Cost-Informed Discovery of New Battery Chemistries - Donald R Sadoway

I'm a program director with the office of corporate relations at MIT our office has a great industry liaison program and active startup exchange platforms through which we connect many of the world's most innovation centric corporations to MIT s leading edge research and innovation ecosystem low-carbon energy transition is a very important initiative in that ecosystem today as a world and definitely our industry is facing on present challenges with this global where is pandemic we at MIT are intensifying our forward-looking agent with the industry in technology development innovation and problem solving that's why the industry liaison program and MIT and you should tip are jointly offering this energy innovation web in a theory on a global basis as the third session today our focus is on energy storage systems which which as you know is the increasingly important area as renewable energy capacity grow rapidly Iran world will have several leading expert from MIT today to share their work and their our insights with us we hope through those active engagements and potential collaboration down the road we'll all be able to better navigate these difficult times for a sustainable energy future after the current crisis now I'd like to introduce our session moderator dr. Rob the stoner Rob is the deputy director for science and technology almighty and the firmly director of MIT Charter Center and he's also the faculty co-director for MIT electrical power system Center and his own researches on energy storage technology policy and the system optimization so really no better person than him to moderate the today's session so with that please welcome dr. stoner rock good to be back in zoom towers overlooking the mighty Charles River the title of our session is energy storage systems the CJ said this is a very big topic that spans everything from synthetic fuel production to pumped hydro to capacitors molten salts flywheels even cuckoo clocks of course it also includes electrochemical cells batteries that we use to store electrical energy to power our laptops and other gadgets and increasingly our cars today we're going to focus on storage devices made for the electric grid why because we're going through a process of decarbonizing our energy system getting rid of fossil fuel generation and replacing it largely with solar and wind solar and wind are getting steeper rapidly and with plenty of financial incentives and mandates we're installing them and our grid at a very high rate and decommissioning dirtier generation sources like coal plants and more expensive ones like conventional nuclear plants the problem is that the wind doesn't always blow and the Sun doesn't always shine it certainly doesn't shine at night outside the Arctic Circle and in some places like Massachusetts the spring it hardly shines at all for days weeks or months at a time this intermittent sea forces us to think about storage saving electricity generated at one time and use later seconds later hours later days later months later and possibly years later timeframe of seconds or less to provide so-called ancillary services voltage and frequency regulation mostly which are very valuable and important for the proper functioning of the system indeed the majority of storage device is now on the grid are there to provide these services but they're relatively small and some so don't really help us on a longer timeframe to deal with internet C they're also relatively expensive which doesn't matter very much because they just aren't that big a part of the system people sometimes refer to the electric grid is the largest machine ever made so when we talk about grid scale storage we're talking about large storage devices and plenty of them one example is the Hoover Dam which like any reservoir hydroelectric facility serves as this sort of energy storage device in fact the Hoover may be outfitted with pumps shortly that will return water to Lake Mead from the Colorado River turning it into a so-called pumped hydro storage system capable of holding enormous amounts of energy that can be converted rapidly into electricity by reira leasing it into the hydro or rather into the Hoover's hydro turbines there's a lot to be said for this technology pump storage is something of a cost benchmark for grid scale storage it just doesn't come any cheaper at least not yet but Hoover dams don't grow on trees and they tend not to grow in convenient places like close to cities and factories expensive transmission lines so needed to shuttle energy between them and load centers which is very often unacceptable to the many people who have to look at or live under them we need relatively compact and easy to cite energy storage for grid powered loosely by windmills solar it has to be cheap or all those cost reductions that we're seeing in solar and wind won't matter the system they're part of will become too expensive overall and our decarbonisation project will stall as a general approach batteries are very attractive candidates for many of our grid storage needs they're modular and so can be made big or small to fit a wide range of timescales for minutes to days they can also have very high energy densities that is they're compact and easy to sight and high-powered densities meaning they can release a lot of energy very quickly despite their compactness and they're getting cheaper all the time and can last for years or they have to get cheaper much cheaper two of our speakers today are MIT faculty members will describe their efforts to develop new types of batteries for the grid I'll introduce them one at a time when I invite them to speak in just a minute our third speaker is an expert on the grid itself leads a group of researchers using advanced modeling techniques to examine the role of storage quantitatively and investigate other approaches for mitigating intermittency that battery storage must compete with each speaker will take 15 minutes to make remarks and then we'll leave in another 20 minutes for questions before moving on to the next so let's go to our first speaker professor Don Sadoway the John f elliot's professor materials chemistry in the Department of materials science and engineering at MIT Don over to you I'm Donald Sadoway I'm here to talk about an approach to addressing the issues that Rob has laid out namely new battery chemistry's that have to satisfy the demanding performance requirements of the grid including cost and my thesis is cost and form discovery and I'll tell you what that means as I move along by way of introduction I always like to show this image of Earth at night but you realize this is a collage of course the earth actually never does look like this when it's when it's dark in Asia it's light in the Americas if the earth ever does look like this it's a really bad day for all of us but let's take the collage for it's a metaphorical value and I like to look at this and be reminded that electricity is tantamount to modernity everything that we take for granted in a 21st century world is predicated on the availability of electricity and in this image where you don't see you don't see the modern world where you don't see light it's either one of two conditions either nobody lives there or the police hasn't been electrified and from my perspective there's no more precious gift that we can give to those people than access to electricity and of course we want it to be sustainable electricity to support a sustainable modernity and that means [Music] sustainable methods of power generation going where there is none and also decarbonization of our modern world the key enabler as we've been told and it's not only the question of addressing the problem of intermittency so that the renewables can be fully integrated into base load generation but also that it would make today's grid more reliable and by better use of assets of generation transmission and distribution so it's it's across the board if we want to look at wholesale decarbonisation but stationary storage is very different yeah we have batteries and yeah the prices have been falling but lithium-ion was invented for handheld devices the requirements for stationary stores are very different long service lifetime not years but decades safety when you get large format batteries many cells hundreds thousands of cells in close proximity there are safety issues in terms of thermal management and power management as Rob mentioned the operations have to be from timescales of seconds to timescales of days and ultimately has to be low-cost and it has to be all of these I mean I could give you a fantastic battery that would do the top three but it would come in at a NASA price point so the ones that so hence the what I call the research paradigm shift what do I mean by that well I I contend that the classical model for research at the university doesn't work in this application and what is that classical model invent the coolest chemistry publish in the highest impact journals and you know maybe something will come of this some startup will turn this into something useful and you will drop the price point with work in the in the private sector but you're not going to hit the price point that competes with heavily subsidized deeply entrenched hydrocarbons unless you start thinking about cost on day one not day 1,001 so that changed the way I approach the problem by factoring cost into the the very early discovery stage so as I mentioned it's not battery versus battery its battery versus diesel battery versus natural gas and so we have to think very differently so what first a blank is confined chemistry to earth abundant elements don't even consider elements that are really high performing but in you know have the earth abundancy of something like a precious metals so I say if you want to make something dirt cheap you should make it out of dirt and preferably dirt that's locally sourced so that you have a secure supply chain to me it doesn't make sense to get rid of our dependence on imported petroleum to turn it into reliance on imported neodymium so we have to think about earth abundance and the other thing is think about manufacturing already the the lithium ion battery is very very complex plants cost billions of dollars and so on and so forth so think about design also at the discovery stage so that you have something that is easy to to manufacture so when I when I started on this enterprise back around old 2005-2006 prior to that I had been doing some work in lithium polymer or lithium metal and so on but I grew tired of that and moved over to grid level storage and so I came to the recognition that you have to ask the right question in other words let's let's not even use the word battery let's talk about inventing a colossal keep cheap storage device so I don't specify if I say the word battery immediately the mind starts to imagine I right circular cylinder and all of these biases creep in I just said I want something that's that's got to be big cheap and it stores energy I also disregarded the conventional wisdom I did not consult with any people in a battery field I said just take a fresh look at this and don't don't be biased because it's won't open your mind so I looked outside the field and my other area of research here at MIT was electoral chemistry as it applies to metallurgy to the electrometallurgy extraction of metals like aluminum magnesium lithium and so on and so for inspiration I look to a modern aluminum smelter and this is an image of a smelter up in Quebec from left to right it's probably about a hundred feet and from front to back it probably goes about a mile and this thing is producing aluminium a 24/7 drawing this this one was probably drawing about 400,000 amperes at four volts consumes vast quantities of electricity to produce metal from dirt and yet it does sold for less than 50 cents a pound so I looked at this thing that I said this is a miracle of modern electrometallurgy it traffic's in huge amounts of electricity it's it gives a product that's very cheap I if I could just teach this thing how to store charge and release it on demand then I know at the end I'm gonna have something big and cheap the classical approach is to say I have a battery it's small how do I make this thing big and I said well it's that doesn't work for me so this is the genesis of the the idea and so out of this came to liquid metal battery the three liquid layers a top layer of a low density metal a bottom layer of a high density metal and in between a molten salt so those bottom two layers are cut and pasted from an aluminum smelter then by getting rid of the gas evolving electrode on top putting a different liquid metal on in its place now we have the liquid metal battery and this is representative I mean we started with magnesium and antimony but today we have a variety of other chemistry's but just for argument's sake let me let me show you how this thing works the museum wants to go from the top to the bottom to alloy with the animo knee but it can't because the liquid metals are insoluble in the molten salt and molten salts insoluble in the liquid metal so magnesium metal turns into magnesium ion the iron traverses the electrolyte the orange zone and then becomes a neutral and alloys with an ammonia at the bottom so the top layer gets thinner the bottom layer gets thicker and hence the battery discharges and then to recharge the battery you force current through and drive the magnesium out of the antimony back to the top and you're essentially making this into an electro refinery so we know how to run electro refineries and they run for decades so I looked at that and said this thing might have a chance and for my team this is my team in the summer of 2010 and of these 20 people maybe two three at most were steeped in electrochemistry I hired people that were fresh bright young people unjaded and so I don't have the team of experts I have a team of anti experts and the first years their work was horrible the rate of progress was imperceptible but after a year - they started to make progress and in year three they came with fantastic breakthroughs and if you take a look at other like large institutions of full of battery experts they're good at optimization and incremental improvements but if you want radical innovation get a bunch of bright young people who most of whom are learning electrochemistry from me and by the way a group of 20 for three years accomplished way more than a group of three would accomplish in 20 years there's a there's a scaling factor non-linearity there and thanks to the MIT Energy Initiative I secured funding the French energy giant hotel was becoming a member of MIT Energy Initiative and through the introductions and conversations eventually they made a commitment for four million dollars back in 2009 and then the first round of arpa-e that Department of Energy in 2010 started us on the way and so with this funding we had this big group work for three years on so what's happened in the interim we've had over 1,200 cells tested many chemistry's different alloys different salt mixtures and so on and a number of them coming in well below $100 per kilowatt hour for the electrodes in the electrolyte and this is not just cooking look in 2014 we published this paper in Nature which is arguably the the premier scientific journal on the planet so the has been recognized as good science as well and then two of my students came to me and said they want to start up and start up is going to be I said I'm not interested in all this money and so on and they said you want you want to change the world they have to get this thing into Commerce so I coined the phrase science and service to society we formed the company liquid metal battery corporation changed its name to came to Ambree because we were invented in Cambridge and our Series A funding came from Bill Gates who was watching my chemistry lectures online and to tell and I'll just show you a couple of attributes of this battery this is the 1:1 chemistry lithium on top and lead antimony on the bottom four and a half years a temperature of 5000 cycles which would be tantamount to 13 years if you did to discharge once a day every day and this is the discharge capacity after this number of cycles its retaining 99% of its initial capacity and the same thing is true with a newer chemistry that we're working with right now calcium antimony again this is data after 20 months now past two years and still retaining full capacity so what's next we're continuing to look at next generation chemistry's on campus higher voltage lower temperature and lower cost and this is one example I went back to the zebra chemistry the sodium nickel chloride and figured out how to make it without the brittle fragile ceramic membrane and again publication in a top journal so I'm gonna wrap it up here as we promised we're going to keep our remarks to below 15 minutes so when I take a look at the whole field of metal molten salt batteries for these massive storage and I look at these performance requirements I would say that this checks all of the boxes and I'm very optimistic that there will be some final deployments here you might say well what's taking so long this is not like making apps for an iPhone this is heavy industry and it takes a lot of detailed work to scale from the laboratory bench to full-scale manufacturing you know with Six Sigma quality assurance etc etc it's a long journey but I think it's worth the effort and I look forward to seeing these batteries widely deployed so with that I'll turn it back to Rob who will curate the question period Thank You Don I guess I'm supposed to am I supposed to turn my video back on I'm not um there are a couple of questions that are coming in and they're very much the questions that I had in my own mind so maybe I'll ask them in my own way and please do keep coming with with your questions one question I think for our hosts at ILP is are these lectures being recorded and can they be accessed later on so all the four answers from you to let us know available after the webinar today about two days after okay great um next question relates to the cost and and you you mentioned the figure of a hundred dollars per kilowatt hour how does that compare with with other competing technologies including Pumped hydro which I cited is this sort of reference and how far do you think you'll be able to go down in a hundred dollars a kilowatt hour is sort of the classical benchmark that comes from earlier deployments of Pumped hydro if you take a look at lithium-ion when it's being deployed at massive scale you you'll see cell costs being reported below two hundred dollars per kilowatt hour but but if you take a look at for example the deployment in South Australia and you look at the total amount of money and it was spent and the total amount of capacity that is online those numbers are up at around five hundred dollars per kilowatt hours those are prohibitively expensive as for where we need to go well below $100 per kilowatt-hour depends on the frequency as well you mentioned weeks or months if you want to have a battery that's going to sit around and only be deployed once a month the price price point has to be down around $10 per kilowatt hour which is probably physically impossible to achieve that's all in because I don't care about cell costs when you're in this application you have to factor in that not only the cell but all of the interconnects the battery management system the all of the inverters and so on because that's what the capital cost is to the to the user it's a little little deceitful to quote a really really low price it's like saying I'll give you a really a fantastic price for the car and then you show up in the showroom and it doesn't have an engine you have to pay for the engine extra you want the whole thing so that you can drive it away sounds like the guy I bought my car from you you use the term stationary storage at one point is that a limitation here are these batteries limited to two cars and a follow-up question which also is just conveying is if if you were thinking of cars or trucks is what technology options are coming along there that could be similarly low cost yeah so when I started this work that led to the liquid metal battery I was purposely thinking about stationary storage and that relaxed one of the constraints because the battery isn't going to have to move and and therefore the energy per unit mass is no longer relevant either because it doesn't have to move so and and i according the phrase don't pay for attributes you don't need the fact that the liquid that the lithium-ion battery can also move in an application that is never going to move so what and so the focus on that and with the three liquid layers there's no membranes no separators you can't subject it to acceleration and deceleration arms it'll mix and so on so it it hopefully but this will give it advantages in terms of cost and and safety and in a stationary deployment now as for the the automotive applications if we want to get something that's even better than lithium-ion that is cheaper and and safer I mean lithium-ion you know it works up to a point but if you start thinking about crash worthiness and so on the story gets a little bit complicated so again I would I would turn to my arsenal of liquid metals molten salts and I have a an initiative right now in my laboratory that is looking at such chemistry and if I want to follow the same rules about think about cost on day one because we want widespread deployment of electric vehicles they have to be cheap or not not not nearly as expensive as internal combustion engines they should be cheap or really really cheap and so I looked at the periodic table and I said well what's the cheapest electric chemically active element I said aluminum it's a third most abundant element the Earth's crust the first two are oxygen and silicon and those things they are kitchens of gas and silicon's and an insulator room temperature so I said make aluminum that's the point of departure make a lumen of solid aluminum one electrode and what's what's the cheapest nonmetal that's sulfur so I put those as bookends and and develop something we have something in play right now so I think that there is room for greater improvement in automotive batteries as well but I got to get this liquid metal battery and stationary applications done first otherwise yeah one thing at a time relates to the the other class of questions that's coming in here in addition I should say to a number of offers to buy batteries from the past those one afterwards the questions relate to deployment and and how these batteries are being deployed and at what scale you know are you imagining them going if they're in stationary applications in people's basements behind the meter or out any enormous fields or it's part of industrial facilities so perhaps even linked to wind farms and solar farms physically what's what's the natural deployment so there's a plurality of options and they'll vary from location to location yeah people people have been asked if they could you know some high-net-worth individuals could they put them into their batteries on their estates or their their islands that they own and so on but we we don't think that that's the right deployment for the liquid metal battery it operates at around 500 degrees Celsius and it gets better as as the scale gets larger so there are several places as you point out it could go in tandem with the wind or solar at the point of generation or it could go nearer to the load centers it could go say at substations or it could go in the basements of skyscrapers in downtown Boston and then the other place that is attracting a good deal of interest is large enterprises that rely heavily on huge amounts of electricity and as they want to green their operations if they cannot handle a situation in which there is intermittency and if in the case of after sunset or when the wind doesn't blow if they have to switch on a diesel generator well then that damages the whole premise of we're gonna go green so I see private enterprises looking at it and then islanded community so you know some of the Hawaiian Islands have expressed interest and a place is in in Alaska for example where you have a community that's a 500 miles from a mainline has to be serviced it would be much better to pair that with wind but without storage it's not going to work so we'll see as we get nearer to releasing this into customer hands this is a specific question regarding factor and weight what really is the energy density yeah we've built we built one that is on the order of one megawatt hour with all of the it was it goes in a 1/4 of a shipping container and energy density for everything this is including the mass of the container and power electronics and the phase conversion all that sort of thing is about 70 watt hours per kilogram well lithium-ion it at the cell level is about 150 it's about half of lithium-ion but when lithium-ion is fully loaded the there there loading is actually worse than ours because we don't we don't have any cooling system lithium-ion as you know if the temperature gets up above about 65 degrees Celsius it can become dangerous with respect to fire explosion whatnot so it has very elaborate cooling system and it's active it's not air it's it's liquid being pumped and to ductwork and so on so at fully loaded lithium-ion it's more like a hundred watt hours per kilogram so there it's it's a minor difference from my perspective yes a kind of interesting physics question here which I think you'll be able to readily answer just relates to the thermal nature of these these batteries you're keeping the metal in a molten stage so they're they're pretty hot is there any thought of using thermal energy itself or in its some way benefiting from that and that fact you've got hot metal commercial through a revenue stream yeah so people that people have approached us about because we we generate the heat as the battery charges and discharges so it's the flow of current that generates Joule heat and that's what keeps the ban so imagine you have four hours of discharge eight hours of rest four hours of charge eight hours of rest you do that every day it'll maintain its temperature you have to put energy and of course the first time to melt everything but after that it becomes stable and and can sustain the temperature so the question is do you want to exploit that waste heat and you know we've looked at thermal electrics and thermal electrics would be fine but cost wise it's it's not there other people have asked could you could you somehow run heat exchangers and bring water to a boil and and make that a source of other generation and quite frankly our hands are full making the liquid metal battery work as a battery by itself and we're not going to this we can't get involved in distractions like that because it's there's just too much I don't I don't think the investors in the company would say until you make the product work as it's supposed to work please don't start putting bells and whistles on it so if there's the thermal character of the battery anyway constrain deployment one question here about whether it will work in the Arctic or in northern Canada yeah so that's a that's a happy to have that question the the interior of the battery is running at about 475 degrees Celsius but it's it's insulated because it's got to be able to run for four hours and keep that temperature above the melting point of the internal components for at least eight hours and the our belief or you know maybe several days um so the outside of the battery is is cool to the touch and so the battery doesn't care it doesn't care you can put it in and in the Arizona Sun it doesn't matter or we you can put it up in the Canadian Arctic it doesn't matter that once that battery starts running it it's fine I mean you have to insulate appropriately obviously but you know the difference between minus 20 Celsius and 475 versus plus 40 Celsius and 475 it's doesn't matter another question is about competitiveness with other electrochemical battery types we're going to hear about flow battery shortly but this one is particularly concerned about nickel zinc batteries how do you how do you fare against nickel zinc batteries to be honest I really don't pay attention to to these other battery chemistries because this is a big market and if somebody can make dick nickel zinc work and and hit the price point of certain applications I'd be happy to cheer them on because we're all aligned at decarbonizing the system but you know that the notion that well which is the best battery chemistry and then let's put all of our money on that bet I think that's a it's not a good idea because as you pointed out in your remarks Rob there are variegated electrical performance requirements on the grid that need to be met and you know the battery that powers a hearing aid doesn't doesn't power it an electric car and so I wouldn't want to say well it's only liquid metal battery and if it's not liquid metal battery it's it's it's no good I think that we're gonna find that there are better batteries for certain applications than than others government support for research in this field is the government just throwing money at you to try to get you to develop improvements and new types of low-cost batteries is it easy no no I would say that the way the government funding is going at it it tends to tends to follow trends so for example I had the really nice package there heard 2010 for several years but getting getting new funding in this area the high temperature liquid metal batteries molten salts and so on has been not successful and and I think that people are looking at this doesn't there's a huge fraction of the community that's that's really trying to rescale lithium-ion batteries and they're talking about lithium air lithium metal and so on and I would say the the portfolio should be much more diverse than it is but there was just an event about a year or two ago where they had this program for long duration storage and there were five awards made out of arpa-e one was two of them were not electrochemical one I think was a variant of compressed air and I can't remember what the other one maybe it was thermal storage but there were three made in electrochemical batteries and all three were flow batteries and I thought don't you want to place your bets on more than one number so I think that I think we could be bolder and more imaginative and and the thing that I liked at the at the beginning was arpa-e in the assessment process it was always about if successful what is the impact that was never a question of likelihood of success but now I see it it's all folded in it's that yeah but this is you know we wanted radical innovation so I proposed something that's radical to say wow that's radical I said yeah that's what it is right it's wrong what's the chances of success I said well if the chances of success are really high it's radical we just running around in circles there's my dog just tutoring us on I heard I heard the sound of mute myself and speak at the same time so policies to the world so so I guess the next question then sort of goes to the entrepreneurial journey that you've been on then is to take technology from a lab form a company and I know you've been at it now for several years developing the batteries and getting them out there where do you see the first commercial deployments taking place where or when we Oh what sort of what's the what's the initial point of entry do you think for ambras batteries in the market III think it's going to be a private deployment either by a power company someone such as on on say one of the Hawaiian Islands where they want to go with intermittent renewables or a large enterprise industrial enterprise that relies on huge amounts of electricity and wants to shift over to decarbonizing their electricity supply that's what I see it so moderate moderate scale gigantic batteries oh yes yeah a liquid metal battery I don't I don't see as working for seasonal or or for things like remote cell towers so to be like a UPS sitting there in case the power goes down these things are designed to be used and the preferably everyday used now if you have two or three days where the Sun doesn't shine you've got clouds of rain and so on or you've got some some two or three days of wind that's still we could we could stretch to that but I business of weeks and months I mean I I'm not a spending time on that I think our our final speaker Derek will probably talk a little bit about this issue of very long-term storage and which is challenging because of the very low capacity utilization that that's implied Matt Lee's assets which drives them to incredibly low cost points correct one final question I can see this is a certain amount of skepticism brewing out there in the in the Q&A list and the question is essentially this what's the catch I mean are these the argument that you make in favor of low cost is compelling the simplicity of the design is compelling are are they are they really a low risk option and guaranteed to happen or do you still face real challenges oh there are challenges the electrochemistry is is better than we had imagined we as I showed in the slide deck the capacity fade is negligible so that's all good so that it it would have long service life time it doesn't degrade the weight lithium ion lithium ion degrades just by cycling it's it's not from some malfeasance or abuse this thing this thing doesn't cuz all liquid doesn't have any memory it clears itself up but where things get difficult is when we go to scale and when we go to scale we have to solve all the manufacturing problems they're these these cells have to be manufactured sealed and there's a break between the positive and negative electrodes and this requires a high temperature seal it's got to be made of ceramic that can endure different coefficient of thermal expansion between itself and the steel case and so that's where the catch is it's in mastering all of the difficulty isn't child facing us in the in the manufacturing of something that's never been manufactured before and there's nobody to turn to I could take the smartest people in the world who can manufacture lithium-ion batteries they take one look at what we're doing and they say I can't help you it's so different you know I look out there and I see who else has disrupted the energy storage you know all the different there was a spate of new battery companies around 2010-2011 and they've all collapsed because this is a really difficult problem to solve and to do so at such a low price point so that's what the catch is it's can we make this thing work at scale all right well I think we've come to the end of our block of time for this talk Don I'd like to thank you very much all of our house this is a fascinating subject and we admire what you're doing very much good luck thank you man thank you for giving me the opportunity to to speak to this audience

Channel: MIT Corporate Relations

Views: 11,652

Keywords: Donald Sadoway, Energy, ILP webinar, Massachusetts Institute of Techonology, MIT, MIT Energy Initiative, MITei

Id: ve3eu0ak2qs

Length: 44min 42sec (2682 seconds)

Published: Mon Jun 01 2020