2015 Introduction to the Inverse Gas Chromatography Surface Energy Analyzer & Selected Applications

Start okay, so welcome everybody to the IGCSEA webinar. As I mentioned a few minutes ago, this is a one way audio so you will only be able to hear me. If you have questions, please use the question and answer box or the chat and we will do our best to answer the questions at the end of the webinar and if there are questions please feel free to email after the meeting. Sciencesuk.co.uk I apologize for the audio. We will have to rectify that for any future webinars. It was supposed to be set up to broadcast so you could listen on your computer, but unfortunately you will have to use a telephone. If you cannot have access to a phone then we will have to listen and you can get a listen to the recorded webinar after the meeting. And I apologize for this inconvenience. So just a brief outline. I will just very briefly talk about inverse GC and then spend the bulk of this webinar toiling over some surface energy heterogeneity applications that's the the predominant use of inverse gc, several applications in a couple different fields and then go over some other additional surface properties that you can measure by inverse GC and some bulk properties. Inverse GC is a gas phase absorption technique. It was not invented by Surface Measurement Systems. It was developed back in the 1950s and been used in various industries from polymers to fillers to aluminas, silicas, large amounts of materials, pharmaceutical materials. It's applicable to powders, fibers, films, particulates and even some semi solids. Inverse GC is what we call a molecular probe technique. We use gas or vapor molecules of known properties to study and investigate surface properties and bulk properties of solids or semisolid materials. We can look at both physical and chemical interactions at very low surface coverages due to the high sensitivity of using these gas or vapor molecules. As I mentioned, depending on the settings or method or application, you can investigate bulk, air and surface properties. One of the advantages of the SMS inverse GC is that we can control both temperature and humidity in the column so we can study things at real world conditions. And I will at the end talk about the range of properties that can be measured and used with inverse gc. So I'm going to get started. Here's a cartoon. If you've ever seen an SMS presentation. You've probably seen this cartoon before, but it's a quick illustration of the difference between an analytical GC and inverse gc. In inverse GC we pack the column with the material that you want to study a Powder, a fiber, a granule. And then we have a helium carrier gas. We inject a single burst of a one component system, a gas or a vapor. We measure the retention time, how long it takes that gas or vapor to travel across the surface, and from the resulting peak that comes out, the retention time, how long it takes that molecule to travel across the surface or diffuse through the bulk of that sample is related to surface and bulk thermodynamic properties of the solid material. Obviously, the longer it takes to go through, the more interaction it has in relation to surface energy or free energy. The longer it takes to go through the column, the higher surface free energy of the material under investigation. Since we pack a column for every experiment, it is important to measure the dead volume of that experiment of that exact column. And we typically do that using a relatively inert molecule like methane. So we inject methane, first get the retention time of methane, and then relate the retention time of all other molecules from methane. So we get the net retention time is the difference from our probe molecule of interest to methane. And that is the net retention time for a series of injections that is done to measure the properties of the solid under investigation. So in general, the absolute retention time is not as important as the relative retention time to methane. And then even within that, it's how that retention time is changing as we change flow rate, temperature, probe molecule, humidity or whatnot. So it's typically the change in retention time as we change some other variable is what the predominant factor of inverse GC is. Let me just see. I've seen some questions about some dial in numbers not working. There should be on the invitation you should see the dial in number and before I get too far, if it doesn't seem to work, I apologize. That's on WebEx to dial into that. So I'm going to go see. I don't know if Majid. You can look for the dial in number and maybe put it on the chat if you see it in the Q and A session, but I'm going to proceed here. Surface energy is the most commonly measured property by inverse gc. It is analogous to surface tension of liquids. It's defined as the excess energy of a material compared to the bulk and in particular to composites or inter particle interactions is directly related to the thermodynamic work of adhesion or cohesion. So when interfacial adhesion or interfacial energy is important, the surface energy is extremely important and a strong contribution to adhesion cohesion phenomena. The surface energy is typically divided Into a dispersive and an acidic and basic, or some people call this polar contribution. Surface energy can have an effect on a wide range of physical or processing or just end properties of the material. For instance, if I have a model surface and I increase the surface energy of this model surface, then you can increase the wettability of a liquid. If you're looking at a powder, and if I increase the surface energy of this powder, I will increase the cohesive forces of and you'll see a higher likelihood to agglomerate or stick together. Typically when I mill or grind or do some other surface bombardment, typically I will see an increase in surface energy because I'm making the surface rough or amorphous or exposing buried interfaces or steps or kinks or defect sites. So typically, but not always, as I increase the surface energy, increase how the disorder, I will increase the surface energy. And finally, if I have a model large particle, and if I increase the surface energy of this particle, I will increase the adhesion to another particle. So as the surface energy of these larger materials increase, they overcome the cohesive forces of the smaller particles and I increase the adhesive nature. This, this could be a fiber in a matrix, could be a powder in a matrix, it could be two powders. But essentially as the surface energy increases, I will typically increase the work of adhesion as well. More people may be more familiar with contact angle. Contact angle is a liquid based technique where you put a droplet of a liquid onto a surface, measure the contact angle, then the contact angle is related to the surface energy. Contact angle works quite well for flat homogeneous films. But once you start using powders or fibers or granules, it becomes much more difficult to get accurate and reliable results. There are other processes like the Wilhelmi technique and powder compaction, but they're not ideal. In particular, when you have a powder where the liquid swells within this little clip here, you can see within basically one second, this liquid completely swells into it. So it becomes very difficult to find a true contact angle. So therefore for powders, fibers, particulates, granules and porous films, inverse GC is really a predominant way to measure surface energy of those materials accurately. In particular to the SMS inverse gc, we've developed the and further the concept of surface energy heterogeneity. Most materials are energetically heterogeneous, meaning they have different surface groups, different sites, different materials, different components. Most real materials are multi component systems, so it's important to not get a single value for surface energy. Much like in particle size, you typically Get a particle size distribution. With this technique, we can now get a surface energy distribution to reflect the true heterogeneous nature of real materials. So the way we do this is if we know the surface area of the column, we know the cross sectional area of the probe molecule, we know exactly how much surface is in the column based on the mass and the surface area of the material. We can inject discrete amounts of the different vapors that we inject to cover a certain amount of the surface. For instance in this cartoon, we know the surface area, we know the cross sectional area of these probe molecules. We can theoretically inject a fixed number of moles to cover a fixed surface coverage. Then we can do this at a different surface coverage and then another surface coverage and then another, until finally we can build up a distribution of surface energy values covering a range of coverages on the surface. And this will become clear in a few examples. So just stepping back a little bit to how we measure surface energy. As I mentioned, we use dispersive surface energy that is measured by a series of dispersive molecules such as in this graph, we're looking at series of injections with decane, nonane, octane and heptane. We get the retention time which is converted to retention volume. And we can plot that versus a property of the probe molecule. We get a straight line and the slope is directly related to gamma D or dispersive surface energy. So a few dispersive or nonpolar molecules gives us the dispersive surface energy. Then we can use some acid base probe molecules to measure the free energy, the delta G. We get that directly. And then by using this specific free energy values, we can apply that to various acid base series, in particular the good Van Os choudhury which is taken directly from contact angle. The same gamma plus gamma minus concept applies in the vapor phase. And we can determine the acid base portion of the surface energy. We can also apply other acid base theories, including the Guttmann approach, which is a direct measure of Lewis acidity or basicity of the surface. And I'll show you a few examples in the upcoming slides. Just a quick example to illustrate the concept of surface energy heterogeneity. If you've seen any of our presentations before, you may have seen this. So I apologize if this is a bit redundant, but what we're looking at here is a mannitol sample and then we silanize that mannitol to methylate it. So we take these oh groups and change them to methylene groups or methane to CH3 groups. Okay? And what we can see here is this is the pxrd. We can see, it's a little bit hard to see, is that this is the untreated mannitol in green, the cyanized mannitol in yellow. And we can see that there is no change. So we're not changing the crystal structure. All we're changing is what surface groups are on the surface of the crystal. And some collaborators in the London Imperial College grew large macroscopic crystals about 2 inches long. And they did contact angle on the individual crystal faces. For the untreated mannitol, you can see that the dispersive surface energy ranged from 44 on the 010 plane down to about 39 on the 011. This is of the raw macroscopic mannitol crystals. Then these crystals were cyanized to methylate the surface. And we go from a energetically heterogeneous material to an energetically homogeneous material. You can see all of these are lower in surface energy and they're all the same value because now all surface groups have been methylated. So this is on large macroscopic crystals. We repeated the experiment on actual powders. These were mannitol powders. And what we can see here, as I'll walk you through the graph, we have dispersive surface energy on the Y axis. Surface coverage, this is a fraction of the surface N is the number of moles divided by the number of moles to cover a monolayer surface. So in theory, this could go from zero, which is no infinite dilution. Nothing is covered out, to one where we can cover the complete surface. And you can see here, for the silanized or the as received mannitol, the surface energy values range from about 50 down to about 35. You can see that the surface is energetically heterogeneous, whereas the cyanide mannitol is energetically homogeneous. We've covered all of the surface groups with the methyl groups, and we can see it's very energetically homogeneous. And we can take the average values using fitting this to a distribution curve. And we can see for the silanized, the minimum and the maximum are basically the same, whereas the as received, the minimum was around 37, the maximum around 52 with the gamma 50, which is similar to a D50 in a particle size distribution, was around 40. So if I go to the previous slide, you will see that the values range from around 39 to 44 for the untreated here we get up to about 50. Most likely because these are on real powders. So we're also looking at defect sites, kink sites, and we're much more sensitive to small deviations in the crystal. And we see we get some higher values here that are not detectable by contact angle. So this is the acid base or specific surface energy. Again, the cyanide mannitol is extremely homogeneous and lower in energy. The AS received or untreated mannitol is energetically heterogeneous. Now, how does this maybe apply in a real world situation? If we take the surface energy, as I mentioned, it is directly related to the work of cohesion and that's done by this calculation here. So as the surface energy increases, we increase the cohesive forces of the powder. So in theory, as we increase in surface energy, you see this little cartoon, you will see an increase in the cohesive forces and that may lead to agglomeration, poor powder flow. So this is predicting from the surface energy value that the silanized mannitol, which is a lower surface energy, should have lower cohesive forces or better powder flow, whereas the as received mannitol had a much higher surface energy across the surface coverage range. This might have problems with powder flow. So this is predicted from the surface energy values. If we go to, we did some powder flow measurements using A Freeman Technologies FT4 powder rheometer and we can see what we have is the energy by air velocity. They flow air underneath the cell and then measure essentially the force required to turn this blade through the powder. And, and with low air velocities, we can see a large difference. We can see here that the cyanized mannitol requires very little energy to turn, whereas the AS received requires more energy. That means the powder is more cohesive. If we go to high enough air velocity, we overcome these cohesive forces and the powders look quite similar. So this is exactly what was predicted in the previous slide. We see that these as received mannitol were predicted to be more cohesive. And if we see on one powder flow measurement, we see that exact result that the as received mannitol was more cohesive. Now, if you went to higher velocities, you could overcome those cohesive forces. But at low air velocities, you can see a large difference in the cohesive nature of these powders. So moving on to how else surface energy may affect material properties, we see this is aspirin. At same particle size fraction 150 to 250 microns. This is 75 to 104 microns. These are milled and unmilled These are not amorphous. We can see here that the milled sample, even at the same particle size, same surface area within measurement, within this particle size range, we can see that the milled aspirin has a faster dissolution rate. So we measure the surface energy distribution. And we can see here in this blue and green, the milled samples have some higher energy sites here. So milling increases the surface energy of the materials, even though it's not amorphous. So it's creating some surface disorder and it could be exposing a different crystal plane, perhaps a 011 during milling. And this increases the wettability, increases initial wetting. And that's the reason why we see a faster dissolution. Here's another case study. We're looking at crystalline, micronized and spray dried budesonide. Here is the dispersive surface energy. The crystalline is in blue, spray dried is in red, micronized is in green. Dispersed surface energy is slightly higher for the micronized sample. If we look at the acid base component, we see a much larger difference. We see that the spray dried has a much higher acid base or polar surface energy compared to the two other samples. We can create what we call the surface wettability or the surface polarity, which is the acid base portion of the surface energy divided by the total surface energy. This is meant to give a representation of how much of the fractional surface is of a polar nature versus a dispersive nature. And we can see if we do this plot, we see that the spray dried has a higher percentage of the surface that's of polar surface groups compared to the crystalline and the micronized. So again, this is predicted. We would predict that the spray dried would have a faster dissolution, a faster wetting, because it's more polar, much like water is a polar medium. It should have a faster wetting. And what our collaborators did using the next generation impactor dissolution method, where they take particles of the same aerodynamic size, this was taken at the stage four. So we're trying to minimize particle shape, particle size, particle area effects, and just look at surface effects. And we can see here it's a little hard to see, but this is the micronized budesonide here. This is the spray dried budesonide here. Here we're looking at the dissolution profile and we do see that the spray dried has a faster dissolution rate than the micronized. So again, the surface energy predicted a faster dissolution due to the higher polar nature of that material. And, and then we see that here with the actual property performance moving on to some fibers. We're going to look at some carbon fibers. These are some carbon fibers that have been oxidized. One carboxyl oxidation and one hydroxyl oxidation. This is the oxidation level just by number. This is the oxygen to carbon ratio, taking the virgin or untreated as a reference. So that's why you get one for the CO for the virgin sample and the other ones are taken relative to the untreated or virgin sample. So as you treat them more aggressively, you increase the oxygen to carbon ratio both for carboxyl and hydroxyl oxidation determined by xps and we measured. This is the total surface energy we're seeing now. And we can see as we go from the virgin fiber. This is the carboxyl oxidation. As we make it more and more oxidated, we can see we see an increase in surface energy. This is for the hydroxyl. Again, we can see this is the virgin fiber in red. As we increase and oxidize them more aggressively, we can see we make the surface higher and higher surface energy. We can then take the polarity, which is the ratio of the acid base surface energy divided by the total surface energy. And we see a nice trend that as we make the surface more polar, as measured by the surface energy, we see an increase in this oxygen to carbon ratio. This is for the hydroxyl, I mean the carboxyl. This is for the hydroxyl. And you can see a little bit more clearly we've divided it. This is the dispersive surface energy in orange, the acid base portion. And as you see, as we oxidize the fibers, we see an increase in the acid base or polar surface energy. This is again for the carboxyl oxidation. This is for the hydroxyl oxidation. Here again, we make it more and more reactive as we increase as we oxidize the surface. I mentioned at the beginning, I won't go into the details of the Gutman acid base concept, but that allows us to get a ka, which is an acid number, and a kb, which is a base number. Now, in theory, if we do a carboxyl oxidation, we should see an increase in the acid number. And that's exactly what we see, this KA number. We see an increase in the surface acidity as we oxidize it with carboxyl functionality. Now we see a hydroxyl. We should see a decrease in the acidity, which we do initially. We do see a decrease in the acidity. This is the acid number divided by the base number. We do see a decrease in the acidity but then it begins to increase. And what we believe is happening is, and we see some other evidence in some SEM images, is that as you oxidize these more and more aggressively, you actually create a high degree of surface defects and surface roughness, and you kind of destroy the carbon fiber. So there may be some other effects going on other than just oxidative effects. And that may be why we see this trend reversed at high oxidation levels. For these hydroxyl oxidation, Looking at some multiple wall carbon nanotubes, we have as received annealed and then oxidized. And again, we wanted to look at some surface properties. We look at the surface area, the crystallinity and the XPS surface oxygen content. As expected, as you oxidize the surface, you see an increase in oxygen content. As you anneal the surface, you will cause it to become more crystalline. So we would expect that the annealed surface might be lower in energy and the oxidized surface would be higher in energy. We can see here from the surface energy results. We see exactly that. The blue is the as received. When you anneal it, you actually passivate it or make it lower in energy and more uniform surface groups. As you oxidize it, you increase the surface energy and make it more heterogeneous. You see a much larger span of surface energy values. That's the dispersive surface energy. We convert that to a distribution which you may have seen in some of our other presentations. This is essentially taking this, integrating the area under the curve, getting a very simplified visual distribution function. You can see the annealed has a very narrow and lower energy, whereas the oxidized has a broad distribution of surface energy sites and a broad span of values. This is the acid base surface energy. Again, this is the as received. Once it's annealed, it becomes very energetically homogeneous. And as expected, if it's oxidized, you would see a much higher acid base or polar or specific surface energy. And again, the surface polarity, which is the acid base surface energy, divided by the total surface energy. When you oxidize it, you see an increase in making the surface more polar in nature. This is the Guttmann acid base concept. Again, this is the base number in dashed lines, the acid number in solid lines. As you oxidize it, you increase both the surface acidity and surface basicity. But bigger effect is on the surface basicity of the materials that become much more surface basic as you oxidize them, moving away from surface energy heterogeneity. I wanted to briefly Talk about surface amorphous content. There was a paper that was done a few years ago we're looking at, with Jeff Broome, looking at dispersive surface energy as a way to quantify surface amorphous content. Most other techniques look at bulk amorphous content, like dvs, xrd, dsc. Here we're focusing, since it's a surface technique, we can look at purely on the surface. How much of the surface is amorphous? You have to apply a concept called effective amorphous surface area. You do need calibration curves of known mixtures. We're going to use some lactose and an API as model materials. So essentially what you need is the surface fraction of amorphous surface area fraction of crystalline. You need to measure the surface energy of the raw materials. This is, let's say, surface energy of amorphous surface energy of the crystal material. If these are different enough in energy, you could use this to calculate or predict what the surface energy would be. And we'll compare that with the actual measured values and we'll show that next slide. So the calibration curves we measured did some binary mixtures of amorphous versus crystalline lactose. We normalize the values based on the effective surface area of the amorphous phase versus the crystalline phase. We get the surface energy values, and then we can plot that as a dispersive surface energy versus the fraction of the surface of the available surface in the column that's amorphous, because we know that by the amounts we put in. And we get our measured value versus the predicted value. The predicted value, just to go back, is from this equation here. And we can see we get a very nice curve. This is for lactose. We did the same thing for an API. Again made physical mixtures normalized the amount of surface in the column that was amorphous versus how much of the surface was crystalline, measured the surface energy, and then again we get surface energy versus fraction of the amorphous surface area in the column. And again, the measured values correlate quite well to the predicted values. So in theory, once we do this calibration curve, we could use this as a calibration curve for unknown materials. I see a question about the error margins. If I do go back, not on this plot, but in the paper and here, you can see that this is the standard deviation for some around one and a half millijoules per meter squared. Here's about a half a meter squared in general, and we can show data offline. We've done a lot of repeated measurements. And in general I would say the error bars are about plus or minus 1 millijoule per meter squared is a typical error bar for most surface energy experiments. But here you can see this is the standard deviation for this particular study. And if I go back to the lactose data, you can see the standard deviation ranges from about 0.6 to about 1 millijoules per meter squared. Again, what we can do is we took some quote unquote unknown samples. We took a micronized lactose, a spray dried lactose and a ball milled lactose, measured the dispersive surface energy, calculated what the effective amorphous surface was, and this is saying it was 28%, 77%, 120%. I'll get to why that's above 100. Then we measured the bulk amorphous content versus by solution calorimetry and we can see here for the micronized sample relatively similar 28% versus 20%. I'm sorry, quite different. We can see for the bulk amorphous that the solution calimetry results said almost nothing or very small below the limits of detection for this. That's because for a micronized sample, the surface, the amorphous material is predominantly and almost only on the surface. So a very few small bulk fraction was amorphous, but it's all on the surface. And that's why you can see around a 30% amorphous fraction by inverse GC, but almost a zero amorphous fraction by solution calorimetry. For the spray dried, where you would expect the amorphous fraction to be more evenly distributed throughout the particle, we see around 77% for surface energy and around 72 for the solution calimetry. That makes more sense because again more of the material is actually the amorphous fraction is spread out through the bulk and the surface of the material. For a ball milled sample, the bulk amorphous around 5%. We got around 120 because the ball mill actually got a surface energy that was higher than our quote unquote amorphous standard. It just means that's why we got a value of above 120. It simply means that this sample was even more energetic than our quote unquote amorphous standard of our 100% amorphous via spray drying. So it's really an effective amorphous surface area, not a bulk amorphous content. Moving away in the last few minutes of the presentation, I wanted to briefly go over some other properties besides surface energy that can be measured by Inverse gc. One of those is an isothermic. And in particular you can measure. You do several different injections of different concentrations. From that you can generate an isotherm. This allows you to determine Henry's constants. It also allows you to determine things like BET surface areas. I would never be a proponent to say that an inverse GC would ever replace a nitrogen or volumetric BET surface area instrument, but you can get accurate results. Here is a certified reference material of 2.95 meters squared per gram. It's an alumina sample. By inverse GC, we get 2.95 meters squared per gram. One novel application that we can do that is not easy to do by nitrogen data is we can do experiments as a function of background relative humidity. Here we see at 0% RH, we get 2.95 meters squared per gram. At 50% RH, we get 2.73. Slightly lower, at 90% RH, we get 1.45 meters squared per gram. Essentially what's happening, and this is just a picture taken not from this study, but just to illustrate. As water content increases, we cover more and more of the surface, so our effective surface area that's actually available to us decreases. So as the surface gets more and more covered with water, we see an effective decrease in the surface area. We can also look at how moisture might affect the surface area when we have a phase change. Here we have a spray dried lactose. This is dynamic vapor absorption data. And we can see here at low humidities, we have surface absorption around 35% RH, it goes through the glass transition. Then around 60% RH, it goes through a moisture induced phase change from the amorphous phase to the crystalline phase. So on the same sample in the inverse gc, we could look at it before the glass transition at or above the glass transition, and then at or above the crystallization point. And that's what we did with this study. We looked at the surface area at 0% RH, this is of the amorphous phase. At 35% RH, this is now above the glass transition, you can see relatively no change. But then at 70% RH, we can see the surface area decreases significantly. We can see in situ, we see an order of magnitude decrease in the surface area due to the amorphous to crystalline transformation. There are some. Speaking of phase transitions, there are some bulk properties we can measure. And this is the last two or three slides before we open it up for questions or to close the webinar, we can look at glass transition Temperatures, polymorphs, essentially any second order phase transition. In this experiment we inject a single molecule as a function of temperature. This is the retention volume, the log of the retention volume versus one over temperature. This is low temperature here, high temperature here. At low temperature, where you're well below the glass transition, you will see surface interactions dominate. As you approach the TG of the material, molecular mobility increases within the polymer. And then you see molecules penetrate the bulk of the sample. And then you can see an increase in retention volume. When it's just surface absorption. As I increase temperature, surface absorption should decrease. But as I go through the glass transition, I will see an increase due to molecules penetrating the bulk. And what you can do is measure the TG of material. Again, DSC are much better for just looking at bulk TGs. But here we can look at things as effective of rh. You see how water can act as a plasticizer and sometimes an anti plasticizer. Here we're looking at maltose. We can see at 0% RH we get 360, whereas the literature values again 360. As we increase the humidity, we can see how water plasticizes the material. We see a decrease in the glass transition temperature. This is a study done by University of Purdue. We're looking at amorphous into methicin. Again, as we increase the humidity, we see a decrease in the glass transition temperature, which is what you would expect. And here you can see the water absorption curve. We see a relatively monotonic increase in water uptake. Now, on the other hand, this is for a different polymer. This is an Utogrit E100. At low RHS, we actually see an increase in the glass transition temperature. So it acts as an antiplasticizer. And then you see a decrease as you go to higher humidities. And what we can see, this has a different isotherm shape. This has a isotherm where you would see a stronger monolayer coverage. So what possibly could be happening is water is binding very tightly to the surface at low humidities. And that may actually see a slight increase in the glass transition temperature. And they also did some moisture diffusion experiments. And as you go to low humidities, you actually see the diffusion rate decrease. So the surface is becoming tighter. And that's why you see this increase in TG as it goes through and starts to plasticize. You see that the diffusion rates increase because the molecular mobility is increasing and the glass transition temperature is lowering as the material becomes softer. So to wrap up, we can measure both surface properties like surface energy. We can do acid base. We can do heterogeneity mapping acid base by Guttmann. We can do isotherm analysis looking at Henry's constants surface areas. We can also look at thermodynamic works of adhesion cohesion. We look at heat absorption. We can measure some bulk properties. I just talked about glass transition temperature. Other bulk properties that can be measured are solubility parameter, both Hildebrandt 1d, Hanson 3d and cross linking density. So with that I'd like to thank all the people at SMS currently and previously and some of our collaborators over the years who have helped us and helped present some of the data. So I would like to see if there are any questions perhaps and see what we can do. What I can do I do see some questions in the chat. I can make this presentation available both via PowerPoint and via the recorded webinar as well. I'm going to open up the Q and A panel. I do see a question on the chemical composition of the Utre Grit E100 and to be honest I'm not entirely sure. Off the top of my head I think it might be a PvP, a polyvinyl parole based material. It's a commercially available material. I had to review the paper that was done by Rodolfo Pinal's group at Purdue, but I believe it's a PVP material. There's another question asking can surface energy be measured via adsorption isotherms using a dvs? The answer is yes. In theory you can measure surface energy by dvs. However the DVS method is based on a surface a spreading pressure. So therefore you have to find a vapor that only sorbs on the surface of the material. So if you have a rigid sample like alumina silica, then you can get very good results. If you have a polymer like hair or PVP or amorphous materials where your vapor absorption can be dominated by bulk absorption, you cannot use those isotherms to get surface energy values. So if I had a a bulk sample where I have bulk absorption and surface sorption you cannot use that by dvs. So I will say for rigid materials you could use those isotherms to generate surface energy and surface energy distributions. But for polymer or for most pharmaceutical materials or food materials it is not practical to do that. Let's see other questions I have here Limited detection for amorphous content that depends on how different in energy you have are your reference crystalline materials and reference amorphous materials. It's not as sensitive as probably some of the other more sensitive techniques. DBS is more sensitive. For bulk amorphous content there you can get down to about 1%. Limited detection and limited detection about a half a percent. Limits of quantification around 1% by DBS. But that's a bulk amorphous content. I would say probably about 5%. May be the limits of detection and quantification. For inverse gc. It really depends on how different in energy the crystalline phase is from the amorphous phase. If there's a 20 to 30 millijoule difference in energy, then your limits of detection and limits of quantification go up. If you're only a 5 millijoule per difference millijouke meter squared difference, then your limited detection and limited quantification go down. Any questions? With any experience with Titania? We do titania quite a bit. We do have some experience with that. For some more energetic materials like titania, alumina, silica, what you may have to do is go to higher temperatures. At room temperature, sometimes the probe molecules take too long to elute and you have to go to higher temperatures in order for the probe molecules to elute at a reasonable time. We typically use weak acids and bases. It's not things that you would use in a titration experiment. We're using things like ammonia or co or things where you titrate acid sites. We're looking at fizzy sorption. So for acids we often use things like chloroform, dichloromethane for bases, toluene ethyl acetate. These are weak acids and bases. We're measuring physisorption, not chemisorption. So we do have quite a bit of experience with Titania. I also didn't mention we've done a lot of fibers, both hair fibers, polymer fibers, cotton fibers. All of these can be done very easily by packing them into the column. We also do a lot of work with films. We do have a film cell, which I didn't mention in this presentation. If you want to measure a flat, let's say metal plate or piece of paper or polymer film, you can do flat films as well. Let's see. I have a question. Will the dispersive surface energy be the same for a micronized API made by the same method, irrespective. That's a good question. Will the dispersion of surface energy be the same for a micronized API made by the same method, irrespective of any IGC method used? If it's the same material you should get the same answer by inverse gc. One of the things that happened in the past with our older systems and with people who weren't very careful about making sure they had an infinite dilution is real materials are energetically heterogeneous. So I'm going to go back to, for instance, an example, just looking at this mannitol example. If I go back to several slides up. If you weren't careful in the past, you assumed you're at infinite dilution. Now, that could mean you're right here on the curve. If you weren't careful, you could be here on the curve. If you weren't even more careful, you could be here on the curve. So you might get different results. So if you measure different samples at the same fractional surface coverage to make sure, whether it's by our IGC or a different IGC, or you have different IGCs in different labs, as long as you're at the same fractional surface coverage, that normalizes the data to make sure that you're accessing the same sites. You should get the same answers for a micronized API, whether it's measured in one lab versus another or one system versus another. In the past, you may see in the literature surface energy values ranging by igc because people didn't take careful time to make sure that either they were sure that infinite dilution or their surface to volume ratios were quite different, such that they may be on different spots of this curve. Real materials are energetically heterogeneous. So it's important to understand that you have to make sure you report values at the same surface to volume ratio to make sure you're comparing apples to apples. Magid, are there any questions that I've missed? No, I think you covered the film cell question and also the results on hair. I mean, we have some additional examples to send to individuals who ask for some results on different materials, which we can email later. It's just a database. If you could mention the wide range of choices for the solvent. Sure, we do have a wide range of solvents. There's a database into the system that you can load a wide range of solvents. The only limitation is the boiling point of the solvent. It should be below around 220 degrees C to get reasonable vapor pressure. But we do have our experience. If you see something here that interests you, maybe, but on different materials, then just email us. We can send you some more information. I do see an inference. What is the surface coverage? I did go over quickly. Surface coverage is N, which is the number of moles that were injected divided by the number of moles that would be required to cover the monolayer surface. Because we know the BET surface area, we know exactly how many moles it requires to cover the complete monolayer, which would be kind of wet the surface of the material. This is a fraction surface coverage. So at 00, we inject zero moles. Obviously, a value of 01 would mean we've injected enough moles to cover 1% of the surface. At 10, that means we've injected enough moles to Cover 10% of the surface. So you can see here that these materials often become energetically homogeneous after you cover maybe 10 to 15% of the surface. We don't always go out to a full coverage. Yes. So you have a question you would always need. For our method, we would always require you to either know the surface area by measuring it by inverse GC or from nitrogen data. We require you to enter the surface area before doing any experiments, because we want to make sure that you compare values at the same surface coverage when you want to compare different samples. So we do require the BET surface area either from our instrument or from another instrument. Before you measure the surface energy, we do require the surface area. Experience with carbon blacks. We do have experience with carbon blacks. Again, with carbon blacks, you would probably have a difficult time getting data at room temperature. You most likely would have to go to maybe 60, 70, 80 degrees C to get reliable retention data. Otherwise you have to use such a small amount of material, it's not really practical. You may only put 3, 4, 5 milligrams, and then that's not really representative of the material. So for carbon blacks, we often go to 60, 80 degrees C. So the probe molecules elute at a much more reasonable time. Otherwise the peaks take too long to come out and they're too broad and too flat. Do you think surface and bulk amorphous material will have different Gibbs free energy? Well, one thing to measure, Gibbs free energy is a surface property. So it would be tough for us to say, would a bulk amorphous content be different? Because by inverse gc, a Gibbs free energy is a surface property. So it's a surface absorption property. So it's really difficult to say, would the Gibbs free energy of the total system be different? Yes, but we're only measuring the surface. That's one thing to understand. If you're measuring by solution calorimetry or other things, you maybe get a Gibbs. At free energy, you're getting a delta G and a delta H, but that's of the bulk here. Just remember we're only measuring the surface by inverse gc. Dan, could you also go over the advantages of SCA over IGC in terms of IGC uses just a mean value whereas SCA covers the whole surface. Well that's I did mention that and this is in our current we made a first generation inverse GC and other home built systems would not have the ability built in to do this hydrogen 80 distribution. This is the reason why we require you to enter the surface area and then you can get this full distribution and the system is calibrated to do injections at fixed number of moles. So that way it makes it very easy to build this heterogeneity profile which would not be possible by other homemade techniques. You can do it by generating isotherms, but it's a lot of calculations. Everything's built in with our particular sea. Okay, well if there's no other further questions, I'm going to close the oh the presentation will be posted. I will send a link via email. It's not on the web yet. We will post Both just the PowerPoint presentation and the full webinar on our website and you will get a link if you registered we have your email. I will send that link out. It may take a day or two to get posted, but we will post that on our website and we will also email you a link to that and you can download both the presentation and and or the full webinar with audio. So to close I'd like to thank you everybody for their time and attention. Again, if you have any questions, please email us@sciencemsciencemsuk.co.uk or you can email me directly at Burnette B U R n e t tmsna.com so I thank you again for your time and attention and I hope we've showed some information that was useful to you. Thank you very much and have a good day. Goodbye.