 This is one of the first in-presence conferences that we're coming back to. You see that the Italian regulations require people to wear masks. I take it off. Speakers on this side of the thing are allowed to take it off because of the sound quality. But inside the building I think we have to comply with that. So just to begin with, ICTP welcomes you. It's been created by these two people over here in 1964. It's been running ever since and sort of improving in its goal to bring scientific knowledge shared between everybody in the world. It's mostly sponsored by the Italian government, but it is conducted under three agencies and our, in fact, our mother agency is UNESCO at the moment. Coming to our subject of nanotribology, we have been going with it for now maybe 25 years. I think the first one that we had was with Bo Person who is here with us today in 1996, I think. And we've had a number of them we hope to continue doing here or elsewhere. In fact, people should consider continuing this kind of activity in other places too. This meeting is dedicated to the memory of our friend and colleague Mark Robbins who has been leaving us unexpectedly two years ago. Some other colleagues here will have to say something about him. The meeting is directed by three people besides myself. One is Ernst Mayer who is here with us, University of Basel, University of Basel who is also providing generous funding for this meeting, we're very grateful for that. The other director is Quan Shui Zheng of Tsinghua University, China who is also sponsoring the meeting. Unfortunately, people cannot come in person from China because of present COVID problems and regulations with quarantine. So we will have him and other people from the institutes in China in remote. The last director is Andrea Vanossiou who did a lot of the work in the planning and things. You will notice that there are problems and gaps in the programs that we shared and everything. Because Andrea has been out of business in the last few days because of some medical controls he has to take and therefore some of the updating did not actually take place in automatic machine very well. Here are our main human resources in terms of helping with this and that and everything. When you have a problem don't come and ask me because I will just fall from the clouds. We just lost these three people, Ali who is here and Andrea is here, Andrea is where, out there and Jin is where, out there, right, they will be helping us. Then of course we have our three guardian angels. The guardian angels are registering us. Victoria is the main one who is on the piece but Adriana and Monica too, they can be reached either by emailing this SMR of the conference or calling this number that's internal 135 or walking up to room 115, two floors above in case of extreme need. That's the people that hold all the information and stuff. This is a program that we are facing today in these three and a half days. We have three and a half sessions as I said. The black symbols here are in presence and the red are in remote. There's some change here. Michael Urbach cannot come and so he will be with us in remote tomorrow. Also I think Jenkins who was speaking on Thursday will also be in remote. The sessions are divided by these coffee breaks that will take place in this building. The afternoon today is not in this building because there is a colloquium going on here. So when the morning session ends we will all go down to Adriatico where your accommodation is. There will be a bus or shuttle taking people down who do not want to walk. One can easily walk down anyway. I think the bus might be at 110 or something to leave some lee space at the end of the sessions. It's very unlikely that we are exactly on time. Then we begin in the Adriatico meeting, in the Adriatico meeting room, this afternoon session and at the end of the session we will have posters right outside of the lecture room in Adriatico and then the reception will take place hopefully if it doesn't rain too badly or something on the terrace in the Adriatico. Anyway, everything this afternoon takes place in the Adriatico. Then from tomorrow onwards everything will be in this building and it will be morning to evening. Tomorrow evening we have a social dinner and the rest of the program goes on like this. We will close lunchtime on Thursday. Thursday is the national Italian holiday. This place is international so it does not go on holiday but still be aware that we are sort of in a holiday kind of situation outside of here. So practicalities. Speakers are requested to keep a rather strict time and the chairman are supposed to take care of that. Allow please the 30 minute speakers five minutes for discussion at the end. There will be discussion also with people in remote and that might take a little time so the five minutes are really needed. The 20 minute speakers may take three minutes instead. Now there is this, we are working in hybrid and so there is complications related with that. Please talk to the technical assistants and to the three angels over here about how to connect and how to prepare and everything. If I instruct you I will confuse you because I don't understand it myself. But this is an important step in order to shorten the interval between the speaker and the next. There will be physical poster sessions as I said today at the article starting six in the afternoon. There will be a prize for best student and postdoc posters posted out there. Even people that are speaking are free. They are welcome to present a poster and then discuss it within say one hour time from six to seven with the participants in the article at the same level as the Castler room which is one floor below the entrance. Again speaking about posters there will be a breakout room session available at the end of the session on Wednesday 1st of June. That is two days from now. This will be managed again by people who arrange these things on the internet. It will be a chance for people to stand besides their poster and explain it to anybody that wants to hear and people will click on whoever you want to hear from. It will be of the order of I think 15 or so breakout rooms. We are sorry that we cannot get everyone to speak properly but that is the best remedy we could find. So this is repeated again today only lunch. Lunch today is not in this building. It's down in the article so you catch your bus or you walk down and lunch will be there not here and then the session continues in the Castler room out there and the shuttle is already I already mentioned it. There will be a welcome reception with food and drinks at the end of today at the article thing. I think it should be enough for people not to have to plan to go for dinner afterwards. Of course you may like or not like what is going to be provided. It's a free free for all kind of thing. All other lunches today, Wednesday and Thursday are here in this building. There is a cafeteria here at the level zero so all other lunches will be here. Only today is down at the article. There's a social dinner on Tuesday evening I tentatively wrote registers with secretaries. In fact I didn't have the time to agree with them about this thing. Every registered participant is welcome to attend. Companions are not included in this invitation. If companions come, they will have to cover their own cost of dinner. That's required by the UNESCO rules. The restaurant is in town. It's called Monte Carlo. I hope it's not as fancy as the name suggests. Last experience we had with Ernst and colleagues was very good. I hope it will be similar this time. It's fish menu. I guess people that are vegetarian or have dietary problems should point it out when we arrive there so that they don't get the wrong food. The three angel secretaries will have accepted the invitation to join us so maybe I will ask them to interview people about any special needs that they might have. This is coming to the end of practicalities. There's a list of chairmen here. So people that have agreed with Andrea Vanossi, I think Rob has agreed for this afternoon. Lars agreed I think with Annalisa Fazolino for tomorrow morning. Annalisa will arrive this afternoon. Annalisa is retired and not active but she's coming to help us and we're very glad of this opportunity. Ernst will be on Wednesday morning, Roland on Wednesday afternoon and I haven't found a substitute for Michael Urbach who was supposed to close. Maybe we'll find one. Now the shuttle hours, today I think there was a mistake somewhere in the planning. I think the bus will come in such a way, I believe, so that you can get your breakfast down at the Adriatico. Now the other shuttles are from here to down. It's 1.10 today, 8.00, 6.30 tomorrow and 7.00 on Wednesday. The breakout room might take a little longer than planned and we have nothing planned after that. So with this I think I've finished this long kind of information session and I quit and I give the word to the first speaker of this morning who's Rob Kalkic here. Thank you. Thank you very much, Arrio, and thank you to all of you for being here. It's wonderful to see people in person for a meeting. It's terrific. Let me make sure I'm set up here. There, okay, you can see it and I hope the people online can see it. So thank you for joining and again thank you for the invitation. What's that? The screen's not being shared. I think if it's up here it's being shared because I know I'm not connected to any other hardware. Okay, no but good to check. Double checking is good. Thank you. So yeah, again, thank you to the organizers, Andrea's not here, Arrio Conchi Ernst. It's really an honor to be here and I was asked to say a couple words as the opening speaker about Mark Robbins. I was very fortunate to get to know Mark over the years. He was located in Baltimore at Johns Hopkins University which is not too far from Philadelphia and he actually spent a sabbatical at my university with some people in physics. So I'm very lucky that I had many interactions with him over the years. I was just reflecting on the different times that I had seen him and one of them was here at the 2011 meeting that was a joint ICTP FANIS meeting that Arrio, Andrea, Nikola and Ogues were co-organizers of and I didn't remember this until starting to look through some old photographs. There was an excursion to this Castella de Miramare during the meeting which was absolutely beautiful location and a beautiful weather that day and just another, you know, example of how the meetings here are so excellent scientifically and aesthetically it's such a pleasure and I found this photograph in fact that was taken there of Mark who you can recognize. Does anybody recognize who the other person is? This would be Lars. I think what's amazing is, you know, this is what Mark looked like. Even, you know, many years later he seemed to stop aging and I don't know Lars if we can say the same thing about you, but what I like about this photograph is you see the smile, the famous smile that Mark always had and I think it was just a reflection of how genuinely joyful he always was about science but also about people and interacting with people and being in different places and working with whether it was students, collaborators, asking questions at conferences, there was this true joy there that he had and that he brought to everyone else's through his interactions. He also would have that same smile when he would ask very, very difficult, challenging questions as well and being on the receiving end of those sometimes I remember that the smile was a double-edged sword but Mark always had the ability to ask penetrating questions in his research and interactions with him to get to the bottom of what the physics, the physical interactions were that we were trying to understand in friction but also many other problems in non-linear, non-equilibrium statistical physics that he tackled over the years and I thought that there's so much that he's done. There are some ways I think to remember and honor him and to continue to learn from him because fortunately we live in an age now where there are many things captured on video and on the internet and as well special opportunities. So for those of you who will be attending the World Tribology Congress this July in Lyon there will be a special session in his honor and I invite you to please attend. It's part of track six which is the track on multi-scale, multi-physics modeling and experiments so very appropriate for one that honors Mark and then I'm going to show you a few other links you might be interested in and I will, well this is being recorded but also share this with the organizers so we can email it out to people. In case you didn't know, Martin Muser guest edited a special issue of Tribology Letters that came out in May of 2021 and there's just a set of excellent articles there as well as a beautiful tribute that Martin wrote about Mark and his life and there's some interesting things online. Mark organized this Kavli Institute for Theoretical Physics meeting back in 2005 that was focused on from the atomic to the tectonic friction, fracture and earthquake physics and as part of that at this institute which it's sort of like the ICTP of America because it's also on the coast it's in Santa Barbara, California with a beautiful view of the ocean and they record all of the presentations including these blackboard talks so this is a talk that Mark gives about friction with no slides it's entirely done with a piece of chalk on the blackboard and you can watch that and it's interesting and you can see here he's drawing some jagged surface talking about friction coefficient dealing with the angle you can see here's Amontan's law right there and as well there's a recording of a lecture he gave more recently in 2017 also on contact and friction for microscopic to macroscopic scales and so that is also available free online so for those of us who knew him well it's it's you know it's enjoyable to watch and for those of you younger people who may not have had the chance to know Mark and his speaking style and his research I encourage you to check these links out because you will learn a lot from them. Mark would encourage you to come to the Gordon conference on tribology Mark was the chair of this meeting a few years back the current chair is Judith Harrison and song Kim is the vice chair and I agreed to help them promote the meeting the there's a pre conference for students postdocs young researchers on the Sunday Saturday Sunday before and then the week of starting Sunday night is the conference and although the deadline for applying was officially May 29th actually they are still accepting applications so it is in person after being on hiatus and I encourage those who are interested to please submit an application to attend as soon as possible it's a unique meeting that is very enjoyable so I'm coming back to this but I'm going to grab my water my dry throat is not from COVID I had in fact I took a test before last night to make sure so with all the travel there's always this fear but I want to spend the time I have telling you about some research that I've done in collaboration with Ashley Martini who's going to be speaking later today most of this actually all of the experimental work was done by my former student Catherine Haas who's about to become a professor at Carthage College in Wisconsin and Ashley student Muhammad Vaziri Suresh who's now at Lam Research did all the simulation work that you're going to see and this is what we've been focusing on over the past several years our 2d materials and layered materials I think this audience needs no introduction to the importance of these which come in bulk form they come in thin film form they come in monolayer form the low the easy shear between the layers of course leads to low friction and this has allowed them to be incredibly successful as solid lubricants you can buy them you just put them on dry you can grow them as thin films you can spray them on a surface in a suspension you can make them additives as particles inside of lubricants and they can help to reduce friction and prevent wear and one of the reasons why I think there's a lot of excitement about these is because in addition to having low friction and low adhesion you can control the friction and adhesion you can it's hard to get high friction to be low but if you're starting with low friction you can often do things to bring it up and so if you think about flexible electronics these there are many opportunities to take advantage of the amazing electronic and thermal and optical properties of 2d materials while also taking advantage of how flexible they are but to assemble and manufacture these you need often to adhere or de adhere or transfer different layers as you bend it you may get sliding small amounts of sliding and shear stress between the layers so that the tribology is important for the electronic applications that integrate these materials and as you may or may not know molydine sulfide in particular that is the solid lubricant of choice for space and we can thank we can be thankful to molydine sulfide for helping the James Webb Space Telescope succeed the moving bearings gears and other components that are exposed to the space environment have a molybdenum and sulfur containing thin film coating on those parts and that is the key to ensuring that they don't seize up in the low temperature low vacuum environment of space so we depend on these every day including for some very remarkable applications so over the last several years we have been looking at how small nanoscale tips AFM tips interact with 2d materials usually we are looking at them exfoliated onto a surface maybe grown by CVD we have worked with a few different researchers we have a very close relationship with Professor Charlie Johnson at the University of Pennsylvania who grows 2d materials in novel and innovative ways and makes devices out of them these are some monolayer graphene islands that he grew using CVD which we looked at this is now 10 years ago or so but I really like this example because what we've got is a copper surface and then just monolayer of graphene on top and this is a friction force map and you can see as you slide across you get an order of magnitude reduction in the friction between the copper it's really there's some oxide on the surface so oxide surface high friction graphing extremely low in the limit of what you know this AFM can detect you know a couple nano newtons of force so this is impressive a single layer of carbon atoms interposed between a tip and a substrate is all you need to reduce friction by an order of magnitude or better and not only that this is quite robust overload different loads so this is the friction force versus normal force for the copper you can see it's up here and it keeps increasing down with the graphene it's much not only are the forces lower but there's almost no slope almost no increase in the friction force with with load which is something that always surprises me because I keep expecting well if you push harder the contact area goes up shouldn't you see more friction but the friction coefficient is almost vanishing in this statistically indistinguishable from zero for this material that's been seen on graphite before by many others and we also see it on graphene so right away just that's one example of how you can have a strong effect on friction force and friction coefficient through the single layer of of graphene now when we slide on these crystals even though it's a single layer thick if it's ordered we often see stick slip friction which I think many here are very familiar with ways to describe periodic stick slip friction using things like the Prandtl Tomlinson model where you take a rigid potential reduce all the order all the degrees of freedom of the tip of the tip to a single point particle and Adam as it were connected to a spring and there's some energy corrugation to that potential energy surface there's some spacing of the lattice and Prandtl and Tomlinson each worked out that you can get these instabilities as you slide so as you pull the little energy minimum you're in gets shallower and shallower until it disappears at which point you'll slip to the next position and that's at zero temperature of course at finite temperature thermal excitations will help you slip even earlier than that and Enrico and Roland Benowitz in the audience are people that helped figure out the right way to to describe that early on so the key parameters in getting the static friction force out of this model is the corrugation of the potential energy surface this Delta E and the lattice spacing actually comes in here as well so this is a very simple relation this is again zero temperature there are speed effects temperature effects but but a very simple relation and this is what I want to focus on is what is it that determines this energy barrier and what happens as you change the conditions and in particular what to what extent does the force you measure depend on how big an area of contact you have or how much does it depend on the frictional shear strength the force per unit area that you have in the single asperity and you can sort of somewhat crudely but I think still meaningfully defied divide these two quantities the shear strength and the contact area into two concepts the shear strength is like the quality of the contact if you have more interfacial bonds if you have a commensurate interface if you have maybe contaminants that can give you higher force per unit area but also if you have just more atoms in contact each one of those interface each one of those interfacial atoms is pairing with something across the surface and that can increase friction just elastic deformation plastic deformation separating these out I think is important if we think about what contact quality it was actually mark robins and martin muser who really first pointed out that you know in commensurate certain transfer surfaces can have high friction in commensurate would have low friction why do we not see vanishing friction all the time interfaces shouldn't normally just suddenly be commensurate well for one thing contaminant atoms will pin the interface so even an incommensurate surface can have high friction and this concept is over 20 years old now mark also showed another way to think about what happens at the interface take the same tip the same radius but make it round make it make it like a bent crystal make it an amorphous structure make it a stepped structure and the contact stresses that you get are completely different down at the nanoscale so and friction will be different at this scale so these are just some examples to say that the quantity of the contact matters but the quality matters to how the atoms are interacting across it interface how they are structured and that's going to be the theme of this talk so I wanted to show three examples I think I will only have time for two and that's okay because this last one you will hear about this afternoon when Ashley martini gives her talk so I'm gonna focus on how humidity affects friction in this case it's going to be for graphite and then how orientation is affected in the case of friction and isotropy so first for humidity we measure friction as a function of the humidity at a fixed load I'm going to show you data at three different speeds there was not a strong speed dependence we measure friction with tips on on graphite frictions very low on graphite when it's dry as we raise the humidity we see an increase in friction this is not too surprising many people have seen you start letting humidity in you get often a capillary that will form that capillary will create an adhesive force between the tip and the sample that will increase the contact area right it's just it's like you're adding load to the system because this capillary has surface tension it has a Laplace pressure it increases the load you get higher friction but then this happens it turns over as you go to very high humidity is closer to saturation the friction comes down again this has been seen with adhesion measurements on various surfaces in many cases and this actually also is still consistent with the capillary model when you have a capillary and you get to near saturation the larger the capillary the less the curvature is very high curvature leads to a high Laplace pressure if it's not very curved you don't have that much of a Laplace pressure and if it's just flooded if it's a hundred percent you're just under water there's no capillary anymore at all so the effect of the adhesion goes down so this actually tracks with what a full-blown theory for a tip on a surface what kind of adhesion you would get we thought okay this is what we're seeing capillary forms and then it floods and friction goes up and down Ashley's group did molecular dynamic simulations using a Grand Canal Monte Carlos scheme where they could vary the part the partial pressure of the water so tip sample and then water molecules that were simulated and they also saw a maximum in the friction response for their simulations so we said okay it looks good but is it really the mechanism or is there anything else going on you know do we really see that the contact area goes up because the capillary forms and causes adhesion that's this idea well they can get the contact area between the tip and sample it was not increasing they saw negligible no no significant change in the contact area and furthermore if there was adhesion you would expect there to be elastic deformation as the tip and sample are pulled together by that capillary and so the vertical distance between a reference point on the tip and sample should decrease but actually it increased a tiny amount only one angstrom net but they can resolve it and outside of the noise of their simulations there's a clear increase in the height of the tip not decrease as the relative humidity goes up so something else is going on the contact area model this adhesion model of of of why friction changes with relative humidity doesn't doesn't work the fact that it goes up the height increases that gave us a hint that actually there must be water molecules in between the tip and sample and this is having an effect and this made us think about Mark's work you put contaminants at interface it can cause friction to go up so we said okay let's take a look at what the registry is between the adsorbed water molecules and the substrate as well as the tip so there's this very useful concept that Odette HOD came up with the registry index when a adsorbate or or it could be an adsorbate it could be an atom on the other surface if it is at a favorable site we say that's a low-energy interaction that's got a low registry index and we normalize that to zero if it's that energetically unfavorable site the maximum most unfavorable site in this case right on top of a carbon instead of in the hollow sites we give that an energy index of one okay so a low registry index is telling you favorable interactions unless possible pinning of the interface so what I'm showing you here are side views that I showed before the tip sample and the water molecules in the simulation in blue and then this is a top view you can see the graphite lattice and then the atoms that I'm showing here are the are the water molecules this is actually a coarse-grained model so there's just a single bump for each water and I can talk about that potential more if you like but it actually does a reasonable job of modeling a lot of physics of water so at low humidity there's very few of these adsorbates and but they are there as you keep increasing a lot and you can see that the red would be that they're sitting in unfavorable sites the green would be sitting in favorable sites okay now why are so many of them in unfavorable sites well some of them are interacting with the tip as well okay so they're finding a balance but in terms of their registry with the graphite not too many of them are very favorable as you add more water you start seeing that there are more adsorbates total and you see that there are more green atoms one that are more favorable in terms of being pinned being low energy and that's more difficult to take out of that site and fine and then as we go to very high humidity's you can see what looks like a capillary and you see many of these many more of these atoms are indeed an energetically favorable sites so what we can do is come up with what we call an effective pinning parameter so first this is a plot versus relative humidity of the registry index in red and we do it as we this is again in the simulation as the humidity goes up as the humidity goes down so ramping up ramping down just to check for hysteresis there's a little bit but the key point is we see that the registry index is high at first so lots of unfavorable sites then more favorable pinning and then finally at the highest humidity the registry index goes back up okay the water is no longer in very many favorable sites so we define an effective pinning as the number of atoms the number of pinned atoms times one minus the registry index right that makes a low registry index give us high effective pinning as it should and so okay we see a max we see a maximum around 80 and that's close to where we see the maximum in the friction in the simulations and and and in the experiments a little bit higher than in the experiments but this is saying it looks like the pinning of those water molecules at the interface is important okay so more pinning sites and we get higher friction what is happening here that's bringing it down this is deep pinning the water molecules as you can see here once you get to high humidity is 80 100% they start aligning with each other you have lateral ordering of the water at the interface instead of being disordered and that gives you now and essentially an incommensurate layer those lateral interactions between the water molecules allow it to create a layer that is now incommensurate with the graphene and allows easier and with the tip and allows easier sharing so I'm going to skip this but I will say or Michael or Bach and Astrid DeWine had a general have a general model that saw something very similar friction versus coverage of generic adsorbates that is consistent with this idea so we see that with this amorphous tip and a graphite surface the pinning of water molecules in other words this contact quality leads to higher friction low friction with few adsorbates pinning sites at intermediate humidities but at high humidities the layer deep pins because of the lateral interactions as the water forms an ordered layer okay and this contrasts with that meniscus idea that capillary idea I mentioned but is consistent with with the work of Michael and Astrid I think I'm down to probably time to time to wrap up I will say I haven't talked to orientation that's okay but I will be happy to talk with you about these topics and you will hear more about them when Ashley gives her talk this afternoon and so with that I will thank you very much and be happy to take your questions at what loads where this experiments and simulations perform right so we typically did so the experiments were done with zero applied load but there was always some adhesion and that adhesion had some variation with humidity so we just had the adhesive load the ease of forces the load effectively I believe those values were around five to ten nano newtons I would have to check that we have that in this paper the simulations we tried to essentially match that but the simulations the tip is quite a bit smaller than what we had in the experiments so so I think a word of caution is that the the mean stresses are the loads are similar between the simulations the experiments but the mean stresses are different they're higher in the case of the simulations is it's graphing right the substrate it was actually graphite we had some experience on graphing and saw similar trend as well we did not see a major difference between we could not resolve a difference between graphene and graphite for humidity it's quite easy to break this hydration layers it's rather easy to break the hydration layers yes yes and also then the coarse-graining model it's really tricky to get the ordered water to describe this ordering process and coarse-graining models perhaps yeah this is the it's the MW molecular water potential and maybe I'd be happy to talk to you afterwards and I we agree there there's limitations but we think that the consistency between the simulation experiments is is encouraging but I you know there's there's much more one can do I was wondering a few years ago there was this controversy about when graphing became graphite and multi-layer rafhing and can you just make a comment about that when graphing becomes actually graphite so tragologically speaking yeah yeah so you know I think that's fairly well-defined from electronic structure standpoint we have lots of data and first principles calculations where it takes so many layers before the electronic properties become bulk-like and I believe it becomes negligible after something like eight or ten people can correct me if I'm wrong but from a friction standpoint we've seen that friction starts going up as you the number of layers goes down so with about four layers of graphene it's indistinguishable from the bulk with three two one the friction goes up and that is an ongoing issue we think that it's largely due to the quality of the contact the graphing the thinner the graphing the more flexible it is the easier it is for it to find favorable interactions with the tip and and the friction goes up but we think there's even more to the story than that so so from a friction standpoint you need four layers you've pointed out very nicely that the radius of these simulations is a lot smaller than most of the experiments and of course most of the old explanations in terms of capillary and so forth are based on quasi fluid continuum what hope is there of bridging that scaling gap yet Jim Belak tried to have multi-million atom simulations where are we right right I think that there's been a lot of development and advances in the fidelity of potentials that describe water and improvements and of course in the first principles methods and being able to simulate somewhat larger things and what I see in the literature and I'm not an expert on water you know modeling but there's still some harsh debate between different adherence to different potentials that said there are more and more examples of simulations that really do faithfully reproduce experimental results and and that the end that they do it down to small scale so I think we're better off but the debates continue yet so if you want to ask a question you can unmute yourself and ask or also I will see if I can find it in the chat I don't see anything in the chat though I see a raised hand from Jermaine Kenmore Kenmo so Jermaine if you can unmute yourself I'm happy to take your question yes thank you for circling for my nice presentation can you hear me yes we can hear you thank you so I saw your paper physical review material 2019 I want to know how can we model in a print on Mr. model simple print on some model the humidity in the model can you say something about it oh how is the humidity controlled in the model so the way it was done in the molecular dynamic simulations was using the the lamps framework for running the simulation and the potential was this so-called MW molecular water potential which we cite in the paper and it's a grand canonical Monte Carlo simulation so what that means is you allow the number of water molecules to vary and reach a steady state you know equilibrium value at each of the different humidities so and that's a fairly well established method and Ashley will be here this afternoon and can answer more about that and also we have the details in the paper but essentially that it's all through molecular dynamics we do not have a way to essentially describe this in a reduced-order model I think I heard you say Prandtl Tomlinson I don't see any way that there's no simple way to accommodate that you could make arguments about okay maybe it changes the energy barrier the potential energy barrier but I think this would be complex good question so I think in the experiment since we were working with graphite in the experiments and the graphene in the simulations was large and covering the whole surface there wasn't an easy opportunity for that but if you had defects or edges water can intercalate between graphene and substrates or between the different layers of graphite so it is possible we didn't see evidence of that but it's something worth looking for