Yiyang Li PhD Defense

bringing everyone it's my great pleasure to introduce ba-leep who is defending this that PhD thesis today so there are many problems and advisors career and this is an exceptionally proud one because IANS one of the first students at Stanford it was there when there was nothing in the lab no blow boxes for example when we knew how to make nothing so Ian was passed with starting research Efrain on batteries in this really put ourselves stake in the ground and make us well known in the field so the work is about to present today so when I look for graduate students when they come as first-year students I often look for two qualities creativity is one of them and also persistence as well so in case he encounter a lot of problems his research and there are two ways to go back one is you you know abandon it and move off the next topic the other way it is - well think about why and as many of you know many of the interesting discoveries come from explaining these you can't explain initially and I think every paper that he has published is exactly that and started with observation that he doesn't understand and over months or years of additional experiment modeling discussion with others and so forth you learned why this is the case so he will present some network so aside from the very prolific research has done over the past five years I also want to highlight his role beyond just in research is being a great mentor there are a number of students that he has mentored undergraduate students Norman and Sophie in the back and a number of graduate student that he had cosa mentor over the years as well he's also an educator holiness graduate students he has th all my classes I think he has taught my class he was kind to take one from a class max I 303 last winter I think in this sense this is really the most comprehensive education you can receive by Stanford by tackling not only the research part but oh so the mentoring education outreach and so forth so without further delay let me let you in get to it thank you thank you very much for the kind of tiny introduction well so over the past five years that the real privilege of working at Stanford working in the field of lithium-ion batteries this is a field that has God has significantly more interest in the last couple years so specifically my topic is trying to understand the mechanism of IO insertion and phase transformation within phase transformation in this material called lithium iron phosphate as we know this is a multi-link scale problem so I ain't you tackle this all cause a variety of a skill of course electro single particle and atomic plane skill I bring up this image here the first slide just to if you can't tell that this is red and green I have a whole turnip green top version they can get from either out here that does not have this red three issues so but if you can tell this is red and green at this presentation so we all know that working slippy amount batteries in our poor boy electronics and now a metric vehicle if you're fairly wealthy the application for lithium-ion batteries going for is that it is enabling mechanism for renewable energy here we have a lot of the price of solar photovoltaics over the past eight or so years since Isis out there's not a Stamper because the United commit is already extremely cost competitive it's already approximately the same cost as sold as electricity from coal and natural gas the problem with solar right now is that it is an intermittent source meaning that the Sun does not always shine therefore if we were to only use intermittent renewable energy it can only about 20% of the electricity grid before it becomes destabilized thus energy storage stability is score did you turn today and to use a trend at night he is the key enabling technology for the widespread adoption of clean renewable energy there are some existing solutions there now the most prominent one being lithium-ion batteries such as this Tesla power pack that was advertised about a year ago and then within each of these energy storage it contains is Panasonic or other manufacturers lithium-ion batteries these cells which connected in series and in parallel can be used to store enough electricity for it so what does a look amount battery yeah give a brief overview thinking for my class that's a 303 lithium-ion battery contains a negative electrode and a positive actual when you discharge the battery when you are taking energy out of the battery and use them for electrical work what you do is you're moving a lithium atom from the negative electrode to the positive electrode the lithium leaves one electrode and enters the other in other words this net kimmel called net chemical or electrochemical reaction releases about 320 kilojoules per mole lithium it can store white a large amount of energy this energy is not that you loss a scheme because I feel the electrochemical circuit and use this energy to conduct electrical work for us in the electrical circuit we have concept known as a voltage which is simply that gives me energy divided by the number of charge passing so note on nomenclature are you when you charge a battery which are doing is I were move removing a lithium from this lithium iron phosphate glass so charging is the same as the theory blue bow exactly studying this electro discharging is the same as lithium insertion well this semantics is simple a fundamental question is how do lithium patterns lithium ions insert into the battery particle where when and how does this reaction happen they look simple but the real lithium ion of macro structure is extremely complex particles have all different sizes and various amounts of connectivity so it's difficult to understand where the lithium atoms go inside of a battery electrode however fundamentally this is a very important topic in terms of both a fundamental understanding as well as pattern performance one example is overcharging for serving battery materials you don't want to charge it too much if you charge it at least the irreversible fabrication my colleague will in the audience here has demonstrated is that even inside of these fabric will actual this is a standard layer oxide you see certain regions that are more charged the green regions than others are less charge the red regions as a result you make deep that your battery is homogeneous li4 you may think your battery is in a safe State but as certain errors of value or more charge than others it could still need you irreversible degradation another example is on the negative electrode side so dr. Steve Harris who is right here in the audience have demonstrated that if you put lithium into a negative electrode too quickly if you charge up the neck and electro too quickly they the lithium were specially start crashing out of the graphic in other words the graphic can accommodate such a high flux of lip here what happens instead is a metallic lithium will start crashing out not only does this lead to tech the radiation capacitive base it can also pose a safety risk so understanding how where and why live lithium insert into battery is extremely important absolutely making this is a multi-link scare problem the first so I will try to tackle this as a variety of links guest the first link skill is the force electro the question here is which particles do lithium ions insert into this fundamental relates to material science concept of nuclear secondly I look up where exactly in the particles the still appearance in certitude so I'll zoom in to each individual particle and look inside the particle this relates to electrochemical reaction as well as of growth as a nucleation growth the growth part of that and finally assuming even further assuming - now that the lithium has inserted into the particle how does it move around how does that go from one part of a particle to another this relates the material science concept of diffusion which is fundamentally an animus they make skill atomistic approach so before going forward how we briefly reveal the double dynamics of lithium ion phosphate has a thermo professor it's right here in August the voltage I said earlier is related to the derivative of the Gibbs free energy of how the DM insertion so this is the change in the Gibbs free energy as a function content however as you insert a moon lithium from a whole structure we take mass in and out of a structure we can imagine that in a single crystal project phase can accommodate such a change in a mass as such a large change in a body that's exactly what happens in this material you see they know that the composition when the crystal is larger devoid of lithium they have one face here which I'll call say the green face here with the crystal is full of lithium was completely full of livia another face may be present here we call that maybe the rest base over here because that's a lower Gibbs free energy I should learn in phase equilibria and thermodynamics we can take a common tension approach you try to look at when these two faces will be in coexistence such as here and then we can drop the faculty kind of like a power we can start really on the face diagram in particular a very low and very high lipid composition we see just one crystal graphic face and an intermediate looking composition we see in other words under equilibrium this material lithium iron phosphate will separate into lithium rich red and looking for green domains even within electro so next we take the derivative of the Gibbs free energy and we arrive at the chemical potential which is shown here and finally the voltage this is the negative of chemical potential and this is experiment to acquire voltage and you can see that the curve is fairly flat consistent with this being separated so next I'll be talking into kinetics talk about diffusion but once again this is a schematic of the phase separation in this material we can see that separates into this lithium rich red face on the outside and looking for green face on the inside this is just the the exact phase boundary is depends on the crystal but generally we can say that is parallel it is in the same direction to face boundaries as 0 1 0 axis this is important because as of having an orthorhombic crystal structure this material is highly and highly anisotropic in diffusion lithium and diffuse allow the 1 0 the size of 0 1 0 Direction extremely quickly it takes maybe a millisecond state direction however if using in the other directions is excruciatingly lis slow to diffuse from this part of the crystal that part of crystal in the fault maintain as much as months actually I shown later why she'll have this that does takes months so for all intents and purposes this is effectively a one dimensional conductor and these so I said my cop my research looks at batteries have different length scales specific relating to nuclear chain reaction and confusion this is a concept that doesn't just apply to batteries this is a very general concept if you apply a general system while the feet but we have a whole chemical species here and then you're in certainly chemical species a via chemical flux into the host as shown here every in certain of chemical species a and they're not miscible you eventually start your face separate perhaps in a pattern that looks like this over here you understand this phenomena is big general chemicals material science phenomena convenient to understand three parts the first is that where does a base transformation begin where the crystal does it begin this relates the concept of nucleation the second is what determines the rate of chemical flux that is a problem of reaction chemical reaction and growth and violate is a question of how the chemical species migration installing fundamental context of the future you can see that the principles in the lithium-ion battery actually a general here are some modern examples of a phase transformation one example is resistive memory where the movement ion can use a filament who used to store information another example done by my friend Tarun in the diode group was looking at inserting hydrogen into palladium crystals inducing a phase transformation in the whole school a year and using that you may be able to store hydrogen gas and finally it can be also be used in nano particle synthesis such as cation exchange but by which you substitute copper into a cameo solar light crystal and they induce a phase transformation that is reversible by changing what species what species your substitute so of course now look at the first bottle specifically a talking about a permutation so no when we think about nucleation the classical concept is this is in terms of solidification you start Wednesday a liquid a liquid metal for instance and you bring that below the melting temperature of this for example higher we don't below the melting sure at the very beginning you're not gonna forming a solid it takes some time for the solid iron to start nucleating you can see the nucleus professional over here however as your lifetime pass as you wait longer what you will see is that these solid nuclei will start growing but the other but at the same time you can form other new class at different parts of this liquid and eventually you will keep growing the new class and new nuclei will be forming a build a new class start start basically combining together and filling the entire so this is a very well study phenomena and this is it is characterized by something known as a run equation which says that the fraction of the solid is can be defined by 1 minus the exponential term that depends on the temperature however this however this type of formalism doesn't really apply to a multi particle lithium-ion battery pack so the reason is that human can start nucleating within each individual particle the size of the nucleus cannot grow to be larger than the size of the particle once exist the size of the particle the particle no longer becomes active so it cannot keep growing as in the solidification example so as a result we need a new theory a new understanding of how nucleation will perceive inside of this multi particle scenario whereby the Lithia cannot grow to be larger than the size of the part so to do this I want to be able to know exactly where the lithium is localized within the particle here we do this using x-ray spectra microscopy at the event at the advanced light source in North Berkeley National Lab how it works is I will have a month we have a fernell zone play concentrate monochromatic x-ray light in the variable spot focus area of all the detector behind it detects a fraction of transmitted or absorb light so can i raster scan in the sample we can create what the same let's like what can also do is we can conduct x-ray absorption spectroscopy meaning we change the energy of the instant of photons and measure what fraction of the light gets absorbed at what energy in other word we measure in other words we're able to conduct x-ray spectroscopy what this does is a sensitive to the oxidation state of the material at the hiren here is the oxidation state of 2 plus vs 3 pod as the distinct chemical signature wind absorption spectra therefore we use this as a finger printing technique you understand where the lithium is localized in the electrode lithium is always paired with the iron 2 plus oxidation state so here are some examples we first take an image at some 107 might rumbles we see that certain particles for example this particle here is more it's darker how about some 114 light rumbles we see now that this particle is a darker one and finally when we finally when we recreate the facespace we will we take the whole spectra we can recreate an image how about something this this green particle here is larger devoid of lithium whereas this red particle sorry the red region there is full so now I'm gonna do start doing this on a real life battery so I am describing this pathway you can see that the voltage is slightly below a big one for both antique and I conduct x-ray absorption spectroscopy spectrum I must be on different areas of the battery as I discharge it because we can take a series of images as shown here so I've shown here we can take the whole different so these are different particles and as you can see it eventually goes from a fully charged fully delineate the green to a fully discharged Lydia get red state this is done in organic liquid electrolyte what is key here is that there exist heterogeneity in the material Alexa doesn't just discharge uniformly fun but we need yellow to orange to red rather certain parts of the particle on board are completely discharged and other parts of particle are still completely charged this significant heterogeneity in their lecture the natural doesn't charge even formally at least for these particles so now I'm gonna zoom in and take a high-resolution snapshot this is this is a student snapshot take care well now you can see what the individual particles look like the red regions are individual particles that are fully lit he ate it and the green ones are particles are dealers needed to see that in this electrode which is normally half charged all of the particles are almost essentially in a fully lithium rich with a fully lithium for state the only exceptions are these two particles here these particles are the ones that are actively circulating while when the sample was being prepared so we look at a schematic view of what the multiple particle instruction packet is we can have two limits in the first limit all the particles are lithium was in a second limit the undergo systemic genius particle by particle mosaic like pathway we can see that our data suggests that a intercalate in this mosaic particle by particle pathway in other words the particles are litigating sequentially one after so onward this insight I conducted this experiment for wide variety of particles I can analyze 1,500 particles inside of this inside access and pop the ring that you are charging for discharging one scene means you're discharged in one hour to see inter discharging in 30 minutes on the y-axis an active particle fracture those are the two particles that are circled on a previous slide the ones that are actively charging discharging what I found is that as you increase the rate of the creation the electron Lydia is much more uniformly in other words it's more uniform more homogeneous at higher risk so this is a little counterintuitive usually think about trauma system faster we starts the in diffusional braking in the system so to try to understand this we have we look we go back and look at the concept of nucleation in this material will force it Sport idea of maybe just having single particle magic lecture with just a single particle or in certain lithium in and out into this particle when we insert enough lithium you imagine go past the so-called nucleation barrier and a passive eventually reducing this chemical potential remember it's easier to add lithium to a particle as are indicated here the chemical potential is not lower what we can say is that we wanna lift the eight this electro slowly everyone that discharges slowly without supplies small over attention we apply small bias and was slowly puckering thank you this part of current and if we want to drive a little faster with a bigger bigger bigger the more the faster you want this particle to charge a larger over attention field so the rate of charging this particle is something that we can control how about fortunately real battery doesn't just contain one particle it contains an ensemble particles working together now let's explore this dynamic we have a hold of all particles now both of them will overcome to the creation area and because they are slightly different one of them will have to be patient and person now you have these particles however we can still only apply one voltage to the system as a result the driving force for looking and the rate of growth is controlled by how high this barrier yes this transformation period determines the over content over a certain lithium here because you can only apply one global voltage on the system the difference in the chemical potential this particle and this particle relates to the driving forces with mr. get back in does we can't use we can no longer use voltage to control how fast we serve lithium thank you a single particle number is another parameter that we can't control we can control how many particles are active and how the particles are inactive so I love rates of our global current is low a small number of particles are active as we increase our rate more particles become simultaneously active even if the particles themselves are living at exactly the same rate eventually all the particles become active and finally had to only have to bury and do all the particles we are not there and then we can change the rate of insertion for a single particle so here's a schematic somewhere of what we found we found that as you increase of cycling array in a very limited bottle the active particle fraction or a fraction of the electrode that is lifting in simultaneously tends to increase traditionally family model issues adhere heuristic of using there's a homogeneous model for trans derivative model whereby the Hat the electrode becomes less uniform anhydrase we shall die nitro it's more uniform a higher rate cycle so we we publish this worked on almost two years ago because recently this year I was excited to see that a group in China has used our understanding to try to engineer to engineer a better battery whatever pulse is that by reducing the height of the barrier by reducing the barrier we can increase the fraction of active link or plenum particle so they're able to do some morphological engineering to try to reduce the very high what they found is that they can make better battery by reducing despair by reducing the size of this barrier more generally the the concepts developed here for lithium-ion battery generally apply to particle nucleation parole system which is driven by an external chemical philosophy at the voltage of the system is externally control the over potential then we can use this these concepts to understand how you can have nucleation growth when the particles are physically isolated from each other so that's a first part of my talk now go to zoom in from the multiple particle length scale and looking to individual particles so sad look at nucleation I'll be looking at chemical reaction and growth the back way related question is where in the particles do but in certainty so like acknowledged by an Inca main partner in this work dr. Joan will if he's been up we work really great as a team together and this work in this part can I have been possible without their for taking that I'll make it today because exact jeopardy taking more these measurements I heard he is our sky brightness so the motivation behind this part is to try to understand why lithium iron phosphate why is this such a good stable high rate material we can see this is taking from a full cell full cell battery what they were able to show is that they can cycle it at a very fast rate can see for charge 5c for discharge for 20,000 cycles without any evidence of degradation this is extremely surprising because the standard model for this material is the space separate is a space separate including phase boundary model in this model because we have this we have two phases coexisting next to each other it builds up a large degree of coherence stress stress stress up to you about one Kingdom one Giga Pascal's due to this coherent phase boundary this the mixing question is how can you do this something this movie and face boundary so quickly but this material being so stable so that's what motivated this this part so the question of where it lithium or halogen insert into an individual particle when how has been studied by others as well this is a standard model the base separating model which selenium only inserts into the base boundaries of this material an alternative model is called a solid solution model whereby lithium inserts into this material very homogeneous thing so the team can insert homogeneously into the material suppressing some recent work done by Professor lama Martha and the Netherlands have suggested and low rates in starting his face boundaries as well as a low fraction of active particle now hydrate you start seeing this diffuse phase boundary did not actually see a taking Purdy's based on the diffraction based on the diffraction patterns here so our goal here is to try to understand exactly where when how lifting inserts enqueue and individual powder particle in other words we're going to be able to describe where the lithium inserts and particle which is they want to take movies of the battery charging and discharging so that's what we've done next here I've synthesized these particles in micro micro planar particles and what we let the particles relax will be discharged them to a certain state and let it relax we can see this very strong base boundaries between lithium bridge and looking for domains now could that take the dynamic picture I'm going to show the movies of this fabric during charging and during discharging first we look at the charging picture I shown here we can see is that different regions of the particle such as here starts litigating before the rest these regions had a more lenient than other places nothing like nothing like that so by doing this experiment dynamically can seat you we're able to capture the nominal liberal state the actual state of the material during the charging so what we can do now is not a Lincoln we look at the image images here we look this is a map of the luteum composition so red being volunteer green being devoid of we can do this we can next take the time derivative of the composition we can take the derivative of the composition and because a composition is related to how much charge you plug into the battery the time derivative of a composition relates to the current or the rate in which of which hardly into this material looking a max of the current density that's on the fireball by these numbers so we have the composition we have the current density on each individual pixel each individual particle and we can also get the voltage from the general voltage of the battery so that's what we can do is we can amplify of a terminal net exchange current density as a function of the lithium composition you can think of the exchange current density as a compact conductivity or compact resistance or lithium to go from the liquid phase to the solid phase this type of compact resistance the higher your exchange current density the higher your compact conductivity the lower your contact acidity that a more lithium you can insert into these particles for a given over potential so you want to really get badly you want a really high exchange current density what we have shown out here is that a very low a very hundred decomposition the exchange current density is very low almost negligible algorithm had intermediate compositions here we'll get a very high exchange fantastic get a very fast insertion of Libya into this material therefore as a result we can as a result we show that as a result the solid solution lithium ion phosphate this yellow regions here are the ones that are very reactive that lithium can be easily instructive material so reason that lithium ion phosphate is a very good hydrate material this precisely because a solid solution lithium ion phosphate as a very high exchange current density but windy district you in the solid solution by suppressing the space separation pathway previously described we can actually get this material it's that much better so this explains why this material cycle so quickly now I'm going to explain why cycle so stable so at low rates again we can see that you see his face boundaries these phase one is create these coherent stresses in the material as you increase the rental if creation has you litigate this material faster faster we can see that compositional gradients starts to diminish other compositional gradients start to diminish in other words this phase boundary is larger suppressant win cycle faster you bring the state inter solid solution do you understand why we look at the picture of a separation here as the material is face separating and lithium can really only insert into the material at or near the base boundary as your cycle of very quickly sorry if you're cycling very slowly that is not a problem because you only put in a little model with you into the material just put in stir again to face boundary is so fish and the material is globally in the thermodynamically more favorable state when it is faced happily however if you want to insert looking into this material they quickly then you overwhelm the capacity of this base boundary to take all the excess looking what you don't have to do is you have to cancel it yeah homogenious li all over the entire article so an entire particle is the active when cyclic a faster rate so high recycling rates in fact minimizes compositional heterogeneity as a result reduces the prospect of mechanical strength as well so that's part two of my presentation the understanding of the difference between reaction reaction and growth inside of individual vadik particles under electrochemical conditions so finally in part three by the park I'll be discussing the migration of lithium within each individual particle so I'm zooming down you know animistic my scale I understand how lithium moves from what part of the particle to another so revisiting some classic ideas of a separation if you imagine you have a cowboy a solid solution of iron and nickel if you cool the alloy down you eventually separate into this nickel rich and these iron mention what determines the rate of phase transformation based separation in this material is only the rate of transport for a chemical species it depends on how fast you can move nickel from this particle from one part of the one part of the material through different part of it here chemical diffusion this has been wavy well studied special metallurgy when you're thinking about both the future with the fuchsias per minute with a bowl you can use the computer equation try to understand this phenomena if you talk about external chemical flux now you're inserting your second chemicals into the first one you can use an alençon formalism I finally when I think about service transport but how are my contribution to this is in the field of surface transportation in particular we have a solid solution shown here and then you can move solid solution shown here so if you have thin separation I propose that you can have movement of chemical species at an interface between asada and a fluid as materials become more nano sized this becomes more important because you have more surface area as a result it is essential to be able to understand the transport of chemical species so now give an example of ion migration the transport of chemical species within lithium iron phosphate to a phase separation this is perhaps a view of a solid solution lithium iron phosphate this is in a meta state metastable state he will eventually to face that break into something like this let in has been moved within the particle has moved from one part of the particle the middle part to end so that the space here's an animistic view of it imagine you have a lattice and lived here inside they want to go for the solid solution picture to the space separating picture we have a space boundary like here looking literally has to move from here here remember as I said earlier Libyan migration within a particle is excruciatingly slow it takes months for this to happen so that makes you wonder how this base transformation so that is the subject of investigation so the question is how do you isolate the lithium iron phosphate particles they separate isolated I mean they are high on eclis isolated the ten extremes look in with each other so if you just take the fabric particle you bring get into not relatively solid solution state and just leave it out in the glovebox for a few days or a few days you'll see that largely the solid solution it's like the solid solution character is smart shape does not change the material still really main enlarged it's not a solution however if you take the steam particles and we've been out in air for another three days we can see that another basic camera is completely based operating from here now this largest solid solution particles becomes like the space separation target or Agra if you leave any hair in our go back to the glove box for twenty days and take another sample leave it in a box for 20 days it still remains a solid solution in other words this material does not seem to be based operator we're putting it in air or couple days significant therefore what we can say is that the environment the environment they put these particles in largely determines the rate and of this using diffraction in our variety of other particles under lecture Agassiz and on the y-axis o'clock the fraction of particles they saw the solution you can see that by putting it in air when moisture the solid solution fraction significantly pops off in other words the soft this material base separates much more quickly we have moisture so now let's revisit our picture of a separation here to see a web propulsive it's copying extremely difficult for a lithium to diffuse along the ball is not only slow crystallographic direction agra by plugin lithium at Zorba's sorry i put in water or hydroxides or based on the surface we can see is that we can so we will suppress the bulk separation or rather folks operations already so present i rather we have surface transport dominating the lithium migration path with these absorbing this molecular absorb experience significantly increased the rate of based separation increases the rate of lucan transport within individual particles dance orbit as a result now we can use the choice of as working we can use the choice of environmental foolish to ultimately control the rate of base operations serviceability is something that can be modified by the choice of the fluid so this is the case of isolated particle we can control the rate of a separate space separation by changing the choice of the fluid molecular absorbance now let's consider the case of connected particle like in a real battery lecture when the particles are connected ironically you can have clean separation in one of two pathways it may happen through surface diffusion as that earlier lithium migrates within a particle but there's still largely localized to his original particles this results in a separation within a particle you can look at animistic picture here it moves on one part of particle different part of particle but it's enlarged to vocalize its own particle because surface diffusion keeps the part keeps the lithium within as original part but you can see phase boundary here however the other way since these particles are ionic be connected you can also have insertion you can also have redistribution do an electrochemical reaction here now we have lithium moved from one particle to another particle as you have this mosaic mystically yeah if you move from one particle good another eventually you get it entirely filled it da dit looking for particle and almost lippy ate it with the enriched car cover so these are the two pathways here you can imagine that this pathway here has a lower energy don't be dynamically this is a stable Catholic with the sphere you build up interfacial energy and you also build up coherent string because these two materials have different lattice now experimenter let's try to figure out which of these pathways do we actually serve so we first look at the baseline case I'm the case of no connection line exposure we can see that behave as a solid solution so we can bring you into this path of the schematic shown which will say to care for a couple hours for eight hours in a 0.1 molar electrolyte we can see that space separated it looks a lot more like this because it phase found reasonable for the particle then it looks like this next we can bring this electro we did the same experiment with a higher molar concentration of the electrolyte so you expect so we are increasing this reaction by a factor of 10 what you can see is that well it looks pretty much the same area there isn't a big hunk dependence no degree a phase separation on the concentration of the electrolyte and finally we keep that in lxy for 20 days restore in there for three weeks now what I find is that we can love at this stage the particles emerge in a multi-state this confirms that this is a lowest energy configuration of the system but this one is kinetically more favored in other words based operation within a particle is activated a surface diffusion because we see these patterns at lower time only after a long period of time does it go into this mosaic so the idea of using surface diffusion to control phase transformation to control you you control diffusion transform can be applied in a variety area I change into molecular orbit which has significantly changed the connectivity of our material this has been used in other lithium-ion batteries this is a recent paper published by Parker cow in the UsWe group in which they use magnesium oxide and lard and insulating material to capture this look employers all high species include the carbon network they were defined that whereas luteum polysulfides those that refused to the ball it could diffuse to the surface of particles which is why these that means the outside can be used to capture and improve the capture side we excite species and prove lithium sulfur batteries another example is that can improve the number of triple phase out rethink electro catalysts typically electric talent reactive is a suit you have not only other triple phase boundary with a electron the ion and the gas to meet each other Harbor by the surface diffusion in this case surface electronic diffusion in is larger insulating material you can in fact have reaction happening a possible much larger regime if you can control the surface diffusion then can improve your electro catalyst by activating the width of these triple phase boundaries finally we can use it to control the rate of phase transformations as we use nano size materials as the amount of surface area starts to increase we can start changing the choice of the fluid absorbing you ultimately control the rate of phase transformation within nanometers so here's a summary of my thesis presentation might write these my thesis I love that phase transformations induced by high insertion and of arriving at my skill first is a multiple particle legs feel another question confusion then I look at a single article links joke when I look at reaction and growth within individual fabric particles and finally I looked at how lithium transports within individual particles held again goes from what part of particle to another and fundamentally this is a puppet about diffusion about surface so now it's the fun part of the presentation I get your acknowledge everyone but Stephanie my PhD so first I couldn't knowledge / professor William to my East my thesis advisor that's been a weird wonderful experience learning how to do science where professor willing true so it's what people have when as a person grasp event if they ask other people is your visor hand soft or hands-on adviser so the great thing about will is that when I was just is I need a lot of guidance will those actually quite hands-on was always willing to help me when I had questions when I have problems at being time at 2:00 a.m. you will say I can just call him anytime you have a lazy wakes him up at 2:00 in the morning however since that since by the time of my third or fifth year when I started to develop my own thinking well they say enable me to just be whatever I want the work about surface diffusion is largely a scientific people I felt motivated by myself and will just encourage we could just pursue that path with it so will I felt has been the idea of Iser that makes all crap that makes rescues with the time the best time so I'd like you to think professor Eric Hauck I 14 the chair of my thesis committee i first i first saw Eric when he gave up material science colloquium I can't use a professor I thank you for letting me sit in on his group meetings this past summer I learned I learned fight a lot and hopefully going to apply that for my future I think professor also lay off work first offered Jimmy thermal I didn't take any thermal as an undergraduate student but I learned it very well very well from Alberto in fact in fact will actually quiz me on thermal windows designing what it does I don't know and we're also for being great mentor I just feel like Weber I could just always talk to her top graph article and it's always about the voltage how do you think dr. John John Alden Elsa lacquer we work together we work together key night in my first sorry in the first jobs in a group working a lot of beam times together and now she has gone on to do greater greater Dean's as black as a scientist so I'm very appreciative that you taught me about hydro microscopy and for serving on this committee thank dr. Fritz Prince we didn't get got too much fun really grateful that you are able to take your time or this thesis I also like to thank dr. expressing its way well unfortunately to not make it in this defense I'm really grateful that he let me use his lab facilities in particular this box right here thank you just some of our experiments before we set up our phone Wow he's always been a great mentor whenever I can talk to him talk to me just chat about advice sometimes I talk to him it involve checking the hotel want to thank you for all the guidance that you have provided so I like to acknowledge our groups I show a series of pictures so that names here is people the new people in their group so this is the first group of people so I'm here that's will we like to thank dr. David Bullock the first postdoc in the group for teaching me all these aspects as you all know David is an expert in promoting so our group has grown since then since then 2013 we got some more people when you're later three months later we've got a bunch of then we got more people than there they they're more people here so we started will foam 81 is group to be 1215 people and at this rate we're going to turn to this week students have privilege to make you to mentor Sophie Norman Peter Sebastian I can take our collaborators and repot elsewhere oh I especially hope : David David another David Steve Flint speak loose in the audience here and funds link or either helping me with the synchrotron work having great discussion which is about science in general I think our flowers at MIT Martin ray and pop has black Mike Anna DJ Sandhya furry Josh lon Kyle Martin University batcycle fun group Michigan State way linkage on and the Slovenian is good use of the chemistry to be around you're insane baffled a and I'll be first scientific Norman Valentine's and it's actually like show these pictures you think that people have working closely with Johanna this is picture class at the teen line oh this is a picture of Kolak the staff scientist at Berkeley and this is a full picture of Professor market Bazaar who is here on sabbatical here this year I have no knowledge of funding sources the funding source for England industry Stamper and Peoria as well as my foundation different a clear stamp are what we said that I really enjoy my PhD at Stanford because I can talk to a lot of different faculty about different different aspects of research life in general career path it's a very grateful that they may not keep any time I touch it I get for example but this email call or email Farkle as always willing to me so I can think members of the twig groups were really helping us get started before we had a lap ever always being here for discussion about different aspects of research like to change make the staff of the share facilities table use here are Stanford the students have material science engineering I don't want to flashes too much because then you start realizing science department the students on masta at three or three from 154 had a really good experience being the instructor as well as a TA for these classes I found some alumni whole length I've seen that have seen specially my former roommate Mithun who is scaring the audience I saw some of the faculty for my undergraduate Mark John John's right so finally like to think like my family so my my mom and dad so I'm getting so mean right here this is women went to Hawaii a few years ago I could date my sister Armenia and aspiring scientist for being for always showing for my family for showing all the love and support I could think my now fiancé Gretchen for being the source of mine love for the last four or five years grad school wouldn't have been the same without you you know we have really great that meant adventurous this is a picture of us in Florence has been a has been a really great time in grad school [Laughter] thank you for being such a gracious audience and for coming [Applause] [Music] so maybe on the tomorrow well we haven't haven't investigated at myself like people in the state who have already demonstrated I think I see just nano-sized materials the surface diffusion I'm using the fluid control migration a solid is actually seen in here there's a presentation tomorrow asked by a scene they were using to change chain the fluid to change the surface conductivity of a knife so high do you think the change the mobility at a surface I think it is a very general phenomenon and more important thank you very much that's okay that's very good point so the question is whether it's not just a lithium-ion board so we have this what we did here is we actually my colleague John Woo has personally looked at these parts these particles to a look so these particles as we know are coated with carbon so to improve the electronic conductivity so that shown is that these parts of a particle highlighted in blue are the ones that are faster these are the domains that are faster and the reason they're faster is that to have a thinner layer or copper coating so that when you ask the question to say electrons or ions if you thin the carbon coating of imagine that electronic conductivity will get worse but this material actually charges faster so as a result it's more likely so I'm so that's uh so I teamed up so let's prove a surface diffusion this one a particles are ionically isolated so when I got it so to have the phase transformation install the solution phase separation you need to have lithium ions migrate in the particle well there are two ways they can happen you can to migrate to the material or you can migrate through the surface molecular hats warpage I find it extremely unlikely to be able to penetrate into the bulk of the material so the fact that we see a dependence on the environment such a yes I believe extremely strong evidence that it is a surface mechanism is a dominant migration pathway because this molecular you stop this isn't a layer of here longer hydroxides cannot actually penetrate yeah so that's a big question we've that's what I've been trying to do in the last month the problem is that lithium is extremely light source very hard to detect and furthermore even worse the XPS peak of lithium is 55 electron volts which is the same as XPS people so it's extremely difficult to see lithium in distant Europe for something we're looking at other compositions with you're making these phosphate sodium iron phosphate you see we can use at each other and I think this is a great argument buy it in the solution though another mechanism sorry talk about the isolated or two either way electrolyte rearranges itself as neither in the ball or the surface but right back so so in the case of DS when I just keep it game are gonna keep it here it's I find it on my date ever so in the case of the ionic a conductive particle actually found so it is definitely possible for lithium when it moves out go back in same particle well what that does some basically a simulation without haven't shown yet but generally that mechanism is it's unlikely because when it goes out of it actually like the lithium is free to go to any part of the chooser and will prefer to go into the board single lithium ion needs a single particle that particle is now negatively charged something immediately has to come right so generally I think with the lithium ion lose the particle do you maintain Chargers how straight it also undergoes transportation so that's a good question so we're still with working with our collaborators that University Act you have to try to figure this out so an electrolyte a sorbet such as ethylene carbonate for post mechanism there is that just changes on migration on landscape of diffusion along the circuit so reduces in the beautiful area we have an ethylene carbonate absorber surface how go water deficit be done MD MD EFT have initial indeed and they found a slightly different mechanism action of water actually could carry the linear how the particle just a very short distance less than Amana be there and then transport it along the surface of course this is it is EFD and IFD in terms of liquid it's not that well at least in this field is not that well developed so i I take these explanations with a grain of salt but they have both seen so that's a that's a big question so I would imagine that if you ring if you have green boundaries in within this direction the vault the futures are ready pretty slow so it probably can't be much slower that other if you have green pouches in the other direction that's up I don't I haven't done the study but I would hypothesize that because the surface diffusion we need an absorbing you have to be surface diffusion so you have a grain boundary it basically acts perhaps like a surface but without a molecular orbit so it's also extremely slow as I've shown you [Applause] [Music]

Channel: Chueh Group

Views: 102

Rating: 5 out of 5

Keywords:

Id: vRJkfHzGI4Y

Length: 61min 42sec (3702 seconds)

Published: Mon Nov 23 2020