 The Archeological Research Facility is located in Weichen, the ancestral and unceded territory of Chochenyo-speaking Olenii people, successors of the historic and sovereign Verona band of Alameda County. We acknowledge that this land remains of great importance to the Olenii people and that the Arche community inherits a history of archaeological scholarship that has disturbed Olenii ancestors and erased living Olenii people from the present and future of this land. It is therefore our collective responsibility to critically transform our archaeological inheritance in support of Olenii sovereignty and to hold the University of California accountable to the needs of all Native American and indigenous peoples, and I always add by our actions as well as our words. Today we're hosting Derek Popple, whose the title is already up so I'll read it. Derek's a 60-year PhD student affiliated with the Department of Chemistry, and beyond his thesis work with the Zeddy Group in Physics, he's focused on nanomaterials and electron microscopy. Derek is interested in the intersection of the arts and science. In collaboration with Professor Stephanie Zaleski from CSU East Bay, they're using the advanced electron microscopy tools available at the RF and other places in Berkeley to study the nucleation and growth of metal soaps within oil paints. Okay, welcome Derek. Thank you. How's it going everyone? Thank you so much for having me. As I stated in the intro, I'm a 60-year graduate student in the Department of Chemistry, although my thesis work is in the Department of Physics, so I'm well versed in kind of hopping around to different departments and seeing what interesting research I can dig up. So the topic of my talk today is going to be about some of the work that I've done applying the electron microscopy methodologies that were developed at Berkeley Lab to the study of oil paints. Again, this is all in collaboration with Professor Stephanie Zaleski, a conservation scientist and chemist at Cal State East Bay, so sort of eternally grateful for her guidance and the ability to collaborate with her. So artworks such as oil paintings have tremendous cultural and financial value within our society, but unfortunately many of these are suffering degradation, as shown here, and so these blemishes or dimples that have popped up on a variety of oil paintings that were painted anywhere between the 15th to 19th century. And so originally this was thought to be sort of an isolated issue, but further research has found that it actually affects I guess the majority of oil paints with some projections expecting that two-thirds of oil paintings will see a similar type of degradation within their lifetime. And so since this was such a widespread issue that affected collections not only here but around the world, a lot of work went into figuring out what exactly was causing these blemishes. And so what was found is that this is from the formation of what are called metal soaps, and so this is what happens when the metal that's contained in the pigment reacts with the carboxylic acids that are introduced from the oil binder in the oil painting. And so here's an example of a painting published in a work by The Met, and in the bottom right you can see sort of this dimpling or blemishing that's present as a result of this soap formation. So if you actually take a cross-section of this degraded area, what you can see is that this is the scale bar is 40 microns, and so you get these crystalline domains of in this case a lead soap that can range from 100 to 200 microns. And as these grow, what can happen is that they'll start to protrude and disturb the surface of the painting. So that's what's giving rise to the dimpling. But then also if it gets too big it'll cause flaking and a sort of delamination of the top surface of the paint. And so not only are you starting to get defects at the surface, but you're actually getting the paint that's giving the quality of the image that's starting to fall out. And then the other thing to keep in mind is that because you're changing the chemical composition of the metal, typically the leads and zinks that suffer soap formation are tip oxides and so they're commonly used in whites. As they change chemical composition, also something that's also perturbed is the electronic properties. And so these pigments can start to become more and more opaque. And so in some cases you can see features from the canvas or the wood substrate that hosts the painting. So not only is this causing degradation to the pigment, but also it creates transparency as well. So since this was such a widespread issue, a lot of work went into studying what metals were present, what types of binders and oils can give rise to this degradation. But there is limited work on the mechanism of formation of these defects. So one of the experts in the field, a guy named Joan Hermans, he's out in the Netherlands and his whole thesis studied this work. I think he's now a faculty member, but he's sort of the world expert in studying soap formation and paintings. He was able to do some clever spectroscopic studies to figure out how this degradation process was happening. And so as shown in the top panel, you can see on the left there's a large particle of zinc oxide. So that's actually the pigment. And then these squiggly lines that are present represent the semi-polymerized oil matrix. And so as ions leach away from the zinc oxide particle, what they found in this paper is that it first forms a small complex. These complexes can start to polymerize into a one-dimensional chain, which will eventually convert into a two-dimensional crystalline region. And so that's what we're seeing in this soap formation. And so as I mentioned previously, there's a lot of established literature on what metals or what carboxylic acids can cause this degradation. But this is one of the only works investigating the mechanism of how this is actually occurring. And so the other thing to keep in mind is that because this is spectroscopic study, it's hard to get an imaging component. They can see sort of chemical traces of the compounds that are causing the degradation, these four compounds that are shown here on the bottom. But it's hard to link that spatially and actually get imaging that shows you how not only do these crystals, how do they nucleate, but also how does the crystal grow and is there anything that you can, any insight that you can gain from the structure of the crystal, they can tell you how to, how these materials are forming. And hopefully if you know how they're forming, you can help to treat them and slow their growth. Unfortunately, I think do the energetics of this precipitation process, the reaction is downhill, so I'm not too hopeful that it'll be reversed, but hopefully we can prolong the life of the artworks if nothing else. So in order to study this, like I mentioned, I'm an electron microscopist by training. Most of my work is focusing on nanomaterials for energy applications. But I'm kind of in the know about some of the techniques that were developed up at the National Lab at the National Center for Electron Microscopy. And so I read this paper about 40 STEM being applied to the study of beam-sensitive materials. I thought, huh, I think I'm working on some stuff that's pretty beam-sensitive. Maybe there's something there, some potential overlap. And so just as kind of a brief background to some of these electron microscopy techniques, I'm sure many people that are involved in research of different artifacts for archaeological applications are well-burst with scanning electron microscopy. And so in that technique, you have a focused electron beam that rasters across the surface of your sample, and then you're collecting the backscattered electrons in order to create an image. And by doing that, you can get below the diffraction limit and actually see things that couldn't be resolved with obstacle light. The other common form of electron microscopy is transmission electron microscopy, where instead of collecting the backscattered electrons, you bathe the sample in sort of a collimated or parallel beam of the electrons and collect the electrons that pass through the sample. So you're looking at ultrasound samples, because you tend to use higher accelerating bulges, you can get much smaller resolutions, getting to the point where you can see nanoparticles or even features on a single nanoparticle. So if you put those two techniques together, the scanning component and the transmission component, you get STEM, which is scanning transmission electron microscopy. So instead of a parallel beam, you focus the electron beam down to a nanoscale probe, raster that across the surface of your sample. And so this can actually give you incredibly high resolution. Some of the other projects that I work on using the really fancy microscopes available in STEM, we can resolve individual atoms contained within a carbon nanotube. And so by doing that, we essentially take a picture of the atomic lattice, and that's again possible because you have this highly focused electron beam, which allows you to resolve the column's atoms. And so in a typical image, what you do is as you raster the beam across the surface of the sample, you'd have an annular dark field detector, that's this ring that's shown right here, and that would collect an intensity just based on the number of electrons hitting the detector. And you can get an intensity value for every single pixel, which allows you to create an image, such as what's shown down here in the bottom left. And so this technique is well established. They still use instruments commercially, but what's come about through advances both in instrumentation and things like detectors, but also in software and data processing is a technique known as 4D STEM. So as you raster the electron beam across the surface of the sample, you not only collect intensity information with your HADF detector, but you also have a detector at the bottom that can take a picture of the entire diffraction pattern. And so that's why it's four-dimensional. You have real space X and Y, as well as reciprocal space X and Y. And so electron diffraction works analogously to X-ray diffraction, where you can create diffraction patterns from your sample, which allows techniques for structural elucidation. But the real power of 4D STEM and why people are getting so interested about it is because you have a ton of information because you have the whole diffraction pattern at every single location, every pixel in your image. And so you can do things like you can sort of decide what you want to be imaging after the fact. And so it really adds a lot of power in your post-processing. You don't have to take multiple acquisitions at different focal lengths or with different conditions. Sometimes you can just change your virtual detector and decide what part of the diffraction pattern you want to look at and change the image after the fact. So it's incredibly powerful. I guess, unfortunately, because it's four-dimensionally, you also collect pretty large datasets. So we do relatively small acquisitions of about the scan is about 40 by 40 pixels. And it's 800 megs of data. It's common to do a 500 by 500 pixel scan. So you're getting to 80 or 100 gigabytes for a single image. They have a lot of work that goes on at the National Lab as well. That's just focusing on how to work with that much data. And in some cases, when you work with the nice microscopes, they have a fiber optic cable that goes straight to the supercomputer to help parse the data and make it small enough to work with. But that's not the focus of the talk. We'll get back to the little things. So one way in which this has been applied, here we have a, or I guess in the panel on the top right, we have a conductive organic polymer and scientists were trying to understand how this material crystallized. And so it was known that this organic polymer underwent or stacked via pie stacking to help with the crystallization process. And so what they did with this 47 study is they looked at the relative orientation of features in the diffraction pattern that come from the one-dimensional polymer chains. And one feature of diffraction is as you twist the crystal in real space, you also twist the diffraction pattern as well. And so by looking at the relative angle of the diffraction pattern at each pixel, you could figure out the orientation of this one-dimensional polymer. And so by doing a lot of data processing, you could actually create these color metric maps where the color is linked to the relative angle of these chains. And so in doing this, you're getting a lot more information that couldn't otherwise be gained through standard imaging. And it allows better visualization that helps scientists to tease, tease apart what's going on in the crystallization process. So since we're working with some sensitive, being sensitive materials, we wanted to apply this technique. We wrote a proposal for the molecular foundry up at the National Lab and it got accepted. And so we've been starting to collect some initial data to proceed on this project. So the project has three main phases. The first is just assessing the suitability of the technique. We had a good hope that this would actually be a good technique, but sometimes the materials can be too beam sensitive or you might not have the sort of the correct resolution of the diffraction pattern, like the materials may not scatter strongly enough. And so we weren't even sure if we could get signals. That was the first check. In addition, we were trying to figure out what the ideal conditions were for imaging. And then as we moved from our reference material, we want to start to look at artificial paint samples. So look at a simplified case of the real thing to see if we can see any insights into crystal growth, especially with respect to artificial aging under different conditions, such as variable temperature, variable humidity, trying to figure out what's, what affects this crystallization and how does this perturb the size of the crystals or how they pack, things like that. And then the final phase of this project is moving to the study of real paint fragments. So real oil paints may have different pigments present. They could have different oils. Artists like to do all sorts of weird stuff, like mixed dirt into their paints or things like that. So these would be incredibly heterogeneous samples to study. But we want to see that as, as we transition to the real thing, do we still see these same insights that we gained in the model system and see if the same findings can be applied. So this is sort of my methodology slide. And what I've learned throughout my PhD is that you need really big microscopes to study really small things. And so how this experiment is sort of carried out, I synthesized the reference material in my lab. This is a zinc steering. So I did a small reaction. And then I dispersed this on a TM grid. So this top image is an optical microscopy image of the copper TM grid with a carbon support underneath. So this grid is only three millimeters wide. And then if you zoom in 50x with an optical microscope, you can get to a single gap in the copper mesh. And these crystallites are sub micron, even into the nanometer range, or I guess hundreds of nanometers. And so they're these small sort of flex they're adding a little bit of contrast. And so they're sitting on the carbon support. So then we bring this up to the National Center for electron microscopy and load it into what's, believe it or not, one of their smaller microscopes. And this is capable of doing TM stem and 40 stem all in the same instrument. So this is really a good microscope to use. We don't necessarily need the atomic resolution for what we're looking at because we're looking at these features that are nanometers to getting into the micron range in terms of the domain sizes. So here's a sample of some of the data that we're able to collect. So on the left, we have the HADF image. So that's high angle, annular dark field. And these sort of spider webby filament like structures, that's an amorphous carbon support that's helping to catch the crystals. The crystals themselves of this ink stirrate are these chunks in here. There's also one on the bottom left. And we take an acquisition within the box that's indicated in red on the HADF image. And so in the data processing using some Python code that's been developed up the hill, what you can do is you can take the maximum intensity at every location within the diffraction pattern and sum that into a single image. And so by looking at the maximum diffraction pattern, you can see all of the diffraction peaks that you have at any point in your sample. And so peaks that are spread sort of equidistant from the undefracted beam, this very bright spot, are going to be the same diffraction spots, but they may be at different orientations again due to the rotation of the crystal. And so in this Python code, what you can do is you can arbitrarily define an artificial detector and say, okay, now I want to back out another microscopy image only using signal from those specific peaks. And so by doing that, you can start to create virtual images. So here you can see some cuts in the crystal right in here where you're seeing the black, which represents the vacuum. So that's no sample. That's just unscattered electrons. And then you have similar black regions in here and in here. And you can see that those same features are recovered in the virtual image. But other features that you can see based on the varying intensity, in certain regions we have more scattering. And so that could be either a thicker crystal or potentially just a denser region of the crystalite that's causing stronger diffraction. And so what we're hoping to do is locate some of these zinc soaps in an oil paint and then acquire a higher res 4D stem scan and start to correlate using the diffraction. So we can figure out, okay, this is actually the material that we're interested in, but then also start to gain insight into the orientation or how there's changes across the scale of a single crystal. Something else that you can do is, and so this was again collected on the same region as the same data set, you can artificially define an aperture and see what regions of the sample were giving rise to that specific diffraction peak. And so here I've selected just one small set of peaks and we can see that it's coming from the upper left portion of the crystal. What I just indicated was a little bit thicker, more strongly diffracting. So that's suggesting that this region of the crystal has a particular orientation, whereas other parts of the crystal may be twisted or oriented differently. And so you can imagine using some of the code, you might be able to carry this out for all of the range of angles from plus or minus 90 degrees and create a similar colorometric map to what I had showed earlier as an example and start to understand how these crystals are oriented and gain some insight into how these materials are actually crystallizing. And so in terms of what's been accomplished and next steps for this project, the first step was just identifying diffraction peaks, making sure that we are looking at what we think we look at. We tend to image these at liquid nitrogen temperatures in order to prevent beam damage. And for about two sessions, we're like, oh no, are we just looking at ice that's formed on the sample? But then we heated it back up to room temperature and there shouldn't be ice at room temperature. So that gave us pretty good indication that it's actual sample. And then other things like determining ideal imaging conditions to maximize signal to noise so that when we switch the real samples, we know what conditions we should go in with. And so we've started some work with model samples. And we're transitioning to zinc soaps embedded in these, it's moving from the reference samples to zinc soaps embedded in these oil paint matrices to help understand are these samples going to be too thick, making sure that there's not going to be any challenges to the actual imaging. And so again, the long term goal is to start to look at age and unaged samples, see if there's differences in the crystallinity, and then start to move to actual oil paints to see what insights we can get from a sample that's actually flaked off and fallen from an oil paint, a real painting. And so the other sort of half of this talk, something that I'm not going to go into too much detail, but I'd like to mention briefly, is the work that I've been doing with Professor Stephanie Zaleski to start the various first conservation science collaboration. And so these sorts of labs are present at places like the Getty in LA, the Met, in New York, and Northwestern Chicago. But right now there's, there isn't a formal research group that does this type of work anywhere in the Bay Area, which is insane, considering we have museums, we have some of the strongest universities, we have two national labs, there's all of the right conditions to set this up. And there's been some initial work to help scope this by some consultants employed by the Getty, but we kind of need some central people to help catalyze this and make sure that this, what starts as an idea doesn't just lose steam and slide back to the position where we're at right now. So the goal of this is to unite some of the fine arts museums as well as local universities and national labs to create a formal research group. And so again, I'm super grateful for Professor Stephanie Zaleski's guidance. She's done a number of high profile postdocs at places like the Met, Northwestern, Library of Congress. And so she's really the residing expert in this collaboration in terms of the art and conservation science world. And so right now we're putting together a white paper and we've assembled a budget, so we're trying to apply for funding. But if you're interested in learning more about this project or potentially getting involved, this could be a great opportunity beyond just conducting research between material science and conservation science. Major goal of the center is actually to provide teaching and training opportunities for underrepresented students, which is a major plus of her working at Cal State East Bay. So we're hoping to not only get more people into the field of conservation science, but also diversify the field as well. And that's it for the talk. I'm happy to take any questions. So thank you again for your attention. Go for it. Well, we're just a national lab. We've got the SCN lab we've got reporting to. And second of all, on your poster, you have a bunch of paintings. Have you worked on any examples or paintings that this technique that you've been working on with looking at how the knowledge. Yeah, so for those joining remotely, the first question is, where is this electron microscopy facility located? So it's at Lawrence Berkeley National Lab, which is just up the hill from campus. And so as part of their user facility called Molecular Foundry, they have a separate building that's electron microscopy facilities. They have five or 10 of these instruments at price tags ranging from one to 10 million. So incredible resources that we have here that are unmatched in a lot of areas. And then the second question is the abstracting which had paintings in it. Have we actually looked at any paintings? So not yet, but that's the ultimate goal. Right now, we've just been looking at reference materials. And then sort of in the next couple of sessions, we're going to hopefully load some paint fragments into the microscope and see what we can find. But right now, we're just making sure that the Yeah, exactly. You've got to be careful what we're shooting an electron beam at, you know, you don't want to degrade it. Oh, what's your Derek? Is water required for initial ionization of the metal pigment for the soap reaction to advance? The implication of that question is that holding a moisture equilibrium in the paint layer can help arrest the process. So that's a good question. I think there's been some initial findings that have suggested that humidity can help to advance the or I guess accelerate the aging process. So I need to look more into sort of the role of the of humidity and of water in the actual mechanism. But yeah, that's that's a good question, something that we're keeping an eye on. Another question. So the question is, what's the role of the oil based paint in the oil painting and in this degradation process? So this has actually been interesting projects for me to work on to kind of get up to speed on the art side of it as well. And I'm a screen printer kind of on the side, but don't do any oil painting myself. So I'm like, Oh, well, I have to learn all about how, you know, how people make these oil paintings and the formulations the pigment. And so my understanding is that you have the pigment particles, which are typically inorganic. And then you grind them in the oil. And so the oil helps to not only disperse the pigment particles, but it also creates this matrix that helps to support them as well. And so that's known as the binder in oil painting. And so what's typically used is like a linseed oil. And that's a mix of several different long chain carboxylic acid. There's like stearic acid or palmitic acid. But there are some like trace components and trace carboxylic acids as well. And then there was a comment in the chat, not all cross linked oils have free fatty acids, the source of the steroids is the huge question. Yeah, so that's that's something that's interesting as well. There's some amount of cross linking in the the oil from what I understand. That's what's helping to make this matrix. And so yeah, it is a good question of if you have something that's covalently bound and semi polymerized, how is that actually kind of leaching out and helping this precipitation process with the inorganics? And so the other question is sort of with connections to the oil paintings or what? No, I'm really pulling stuff out of thin air or with the cave paintings. I mean, um, so this has been noticed mostly in lead and zinx. That's those are kind of the main suspects in terms of the metal ions. But then there's also some evidence of other metals like aluminum soaps forming, although the I would suspect the energetics that reaction arm is, it's not as downhill for the reaction proceed. And so it may happen to a lesser extent. But yeah, it'd be interesting to see who knows, maybe these cave paintings were like a van Gogh or something and now integrated into what we see today. Yes. Yeah, I would think of actually two questions that you mentioned already and one relates to looking at these. Like, have you done any work or has anyone done any work at looking at how particular, you know, recipes would either be encouraged or inhibit the creation of the mineral soaps? Yeah. And whether it would be comparatively, or colors that show that. And the other question I wanted to ask was, so you talk a little bit about looking at different conditions for energy by raising them lower in temperature in particular. And I was just wondering, what have been done, like that seems like the most relevant in terms of if you can't reverse the process, but if you can inhibit it, understanding the conditions, you know, more certain temperature conditions under which the soaps form, if there's been work done on knowing what those conditions are, or it seems like a good place to experiment. Yeah, definitely. So the two parts of the question were what works been done on the actual artist formulations to sort of accelerate or inhibit the growth of these degraded regions. And then also has what work has been done on things like temperature and humidity. So, there has been some work when people were looking at these, especially across a variety of different paintings. People started to be interested in what metal ions would cause them this or what potential oils. I'm coming at this more from the, as the microscopist, so I'd have to defer that to my collaborator, she would be able to tell you both the art side and then how it starts to feed into the material science aspect of it. That's a good question. And then, yeah, these conditions such as like temperature and humidity, in general, a few cool things down, chemical kinetics, slow down. So I'm sure if you had these paintings cryogenically frozen or something like that, that would be excellent, but you're not going to, yeah, not very practical. So that's kind of the whole thing when you're thinking about degradation is what can actually be done because these paintings, if it's in a collection or on display, it may belong to the public or may belong to the collection and they generate revenue and keep the facility alive by showing it to the public. So it's like, how do you keep these in the best state possible, but sort of in a reasonable way as well. But yeah, I think the role of temperature, so artificial aging of these in the presence or in absence of humidity would be our sort of interesting directions that we're hoping to probe and with our reference materials. Yeah. So I'm excited about the kind of sampling forever. So you're talking about paint chips, for example. What other types of materials does quality stem provide? For example, if I understood in a way that stone is damaged by wildfire, and I want to quantify that, that's a respectful change from that. Is that the kind of thing that a sample of a better off order set to you as a tangible chip could be, you know, evaluated through quality stem? Because I don't know how to do that with XRB, but I don't know how to do it with quality stem. And I'm curious what would be the better part taken of the amount of damage to such a thing or something like this. Yeah. So the question is commenting on sampling and sort of what's the power of 40 stem beyond studying these metal soaps. So with any TM sample, it has to be ultra thin. If you get above 100, 150 nanometers in thickness, it's too thick, like you won't get the, it's not transparent to the electron beams, you don't get signal. And so what we do is we'll, for the reference materials, I use a sonicator to disperse them in alcohol, and then I drop cast them on the sample. So they're these really small crystallites. But then what my collaborators done for her initial studies, when she was just doing imaging via stem was you can take these sections, embed them in a resin, and then use a microtone to slice ultra thin samples. But then you're like, okay, well, you have this resin and then you have ultra soft material. And so it can be hard to cut. So with respect to stone, some things that are interesting that you can do with 40 stem is math lattice distortions. So you can actually look at the atomic lattice and start to figure out where you have strain or things like that. And so you might be able to see over sort of say several crystalline domains, like if you got say oxidation of your inorganic materials due to the elevated temperatures, that may cause changes to the lattice and that might be able to be picked up via 40 stem. Yeah, exactly. You kind of need like two different samples or because the samples are so small, you'd either need to, it's kind of hard to see changes to it over say like millimeter scale. So you'd have to take two different samples and compare it. It's a cool technique. I think with any technique, you have to be careful not to over apply or you know, people get into their specialty and think it'll be the best technique for everything. It's still a lot less destructive and a lot more applicable to the broader range of material dynamics are me, which I had to grind everything up and compress it. It's down like at least as far as conservation is thinking about, you know, travel conservation efforts on precious and different places, you think about the minimal amount of handling, you're giving me some sort of qualification of the damage, but I hear the fire regimes would be useful. Question right there? Are modern, the real things subject to this kind of degradation? And there's also an interesting comment on the chat that I didn't quite get. Um, yeah. So the comment was, aluminum stearate was added to commercial paints as a dispersing agent starting in the 1920s. It's almost all 20th century paintings will develop soaps, while aluminum stearate is still used today. They're used in much smaller proportions than in 1920 to 1950. Thank you for the comment, Dale. Wait, next time we'll have you coming up and giving the talk. You're obviously an expert on the form, sort of the formulations and that, that sort of thing. So, yeah, I mean, if you have the, like, so say for the reference materials, I started with, I think like zinc nitrate, and then I added in the steric acid. And if you stir it and heat it, you form these precipitates. And so if you have these metal ions that are particularly susceptible to it, like a zinc or a lead, and you have the right carboxylic acid, the reaction is downhill. And so that's why you see this precipitation process happening spontaneously. So, yeah, if you're not careful about your materials, that will start to be an issue. But the sort of the goal of a lot of the conservation science work is to help inform conservators, but also inform materials, or artist material manufacturers, so that you can reveal sources of degradation that people didn't know about. And hopefully paintings that are created with new materials moving forward will be more archival and less susceptible to this form of degradation. Yes. Thank you so much for the talk, it was really interesting. Apart from the painting itself, too, I would be interested to hear you think about the airplay of varnishes along with this effect, especially to these varnishes have fallen in and out of style throughout the screen. I've been made of so many different things, anything from climbers and official acts, and now we work in vetic varnishes. I would just carry it to see how that kind of looks to effect the painting. So the question is, what's the role of varnishes in this sort of degradation? So from the literature, people are like, they think this degradation can come whether or not, like it's from the binder or from the, like if you're using a linseed oil or something like that to help varnish the painting or as a protective coating, it might not be as protective as one would hope. So really, if you have the metal ions and any form of these carboxylic acids, the conformance type of degradation present, you're going to see the crystalline soap start to form. So yeah, you have to be careful, but then it's a balancing act. Maybe if there is a strong role of humidity and you can keep humidity out or something like that, you might be able to slow the process, but again, you don't want to trade one problem for another, introduce it through another material. Here's a question over here. Sorry, this is not the most insightful, but I think here is, since you're getting a different amount of resolution, how many places can actually compare those samples that have that kind of white property power and how much of an active talk that you've done, for example? Yeah, luckily our stuff's supported by taxpayer dollars, because it's done at the National Lab, so we don't pay any instrument fees, which is nice. And so if you're just looking at the reference materials, sonication and dispersion, and isopropyl alcohol is fairly standard, so that's a common procedure anywhere you're doing TEM. The ability to do 40 stem is becoming more popular, but it is restricted to certain facilities. So there's big groups in UCLA that do this, the National Lab in Berkeley does it. But it's sort of key at strong microscopy schools, but even if you're at an R1, but it's not strong microscopy, you may not have the right detectors to conduct these studies. But luckily, there's a lot of user facilities where people can gain access to this and with a couple sessions, you can get some good insight. In terms of the price tag and actual analysis, I mean the machines that we're using are probably like a million dollar machines, they're pretty nice. I just hosted a professor who does this stuff from UCLA, and when he was launching his lab, he was interested in doing this type of work. And he placed some interesting bets on like, you know, maybe you don't need the super fast detector to do, so he kind of like optimized his machine and bought things secondhand in order to drive the price down. But unfortunately, it's all TEM, so driving the price down is like, oh, it's only a half million dollars, not a million or 10 million or that sort of thing. Yeah. Yeah, so I was just thinking with the varnish question. I mean, most of you, even if something is varnished, you know, they're on forest. So you might still have the reactions. Oh, from the back. Yeah. But have you ever done any work on looking at layering? Like I was just thinking, you know, many of you, well, if we can layer on there, we need to go over. So I wonder if anyone's done any work on between layers, are more or less susceptible to this kind of formation on the vague or on the elixir? Yeah, that's interesting. I'm not familiar with anything off the top of my head, but yeah, definitely. Yeah, some of my sort of non conservation science work is working with Palmer scientists and studying interfacial activity of different materials. So I guess that's an interesting question because it feeds into something that I think about for sort of my main thesis research. But yeah, the energies of surfaces at different interfaces is completely different. Like if you're looking at an oil-air interface versus an oil-oil interface, you can see a completely different sort of migration or movement of materials just based on the relative energies or say like surface tensions between the two. Yeah. Yeah, that's interesting. And then something that comes to mind is like if you had sort of a poor combination of two materials that were really susceptible to this, you might get migration of these or formation of these metal soaps at the interface and that interface may be more mechanically weak and therefore more prone to sort of disastrous flaking. Whereas like if these things form but they're embedded in material, maybe that won't have the same mechanical effects on the painting as if it was at an interface where you have poor adhesion or poor binding between two different pigments. So yeah, that's really interesting to consider. Any other questions? Check the chat one last time. One last comment from Dale. He had mentioned that rapeseed oil was also added to commercially primed canvases in Europe since the 19th century to keep rolled canvas flexible at the art stores and that and it's nothing but free fatty acids. And so even the you know maybe the canvas manufacturers are undercutting the artist to some degree without realizing it. So sweet. But linseed is commonly used as a varnish or as a binder. So thanks again for everyone's attention and for all the simulating questions. I really appreciate the opportunity to present this work to a broader audience.