 It's already ready, so we can announce it's Plotomiro Caffolla from Durham University, and his talk is dissipation and lubrication of solid-liquid nanointerfaces, complex balance of substrate topography, molecular diffusion, and environmental conditions. So it's a long title, yeah? Yeah, it is. So it's always yours. So thank you very much to the organizers for the invitation. It's a real great pleasure and honor to be here. And as the title and so was suggesting, and above all, as we have seen over the last couple of days, lubrication and dissipation are really complex problems. So today I would really like to focus on the importance of molecular ordering on lubrication, by our defects in the fluids, in the surface, and microface separation impact the molecular ordering and hence lubrication. And finally, a very, very quick look at the potential smart pathway to control ordering and lubrication. So before actually going to the core of the presentation, a little bit about myself. I'm a postdoc at Durham University. My main research interests are in the area of nanoscale interfaces, particularly solid-liquid interfaces and transition metal oxide interfaces, and also some interest in developing new technique biomedical studies. So going to solid-liquid interfaces and lubricated friction, we all know the crucial importance of lubrication in a wide range of systems, our joints, engines, electromechanical systems. And we also know that this is a very complex problem. So for example, if we zoom in in one of these contact, and we consider just one molecule, looking at a perfectly flat surface, then we can relatively well understand the system. The problems arise when we go actually to the real case scenario, when the surface is not perfect, there may be contaminants in the fluid, and so we have to worry about solid-liquid interactions, liquid-liquid, potential jumping, geometrical effects. So to tackle this problem, I used a combined approach, mini-experimental, based on imaging the liquid-lubricant solid surface with atomic resolution. This is combined with nanorealogy, with highly localized shear measurements. And what possible, I did also some molecular dynamic simulations so as to get our better insight into the dynamical evolution of the lubricant molecules. Looking at the details of the experimental techniques, they are both based on atomic force microscopy, AFM. Imaging is done in amplitude modulation, so we oscillate the cantilever with a constant oscillation, sort of probing the interface like the stick of a blind person would do, basically. This is complemented with using the AFM also as a nanoscopic shear rheometer. So what I do in this case is that I laterally oscillate the sample and I record the induced torsion of the tip. And the magnitude of this torsion gives us information about the magnitude of the lubricated friction force. And the phase lag between this torsion and the oscillation of the sample gives us information about the viscoelastic properties of the sample, with a phase of zero for a perfectly elastic coupling, conservative system, and a phase of 90 for a perfectly dissipable viscous coupling. So let's see how we can use this combined approach to study the ordering of the lubricant molecules. And the first case study is based on axial solutions confined between the AFM tip and mica. Yeah, we have a pictorial representation of the lattice. And if we have just pure water due to the nano confinement, epitaxial effects, the water molecules get well-ordered in a nice, ice-like structure. But as soon as we introduce point effects in our fluids, in this case ions, because of the electrostatic interactions, they tend to disrupt the hydrogen bond network of the water molecules. And what we observe is different absorption profiles depending on the charge density of the ions. And they have a huge impact on the dynamic response of the system. So it's very clear the response of the system just in pure water. And when we add ions, so when we add ions, the system is more disordered. It's less able to resist when applied load. We see the decrease in the shear force. And we see that the shear phase suggests a much more liquid-like behavior. Still, the linearity of the friction force with respect to the applied load allows us to extract an effective friction coefficient mu, which we can model as a function of the charge density of the ions rho, and an empirical parameter lambda, which captures the thermally-activated motion of the ions. They need to overcome an activation barrier per unit charge, Ea over Q, with the thermal fluctuation over the megalattis favoring the jumps of the ions between adjacent sides. And both Pietro and Nicola gave really nice talks about Prant-Tomlinson model. So to a certain extent, this is an adapted version of the model. The FM tip provides energy to the ions, and they need to overcome an energy barrier represented by the interactions with the surrounding fluid molecules and with the surface. So take a message from these first case studies that point defects in the fluid that disrupt the order and hence reduce the friction force. Let's look at another type of defects. In this case, defects in the surface. And in this case, we look at a system that is quite relevant for industrial application, since our fluid molecules are made of squalene, commonly found in lubricants. And the surface is graphite, and we know that carbon-based surfaces are quite common in contacting parts. And the type of surface defects that are considered were step edges. So here we have a classical step edge. And if we are cold enough, what we see is that the molecules tend to self-assembly in sort of lamellar structure, very similar to what Roland was saying on Monday. And this is very clear. In particular, if we take the derivative of the topographical image, something that we call the amplitude channel, it's very sensitive, emphasizes the local variation. And we see that when we adopt the system, then the order progressively vanishes, except in a region in close proximity of the step edge. And then when we further ramp up the temperature, then the order progressively vanishes. And finding disorder at the top and the bottom of the step edge seems a bit counter-intuitive if you think that the order persist in particular close proximity of the step edge. So you would expect actually the geometrical singularity to disrupt the order, not to promote it. Molecular dynamic simulations run both on the top and the bottom of the step edge, seem to suggest that there is a loss of mobility in proximity of the surface defect. So this is why there is an increase in order. So this is a completely different scenario from the case when the ions were actually disrupting the order. So what happens here is that in proximity of the step edge, the molecules have less configurations to stably absorbed, and so there is a reduction in entropy. And this, of course, comes with a cost. So if we take our nanorealogical measurements in close proximity of the step edge, we see that the friction force is much higher and the system behaves much more like a solid-like way. The molecules are stuck. So being really the relaxation dynamics of the lubricant molecules dictating the lubricated friction force response, what we can do is to examine the evolution of the system when we change the shear velocity and the temperature. And what we find is that increasing the shear velocity, the molecules have less time to explore available configurations. So they are stuck in their position and explore a greater friction force, something very similar to happens when we decrease the temperature. And actually, I use all this data taken on different shear velocity and temperature to develop a quantitative model to describe the dynamical response of the shear lubricant. And overall, the response of the shear force curves is this. So we have an initial ramp-up corresponding to an effective yield stress than a plateau-like region. The yield stress presenting a number of irregular features I actually focused on the linear region for the fitting. And in this linear model, the intercept correspond to a yield stress and the gradient to an effective dynamical friction coefficient. And what I found is that it's really the yield stress that depends on the velocity and the temperature. When it comes to the velocity, the dependence is a power law. With an exponent that is classical for a system where the thermal fluctuation are lower than the activation energy, but still able to promote some molecular motion. And what is really interesting is that in this firmly activated motion, the activation energy, EA, is basically just four times smaller than the latent heat of vaporization of squalene in the bulk. So at the interface, of course, the molecules have fewer neighbors, but still this EA is very much the energy needed to break the intermolecular bonds and get a molecule free from the cage of his neighbors. So it's a penalty cost that we have to pay, but once we have paid that, we are on a free ride, and so the friction coefficient does no longer depend on the velocity and temperature. And take on messages from the second scenario is completely different from the point defects in the fluid. Surface defects here actually promote order, and they also provide us with a complementary framework to better understand the usual roughness increase of friction that we often experience in contacting parts. So far, what I have been examining are relatively ideal case scenarios, but as Rob was mentioning on Monday often, we have in our system an amount of undesired particles, not just point-like defects, and for example, water molecules. They are undesired in many car engines' applications. So what I have here is an hydrophilic alumina silicate, MICA, very common in engines, no MICA, but alumina silicate, and on MICA what you have in ambient conditions is that there is a nanoscopic film film of water absorbed on it, and then I use the hexadegan as a model lubricant, and what we see if we track the evolution of the interface with temperature is that when we ramp up the temperature, the water nucleates forming these water nanodroplets, they increase in size and number with temperature and they coalesce when we cool down the system. And these nanodroplets have a huge impact on the friction force. So if we compare the friction force on one of these nanodroplets and on an homogeneous area of the interface, well, there is a twofold increase in the friction force. So this is very much related to a penalty cost that we have to pay because what we are trying to do on the water nanodroplet is that we are trying to mix two separated phases. And of course, this anomalous temperature behavior can be well rationalized because here the temperature effectively shifts the probability of water nanodroplets nucleating. So the higher the probability, of course, the greater the number of water nanodroplets that you find on your surface, and the greater the friction that we experience. But we can shift the equilibrium of the system also playing with the water content of the system. So we can sort of dry out the atmosphere, decreasing the probability of the water nanodroplets to nucleate even at a high temperature and get a relatively low friction force at all the temperatures. Or we can actually saturate the atmosphere with water. We increase the probability of water nanodroplets to nucleate and we increase the friction force. And what we have seen experimentally can be rationalized with a simple thermodynamic model with water and hexadegon. Of course, we know they don't like each other, but they have a similar work of adhesion when it comes to the mega-surface. So they compete against each other for the substrate. And once the water nanodroplets get the size that we have seen experimentally, then they are thermodynamically favored in their spherical shape as they also minimize the interactions with hexadegon. So this is very much a liquid-liquid problem. So what we can do is that we can add a surfactant model. Now, if I work in the UK, I'm still very much Italian, so I use oleic acid, the main component of olive oil. And basically what we do is that with this surfactant molecule we reduce the interfacial energy between hexadegon and water. So we decrease the probability of the water nanodroplets to nucleate and we decrease, hence, the friction force at any temperature. So take on messages from these free-case studies is that the ordering of the lubricant plays a very important role in lubrication. And we have seen two different types of response when it comes to very localized defects. So defects in the fluid ions, they disrupt the long-range order of the water molecules and reduce the friction force. Whereas in the case of surface defects, they result in reduction in entropy and then increase in the friction force. And finally, we have seen how the microface separation affects also lubrication and an effective way to help with this is to use surfactant molecule. So what is really cool to me is this controlling of the lubricant ordering. And we have seen that temperature and humidity can do that, but probably they are not the tools that we would like to use. So something that I have started recently doing is to investigate how we can control the order of the lubricant using, for example, magnetic field. And so here I have used... These are preliminary experiments where I used a magnetic ionic liquid. Here there is a steel interface with some iron oxide particles deposited on it. And when the magnetic field is off, well, the magnetic... And we see that there is basically no magnetic signal. The friction force is pretty much homogeneous with the magnetic ionic liquid helping also with the roughness of the iron oxide particles. But as we switch on the field, well, in this case, this ionic liquid can perform these domains, microface separation, Nicola published a paper a few years ago about this ionic liquid, not this one, but generally speaking that they can give rise to microface separation. And hence we have again an increase in the friction force because probably what is happening here is very similar to the hexadec and water case, when we try to mix two differently separated phases. This is pretty much stuff. Hopefully maybe if there will be again this conference next year within two years I'll be able to show you more data on this. I would like just to finish off highlighting this conference in Durham that I'm organizing. So thank you very much already to a number of invited speakers who have confirmed their presence. So I hope I'll see you many of you there. If you'd like to come just drop me an email. And thank you very much again for your attention. When I was talking, there's already very nice work. I'm wondering about for the squalane on the graphite was the surface. What was the tip material? The tip material was made of DLC, diamond like carbon. I really expect some interactions, significant ones I think between those molecules in the tip too. And so I'm wondering if you've thought about, I think this is a hard question to answer without simulations, but could there be some organization and influence of the tip, organization of molecules interacting with the tip? And the reason I'm asking is, is the shearing, I'm wondering if the shearing is between the tip and the molecule, or is the shearing you're measuring between molecule adhered to tip and the molecule adhered to surface, both of which are interesting. Yeah, so definitely that is a very much a possibility. What I'm trying to do is to replicate also those experiments using different tips. But at the moment I don't have any definite results, but it's a very nice idea and thanks. Yeah, yeah, yeah. Basically, I don't have a definite answer, because I don't have yet the complete data set for that. And above all, as Rob has you were suggesting, there will be any dose for some mega-dynamic simulations, and I have some limited experience with that. I was curious about the nanodroplet experiments that you showed, because one imagines that if the contact angle of the nanodroplet is sufficiently low, they actually should act as an additional lubricant, so as if you had a liquid-infused surface. So I was wondering whether you can tune the contact angle of such droplets, because in the end, not always they should act as defects, but on the other hand, they could be. Yeah, so we, in the experiments that I run, I couldn't control the angle the nanodroplets were at. Also remember that when we run our experiments, they run completely with the tip immersing the fluid. So basically, imagine that there is this huge drop of hexadegon, which is the dominant liquid, because we were talking about 200 microliters of hexadegon with the tip completely immersed, and then this nanoscopic film is like two nanometers. And so what you are doing is that basically you go, this is what I have in my mind, is that you go on one of these droplets, and that doesn't just act as a pinning point, but there is also, you are trying basically to remix the water with the hexadegon while you shear it. So definitely anyway, the point of contact angle is a good one. Try to do some more experimental, the macro scale for the paper. That was more related to work out, to the work of addition. Yeah, controlling the contact angle is definitely something to explore. Thanks. We have to proceed, so thank you again for a nice talk.