Liquid and Gas Chromatographic Methods for Studying Solid and Solute Properties and Interactions

Foreign.

So, let's start. Welcome everyone and thanks a lot for joining in. Today we will be discussing liquid and gas chromatographic methods for for studying solid and solute properties and interaction. Today's workshop is organized in collaboration with Collaborative Research Center 1411, design of particulate Products at University of Erlangen. Before we start, I would like to mention that on your right side you have possibility to post questions during the session. Afterwards we will have a live Q and A session where I can pick up your questions. If you like, you can also raise your hand and I will unmute you so that you can ask directly. This speaker myself is Ravi Z. Ahmed. I am Sales and Business Development Manager at Surface Measurement Systems. My background is particle synthesis and characterization. I will take this opportunity to introduce also some of the key persons involved in today's workshop. And the first one is Dr. Monica Destazo. She is the Scientific coordinator of the Integrated Research Training Group as well as group leader at the Department of Chemical and Biological Engineering. Some years back when I was doing my research work under the supervision of Professor Poykit, I had the pleasure to enjoy her guidance and supervision in person. She has been a great help in organizing this workshop. Thanks a lot, Dr. DiStasso. It is an honor to have you with us today.

Thanks a lot Ramis. And for initiating this collaboration and for having this great idea to organize the workshop. And I am also very happy to see that so many people have also great interest in this topic. And of course that's great. And I look forward to hearing the lesson of Professor Derry Williams.

Thank you, Monica. So, talking of Professor Darryl William, he is also with us today. He is founder and Managing Director of Surface Measurement Systems and the professor of Particle Science at the Department of Chemical Engineering at Imperial College London. I have few more words to say about him, but I will say them before his presentation at this moment. I thank you for being with us today. And our first speaker today is actually Professor Dr. Martin Hartman. We are also honored to have him with us today. He is professor and Director of Erlangen Catalyst Resource center as well as vice Sports spokesperson of CRC 1411. In 1993 he received his PhD at the University of Dortmund. His work was focused on state of oxidation and distribution of cations in copper exchange Fujicides, a multinuclear NMR investigation. Since then he was active at various universities, namely University of Houston usa, University of Stuttgart and Technical University Kaiserlautern. In the year 2005 he joined University of Augsburg as the professor of Advanced Material Science in the Department of physics. And in 2009 he came to Friedrich Alexander University Erlangen to become professor and Director of Erlangen Catalyst Resource Center. His research interests focus on the synthesis and spectroscopic characterization of porous materials for application in catalysis separation and energy storage. He has published over 190 papers, reviews and book chapters in this area and gave more than 160 invited and contributed talks at the international conferences and research institutes. He will be presenting a short introduction of CRC today. Professor Hartman, thanks a lot for your time and I would like to invite you to take the stage now.

Yeah. Thank you Ramese very much for your extremely kind introduction and I would also like to welcome you to our online workshop on chromatography, which is kindly hosted by Surface Measurement Systems. Thank you very much for that already. It's also a great honor for us that Professor Darrell Williams is willing to share his insights into liquid and gas chromatography us today. First, however, as it was announced, I will give you a few insights into our CRC 1411, which is sponsored by the German Research Foundation. And I hope you can see my my screen.

Yes.

So let me start with this famous calligraph or painting that's still under discussion from Sengai Gibbon, a Japanese monk who inspired the logo of our crc. It's called the Universe in English and actually painted with ink containing nanoparticles. And as you might guess for this reason nanoparticles are also the main focus of our CRC. And we will definitely deal with the synthesis, targeted synthesis and the processing of these nanoparticles for certain applications. And as many of you know, the properties of nanoparticles are certainly a function of their dispersity, so their size distribution and also their composition. And to synthesize nanoparticles with a with a certain narrow dispersity and a narrow composition is quite challenging. And therefore we were thinking about whether we have some process variables later on to deal with the to narrow the particle size distribution. This is the reason why we turned our heads into the direction of chromatography and the particles we are looking at will have properties which are mainly optical properties. And we try to tune this optical properties not by synthesis but also by the pre processing. And when we started the CRC we that you realized that there are certain challenges in the design of nanoparticulate products and in particular that there is a large gap between synthetic protocols and for the synthesis of let's say a rather small amount of nanoparticles typically and a technical large scale synthesis and also the technical application of nanoparticles. And furthermore, we miss process for the further processing of nanoparticles for, let's say the size or shape separation. And also what is also very important from our point of view, that we want to do a predictive design of nanoparticles. This is actually what is at the heart of our sea to explain it in a nutshell. So the CRC starts with the formation of functional particles. As I said, the function is typical, an optical function at the very moment. But it could be of course any other function, also catalytic function or something else. And then the other area where we are interested in is the classification of nanoparticles. And we use and will use preparative chromatography for the classification. So these are the two main research areas, A and B, that's the rigorous design of continuous particle formation and also preparative chromatography for classification of nanoparticles. This is supported by two more research areas, namely C and D, which deal with in depth characterization of the nanoparticles and also with the model based optimization of the synthesis of particles, but also of the separation of particles. In order to illustrate that, let me briefly just show a brief process description which we have in mind at the very end of this crc and we are trying to work out the steps in between. Let's assume that we would like to synthesize nanoparticles with a certain size or narrow size distribution and maybe also a certain shape like spheres or squares or something like this, as you have seen in our logo. So therefore, what is our process chain that we have in mind that we have to define the target? So what we want to get and to do this with a dedicated synthesis, but also with the further processing with chromatography. So in order to do so we need, after the definition of our target, we need to design a stationary phase material with the targeted separation properties. Then next step is certainly to synthesize this SPMs and characterize them in depth. And then of course finally test the separation of nanoparticles. So this is similar to also chromatography in the liquids or in the gas phase, where you design a particular stationary phase material for targeted separation. And we'd like to transfer this concept into the separation of nanoparticles in order to show that this is actually feasible or working. I have say two or three examples to show you that this is in principle working. Let's start with rather large molecules like this polyethylene glycol standard, which is separated by, let's say, sec column, a size exclusion column. And you can clearly see that we can achieve a baseline separation into single ethylene glycol units. So in principle, for large molecules, you see the average molecular weight is 1400 square meters gram per mole. This is actually working quite well. If you transfer this now to larger molecules, for example, here, this fullerene mixture, C60, C70 mixture, you can also separate them, classify them by using chromatography, and you can even go further into small nanoparticles like this zinc sulphide particles, as well as the gold nanoparticles. So essentially you can transfer this concept of molecule separation by liquid and gas phase chromatography into nanoparticle chromatography. In addition to that, we'll have a very complex characterization toolbox at hand. So that includes electron tomography, that includes solid state NMR spectroscopy and in situ adsorption calorimetry. We can look at the diffusion in confinement, which is of course very important for the separations. And finally also we employ analytical ultracentrifugation to look at the sizes and shapes of the nanoparticles we have been synthesizing, hopefully at the end of this project. And to summarize basically what we want to do, the goal is that we have a predictive modeling and optimization of continuous particle formation and classification that also includes quality control and process stability. We will demonstrate, or we hope to demonstrate and realize particle chromatography as a novel method for nanoparticle classification. And in order to do so, it's necessary to understand the transport of nanoparticles in SPMs and also their interaction with those materials in a confined space in an elevated concentration. And there here, of course, this meets where, let's say nanoparticle chromatography will meet liquid and gas phase chromatography, as Professor Williams will certainly demonstrate in a minute. And so essentially the goals are to design particles with optimized functional and structural properties with a knowledge based design, not by trial and error, but a knowledge based design. And we will try to close the gap from synthesis to application of nanoparticles to introduce new processing steps. And we will use digitalization approaches for the product design. This will become increasingly important also today or in this years. So with this, let me thank Surface Measurement Systems again for hosting this interesting online webinar. And I'm really excited to look forward to this interesting seminar on chromatography. And thank you very much for listening. And back to you Amir.

Thanks a lot Mr. Hartman for sharing few insights on CRC. I will give the audience one or two minutes. If they have any questions about CRC, you can type it in the chat or raise your hand. So maybe when it's. Maybe when the audience is typing, I can ask one question. So how many institutes are taking part in this collaborative research?

Basically we have more than 20 different projects and we have mainly, let's say three main institutions. So that's the Department of Chemical and Biological Engineering which shares the large part of. Then we have participants from Materials Science department and also from Natural Sciences, mainly from the Physics department. So that's the core group which is supporting this. So it's quite an interdisciplinary research effort that we are doing here.

Thank you. Okay, I will actually move on to our speaker, Professor. As I mentioned earlier, he is founder and Managing Director of Surface Measurement System and Professor of Particle Science at Imperial College London. His research interests include the surface and bulk characterization of complex organic solids, including specially pharmaceuticals, foods and polymer, as well as their synthesis using spray drying, crystallization, freeze drying, milling and granulation. Professor William has published over 100 papers in peer reviewed journals and has been granted five international patents. He has invented one and has led the commercialization of two standard techniques for material characterization, namely the dynamic vapor absorption and the inverse gas chromatographic methods via Surface Measurement System. His work has been recognized by MPL College with the President's Medal in 2017. In 2018 he won the EEF's Future Manufacturing Award in Innovation. And in 2018-19 he was awarded the Institution of Chemical Engineers Geldard Medal for his major contributions to particle technology. Without further ado, I would like Professor Williams to take the stage.

Thank you very much for that kind introduction. Ramese, let me see if you can see my screen. Can you see my screen now?

Yes, I can.

Okay, great. Well, thanks everybody for joining this afternoon's presentation and I'm going to spend some time. Let me just move it out of the way here. This afternoon's presentation is really going to be about the use of liquid and gas chromatography for studying really solid state materials and in the process understanding the interactions between solutes and solid state materials and also interaction between solids and themselves. So we're going to start by spending a few minutes just talking about chromatography and just giving a perspective to the sort of chromatography I'll be describing today, which is not the sort of chromatography that you will be familiar with. I think we probably all in our own way are very familiar with gas and liquid chromatography, but we probably most commonly know them as key reference techniques for analytical science. So if you want to identify and quantify unknown species in a mixture in the gas phase or the liquid phase, gas and liquid chromatography are really the go to techniques. And indeed, more recently we're seeing chromatography moved into manufacturing. So the industrial purification of biological therapeutics using reverse phase liquid chromatography and protein A chromatography has become a major advance in the manufacture of biotherapeutics and are now standard industrial processes. So really analytical and process chromatography are used. There must be a million chromatographic instruments in use at any one time in the world. So although those techniques are very well established, we'll be talking about something which is very closely related to but slightly different. We're going to be talking about physiochemical chromatography today. So let me start by just reminding you about some of the key elements in chromatographic methods. And one of the key points about sort of physiochemical chromatography that I'll be discussing today, that in our chromatography we actually pack our own columns. And that's the first thing and the most substantial difference between the chromatography I'll discuss and the chromatography you might have come across. So if you're familiar with analytical chromatography, you buy columns from manufacturers that are designed to separate specific molecules in specific mixtures. So the standard columns that you buy analytically specific to analytical chromatography. But in physiochemical chromatography we're going to be making our own columns. But the actual instruments in many cases are the same, except for the columns. So if we're talking about experiments involving liquid phase chromatography, but here's a schematic of a typical liquid phase chromatographic instrument. We're gonna have some solvents over here, we're gonna have a pump, some sort of injection system which allows us to inject a solute, which is called here as a sample. And of course the HPLC column. In the chromatography I'm gonna be describing today, we are going to pack our own column. That's a key part of the method. And at the end of the column we have some sort of detector. And that detector will allow us to actually see. Let me just pick up the pen here. Drawing tools?

No.

Is it a drawing? Let's look that's going to give me. Oh, I think we'd like a spotlight, I think. Okay. Right. So we've got some sort of detector, and that will allow us to work out when the solid or the samples pass through the column. And this will be seen by our data acquisition system. So, as I said, in the case of analytical cryptography and the special physiochemical chromatographies, they look to be the same equipment, except the columns are different. Let me just explain the methodology, because it is a bit different. In analytical chromatography, we in many cases are fingerprinting the contents of our unknown, typically using reference chromatograms. Whereas in physiochemical chromatography, we are very focused on the retention time and retention volume for the species that are eluting through the columns. Typically in a physiochemical technique, we are going to be injecting a known sample, and in fact, the unknown material is really in the column. So this is actually a reversal of roles in analytical chromatography. The column here is known, but the sample is unknown. But in physiochemical chromatography, generally we don't know much about what's in the column, but we know what the molecule is we're going to be using to probe it. So molecules will probe our unknown solid state materials. So when we inject a molecule which does not interact with the column, that allows us to determine the dead volume of the system, and we need to know that. So, typically, if we're using liquid chromatography, we might inject a solvent molecule which will sweep through the column. It will not interact with a stationary phase, and that will allow us to measure a dead time. And that dead time, once we multiply it by the flow rate of the liquid phases, we will give us actually a dead volume. From there, we're then going to elute other probe molecules, typically one at a time. In this particular experiment I'm showing here, we're going to elute hexane, heptane, and octane. And the time it takes for those molecules to elute will increase as the molecules get bigger. In this case, they're absorbing on the surface of the stationary phase, which is our sample. And the time difference between our reference material here, our dead volume marker, and these probes. This is going to give us what's called the net retention volume. And net retention volume is a measure of how strong the probe molecule is interacting with the stationary phase. And remember, the stationary phase here is your sample, and we're going to see A number of examples of that. Today we're not going to spend too much time on equations, but just to remember that we are particularly interested in the net retention volume, which is the retention volume we measure for a probe minus the dead volume for our system. And this net retention volume is a measure of the strength of interaction between the solute in the mobile phase and the actual stationary phase. And it's also directly proportional to the partitioning coefficient or the slope of the absorption isotherms. And whether it's a liquid phase or gas phase, these equations and these principles are also applicable. Now one of the things, of course we're interested is the partitioning process. And in chromatography, be it gas phase, liquid phase, there are a range of different physiochemical processes which involve or control the way a solid interacts with our stationary phase. So very commonly it's just simply adsorption phenomena. And sometimes if the, if the, if the stationary phase or the sample we're studying is very hydrophobic, we might talk about being called normal phase adsorption. But these are essentially surface absorption processes. We can have reverse phase processes, hydrophobic interactions, size, exclusion, ion exchange, affinity, or if you like, chemical reactions. These are all the different processes by which a molecule in a mobile phase can interact with a solid state material in the stationary phase. And it turns out each of these methods is a potential basis of a physiochemical characterization approach. And we're going to look at two or three of these today. So the first example we're going to have is actually based on gas phase chromatography. So we're going to look at how we can measure the surface area of a freeze dried biological powder using inverse gas chromatography. And I'll give some explanation for why we're doing that. And this is a part of a collaboration we did at Imperial College with the National Institute for Biological Standards and Control in the uk and they manufacture a freeze dried biopharmaceutical reference standards. And these materials they need to understand the physical properties those, that's an important part of the QA for the products that they manufacture. So just a quick, a quick summary here because I'm going to be talking about freeze dried products. Some of you may be quite familiar with those, some of you may not be familiar. Let me just spend a moment discussing those. Freeze drying is a way of stabilizing typically a biopharmaceutical material. If I've manufactured a protein based therapeutic or a peptide or monoclonal antibody, if I leave that Material in solution, there's a very high probability that material will lose its biological activity. It will aggregate, it will go off. So one of the ways we can avoid materials in solution losing their activity is to turn them into solids. Generally speaking, solids have much more stability. However, we have to transform our solution into a solid state material and freeze drying is one of those approaches. So you take the material of interest, you dissolve it into a solvent, typically water, and there are going to be some other additives and stabilizers often present. Then this particular material is then frozen, and the frozen water is then sublimed under vacuum, which leaves us with a highly porous dry cake. And you can see in the vials down there on the right, you start with a solution. We've got what's actually a very low density powder cake. It turns out this powder cake, if we keep it dry, has very good long term stability. It can actually survive for months or years, depending on the storage conditions, much, much longer than if we leave it in solution, in which case it might only last for a couple of months. The actual process is not a fast process. It might typically take a few days to turn these solutions into solid cake. But once we've got them, we then need to characterize these materials so we can understand their performance and optimize them in terms of sort of materials that we can freeze dry essentially. Most biological materials are commonly freeze dried these days. As I said, this is because their stability in liquid solutions is generally quite poor. And the freeze drying methodology is now very well established, is used by all major manufacturers throughout the world for stabilizing biotherapeutic materials. Although it is important that once you've made the material, you end up with typically amorphous solids and you must keep them dry. If they get wet, then they can start to misbehave and cause other problems. So a freeze dried material that's kept dried can actually survive for many years under refrigeration conditions if required. So let me talk about the first chromatographic approach. This approach is called inverse gas chromatography and it's based on traditional chromatography, except that as I said before, we're going to study your material and we're going to pack it in a column. So in this particular study, we're going to typically take 50 to in up to a few hundred milligrams of the sample, going to pack it into a small stainless steel column, and that column goes inside this machine here, which is basically a special type of gas chromatograph in a Gas chromatograph, you're going to have some type of carrier gas, typically helium, or it could be nitrogen. We're going to control those flow rates and we're going to choose inside the instrument which actual solids we want to eject. So typically we might use octane, ethanol, chloroform. We can choose different organic solvents to interact with the solid state material you want to investigate. And we'll choose the solvents partially based on the sort of experiments we want to do. In this particular project we are interested in measuring the surface area of these powders. So to use those we'll typically use something like octane or cyclohexane. That's a very good molecule for doing these sorts of measurements. And just to remind you, what do you do if you don't do inverse gas chronography? Most people would measure the surface area of a powder by using adsorption of nitrogen at 77 kelvin. So nitrogen absorption is a standard done at 77 kelvin. But we'll see today there are some benefits for doing these experiments at room temperature. So essentially we're going to pass various sorts of adsorbates through the column here. We have a flame ionization detector at the end which, which is a standard detector that you'd find inside most gas chromatographs. And we're going to measure how long it takes for the solvent molecules we choose to elute through the column. And of course, as they pass through the column, these molecules will send some of the time absorbed on the surface of our powder. So essentially we're going to be studying the adsorption or an organic molecule on the surface. So from the chromatographic experiments we're going to get these sorts of chromatograms. We will turn this into adsorption isotherms and will then turn us into properties. And today we'll turn it into the surface area, the BET surface area. So that's the general approach we're going to be using today to characterise our freeze dried materials. But before we do our freeze dried material, it's always good, or in fact good practice, if not perfect practice for us to look at some reference materials to start with, just to make sure that our method is doing what we want. So in this particular study that we reported about three years ago, now we started by looking at some BET reference materials. These are materials you can buy from accredited agencies at CRM170. This is a well established material with a well established BET surface area of about 1.05 meters. Squared per gram. And the reason to choose this really fairly low surface area is that freeze dried materials generally have low surface areas. So we wanted to check a reference material with a similar surface area to the materials we're going to be characterizing. So what we can show you here by doing a whole series of injections. In this case we're using octane. These are octane isotherm. So the first figure here shows us this is the partial pressure here, this is the amount absorbed. So this is the actual adsorption isotherm. And from that we then going to analyze that using the BET model. So here is the linearized version of the BET equation and we calculate the surface area and we calculated about 1.012 meters squared per gram against the reference figure of 1.05. So quite satisfied that that's a good consistent result for actually that particular reference material, which is actually, I think it's an alpha aluminum. Here is another, here's a different, a different reference material, a slightly different area. It's also an illumina. This is CRM171, has a slightly higher surface area. Its surface area is about 2.95 meters squared per gram. Again we measured it using IGC at 25 degrees, not 77 kelvin, a bit at 25 degrees. And we got a variable factor, 30 degrees. As you can see here, a very good agreement between, between the nitrogen data and our IGC data. And also one of the things which is interesting about this particular inverse gas chromatographic method is not only can we do it at room temperature, we can also do it in the presence of humidity. And of course in the real world we very rarely have samples at 77 kelvin and we very rarely have samples that are also exposed to water vapor. So in this particular experiment we're able to look at a sample at both room temperature and in the presence of humidity. And what you can see here, we've got some data at 50% humidity. And you can see here the surface area is just a little bit lower. For 0% humidity the surface area was about 2.95. But at 50% the apparent area is slightly lower. And of course the water molecules are competing for some of the sites on the alumina particles. And the result of that of course is that we don't quite see the exact same surface area. And that's also useful. If we were doing experiments at 25 or 30 degrees at 50 cent humidity. It might be useful to know that not all the surface area is available for gas phase adsorption. So these are some studies, these are reference studies. So let's move on I think now to the real samples. Before I do that, I've just got a summary slide here talking about the surface area of common materials. So here is a log scale of the BET surface areas. Here is a commentary on the techniques that are available and also some classification of the very common sorts of materials you find with the areas. We do a lot of work down this end here. Excipients, hair fibers, drugs, biopharmaceuticals, they often have low surface areas of a few meters squared down to 0.1 meters squared per gram. And as we'll see in a moment, although nitrogen is the most commonly used approach, it's not the most accurate approach. We'll see in a few minutes that it turns out that the IGC methods for low surface area materials, typically up to about 100 meters squared, are superior. But if you've got high surface area materials, the traditional nitrogen methods are superior. So as you know, this presentation is about the surface area of freeze dried cake. So why would we want to do this? Why do we want to measure the surface area? Not everybody does that. Well, there are a number of reasons for measuring the surface area and one of them is it may correlate with biological stability. Because the whole reason to freeze dry the materials is to prolong their shelf life. And we want to know when the shelf life is lost and it may turn out to the surface area helps us do that. It could help us in predicting the reconstitution time of these freeze dried materials. Once you freeze dried a material, you then when you want to use it, you have to add water to it and dissolve it. And sometimes that can prove problematic. Some materials that are freeze dried over a long period of time after storage, they sometimes don't reconstitute. And again, surface area might give us be a proxy for understanding reconstitution problems. Also, anybody manufacturing pharmaceuticals is in a highly regulated regime. So having quality attributes that my later tract, batch, batch variations and product qualities is potentially quite important. Of course the surface area relates directly to the average pore size distribution and that tells you about surface area, which also tells you about the probability and the ease of reader solution. And also sometimes people are concerned about the mechanical performance of freeze dried cakes. This is to do with their transport and damage. Again, the surface area might be helpful there. So just going to briefly comment on the protocol for making the samples that we'll be looking at. We've got sucrose and Trekalose samples, which are basically a 5% solution. So 5% sucrose, 95% water. And we've got a range of commercial formulations which include some antigens, it's a flu antigen at a much lower concentration. And it also influenza antigen, again at different concentrations. So some of these are simple model samples and some of these are real samples that are used in clinical manufacture. This is just the temperature and the cycle conditions. But you can see here these are quite long manufacturing process. It typically takes two or three days to make a batch of vials. And a batch of vials depends on how big your freezer is. In my lab we can make 200 vials. In some of my industrial partners, they can make a million vials. That involves a very large freezer. So in terms of the method, just a brief comment on that. Because the materials we're looking at have relatively low surface areas and we'll see soon that they're up typically about at 1 meter squared per gram. We typically use more sample than normal. So we typically use a few hundred milligrams of sample. We very lightly crush it and we pack it into a column. That's a glass column actually, which is 4 mil in diameter. In our study we used helium as a carrier gas and our dead volume reference marker is methane gas. All the experiments, 20 degrees, 10 cc's of the carrier gas. And most experiments were done at 0% humidity unless indicated otherwise. And we use a peak maximum method for analyzing the data. When we've got the isotherms, then we're going to use the BET model to fit that between 0.05 and 0.35 PvP 0 to calculate the monolayer capacity, which will then give me the surface area of the materials. Sometimes we have to change that range a little bit. The non linear to the isotherms and other factors. Sometimes we might shift the region which we analyze the data to improve data quality a little bit. So something simple to start with. The most basic thing you can do is reproducibility. So this is simply an exercise in taking the same sample, running it three times sequentially and seeing what sort of reproducibility we get. This is an IGG 5% sucrose and also a flu antigen which has got a higher composition with sucrose. And we can see here the materials have areas of typically about 1 to 2 meters squared per gram. So they're quite low. And the errors, the relative standard deviation, typically about one and a half percent, which is, which is quite good actually for surface Area measurement, that's pretty good. And for that low areas, that's particularly good. And this is the reference method. This is the standard technique that most people would normally use, which is nitrogen absorption at 77 kelvin. Don't worry too much that the results don't agree. We'll discuss in a moment the fact that there is quite a bit of vial to vial variability. So if you've got a thousand vials and you draw two vials that they often have different structures and different surface areas. So that is not a manifestation of the measurement, it's a manifestation of the manufacturing process, which is not always as good as you'd like. But if I look at a series of samples which are tested using nitrogen, essentially we get numbers in the similar ranges. Obviously the synthesis involved a viral variation, but. But the main point here is that the typical errors are about two or maybe a bit bigger, two times larger. So we don't quite have the accuracy with our nitrogen method. So that's sort of. And therefore that's something to be aware of. And here's just a summary. Between the nitrogen absorption and this inverse gas chromatographic method for determining surface area. The amount of sample red required about the same. The errors are a bit lower in the octane absorption. The run times are actually quite similar. A big difference here is we can do the experiment at room temperature and there are going to be a lot of cases where data at room temperature may not agree with data at 77 Kelvin. And there'll be a number of applications where actually 77 Kelvin is actually not very interesting to us and we really want to look at it at room temperature. So one of the things about this particular chromatographic technique allows us to do that here we've got some more information on basically there's a batch to batch variability. So this is really telling you about the uncertainty in the manufacturing process. It's not telling about uncertainty in the measurement process. The measurement process is very accurate. What we're seeing here is there's quite a bit of variation inside the batches and the techniques can see that. So first thing you can see is that the surface area is a potential metric for looking at the variability within a batch of products. And that's something which manufacturers are quite interested in because it is a concern, because that variability in area probably means there'll be variability in other properties as well, including the rate of dissolution on the materials and their performance. Again, looking at some more batch to batch variability data here as well. And the final bit of data to Talk about in this particular little study is what happens when we measure the surface area at different humidities. And this is for a flue antigen. What you can see here is that as we do our experiments, we get about one and a half meters squared per gram at zero percent. But once we get above our 40%, we start to lose surface area. And that could just be the water is blocking some of the sites, or it could be more profound changes in the stability of the material. But at least with the IGC technique you can get an insight into surface area effects with humidity, which is something that's traditionally very, very difficult to do. So here are some conclusions for this particular topic and it's one of the three topics I'm going to talk about today. We can use IGC to determine a specific surface area of freeze dried materials. The numbers agree with standard notching methods. The reproducibility is a little bit better with the IGC method. There is a lot of volatile variability in the data sets. The IGC is a little bit faster to load and the method times are a little bit better. But we can also look at samples of both room temperature as a function of humidity. And I've included here and I'll share the presentation at the end of the meeting. This is the paper from which the data is taken from. There's a lot more information in the paper. I'm only paraphrasing some parts of the work because of the time constraints today. So we're now going to move from gas phase. We're now going to move on to liquid phase. Yes.

Before you move, maybe it's the right time to ask an IGC question I received.

Sure.

I'll go ahead, please.

Yeah. Does the quality of column packing affect the IGC testing results?

The short answer is no. We've packed hundreds, indeed thousands of columns over the last 20 years now and I would say there's probably about one in 500 samples where a little bit. It's a bit more difficult to pack the samples. But for 99% of the time you. If I pack a column and you pack a column or a mise, you pack a column, we're all going to get the same result. So the column packing is not a challenge and is not an issue for anybody.

All right, thank you.

Okay.

Any other questions on it? Any other IGC questions from this?

Yes, one more. Is this applicable for materials?

Is this applicable to what? Sorry? To what sort of materials? Layered materials, laid materials. The only constraint is can you pack the material in. Into a column? If you can pack your sample in a column. Potentially we can do igc. So we've done fibers, we've done films, we've done granules, I've done pellets. If I can pack them in a column, I can do igc. That's the only constraint. From time to time we get samples that are like thin films or even thick films. And there are accessories you can buy that allow you to study larger film or larger monolithic samples as well. So even that can be done.

There is one more question. The freeze dried samples were crushed before measuring the surface area. Does the surface area affected by crushing?

In short, no. They were very, very gently. They're very soft materials. So they were just very gently broken. So we could get them into the columns. Not enough to actually affect the surface area.

Okay, thank you.

I mean, just to make a further point there, if I tried really, really hard and I put him in a mortar and pestle and I ground them up. Yeah, I could change the area. So we have to be slightly careful about that. All right. Yes, go ahead. Ramis.

Can we implement IGC technique to our regular APIs.

On APIs? Yes, absolutely. When a characterize APIs. Yes, we, we characterize. I mean most, most major pharma companies do igc. Now materials to measure two properties. The surface area which I've just spoke about and the surface is the surface energy which I did not speak about today. So Those are the two popular things to measure on APIs and excipients is surface area and their surface energy.

Thank you. In connection to the question of layered materials, the participant is asking what is the mesh size used?

So let me, let me turn the question around slightly. We have a range of different columns that go in the instrument and they have IDs of 2 millimeter, 3 millimeter, 4 millimeter, and I think 10 millimeter. So it has to fit inside those dimensions to go inside one of the round glass columns. 2, 3, 4 and 10 millimeter are IDs. Okay.

Okay, thank you. Next one is the. Does the results obtained from IGC pore sizes and volumes differs with the nitrogen bed method.

This. Well, well, okay. The surface areas, as we discussed before, agree, but in terms of pore sizes, we need to take into account the fact the molecular size. So if I'm doing a adsorption isotherm with cyclohexane and you're doing an absorption isotherm with nitrogen, we won't see the same micro pools because my molecule may not go in the micropore because it's Bigger, but actually it might turn out in the real world that's more important. So you will get some differences because the molecules that we're using for the adsorption experiments, they may not be able to access all the pores in some cases, but that sometimes can be helpful. So I would suggest you that's not disadvantage in many cases. That is an advantage because by using different size molecules, you can get an understanding about the porcelain distribution in a more subtle way than you can get simply by just analyzing a single isotherm.

Okay, thanks a lot. I think we can continue with your talk and then if I have more questions, I will jump in.

Great. Okay, so we're now moving away from gas phase chromatography. We're now moving to liquid phase chromatography. And this particular study, we're going to start to understand how macromolecules interact with each other. And we're particularly interested in protein aggregation. So let me talk to you a bit about protein aggregation and how we're going to characterize that. So we've been working on monoclonal antibodies and proteins for many years in our group, really around stability and aggregation behavior. And of course, monoclonal antibodies are used for a whole range of diseases these days. Cancers, autoimmune diseases, infectious diseases, cardiovascular diseases. And it's a big growing segment. But they're quite complicated and difficult and expensive molecules to make. And they unfortunately have a propensity to aggregate. And if they aggregate, that has a number of problems. It can cause immuno, it can come cause immunoimmunogenesis problems. So it can cause reactions with people with the actual aggregates, as well as the loss of material in manufacturing. So aggregation is an undesirable state of affairs. And we're very interested in working out how can we avoid aggregation occurring. So what tends to happen in aggregation is that the very large molecular species, which are the monoclonal antibodies, will start to associate together. So the monomers will fold, dimers and trimers, and essentially these stable folded proteins will start to misfold and you may end up with an aggregate. And these aggregates are subject to a range of external variables that can affect the probability of aggregation occurring. So the ph, the protein concentration, the temperature, mechanical stress factors, ionic strength, the sort of counter ions present in solution. There's a whole range of solution factors that can control or define whether an aggregate will occur or will not occur. And we want to understand what is the optimum solution conditions so that our protein will stay stable. And alternately, what are the conditions, the PHS temperatures that we want to avoid when we're making our products so that we don't have a problem with aggregation. That's the basic background of the problem. So let me explain. This technique I'll talk about is known as self interaction chromatography. And what we're going to do is we're going to take some particles, these are the sort of particles you'd find used in liquid chromatography columns, and we're going to immobilize on the surface of those particles our protein. Okay, so it's, so it's, you know, it's a slightly involved process. Before I described packing some particles in an IGC column. That is very easy. I can do that in 10 minutes. This will not take 10 minutes, this is going to take you a day or two. So we're going to have to take inert particles which have got well known stationary phase chemistry and we're going to graft onto those, we react the proteins onto these particles. So we're basically going to have, going to have protein immobilized particles. Now once we've created these particles typically in solution, and we're then going to pack them into a column. But as I said, this is a little more work because this is actually an immobilization reaction which takes a little bit of effort. So once we've immobilized the protein of interest or the monoclonal antibody of interest, we then are going to pack a column and we're going to start passing through that column. Our mobile phase is typically going to be water. The mobile phase and self detection chromatography is typically water and the water will have in it various buffers and other chemicals present as well. But in this experiment we're going to inject through the column the same protein that's been immobilized. Okay, so let's just make sure we've got that. So here is a particle we've got on the outside is we've got our proteins that have been immobilized and we're going to elute through the column the same protein species. And what we want to basically understand is when the protein in the Merbore phase sees a protein coated particle, is it attracted or is it repelled? And that attraction or repulsion is a measure of the molecular forces of attraction. So we're measuring the second viril coefficient, which is a very important way of understanding the strength of molecular interactions. So we're going to look how, work at how long it takes for My injection protein to pass through the column. We've got a detector here and we're going to use the chromatograms to tell us whether there are attractive or repulsive interactions between the peptide or protein molecules in the stationary phase versus the mobile phase. So this is this cartoon I hope will make it a bit clear what's going on. So if it turned out that the protein species in the mobile phase are strongly attracted to the stationary phase, then the chromatogram will be fairly slow to come out and it'll be retarded by those attractive interactions. And we can also sometimes have repulsive interactions. If we have repulsive interactions, the peaks come out a bit quicker. So what we're doing in the experiment is we're comparing the elution time at different solution conditions. So what's the retention at Ph3, 4, 5, 6, 7. So as I change the solution in the mobile phase, it passes through the column. I'm going to look to see are the chromatograms coming out faster or slower. And from the retention times from these chromatograms I can measure the second viril coefficients. That can tell me whether I've got net attractive interactions. There's neither net positive or net negative or the net repulsive interactions. So what I can do is relatively quickly is I can look at the attractive interactions in this system quite quickly. I can typically run 100 different solution conditions in a day quite easily. So the whole experiment, once I've immobilized a protein, the whole experiment can be done typically one day. So what's it look like when I do this? So just just a bit of a recap on the experimental setup. Again, here is the column. Here in the column I've packed my specially immobilized peptide or monoclonal antibody particles. I also, for various reasons I have a detector pre column, a detector post column. I just like to check that the number of molecules that go in the column come out again to make sure nothing funnies happen. So once I've got. So I immobilize the peptide or protein onto the stationary phase, I pack the columns and then I'm going to map the formulation space by injecting the protein at different pages and at different conditions, solution conditions. So here is an example of the screening. So every dot is an experimental data point. Okay. And we're screening against the ph in the x axis and sodium chloride concentration in the actual Y axis here. We probably have about 60 or 70 data points in there. And we are measuring B22 at every data point. And then we're constructing this false color image of stable and unstable regions. So what the experimental data says, and this can be done in a day, it says that at pH about five and a half and that pH 7.5 for high soil concentrations, I have a very stable solution. I do not expect aggregation to occur under those circumstances. Now if I actually look here for low ph at about 4, over quite a big range of salt concentrations, or down here at very high ph, but no salt, I've got two regions of instability. So I would expect that particular condition to give me potential problems around aggregation. So this is a screening experiment we did in a day. So now what we do is we make up these solutions and we test them over 1 week, 2 week and 3 weeks using dynamic light scatter to see have they aggregated or not. So this is the fast screening experiment. Now we're going to look at the slow trial to see how they compare. So here is the SIC data done in a day. And here is the monomer data that we've done after three weeks. And you can see here quite a bit of, quite a bit of agreement between what we saw in the SIC and what actually happened experimentally after three weeks. We measured the percent monomer present from either size exclusion chromatography or dls. So we see the self interaction chromatography as a screening technique allow you to make quick assessments without going to all the trouble of running one month or two month aggregation studies and encouraging data in this study. So our conclusion for this section was that self interaction chromatography can be used to map the stability place for monoclonal antibiotics using just milligram quantities of materials and the SRC results were in good agreement with the data obtained in a three week aging study using dynamic light scattering and size exclusion chromatography. And full details of this paper are published in Analytical Chemistry. So you can look at that. There's a lot more detail than I've given you today. I've just given you sort of a very brief summary and the final topic today is looking at diffusion. And particularly we're interested in looking what are called count diffusion constants in liquid phase catalytic reaction issues using a technique called zero length column chromatography. Ramese, are you happy for me to carry on or do you want to pick up questions?

Yeah, I have a question, thanks.

Sure.

Is it possible use this technique to study the affinity of some protein molecules to some solid polymers?

Yes, you could do that. Yes, absolutely. Yep. Yes, you could do that. You'd pack the polymer molecules or the polymer particles or polymer fibres into the column and you elute your peptide, your proteins, and you would be able to measure the adsorption potential of your peptides for those particular polymers.

Okay, thank you. The next one is, are the proteins acting in the stationary phase and the mobile phase is the same in the same conformation? If so, then how can you ensure this.

The actual confirmation? And it's a good question. It's difficult to be exactly certain whether the peptides or in fact, whatever you've immobilized, it's sometimes difficult to be sure whether you've got the same sort of confirmations in that. So we're doing a project at the moment where we're trying to get a bit of understanding as to using different immobilization methods. If I immobilize this case, there are peptides two or three different ways, do I get the same result? So we're doing something on that at the moment. But the confirmation of the particle in the immobilized stationary phase is. It's difficult to be certain how well that mimics, if you wish, the confirmations in the actual mobile phase. It's a good question. I don't think we've got a clear answer to it at the moment.

Okay, there's one more question in connection to the first one Naveen is asking. So in this case, the column will be packed with powder polymers, for example, and maybe water that contains protein can be fed.

So maybe I have to repeat that again. Are we talking about inverse gas chromatography?

So the first question was that can we use solid polymer as a filling media in the. In the column and feed protein.

Okay, let me answer the first question. The only constraint is you need particles that aren't too large because we need some surface area. We're doing an absorption experiment. So we don't need particles that are very large. They don't have much surface area. So I. We need to either have particles that have relatively well have a surface area of maybe a meter squared per gram, or we might need to coat an inert particle with a higher surface area with the polymer. So we need surface area for the. For the method to work. The second part of the question.

So she's asking about how the experiment will be constructed. So in this case the column will be packed with powder polymers.

Yes, correct. That's fine. We do that.

Yep.

And. And then proteins which are contained in. In water, for example, can be fed to it.

Well, well, not, not in igc. Won't be. But in, but in inverse liquid chromatography it could be. I mean, I mean it sounds like if you're interested in proteins interacting with polymers, I presume that's happening in some liquid phase system. So if it's a liquid phase problem, we would want to do it in liquid phase chromatography. All right, maybe, maybe the person in question could email me and I could give them a more detailed answer offline. That would helpful.

Yeah. And Naveen, if in the end we have time, you can always unmute yourself. Just raise your hand. And you can also ask us.

We never proper chat about it and no problems. Yep. Okay, so the final topic for today. Yep. The final topic we'll talk about is actually doing some work on diffusion constant measurements. And this is some work using some liquid phase catalytic reactions. So in the problem we're going to discuss today where interested in characterizing how reactants and solvents interact with porous particles. And the porous particles here, some of them are actually catalysts and some of them are just porous particles. And we're just understanding how small molecules diffuse out of the pores in catalysts and also understanding how solids interact. So this is a liquid phase process and we're looking at some reactions which I'll describe in a moment. So this is a technique that's been around for a little while. It's called a zero column length method and it's based in this case on liquid chromatography. And we're going to calculate counter diffusion coefficients for different reactants in some porous solvent swollen catalytic pellets. And these are also sometimes seen as gate opening support materials. And what we're looking to measure is the diffusion constants for these porous materials under industrial react conditions at about 80 degrees Celsius. So here is the actual reaction. It's a heck reaction. So it's a synthesis of stilbene from iodobenzene and styrene. It occurs in DMF at 80 degrees C. And there's actually a cat, a catalyst, a supported catalyst there which has been used. So what we've done is we've done experiments with packed columns. Some of those columns have different sorts of standard absorbers and some of them have the catalysts that are used for this particular reaction. And we'll see sort of information we can gather from that in a moment just here. Right, so let's press on. So this is an interesting technique because we can actually characterize a single sample palette. So if you give me a Single catalytic pellet, which typically might weigh between 5 and 50 milligrams. We can actually characterize the diffusion properties of a single catalyst pellet, which is, which is quite interesting. So in our problem today, we've got three molecules of interest. We've got stilbene, which is the biggest molecule, iodobenzene which is a reactant, and toluene, which is essentially an inert solvent molecule. We've also got a range of different porous materials. The palladium NCAD 30, that is the industrial catalyst, you can buy that commercially in this project. We bought that. We've got some ideas what the pore sizes are here, about 3.8 nanometer pores, quite a high pore volume. And we've got some other porous materials that are not catalysts, but they're certainly porous materials. And we'll see later on that we've got the catalyst and we've also got a deactivated version of the catalyst. Okay, now what are we interested in? So, so here, if you like, some porous particles in my column, and we're interested in what happens to solid molecules that they pass through the column. Now some solid molecules or reaction reactant molecules will simply pass around the actual columns. So we actually get inter particle countercurrent diffusion. But of course, if it turns out the molecule is quite small and the pores are large enough, we will actually get intraparticle counter current diffusion. So we're quite interested in understanding can the reactant molecules diffuse inside the pores or not? And if they do diffuse inside the pores, can we estimate the diffusion constant? So that's what we're going to try to do in today's little project. So the ZLC theory is based on some classic diffusion kinetics and it's described in the classic book by Crank. It assumes the pellets are spherical in geometry and uniform in diameter. It assumes ficient diffusion applies to the system and at the region of the absorption isome is essentially linear. And there are some standard equations in the literature for analyzing the data. And I'll show the data in a moment. But essentially we're going to look at the desorption behavior and there's some fairly well established mathematical equations which relate the mobile phase concentration as a function of time and the starting concentration allow us to actually estimate what the diffusion constant is when there's diffusion constrained desorption from the actual porous particles. As I said, we'll see a bit more in a moment and hopefully become clear what we're doing here. So how we do these experiments. So we're going to pack the column with catalyst particles, which could be one larger particle or a few small ones. Okay. Now in the mobile phase in the liquid Chemox system, we've got a solvent dmf and we've got solutes and reactants, got iodo, benzene, we've got tolium, we've got still beans. So we're going to set up a constant concentration of this reactant in dmf and it's going to be pumped through the column continuously. And after a period of minutes, the catalyst particles and their pores will become saturated by the reactant, whether it's steel bean, whether it's toluene, whether it's idebenzene. And once we've got that established, we then going to swap from a mixture of reactant and solvent to a 100% DMS solvent and we're going to look at the desorption chromatogram. So this is a study of desorption, okay? And the desorption chromatogram is going to be analyzed using this ZLC theory. Okay? So let's have a look at what this means. So what we're doing here, this is the concentration, okay, as a function of time for the desorption chromatogram. So what, as I said, so what's happening is we've got still bean molecules. They potentially are inside the pores and if they are, they'll slowly dissolve out. So let me just take you some of the results here. So the first thing we want to see here is, let's look here. This is toluene, okay? And what we can see here with toluene is that essentially the toluene molecules come out essentially very, very fast. Okay? So that's a very different behavior to what we're going to show you on the next one. So something like toluene, it's not going to react with the catalyst because it's not reactive in that sense. It's also a very small molecule. So if it does go in the pores, it will diffuse out quite quickly. So in this case we've got fairly clean diffusion and no strong interactions. And we're predicting in these lines because as you'll see in a moment, some of them don't do this. And this is a log scale, by the way. So although you might say it's not linear, this is a log scale. So the concentration has dropped by a lot by the time it gets down here. We're at 0.1% starting concentration. So let's look at. Let's go over here and look at this species here. This is iodobenzene. This is a reactant, okay? It wants to react with the catalyst. That's why it's there. And you can see here that for the porous particles here, there is very fast desorption. But for the catalyst particle with iodobenzene, there's very slow desorption. Okay? So what that says is that this reactant has a preferential interaction with my actual catalyst particles. And if we go up here, we can also see some strong interactions here. This is the product from the reaction. And that product also has quite a long tail to it. And the slope of these lines we can analyze for diffusion. But there's also some strong absorption interactions here. That's partially because the molecules, much bigger and larger molecules, have strong absorption. So we're seeing here. So. And also it's a larger molecule, so it's slower. It's slower to desorb or diffuse out of the pores. So we can see quite different behavior. So the behavior here for the large molecules is quite different to the reactive molecules. In this case here, we're looking at some chemistry between the iodobenzene and the catalyst here. It doesn't vary that much. So this is probably looking at some slow diffusion and, or some strong absorption on the surface of those particles and those pores. And it's just, I've captured that there. And in this particular case, we're looking at the catalyst which is in the red lines or the yellow, and also the deactivated catalyst. Okay, so what can we see? So the first molecule is toluene, and essentially there's not much difference. So whether the catalyst is activated or deactivated. Not much difference. So the second molecule is stillbene and actually doesn't make much difference here. Activated or deactivated. We've got some tailing here, so there is some slow diffusion, but it's the same for the activated and unactivated. In the third case, we've got iodobenzene. That is our reactive molecule. That's the one that does the reaction. And we can see here that the absorption behavior is quite different for the activated where it's in a catalytically active phase and here where it's not catalytically active. So we can see the difference in some of that chemistry due to the catalysis process. So, and from these experimental data, we can can estimate the diffusion constants for the molecules inside these porous catalyst particles. So it's only a brief introduction to that particular. That particular topic, but it gives you some idea of what you can do by using liquid chromatography with porous particles and looking at the rate at which particles can actually escape, or in this case, molecules can escape from the porous structures with constraints considering also reactivity of the materials. So DDLC can monitor liquid phase reactions with porous particles. The differences in diffusion processes can be seen to relate to molecular functionality and size. The active versus deactivated catalyst particles exhibit different diffusion dissorption behavior. And single catalyst particles can be assessed using this approach. So it's been. And these are the people in the group to do all the work. So Arnold over there, his work was on the IGC study. Sarah did the work on the actual monoclonal antibodies in B22. And Sean, where is Sean? Sean's not in that photograph. So Sean, who did the work on the actual ZLC work, he's unfortunately not in that photograph. So I thank you for your time and attention and happy to answer any questions you might have.

Thanks a lot, Daryl, for such nice presentations and your time. I have one question. I think it's about igc. How would you measure the dead volume for a sample and column when analyzing a sample that interacts or absorbs methane?

There aren't many samples like that in my experience. There is another approach. There are some other ways we can do it if we have to at room temperature. Methane doesn't absorb with many things. But if that was a problem, there's a couple of other tricks we can deploy to measure. So there are some other ways of doing it. But I would be very surprised if your material was a problem. Problem.

Okay, thank you. So for the audience, you can submit questions using the question tab or you can also raise your hand and I will unmute you and you can ask your question. Okay. For ZLC experiment, how to make sure the desorption is completed.

Sorry, Mies, could you say that again, please? Sorry.

For the ZLC experiment, this option is completed.

Well, what we're doing is we actually. What we actually do is you actually fit. You. We fit. We fit a line to that. That line there. We fit a line to it. So it does. It doesn't have to finish. We're going to fit a line and use a theory so we don't have to go. It doesn't have to be totally absorbed. We need to just fit a line to the curve to actually get the diffusion analysis. So we don't require 100. We don't require all the models to dissolve.

Okay, the next question is from Mikael Auer. He would like to ask. Michael, you are unmuted. You can ask your question. Yeah, thank you. Can you hear me?

Yes, yes, clear.

I would like to know. You've mentioned that for the IGC also nitrogen could be used as a carrier gas instead of helium. And yeah, there might be some shortages of helium ahead. So this could be interesting. But I guess there are some. You're right there some drawbacks too. What would be drawbacks? And how could we work with nitrogen anyway?

No, I mean honestly, I mean it's a good question that. I mean people have used helium for a long time. But we started using using nitrogen about two years ago. And actually we have. We haven't found any disadvantages. In fact we found some advantages using nitrogen actually get. You actually get better. We get often get better data with nitrogen than with healing, which is a bit of a surprise, but that's true. So I don't think there is. I'm not aware of any disadvantages at the moment using nitrogen.

Okay, great. Thanks a lot.

Sure.

Okay, so I will now give. Now I will unmute Naveen. She also wants to ask question. Naveen, would you please try to unmute yourself?

Yes, hello, can you hear me?

Yes, yes, Berkeley.

Hi Dr. Williams. Thank you for the presentation. It was very nice. Regarding my question that I asked already when you presented the self interacting chromatography, I was wondering if I have a number of polymers and I would like to know the affinity of certain proteins like let's say bsa, humic acid, any type of protein. Know what is that interaction or the affinity between a certain polymer and a type of a protein. If I packed for example the column with polymer powder and then I allow water with a certain concentration of a protein. This is going to allow me to get this kind of information.

Yes, yes, that's right.

Yeah.

So that. Yeah, that. That's correct. So if you were to pack a column with the appropriate powders or in fiber form, whatever form the polymers in that we find you would elude that column. Aqueous solutions with your proteins or peptides in there, then the retention behavior would give you a measurement of the affinity of the peptide or the protein with those. With those polymers at different temperatures, at different phs, at different ionic strength. As you see. But that should work. Yes. I mean there's always a bit of work to develop the method but in principle what you described should work.

Yeah.

So if I have the polymer, for example, in the form of a film, so I can fold it inside the column. Is it going to.

Yeah. So let's, let's talk about a film. The, the only constraint for any of the techniques I discussed today is you need to have enough surface area to see something. So I would probably suggest a powder or fibers would be better. If you've got films, do you know how thick the film is?

I have no idea. Just for example, a certain polymer film. I'm not certain about the thickness, for example. You mean the thickness of the polymer.

I'm trying to work out how much surface area you're going to get in the column. Do you know what the polymers are at the moment? Moment? Are they, can you, Are you able to say what the polymers are you might be interested in?

For example, like pvc, cellulose, acetate, polysulfon, like different types of polymers?

Well, right. Okay, fine. Okay, right. Well, quite a, quite a few of these materials will be available in thin film form. Okay. So if you, if they're available in thin film form, I would chop them up in thin films, I would chop them up and I'd put them in a coffee column. That's what I would do.

Okay. So it's better.

I would, I wouldn't try to fold it up. If you want your book, you want a random, you want a randomly packed column. That's what you want. Okay. And if you fold them up, you won't get random packing. You'll start to get flow channeling and you be complicated. I would just. I would chop them up into very fine pieces and I'd pack them in a column and I would go for films at no more than 20 micron thick. You should, should. I think those materials would be, should be available.

Okay. Okay. Thank you so much.

Sure. No problems.

There is one question that I received really in the beginning. I think that was for Professor Hartman and it is about will fluorines of different size create different signals due to the difference in surface area or volume?

Can you just repeat that once more, please? Ramese?

Will fluorines of different sizes create different signals due to the difference in surface or volume? I think this is about NMR that Professor Hoffman presented.

If it's an NMR question, I'm not the person to help. If the questions about, if the questions about flow rates in chromatography, there is a. There's a good question in chrome. If you want to ask me a good question. It goes like this. Are the results I determine using any of the methods you described today affected by the flow rates. And the answer is in a method like zlc, they are affected by flow rates. But that's built into the analysis in the IGC methods. The different flow rates will give you the same results and the same for the second vero coefficients. So if I change the flow rate from 1 cc to 2 cc or whatever, that should still give me the same net retention volumes. So that wasn't a question I was asked, but a question I'm happy to answer. But if it was the other question about nmr, I'm not the person to ask it anything.

Okay, thank you. I think I can move to the next question and I don't see Professor Hutton in the attendees right now, so let's move on to the next. Thank you for your presentation. If I understood correctly, the only measured parameter from IGC is retention time. I am just wondering to know whether there is any mathematical model of the column is involved in between in order to correlate the relation time to the isotherm. And how accurate is the model based on your experience?

Okay, fine. So the core measurement you're measuring is the retention is the retention volume. Okay. It's a fundamental physiochemical property and you'll use that to construct the adsorption isotherms. There are some mathematics in the analyzing of the chromatogram to get you the isotherm, but that's. That's all quite tractable. The isotherms measured using IGC and traditional approaches generally correlate quite well. Probably only constraint is if you want to do isotherms at very, very high surface coverages, IGC is a bit more challenging. But if you're doing surface areas and things like that, where you don't tend to go, you don't need to go to very, very high partial pressures. The technique works pretty well. That probably answers most of the question. But if I missed anything, please come back to me and I'll try again.

Thank you. The next question is could the theory of IGC be applied to liquid chromatography? Could surface area and surface energies be measured by using liquid probe molecules?

The short answer is yes, certainly. We've done some work on determining surface areas using inverse liquid chromatography. The but that there's no reason why you could not do surface energies as well. I guess the difference to be aware of is that when you do IGC and I inject an octane molecule in nitrogen, all the substrate really sees is the octane molecule. Because the nitrogen molecule effectively doesn't Absorb. If I try to do the same experiment in liquid phase, it's more a competition between the solvent molecule and the solid molecule. So it's a little more subtle as to what you're measuring, but in principle you should be able to do that. Yes.

Thanks a lot. The next one is how to quantify the number of OH groups present on the surface of solid phases and to access acidity constants.

Okay, so let's start with the first question. Number hydroxyl groups. There is no reason why you cannot do titration type experiments inside an inverse secret chromatography experiment. So what's that mean? Okay, so I packed the silica powder into my system. I elute water through the actual column or the mobile phase. I've got may not be water, maybe it's going to be octane or something. In the configuration instruments we have, I have the detector pre column and a detector. I mean just enough. Can you still see my screen?

I think no, it doesn't matter.

So I just kind of show something. So basically, if you were to pack your column with a detector, pre column and post column, you could basically do a base titration. So you could inject something, I don't know if you're extreme like pyridine, and the pyridine would complex with the hydroxyl groups and then you could keep injecting them until the pyridine started coming through the column. So you could do a titration and use that to estimate the amount of silica or hydroxyl groups present, for example. I haven't done it, but that should be quite doable. I think there was a bit more to the question than that. But if the person wants to rephrase it, I have to pick up the second part of the question as well.

Dr. Destazo, it's your question. Would you like to unmute yourself and ask in more detail?

Yeah. Yes. Thank you very much for your presentation. First of all. So the question was the second part of it, it's about the measuring in principle of the acidity constant of these OH groups Y because according. So in some materials, especially in oxy hydroxides, the acidity constants of OH groups is not equivalent. So there are groups that are bound to true three metals and OH groups that have only one metal bound and they have different acidity. And I was wondering whether these techniques can shed light on this very complex problem.

Right, okay, Right. So we're talking effectively about heterogeneity of acidic groups in. On a surface, my View is almost certainly, yes, we need to think a bit carefully about the design of the experiment, but I'll give you an example as to how you could do it. And I think this would work. And we've done something like this before on carbon fibers. So what you do is you would pass through the column a particular basic. Okay, it could be pyridine, could be an amine. Then what we would do is we would allow that to essentially irreversibly absorb on the surface. Then we do a bit like a thermal desorption experiment, but we would do that probably with a second solvent or with temperature. So basically, we would look to dissolve the base molecule from the surface and look at the shape of the desorption curves to try to calculate from that the population of sites they're dissolving from. So think about a. Sort of. Think about a thermal desorption experiment, but done in liquid phase. That's what I would. That's what I would try and do.

Okay, thank you.

Okay, thank you.

All right. I do not see any hand raised and no new questions.

Ramirez, Happy to take any questions offline by email. People may have questions they don't want to share publicly. And do let people know if they want to email me or email you and you can send it to me. I'm happy to answer them offline anyhow.

Yes. So let's do like this. I would say that we had a very interactive audience today. Thanks a lot. And if you have more questions on the webpage, you can see an email address. Professor Darren Williams Address is also public everywhere, so do get in contact if you have more questions. And I think we can, for today, close the session. And I thank you and everyone else who is actually from all parts of the world. We got registrations, so thanks everyone for being here.

Great.

Thank you very much for hosting this very interesting workshop from my side.

Great. Thank you. Well, thanks everybody for attending. And thanks for the very interesting questions as well. Goodbye, everybody. Good afternoon.

Good afternoon.

Bye, bye, bye, bye, bye.