronic humidity probe, a photoionization detector for VOCs, and an infrared probe for CO2. So we carried out some dilute toluene adsorption experiments. So these were carried out at 10 ppm of toluene, which is very low. This is much more representative of indoor air, and it's around 18 to 20 times lower than what we were using in the DVS endeavor. And we found that all materials, if you can, you can. If you basically integrate downwards from a relative concentration of one, you work out your capacity. The capacities were generally in the range of 50 to 100 milligrams per gram. This is because all the materials we're working with had a high level of microporosity. And we can see that certain curves have sort of a steeper breakthrough curve. Generally, the steeper the breakthrough curve the better the mass transfer through the column, and that generally represents a higher column efficiency. So we found that hierarchically porous materials like zeolite Y that they charcoal had better column efficiency. But the more interesting data is when we obviously expose it to humidity. We see that with zeolite wide, there is an initial drop in capacity as we go to 40% relative humidity. This is likely due to these surface defects I mentioned previously. So there will be some overlap between where water and toluene can absorb. And so as water occupies those more hydrophilic sites, toluene cannot. So you see some drop in capacity. But as we increase to higher levels of relative humidity, zeolite fire was relatively unaffected, maintaining around 80% of its dry capacity. In the presence of water vapor with activated charcoal, we actually see similar trends to what we saw in gravimetric experiments. We see that these breakthrough curves shift towards the left. This means not only that we're basically detecting these vapors earlier, but also that the capacity of the bed is decreasing. And this is Fairly significant of 40% and very significant to 80% relative humidity. However, in both of these cases, we do not see displacement, which I'm going to be going over in the next slide. Which means that water vapor is likely competing by saturating the adsorption bed before toluene can. So it's able to essentially pass through the packed bed much faster due to superior mass transfer. An interesting data set here is molecular sieve. So we can see that in the blue curve, the outlet concentration of toluene is actually exceeding the inlet. So it's getting to relative concentrations of around 1.8, almost 2. What this means is that as water vapor is passing through the bed, because of the hydrophilic chemistry of the zeolite, water is able to preferentially absorb onto these adsorption sites and kick off the toluene molecule. So this displacement interaction corresponds to very significant competition. And we found that molecular SIF had pretty negligible performance. So you can see that in the normalized time values, the black curve, it took around 1500 milliliters per milligram. So this is around a day. Actually, if we, if we normalize based purely on time, took around a day to actually break through, Whereas we can see that it's around almost a thousand times. Sorry, almost. Almost 100 times lower in the human experiment. So we believe this may be due to the, the poor filling mechanism of water vapor leading to significant competition. And we wanted to investigate this. And for this we designed two monolithic metal organic frameworks so metal organic frameworks or MOFs are porous materials synthesized using a metal center and some sort of organic ligand which goes on to attach to the metal center. And this continuous reaction leads to a three dimensional porous structure. So we synthesize xiphate, which is a zeolite like moth with small apertures and very hydrophobic chemistry, and also MIL 100, which is a micro mesa porous moth. Both materials have a high degree of micropore volume above 0.5 cm cubed per gram, and surface areas in the range of 2000 meters per gram. We also synthesize these as monoliths. So this is a brand new procedure we came up with. So as I mentioned, the MOFs are synthesized by organic ligands interacting and bonding to metal centers. By introducing another ligand, a competing ligand, in this case butyl amine, it's able to interact with the metal centers and essentially restrict or limit the moth crystal growth and the final size of these crystals, which results in a sol gel macrostructure, which when dried very gradually or very slowly, results in high density monoliths, which you can see on the bottom. This does reduce the crystallinity and also the porosity of the final material. But when you take into account the superior density of the material, so in some cases the density increases by three or four times the volume relative porosity and the volume relative performance can often be either the same or even higher than the powder. So these are the water isotherms for the two moths. On the left we see XIF8, which is described as being hyper hydrophobic. So this has a, quite frankly ridiculously low capacity for water vapor below 30 milligrams per gram in all cases. And then in the case of Mil 100 we see a very interesting water isother. So these inflection points here represent essentially pore filling events. So initially water absorbs purely by surface absorption. It's then able to penetrate the hydrophobic barrier of the smaller micropore, followed by penetrating the barrier of the larger micropore and able to fill the small and large MESA pore apertures, respectively. So these pore filling events we believe will give us a lot of insight onto how the absorption mechanism for water vapor will actually influence the competition, the absorption capacity for VOCs. So humid breakthrough curves with ZIF8 we see similar to zeolite y, very limited competition occurring. But we also see a. You might notice that we have a much less steep breakthrough curve and so this reduces the overall column efficiency. This is likely due to the slow diffusion of aromatics into the small micropores of XIF8. So although it has a flexible pore structure, its pores are still small enough to restrict adsorption and again, negligible loss in capacity due to this hyperhydrophobicity. And this is probably ultimately, probably the most interesting plot out of my entire PhD. This is the Toluene Adsorption Breakthrough Curves on Mil 100 at different levels of relative humidity. So this is really interesting because it tells us quantitatively that the adsorption mechanism directly affects the competition mechanism. So at 20% relative humidity, again this is before pore filling occurs, Water is only adsorbing via surface absorption onto uncoordinated ion defects and the micropores and mesopores are largely unfilled. So we see the breakthrough curve shift to the left, but we don't see any displacement occurring when we move to 40% relative humidity, which again is when we're filling the smaller of the two MESA pore apertures. Water can access the pentagonal micropore windows and it leads to displacement of toluene. So we see the concentration exceeds 1. And at 80% relative humidity, water is able to access both cages and leads to negligible toluene uptake and also much more significant displacement. So this was really interesting. It basically showed us that our hypothesis regarding the adsorption mechanism was correct and that pore filling is an essential adsorption phenomenon. So, conclusions GRAV metric experiments. So experiments using DBS can provide really useful information regarding hydrophobicity. It's easy to understand, especially when calculating these hydrophobicity indexes I mentioned earlier. And we can also generate wet adsorbent selectivity, albeit semi quantitatively regarding the mechanism. Fixed bed experiments allow for determination of the wet adsorbate selectivity and also really importantly, lets us understand what kind of mechanism is happening in the competition between toluene and water. Hydrophobicity, which is often sort of assumed material property should not be assumed. It's not immutable, and it's dependent both on the material that you're working with and its physicochemical properties and the environment. So as you increase relative humidity, you can't assume that your material will stay hydrophobic. And finally, poor filling by water vapor appears to have a significant impact on the adsorption mechanism and also on the competition mechanism compared to simply surface saturation. So I'll quickly go through the acknowledgments. Thank you to Daryl and Ivan, my supervisors at Imperial College and ucl. Thank you to the spell group at Imperial and surface Measurement Systems for Helping out with a lot of the experimental design and also for helping to build the breakthrough curve analyzer. And a special thanks to Paolo for helping so much with designing and building the machine. Thank you to the CDT and Advanced Characterization of Materials and Procter and Gamble for funding the project. And thank you to all of you for listening. And if you've got any questions that I don't manage to answer in the session, you can contact me at my email there. Thank you.

Brilliant. That was fantastic, Elwyn. Thank you very much. We do have a question or two submitted and we've got a little bit of time, so we'll just get through a few of those. So, Elwyn, David Farasang asks. Well, first he says great job. And he asks for the breakthrough measurements, are the powders sieved? And if. Yes, by what fraction or are they used as fine powders?

Yeah, so good question because obviously if we're working with powders, we want to make sure that we're not using fine powders. In the case of the. So with molecular Civ 13X, it comes already as a. I think it's. I want to say it's 45 mesh, but generally we sieve all the powders I've sieved to be between. I think it's 250 micrometers to 180 micrometers. With the monoliths, they're much larger, so we can actually generally go to much higher sizes. But yeah, with the powders, we do sieve them to between 250 and 180 micrometers just so we don't have too much pressure drop in the system. Yeah.

Thank you. We'll just, we'll just wait a moment or two to give anyone else the opportunity to submit a question or raise their hand.

May I ask one question?

Of course.

It's regarding the dielectric constant that slide when you show the isotopes with the dielectric constants. I wanted to know if you had you performed any correlation between the isotherm and the permittivity of the materials and what the dielectric constant tells in your experiments in the end.

Sure.

So in the initial stage of the project, we were planning to try and relate. We weren't necessarily looking at permittivity. We were going to look into essentially sort of an activation energy for diffusing into the pores. We were going to try and do some modeling work with some people in our, in the chemical engineering department. Nothing ended up happening with it, which is quite unfortunate. And we, we are interested at some point to try and look into this because some of the experiments we did with Zif8 have similar results. So we do find this, this sort of poor sieving effect based on polarity. The dielectric constant was really just used as a metric to say we're increasing in polarity and it requires a larger driving force. But if you know anyone who can work with permeativity modeling or permittivity calculations with an influence of polarity, that would actually be really interesting because it is something we want to explore.

Yeah, it's something we, we want to explore too, in our laboratory.

Okay, I think I see a question in chat. Can you explain the coadsorption with the breakthrough analyzer? Yeah, yeah. So I'm not sure what you want to know exactly, but essentially because we're exposing. I'll go through to the. So because we're able to monitor these breakthrough curves for multiple species simultaneously. So for example, the PID or the photo ionization detector we're using is humidity resistant and at low concentrations the humidity probe doesn't detect the voc. And so because we're able to generate these breakthrough curves for multiple species simultaneously, essentially every curve is independent. So we can actually integrate downwards and calculate the capacity in for each system individually. When we're looking at these competitive curves, what we need to do is we basically, we integrate downwards from an inlet concentration of 1 and we subtract this displaced, this basically displaced area. And so basically by. I haven't shown the water breakthrough curves on here as well, but we do have those. So essentially what you'll end up having is you'll have a toluene breakthrough curve that has this displacement occurring and you'll also have a water breakthrough curve. And by integrating the areas of these curves separately, you can work out the sort of co absorption capacity of each material. But they're essentially treated as separate systems because they're calculated using separate probes.

Thank you. I think we have time for just one last question coming from Alan Connolly and he asks the final application is to remove volatiles in a building. What impact would changes in ambient relative humidity have on volatile absorption desorption?

So I guess if you're working with a hydrophobic material that's got high resistance to relative humidity or partial pressure. So if you're working with something like Zif8, you'd hope that a change in ambient humidity isn't going to have a significant impact. If you're working with a material such as MIL 100 or something that's sort of supposedly Hydrophobic, but you may see pore filling events or you may see changes in hydrophobicity with humidity. You'll likely see some desorption of VOCs as you increase humidity. So we can see that fairly clearly in mil 100. But also relative humidity has like a variety of effects on indoor air. So it's not only going to affect your adsorbent so what's meant to remove the VOCs, but as you increase relative humidity or decrease it, the emission rate of VOCs from things like furniture or paints or varnishes is also going to change. So as you change relative humidity, it's not a simple case of a fixed level of voc, the VOC concentrations will also change. So it's a very difficult environment to understand. But in terms of the porous materials adsorption capacity and its selectivity, you'll definitely see changes. And if you see certain increases in relative humidity, you may see desorption from the porous material, which is obviously not what we want to see, which is why we're sort of emphasizing having a material with hydrophobic pore structure over a wide range of relative humidity. So you can kind of, it's kind of resistant to these unpredictable changes.

Brilliant. Thank you Elwyn. We do have a few more questions, but unfortunately we are going to have to move on just to make sure we're not running too over schedule. But we will endeavor to get back to you on your queries after the, after today's session. So thank you once again, Alwyn.

Thank you.

Moving on to our final speaker for today, we are very lucky to have Paola Cerniz Cavazos Cavazos with us. Paula is a PhD student at the Imperial College London in the Department of Chemical Engineering. And Paola works on carbon capture and utilization as one of the potential solutions for climate climate change. After pursuing a degree in chemical engineering with a minor in Research and innovation and completing several internships focused mainly on lab work and related to sustainable topics, she decided to pursue her interests in more depth. Paola obtained her full Seneca Konak City. Apologies if I mispronounced that. Scholarship to do a Master's in Advanced Chemical Engineering at the Imperial College London. And her Master's studies gave her state of the art knowledge in the areas related to energy sustainability, climate change, carbon capture and product characterization. One of Paola's main interests is to develop research with potential commercial outcomes that can have a positive impact on the environment and the quality of life. And may I add on my own note, Paola has been very helpful with myself from a scientific perspective on several projects here at Surface Measurement Systems, so I'd like to extend my own thanks.

To Paola on that.

So without further ado, I'll hand over to Paola to give her presentation.

Okay, excellent. Thank you for the introduction, John. And I'm going to talk to you today about novel porous materials for carbon capture, utilization and storage studied under relevant industrial conditions. So, so let's start with what is carbon capture? So carbon capture and storage has long been seen as one technology with the potential to reduce greenhouse gas emissions, and the basic idea is to collect carbon dioxide captured from fossil or biomass fuel, power stations, industrial facilities, or directly from air, transport it, and then confine it underground in geological formations. Alternatively, the Capture CO2 can be used as a feedstock to create commercial products such as fuels, concrete, materials and chemicals. So the technologies used for capture, transport, storage and utilization are outlined in this table, which provides a snapshot in time of the level of maturity of a given technology with a defined scale, which in this case ranges from 1 to 9, with 1 being the less mature and 9 being fully commercialized. This method was originally developed by NASA in the 1970s and it is now widely used by research institutions and technology developers around the world to set research priorities. So, as you can see, so as you can see, the post combustion capture technology which is commercialized nowadays it's only amine adsorption and post combustion absorption which uses solid sorbent is currently in a demonstration stage. So that's TRL7. And the large scale projects that use adsorption technology with solid sorbents are very few compared to the widely used liquid absorption technologies. And one of the main reasons for this disparity is the wide range of potential solid adsorbent materials. So this makes the screening process tremendously challenging. So, just to name a few, there are different types of solid solvents. The most typical ones include zeolites, carbonaceous materials such as activated carbon or thermapor silica, and metal organic frameworks. So metal organic frameworks are considered like the novel family of soil sorbents used for carbon captures and other applications as well as Elwin mentioned before, and they show high porosity and they can be tuned or functionalized in order to ameliorate their CO2 uptake capacity and their selectivity. What is more, MOFs can be used as supports for catalysts or act as catalysts themselves to further convert CO2 into other valuable products such as fuels or chemicals, and therefore understanding uptake selectivity and transport diffusion Limitations could very positively influence this technology in advancing their technology readiness level. So currently MOF based catalysts are in proof of concept. So lab test is that technology. That's technology readiness level number three. So really the question is how do we move forward and how am I applying my research in order to do so. So for adsorption technology we have three different pillars. First we have process scale modeling which is necessary to assess the techno economic feasibility of the process of industrial scale. Then we also need molecular simulation which has shown to greatly accelerate the piece of identifying promising materials for CO2 capture. However, most of the models that currently exist only use a single material property at a time and most of them don't take into account impurities present in the flue gas such as water vapor or acidic gases. Which leads us to the bottom pillar and the one that I am focusing on. It's fast and accurate characterization techniques that provide crucial experimental data needed in order to help build and validate both molecular simulations and process scale modeling. So today I will present to you how we use dynamic perception technique as a single and rapid characterization technique to screen novel materials. And that it also considers several real world industrial conditions altogether. So in order to highlight the tunability and bifunctionality of MOPS, we chose a fluorinated MIL101 as our material of study. This material was synthesized by our collaborators from Mexico City and it is a result of the successful partial fluorination of MIL101 chromium. So here we see the BET surface area, the pore volume and the selectivity for both the pristine form and the fluorinated mil 100. As you can see, the BET surface area and the pore volume was reduced by the incorporation of fluorine atoms. However, the CO2 selectivity increased also. By incorporating fluorine atoms they effectively increase the hydrophobicity of the material which I will show you later in the isotherms. So we used a DBS resolution from surface measurement systems to measure and analyze adsorption and CO absorption of low concentration CO2 and water on the enhanced material and the pristine form. All the experiments were done at ambient pressure and temperature in an attempt to replicate real industrial conditions. So the DDS resolution is a gravimetric instrument that uses an electronic balance which is fitted to a stainless steel housing and everything is inside a enclosed temperature system maintaining the temperature uniformity and a series of mass flow controls allows gas mixtures containing dry air, CO2 and water to be prepared and flow over the absorbent sample. And the relative humidity and CO2 concentrations are measured in real time with capacitance and speed of sound sensors. So we reactivated the samples. We used around 10-30mg of samples, we placed it inside the sample pan and sufficient time was allowed for these samples to reach mass equilibrium. At each of the user selected gas phase concentration until no significant mass change was detected. So basically until equilibrium was reached and at the same time the equipment recorded the change in mass versus the time, which is what you can see in figure number one using the DBS software. Then for the multi component experiments or the coadsorption experiments, which is very similar to what Elwin did, it's only that he did it on the high throughput machine. So in the endeavor we used the dual solvent mode of the DBS resolution and after activating the samples, they were exposed to a selected range of pressure of water vapor to simulate the flue gas composition. So we choose 0.5, 0.1, 0.15 and 0.2. And while this pressure was held constant, then we introduced a step in the CO2 concentration of 5%, so 0.5 bars. And then this was followed also by destruction of CO2 and finally by the desorption of water vapor. So some of the, just to name some of the advantages of using DBS is that we only need a very low amount of samples. So usually 10 to 15 milligrams are needed. We can use ambient pressure and temperature conditions and we can replicate flue gas, CO2 and water composition. It's also a relatively fast experimental measurement and it it's close to real world conditions or tries to imitate that. So these are the single component results. So Figure 3 shows the water absorption and desorption isotherms for the functionalized material. We can see that the shape of the isotherm corresponds to that of a mesoporous solid or a type 5 where multilayer adsorption is observed at low partial pressures. So from 0 to 2 followed by a clear hysteresis loop between 0.2 and 0.6 that can be associated with capillary condensation taking place in the mesopores. And the maximum water uptake for the fluorinated material was around 50 millimoles per gram, which is lower than the one reported for the unmodified material, which is around 55.5. So proving an effective augmented overall hydrophobicity that can be attributed to the incorporation of fluorine atoms. Figure 4 shows then the results for CO2. So those are the isotherms for both materials. And as you can see the uptake is quite similar up to pressure pressures 0.6, but at higher partial pressures the pristine material shows a slightly higher CO2 uptake and this can be attributed to the lower pore volume of the fluorinated material compared to the normal mil101. So but however since we're looking at post combustion flue gas applications, we're only interested in the partial pressures ranging from 0.05 to 0.2 where basically the uptake of both materials is the same but the fluorinated material exhibits a higher hydrophobicity for the multi component results. As you can see, figure 5 shows the CO2 absorption capacities of both materials. This was at 5% concentration of CO2 with different water loadings that replicate the amount of water that is usually present in the flue gases is ranging from 0 to point 2 or 0 to 20% relative humidity. So for the non modified mil 101 you can see that the percent of water is completely detrimental. So as you can see the CO2 capacity decreases with increasing amounts of water vapor up to the highest value studied. So in uptake decreased around 18% at a relative humidity of 20 compared to the dry material. On the other hand, the fluorinated mil 101 or the functionalized material shows a CO2 optic enhancement at low and moderate water loadings reaching a maximum uptake of around 0.1 millimoles per gram at a partial pressure of naught 0.15. Nevertheless, at the highest partial pressure study the CO2 uptake of the functional as a material decreased 22% compared to the dry material. Now this could be because during activation of the material some terminal water molecules might be lost. However, the complete loss of these water molecules means that some favored CO2 absorption sites may be lost to well and by increasing the relative humidity values some of these adsorption sites are recovered and therefore the uptake improves. However, there's a value of relative humidity where the presence of water molecules has no longer a positive effect and rather these molecules compete for adsorption sites with CO2. This phenomenon has been previously reported for other MOFs such as MIL100OIO66. So it's important to mention that the CO2 uptake of the materials studied is not exceptional under real industrial conditions. However, they are both very stable when exposed to moist environments compared to other commercial materials that are usually used for CO2 capture. So for example we have HKURST and MIL53 which is also a flexible MOF as you can see in figure 6 and experiments were carried out exactly under the same conditions. And you can see that their CO2 uptake at the highest partial pressure, so 0.2 decreases by 99% and 80%. So 99 for age cast, so almost nothing and 80% for mil 53 compared to the dry pore, obviously. So then we use the DDS data. We can also use the DBS data to calculate the diffusion coefficients. However, because we are using a gravimetric instrument, we need to be very careful in order not to make mistakes. So that's why we set the instruments. During the absorption and desorption experiment, specific settings were selected in order to minimize heat transfer and external mass transfer resistances. So we have to do like three checks. So first we have to verify that the uptake experiments were slow enough so as to not mistake mass transfer with instrument response time. So here Figure 7 shows the gravitational measurements of 5% CO2 in the fluorinated MIL 101 for different sample masses. The curves have a time scale in the order of 250 minutes and according to the manufacturer. So in this case surface measurement systems, the response time of the instrument is less than a minute. So mass transfer kinetics can be measured accurately by assuming the drag force and the buoyancy effects reach steady state almost immediately after that. Extra crystalline diffusion and heat transfer control can also be avoided by changing the sample mass and obtaining consistent results, as you can see in Figure 6. And finally, small amounts of samples were used to avoid bed resistances and heat transfer limitations. We assumed spherical particles for both materials and determine the CO2 diffusion coefficient in the presence of different water vapor concentrations using Fick's second law of diffusion, which is generally a good model and a good approximation for slow diffusing systems. So in this slide you can see the diffusion coefficient results a small increase in water. So when we go for the dry form to 0.5% in relative humidity, this results in an easier transport of CO2 because water molecules absorb on the most attractive sites for adsorptions, e.g. the open metal sites of the MOF, providing a more homogeneous surface and consequently allowing CO2 molecules to move more freely. This phenomenon is clearly observed with the diffusion coefficient values increasing in two orders of magnitudes. For both materials, yes. And then for partial pressures between 0.05 and 0.15, the diffusion coefficients of the two materials remain almost unchanged, while for high water partial pressures. So at 0.2 the diffusion coefficient drops. So very interestingly, very interestingly, the decrease observed is more prominent on the fluorinated material and this could be related to also the dramatic loss that we see in CO2 uptake in the modified material. However, the lowest diffusion coefficient for the fluorinated MIL101 does not coincide with the partial pressure that has the highest CO2 uptake. So this indicates that in order to promote the interaction between the CO2 molecules and the porous material, an optimum rate of transport must be achieved. This means not too slow that the trapped molecules can block and thus impede other model to pass, but also not so fast that the molecules are not able to interact with the framework. So usually a trade off between uptake and kinetics should be considered depending on the specific process application. So in order to conclude this presentation so fast and accurate characterization techniques together with molecular simulation and process scale modeling are needed to move solid state absorbance onto the next technologies readiness level and DBS is a powerful tool for scaling creating novel materials quite fast and accurately. We can also we have also done some radio curve analysis on the same materials which are obviously more point to point and can describe more as Elwood mentioned in his presentation, how relationships between pores and size of pores and axolotls. That's obviously further characterization. What's important is that we need to prove fast if these are going to be good adsorbents for CO2 or if they are not going to work. So the fluorinated MIL101 proved to have an augmented overall hydroplasmicity that was attributed to the incorporation of fluorine atoms and the CO2 uptake of both materials and were pretty similar up to a partial pressure of 0.6. However at higher partial pressures the non modified materials showed a slightly higher CO2 uptake attributed to the lower pore volume of the functionalized materials. So we can see that at low and moderate water loadings improved the CO2 total capacity for the fluorinated material with the best uptake being around 0.01 millimoles per gram. But at higher water vapor concentrations such as 20% seem to inhibit CO2 uptake. So the presence of water is detrimental for the non modified MIL101 performance which shows a decreasing CO2 capture with increasing relative humidity or partial pressure and ultimately a trade off between high absorbency capacity and fast kinetics needs to be considered in moth engineering and actual requirements will depend on the specific process application. Lastly, I would like to thank you. I would like to thank you for listening and conaceet and and senair for the scholarship that they provided me to study at Imperial College London and I Would like to also to thank Eliji Barra and Mariana from UNAM at Mexico City for characterizing and for synthesizing these materials and collaborating with me and surface measurement systems to be able to characterize them. And I would also like to thank Elwin because he's the one who showed me how to use the DDS machine and get the most data out of it. Lastly, I would like to thank Daryl, my supervisor. And yeah, thank you all for listening.

That's brilliant. Thank you so much. Pawa. We just have a few questions submitted and we've got a little bit of time left, so think it'd be great for us to get into those. So Dr. Giulio Santori has asked on diffusion measurements, how did you get rid of the instrument response? And is not the instrument response going to dominate the kinetic data, leaving you with a value which is lower than the experimental accuracy?

So that's why. So I explained that in the. I think it was a slight number. Let me go back. So. So yeah, that can happen for systems that have very fast kinetics. So that diffuse very fast. It can be mistaken with the system response time. However, as long as your system response time is 10 times lower than your adsorption kinetics, you should be fine in order to calculate the diffusion coefficients with a gravitational technique.

Thank you.

Let's just see if we have any more questions here.

So just a reminder to our audience, if you do have any questions for PALO Alt, you can submit these in the questions panel or you can, you can raise your hand, then we can. And you can ask the question yourself over the microphone. So we do have a few minutes left. So if anyone would like to ask a question, please submit it now. Okay, we can't see any questions coming in and it's been a long session today, so maybe we'll, we'll wrap up there. Thank you so much again, Paola, for being here today to present with us.

Thank you, John, thank you for the invitation.

No, that's perfectly okay. And of course a big thank you to all of our speakers today. Dr. Giulio Santori from the University of Edinburgh, Elwyn Hunter Sellers from the Imperial College London, and Paola Ceniz Cavazos from the Imperial College London. And of course, a huge thank you to all of our attendees today. I would like to remind our audience that if you do have any follow up questions for our any of our speakers or on the topics discussed in general, you can send these through to scienceurfacemeasurementsystems.com and if you have any queries about the webinar events in general, or today's webinar from a logistical standpoint, please don't hesitate to contact myself at marketingurfacemeasurementsystems.com our next educational event is another online workshop taking place on 22 June, which is a Tuesday in two weeks time. And this this session will be exploring surface characterization of solid state materials using gas absorption techniques. We've been very fortunate to organize this in partnership with the Eindhoven University of Technology, and while this is traditionally billed as a Benelux targeted event, we of course invite anyone who is interested in this topic who's attending today to sign up to this workshop on our website and join us there in two weeks time time. So thank you all again for joining us today and we will speak with you and see you soon. Thank you.

You.