Ron Larson | Shape Shifting Materials

Transcript

well thank you so much I'm I'm so delighted uh to be here and to to see you um attend my lecture uh it's a a terrific honor um I don't take it lightly at all it's it's it's uh just fabulous especially considering how many uh outstanding colleagues I have here I I was truly surprised when I was informed of this um and uh did did not expect it at all but I'm delighted to receive it and and delighted to uh tell you about some of the work that uh my group has been involved in uh over the years um I'd first like to uh say a little bit about um Alfred Holmes white For Whom the professorship is named so I chose him uh primarily because of his long service to the University of Michigan uh and in particular the department of chemical engineering he was there at the very beginning in 1898 and served the department for 40 years uh for 28 years as chairman of the department uh I thought I served a long time as for eight years uh and uh and he brought the department uh to the first ranks of of Excellence uh in the country um he also uh his name was used as well by Donald Katz who was a a pioneer and one of the outstanding figures in chemical engineering uh Donald Katz was the first ah white uh Professor so what I'd like to um talk about today is a shaping of materials uh and the shaping of materials goes back to the dawn of human civilization and it's been critical technology ever since the beginning but it's only been in the last 100 years or so uh that we've begun to shape materials that are somewhat similar at least uh to the materials that are used in biology a very good example uh is uh the silk threads that are spun by a spider um to this day as I understand it uh silk threads spider threads are tougher than any known material on a per weight basis and yet they're processed at room temperature with water as a solvent so it's it's truly a green processing operation and uh we are beginning to mimic some aspects of that uh as we see processing large quantities of of polymeric materials for everyday applications but there are many problems that are encountered and defects that occur uh in the processing of uh polymeric materials uh films can tear um uh there can be defects on the surfaces such as the shark skin or even worse uh instabilities that occur uh in shaping these materials and we'd like to understand these so we can control them and process better uh so because these are complicated materials uh one strategy is to look first at simpler materials uh and for which the instabilities and flow properties are somewhat better understood and we can look around in nature and see many instabilities that occur in simple fluids like air or water or in the kitchen uh where you see uh instabilities uh that occur uh one of the best studied of these uh is in the flow between uh two cylinders so you have an inner cylinder rotating and fluid contained between it and an outer cylinder and if you spin the inner cylinder fast enough uh an instability develops that's Illustrated here uh and it's caused by centrifugal force which tries to throw the fluid from the inner cylinder to the outer cylinder but it has to push out of the way the fluid that's already there and that creates a pattern like this and there's a whole stack of these so-called roll cells that appear and if you spin the inner cylinder faster and faster you get additional instabilities that eventually lead to full-fledged turbulence so polymeric materials are way too viscous for these kinds of inertial or um centrifugal instabilities for example here's an especially viscous polymeric material it's a it's a toy but it's uh it's called Silly Putty but it has the properties of a lot of these polymers so what can cause instabilities in processing polymers well my colleagues and I um uh several years ago um back at Bell Labs discovered that this uh rotational U flow between uh two cylinders leads to an instability even with polymers that are very viscous so here's what we found with polymeric fluids and we know that this is not a centrifugal instability it's not due to inertia because the fluid is way too viscous and moreover if you rotate the outer cylinder instead of the inner cylinder you still get this instability which you would never get uh in the case of the centrifugal instability with simple liquids so we want to understand how this forms and how ubiquitous it is with other polymers uh so this was done with my colleagues um Susan Muller and Eric Shak so uh in explaining this I'd like to take a bit of a uh first I'd like to show another instability that's related closely related to the instability between uh cylinders and that's if you have two discs two parallel discs one of them rotating um you can also generate a Cascade of instabilities that are shown here uh these images were produced by my graduate student Bruce shamberg leading to very irregular behavior that you see here that's uh come to be called elastic turbulence so how does do this how do these kinds of uh instabilities uh occur so I'd like to take a a tour of four principles that are used for controlling uh and shaping polymers and other materials uh along the way in explaining this and these are brown in motion elasticity self assembly and then we'll talk about some futuristic work in self-propulsion so we'll start with Brownie and motion uh Brownie and motion was uh discovered by Robert Brown uh he was looking through a microscope at little particles that came from um from pollen grains and he saw they they perpetually danced around like this these are images that Mike Solomon took in our in our lab in our department uh and uh Brown got very excited because he thought he had discovered something called the vital force and at the time uh it was believed that living things had a special force called the vital force that would animate them and he thought that's what he discovered so he's quite disappointed to learn that dead things do this exactly the same thing as long as they're small enough and suspended in a liquid they will move around like this so that may be why he looks so grumpy here when he discovered this um so the the true explanation for this Behavior actually uh was first articulated 2,000 years earlier by this guy lucretius he was one of the early believers in the existence of atoms uh and his Insight was so great that he he said that these atoms would be in Restless movement they would be driven on in ceaseless and varied motion uh and in fact that's exactly the case that motion is caused by Collision collisions with solvent molecules battering that little bead from all sides pushing it around in a random in a random motion and Einstein was the first one to analyze this uh and show how to interpret it and to explain it as being the result of atoms and molecules at the time of Einstein actually um the existence of atoms was still in dispute there were holdout physicists uh like Ernst Mach who uh opposed this notion and in fact uh Einstein's analysis became one of the most definitive proofs of the existence of atoms at the time okay so uh it turns out then Brown and motion is ubiquitous it's really nothing other than waste heat which will exist in all fluids um and it's taken advantage of especially by the cell because Brown and motion is the primary means by which molecules in the cell find each other and bind and carry out their uh day-to-day activities and a particularly good example occurs in the nucle nucleus of the cell where the human cell has 2 meters of DNA packed in the tiny space of a nucleus and it has to be packaged uh in chromosomes just to just to keep keep some order but you can see all these levels of structure that exist down to the individual DNA molecule and if you digest away the packaging a single chromosome looks something like this if you zoom in on this little rectangular area here's what you see and some somehow the cell has to read this and find the information that it needs moment to moment in order to continue its operation so how does it do that how does it find the information it needs how does it read the right part of the DNA well it does it by Browning motion using proteins that will find the target site by moving about in this jittery fashion uh and it's believed that uh this is aided by the protein first finding the DNA and then moving along the DNA to the Target site so it would be much like waking up in a dark room trying to find your dresser and you first find the wall and then move along the wall and so uh we uh were one of the groups that tried to actually image this process okay so we weren't the first to do this but we used a method in which uh we aligned DNA molecules onto a glass slide and this was done uh by graduate student jiun Kim and it was important that the molecules be attached only at the ends to this slide so that the rest of the molecule is free to interact with protein molecules that are brought in by a flow in a perpendicular Direction and then we we image them and watch to see how they interact with the DNA he so here uh I'll show uh a video of this uh and the DNA in this case is invisible you can't see it uh but you'll see that it's there because when we bring proteins in from the side they begin to bind and you see the individual binding events and there are many places uh for binding of these proteins on this particular strand of DNA uh there's another example here where the protein concentration is less so it takes longer uh for uh uh proteins to find the DNA but they eventually do now to look for this process of searching the the Brownian process of searching r rly we replace this DNA with a DNA strand that had no binding SES and when that's done um you see the uh uh protein moving back and forth in this characteristic random fashion known as Brownian motion and we can analyze that motion by looking at a time series like this you see it going up and down uh and if you take the distance traveled in in a given interval of time and do many experiments and bend up all the results you get a curve that should be familiar to many of you this is called a normal distribution or gaussian distribution and it's a characteristic of a random process the width of this when you square it is linear in time and the slope of that gives you what's called the diffusion coefficient this turns out to be much slower much smaller than is the case of the protein moving in bulk solution so why is it so slow well the reason we believe is because the protein actually has to read the DNA it's looking for a particular sequence so in this experiment the DNA is Tethered at one end onto a surface and then flow stretches it out again you don't see the DNA but the proteins that are brought in stick to the DNA and they're pushed along in a kind of um irregular Herky jerky fashion and we believe this is because the proteins are looking for their binding site they're looking for a particular base sequence here of 8 or 10 bases and when it finds something that's close it's sticky it tries to stick and then realizes it's not the right spot and it has to escape from this trap to move further um and so this will greatly slow the motion of this of this protein uh as it searches along the DNA and so we've analyzed that uh and we find this sort of hery jerky motion can be uh predicted because there are long periods where it's trapped temporarily and so the diffusion coefficient goes down rapidly as as is predicted if you make the depth of this uh of this rough uh surface so-called free energy surface greater okay finally um we looked at a transcription so the DNA is um on a surface and we can add in what are called nucle uh nucleotide triphosphates uh these are you used to build RNA chains these are the transcripts which are like photocopies of pieces of the DNA and those transcripts are imaged here and this video here is showing you in real time the motion of the protein which is moving in this direction along the DNA transcribing making this chain as it goes so here it's moving in Only One Direction there no brownie in motion uh and the reason it can do this is because these uh bases these nucleotide triphosphates carry little power packs with them and that energy source is used to grow the transcript chain and push the protein further in One Direction so that it can uh it can convert the DNA information into the into the RNA transcript okay um we tried to mimic this process and found it to be very hard so um the idea here in this grant that involved a number of of uh researchers including Jim Baker and his his group was to take a molecule called a dender which is proteinlike in its size and shape and Carries positive charges so it should be able to stick to negatively charged DNA and we hope to see it move diffusively along the DNA proving that the interaction was mostly due to charge uh but we found it didn't do that and in fact in solution the dender uh causes the uh DNA to basically collapse into these little bundles um and uh when we simulate that we simulate the dender interacting um with the DNA uh we find that the DNA is wrapped around uh the dumer uh like string on a spool and we believe this interferes with its ability to move along the DNA but this is related to other problems of Interest including uh what are called histones These are structures that are uh part of the chromosome structure it's the wrappings that you get in packaging in packaging DNA into the chromosome so what would it take to get a an artificial protein to slide along DNA uh and looking at actual proteins you find that this is an example here where the protein is shaped so that the DNA fits uh like a hand and glove and the charges on the protein seem to be strategically located in order for the DNA uh to be able to configure itself um so we hope to be able to uh learn more about this by designing uh ders that have specific shapes and charges okay so now I want to move on to elasticity um and to do this we'll start back with this problem of coding DNA onto a glass slide so I mentioned that we wanted the DNA to stick only at the ends and this we believe happens because the ends of the DNA a have bases inside that are greasy and will stick to a greasy surface or an oil loving surface so the surface that we use the only kind of surface that will work for this is a surface that doesn't like water doesn't like the water droplet but likes oil so we believe the ends of the DNA stick but not the middle and to demonstrate this um my student Jun deposited a DNA uh chain on the surface and then cut it with a beam of light and you'll see that the DNA snaps back to two positions you can see it repeated here indicating that the DNA indeed is just stuck at the two ends leaving the rest of it free for the protein to interact with you can also see here how elastic the DNA is it snaps back like a rubber band and this elasticity is again caused by Brownie and motion every little piece of the DNA is trying to move in all directions randomly and so this will lead the DNA to very quickly coil up into into a blob because of this random motion so if we look at free DNA long DNA in solution it coils up into little blobs and then these dance around by Brown in motion but if you put flow on as Steve Chu and his group did you can see that the the DNA will stretch out if you turn the flow off it will collapse again so this becomes a basis to understand polymer elasticity and we can begin to get at the problem of how polymers are elastic and how that affects their flow properties um so we've simulated this in in in very clean flows where the experiments were done at Stanford and you can see the DNA stretching here and using Brownie and motion and fluid mechanics we predict the way the molecule will stretch out in the flow this is called a dumbbell mode of stretching uh and it agrees extremely well with uh what's measured but there are other modes I'll show you just one other which is called the fold mode where the DNA acts like rope over a pulley and you should see here that's brighter on the top half uh this U simulation shows you in more detail What's Happening Here uh and this is another way that the DNA can unravel so by looking at this in great detail we can understand how uh long polymer molecules are stretched in a flow and then we can apply that to other polymers such as the one ones that are used to make products so an example is here where a collaborator in Australia is using this device to stretch out um drops of polymer into long filaments the way perhaps a spider would stretch out a liquid into a filament and you can see how you can stretch it out and it doesn't break you'd expect it to break up into little droplets it does not break up in the droplets because of the large elastic stresses that develop as the polymer molecules are stretched and in fact fact in this device you can measure the forces and determine what the stress is as a function of time after you turn on this flow and you can see as you stretch the filament these stresses get to be very very large especially if you go at high rates in fact you can get to 300 times the stress that you have in uh the fluid without without the polymer even though there's only .1% polymer present so it's a huge effect and this is the basis for spinning fibers that we use in fabrics for example and if you get bored with my talk you can demonstrate this to yourself by taking a little bit of your saliva and if you pull it out between your thumb and forefinger you will see a little filament form and it's remarkably stable okay if you do it right which I learned to do in grade school and I guess I kind of got bored but uh maybe it was a precursor to the work I'm doing now um uh but anyway this this is the basis for the large elasticity that enables processing um polymers into sheets large sheets and into uh into uh fibers uh and the lines here are the predictions that we made of these forces based on what we learned uh initially from the DNA molecules and how they are stretched out now here's another application of the elasticity uh and those of you who work in the kitchen have seen this probably uh if you try to a cake batter you'll see the batter climbs up the shaft of your egg beater and that's what's happening here this is a polymer solution uh and this is called Rod climbing uh the reason this occurs is the polymers are stretched out now and they wrap partially around um this shaft of course they're too short to wrap all the way around but uh the polymers are are are curved uh and this curved stretched uh shape of the polymer cuses it to pull the fluid inward uh and it drives the fluid up up the shaft so this is like the opposite of what we see with a centrifugal instability or centrifugal force where you expect rotation to be throwing the fluid outward here that effect is very weak and what's strong is the elastic Force pulling the fluid Inward and pumping it upward so now this is the clue that lets us explain the instability that I showed earlier that is if we have fluid between in the gap between two cylinders and subject it to flow uh the stretched polymers want to push fluid inward that pushes other fluid out of the way and it has to cycle around now it turns out when we solve the equations for this the fluid mechanical equations it's quite a complex flow that's not even steady it's it's time periodic with these vortices that appear at the inner surface and actually move across and then repeat themselves in an cycle and this is all caused by these elastic forces in the polymer uh and this basic instability shows up in many process processing applications of polymers including the parallel disc flow that I mentioned and even flow in channels uh is now suspected to have instabilities related to this phenomenon okay now we can try to use this elastic force uh for good purpose as well and this is some work that we're just starting uh with stha NR who is developing little microfluidic devices to separate cancer cells from normal cells uh and they can be separated based on the different stiffness of the cell so here this is a simulation showing uh a flexible cell and a stiffer one and you can see that in this curve geometry the interaction of the flow and the stiffness of the cell can cause a separation so this would be useful for detecting uh cancer cells and we'd like to be able to to uh to understand this process okay now I'm going to go briefly through uh self assembly um and self propulsion so self assembly is very important in biology is only be beginning to be utilized in uh in in commercial applications but uh of polymers but one simple way to get a kind of assembly is to make the polymer self-attractive using charges for example and stiff and then as you turn up the attraction the polymers will collapse and they will collapse into different shapes either these bundles or toruses or globules and these are uh reminiscent of what's seen in biological polymers for example in DNA uh or or proteins another kind of self assembly occurs uh with um what are called lipids with self ass which self assemble into these Bayers which uh are the membranes that uh cover every cell and this can also occur in self assembly also involves proteins and peptides which will take on specific orientations um in the interior here of a lipid bilayer and these are very important in transporting ions uh and information across uh across lipid membranes uh another application of self assembly with um with lipids is in what's called lung surfactant um so when you breathe you have little air sacks that expand and contract and the work of expanding them is so great that you couldn't breathe without a surfactant that decreases the surface tension and lets these sacks expand but then when they contract the lung surfactant has to go somewhere so in these simulations during this phase when you're exhaling um we find that the lung surfactant will uh form these little pouches um they little folded regions facilitated by peptides that are present in the lung surfactin and if the these are absent uh as they are in very premature infants you can't breathe so that's known that you need this in order to enable uh proper breathing um so uh one of the interesting applications here is to design artificial um lung surfacant components that would could be of use uh therapeutically finally I'll mention um particles that can self assem assemble because uh they are not homogeneous and an example is a so-called Janis spheres um this is the this is the god the two-faced God Janis uh and these spheres um have an attractive side and a repulsive side and so they will assemble into different patterns as groups such as the granite group have shown uh and also there's very beautiful experiments by um Mike Solomon and Sharon glatzer glatzer is doing the simulations of these um so uh in our group this is worked by Daniel Beltran in collaboration with uh Mike and Sharon we find that these kinds of particles can self assemble into into Lamar structures that are shown here and these these are like very a poor man's um proteins so finally I'll get to the last topic which is uh self-propulsion and this is something that's quite a hot topic now to try to make artificial self-propelling materials uh so large animals swim by using inertia so basically in one way or another uh uh you you throw water behind you propelling yourself forward using the um equal and opposite uh for every action there's an equal and opposite reaction so use the inertia of the water that you throw behind you to propel Propel yourself forward uh but tiny organisms like eoli are too to small to have any significant inertia so they have to find another way to swim uh and the way they do it is they have little quk quk screw tails and they literally screw themselves through water they rotate these Tails propelling themselves through water and they have multiple of these Tails called fella uh and they can use one of them to steer like a like a Rudder and you can see these little guys here from Howard Berg's website uh steering themselves around uh these are quite fantastic uh machines really this is a a video that shows what's really going on there's a motor here which looks like an outboard motor but it's not powered by gasoline it's powered by protons and that rotates this shaft around and it goes around 100 times a second um and then drives this uh flagellum and these pellums spontaneously bundle together and they do that because of the stiffness and because of the fluid flow that they generate so we want to understand in detail how this worked um and so uh my student Nobu watari developed the first whole body model of an eoli um bacterium that includes the fella uh the body which is built out of little beads and springs kind of a tinker toy model that has the mechanics in it as well as the flexible hook that connects the um felm to the body and then there's a torque balance you have to satisfy uh again Newton's law is at play uh and this allows prediction of the swimming and the correct swimming speed and here you can see uh the bacterium tumbling and changing its direction so it changes its direction by counter rotating one of its flagella uh and that causes it to Tumble and then take on a somewhat different direction once it starts swimming straight again uh and we find that the uh tumbling the angle through which the uh bacterium will change its direction is very sensitive to the location of the fella on the body so you can get large tumbling angles or very small ones and we're comparing these now with experiments from Howard burgs group that also show a very large distribution of angles that you get so there's some very efficient um uh rotations or changes of Direction in the bacteria and some that are very inefficient bacteria actually can smell food and they can direct their turning so that they they they rapidly find the food source okay so I've given you a little bit of a snapshot of what goes on in my group um and I wanted to thank thank those who really did the work this is my current research group uh and you've seen only a little bit of the Fantastic work that that they're doing uh previous students in posts are are as many of them as I could find anyway they've they've been scattered to the Four Winds but um they're at some of them professors they're working at companies um startup companies some of them are Consultants some of them work for uh government agencies uh government labs and so forth uh and I'm very proud of of the tremendous work they've done also I I I would be uh doing practically nothing without the collaborations that I have almost all the work I do is in collaboration and uh uh these these are U um many of the collaborators that I've worked with I want to especially mention Mike Solomon and Sharon glatzer with whom I've worked uh uh on on a number of projects um okay uh I also need to thank the University of Michigan uh which uh has a a glorious history uh and u a very bright future I believe uh also of course I wouldn't be doing anything without the funding sources uh so I want to uh thank them uh for supplying me the resources to be able to continue this work um and uh my family very very supportive um um BB's here I think I saw my daughter Rachel come in as well others are too far away one's in China Emily's in China right now teaching English uh so she couldn't join me uh and finally I hope you'll indulge me uh for a moment um um um because of my convictions I feel I should uh thank God for uh all the good gifts that he gives me and finally I want to thank you the audience um for coming to hear what I had to say and uh thank you very much I'd be glad to take questions from any of you [Applause]