EESN | Electricity Storage at Scale: High Added Value
SARAH SIMON: Good afternoon, everyone. Welcome to the MIT Alumni Energy Environment and Sustainability Network's monthly webinar from May. Today we have a topic of electricity storage at scale, high value added, with two excellent speakers, who Ramon will introduce in a moment.
Although they're already on your screen. The EESN is dedicated to connecting alumni from MIT, and other people in the public, on the big issues that are going on today, setting up a clean future, clean energy, preserving our environment, and making it sustainable. Basically, life cycle kinds of questions. The webinar today is a webinar. The chat function is disabled for the attendees but you can put questions into the question and answer function on Zoom. The chat will be open, though, for panelists to provide extra information for the audience.
So you might see a little dot there, that'll give you a URL or something. And we are pretty strict about the one hour time frame. But we're going to get going pretty darn quick here.
And we welcome you. It's certainly a very big issue, this issue of battery storage. I just heard it, the phrase-- and I don't know where it came from-- that battery storage is going to be the Swiss Army knife of the electric supply system. The power system and our energy systems are changing so rapidly. We've got to do-- energy efficiency is one thing, but the battery storage is going to help us use all kinds of energy from everywhere.
The distributed energy that most of the industrialized nations' grids do not recognize yet, they're beginning to be changed and modernized. So we can do that. So with that, let me hand this over to Ramon, to introduce our speakers today. RAMON BUENO: Hi.
Good afternoon. Welcome, everyone. I'm going to introduce both of our speakers. I think it's an interesting topic.
It's a complex topic, so we're very fortunate to have them help explain this to us. So first we'll have Matt Harper in the class of '11 here. He's a former founder and President of Avalon Battery, and has been developing and producing vanadium flow batteries for heavy duty applications, for over 15 years. He's currently the Chief Commercial Officer at Invinity Energy Systems, focused on accelerating our transition to a low carbon grid.
For over 25 years, he's been doing pioneering work in electrical energy storage, wastewater treatment, hydrogen generation, and fuel cell vehicles. His holistic approach to industrial technology development has balanced the technical and operational development, with commercial and organizational lifecycle characteristics. He's the inventor of several patents related to clean energy and industrial technologies. So he'll speak first then. Dharik Mallapragada-- if I pronounce it correctly-- is a research scientist at MIT Energy Initiative, MITEI.
His current research focuses on advancing energy systems' modeling tools to study the implications of renewables, integration into the power sector, economy wide electrification, and the assessment of the emerging energy technologies. Prior to that, he spent nearly five years in the energy and petrochemicals industry, working on a range of sustainability focused research topics. Most recently at ExxonMobil Corporate Strategic Research, working on research on power systems' modeling technology life cycle assessment, and leading a research program to study energy challenges in developing countries. He has an MS and PhD from Purdue University, and a bachelor's from Indian Institute Technology in Madras. So we welcome both of you.
And without any further delay, Matt, the floor is yours. MATT HARPER: Thanks very much Ramon. I am just checking to see if everyone actually sees my slides. Great, well look, thanks so much.
Look what I want to do here today is talk a little bit about of the case for energy storage. And because-- one of the things that I often find is, there's so much conversation around energy storage going on in the marketplace right now. What we really want to do is to-- what I always think is useful-- Oh hold on I'm just going to switch-- SARAH SIMON: Presentation, there we go. Thank you.
MATT HARPER: There we are, perfect, great. I always find it useful to go back to first principles and explore a little bit why it is that we care so much about energy storage. The thing that I always go back to, when I talked to people for the first time about storage, is you think about how humanity uses resources. And we've these incredible inflection points in human evolution that have come about by how we use our resources, right? We were able to come off the plains and hide out in caves, protecting ourselves from scary animals because we were able to transport water. Later we were able to transport that water over a great distance into cities, leading to some of the inflection points and social development that came about 2,000 to 3,000 years ago.
In the last couple of years, we've seen tremendous inflection points, in terms of, how we manage energy, especially around liquid fuels, right? We're able to store a huge amount of liquid fuels so that they can be used as needed. We're able to transport those liquid fuels around the world with tremendous efficiency. We've never been able to do that time function with electricity. I often say to people, the way the electric grid operates right now, it may be the biggest, and most complex, and most amazing machine that's ever been built.
But imagine if water needed to be consumed the instant it fell from the sky. That is effectively the situation we have today with our electricity system. Now the thing is that, that hasn't mattered up until now because we've always been in a mode where we were generating electricity, either by burning things, or by splitting atoms, or by making water fall through dams. But as we move to the renewable space, and increasingly generating electricity from renewable sources, we lose that control.
Why that matters is that-- and the question is why that matters now-- is around the world, renewable sources are becoming the lowest cost source of electricity anywhere. And I think if you think-- if you look back 5 or 10 years in history, people were talking about solar or wind generation from an ecological perspective versus a sustainability perspective. But it's now unequivocal that the lowest cost way of generating electrons is from renewable sources. What that means is that, obviously, there's a huge incentive to build those sources over fuel based alternatives, both for, again, sustainability reasons, but also because of the basic economics of that. The question is, what does that do to our electricity system? And one of the ways that I like to visualize this is to think about how a local energy system is dispatched to serve demand.
And what this is-- what this is showing you-- this is a hypothetical dispatch curve from one of the energy systems domestically. And what this graph shows you is, this is essentially a decision making graph for how the energy system operator brings certain resources to bear, to be able to meet the demand that's required at any given time on the electric grid. And what you can see is that all of the renewable sources are at the far left hand side of the graph because they essentially have zero marginal cost of production, right? People are bidding those resources into the electricity system, essentially zero price, because they know they're going to get the market clearing price anyway. They want to be sure that they're the first resources that are dispatched. As you go off to the right hand side of this graph, you're looking at the more expensive variable sources, the peaking plants. The plants that come on 20 to 50 hours a year to match that absolute peak demand, as and when it's needed.
The challenge is that if you think about what happens to this graph when those low cost renewable sources in the left hand side become intermittent, this entire graph will shift right and left, as that intermittency manifests itself. And that-- because of the relative elasticity of price on the right hand side of this grid, that means that you end up with these tremendous fluctuations in electricity prices. That is a huge challenge and one that, not only from a technical perspective, in terms of how we ensure resource adequacy, but also from the perspective of, how do we make sure that we've got a grid that functions in an economically viable way, is becoming a very large problem. So where do we see that problem manifesting itself in the electric grid system? This is a screenshot showing electricity prices, recently, from a couple of years ago in April, in California. What you're seeing here is prices that, for a couple of hours a day, are at 50 times the average cost, again, because of that fluctuation driven by intermittency. Similarly, you've got negative pricing on the other side of it.
Because once you have those intermittent sources ramped up in generating, when all of a sudden a whole bunch of the solar comes back online, you start to get negative prices, you need to do something with that electricity. And the grid operator will actually pay people to absorb that electricity, to maintain the balance of voltage and frequency. We start to see-- we're seeing this in electricity systems all over the world. This is a view, more recently, from the California spot prices for January last year. You see four or five hours a day where the spot prices are, in fact, negative.
California is paying people to take energy away from their grid system, in order to maintain balance. We see this in-- starting to crop up in a lot of jurisdictions that have a huge amount of renewable penetration, for example, Europe, a very similar situation. You see negative pricing regularly starting to happen. We've seen this, especially over the last year when there was decreases in electricity demand based around the pandemic, springtime of last year.
Finally, just by way of example, we do-- my business does a lot of work in Australia. And you see very, very similar situations in Australia, right? The amount of renewable generation through wind and yellow, which are the yellow sections of that graph at the top right, are in excess of the absolute demand in a particular region. That drives, not only instability and inefficiencies in how the generation happens, but it drives negative pricing that can be very challenging to manage for the grid operators perspective. So with that as sort of a problem space, the question is, How do flow batteries fit in on that storage landscape? Well when we think about how we describe the overall storage landscape, we usually think about it on a series of axes that looks like this. We think about the number of cycles per year where a battery needs to charge and discharge, and then the number of hours that the battery needs to charge and discharge for, every time it's called into play.
Within those two axes, you can plot a number of different applications where storage can, in theory, come and make a material difference to how we operate our grid. Everything from the operating reserves, which are the very, very frequent but short duration services that, essentially, make sure that, over a very short period of time, there's sufficient generation to cover demand. Also maintaining the voltage and the frequency on the grid, within the specified range, falls into that far left category. In the middle of this graph, you see a lot of shifting of renewable power, right? How do we take wind and deliver it through the course of the day. How do we take solar and use that to generate baseload power overnight. And then finally, all the way to the right of this graph, we talk about resilience.
We talk about, can you take renewable power and use that to support grid outages or regions in the world where you only have 18 hours of good power on the grid, over the course of the day. Can you use storage to mitigate some of those gaps. And the answer is you absolutely can.
What we see is that there are-- if you look at where storage has played on the electric grid to date, really, lithium ion batteries are tremendously capable at serving the bottom left hand side of this graph, peak or replacement. Making sure that the grid, the elements or the batteries that are serving that right side of that dispatch grid, dispatching maybe 100 times a year for a couple of hours at a time to make sure that the absolute peak requirements of the grid's demand are met. That's a phenomenal application for lithium-ion batteries.
The challenge, of course, is anyone who has charged and discharged their cell phone every night, for a couple of years, will recognize is that lithium ion batteries are limited in their ability to cycle indefinitely, right? After a couple hundred or a small number of 1,000 cycles, they're only providing a portion of what they were doing on day one. That's where we step in. Our battery is-- the vanadium flow batteries that we make are designed around either very, very long storage durations or very, very high cycle counts because the inherent in a flow battery is that you don't see the kind of cycle based degradation that you see in a conventional battery.
People often ask us, what does that mean, in terms of your competitive position? Are you trying to displace lithium? The graph we always love to-- the graphic we always show is this. Both of these devices that you see on screen here are devices with internal combustion engines and four wheels, right? They serve a very, very different purpose. And although lithium has enjoyed tremendous commercial success, in terms of serving that one portion of the storage landscape, our view is that as storage evolves, as a part of our electric grid, that there will be a split in terms of the different applications that different technologies can serve and how those benefits can be delivered. In terms of what our flow battery is and what it does, the six benefits that we usually talk about are fundamentally based around that lack of degradation.
We don't see a degradation in the performance of the battery with a number of different cycles. That means we can hit very, very high utilization. We want to drive these batteries as hard as possible, as many hours per day.
What that does is it delivers very compelling economics. In terms of the absolute cost of delivering a megawatt hour out of a battery, over its life, is quite a bit lower than comparable technologies. One of the misnomers about flow batteries is that we can respond very, very quickly. Even though we're pumping a liquid electrolyte, in order to have the charge and discharge reaction happen, it's still an electrochemical device where we can respond to changes in demand, within milliseconds. Finally, we have some advantages, in terms of the sustainability of materials.
Most everything in our battery is fundamentally very recyclable. The liquid electrolyte that stores the energy is infinitely reusable because, as I said, it doesn't degrade. The rest of the system, 95% of it could literally go in your curbside recycling bin. It's very, very conventional materials that we work with. What does that mean in terms of operation? Going back to this point about a lack of degradation, well this is just a quick graph that drives the point home a little bit, in terms of, what you would get out of one of our 220 kilowatt hour modules, by comparison with a lithium battery of similar capability.
Again, it's because of that-- not having a degradation capacity, but also being able to deliver that non degraded capacity over 25 years. We're able to deliver a lot more kilowatt hours out of a single unit than other solutions. Ramon, you mentioned at the top of-- actually I think it was Ramon, maybe it was you Sara-- talked about storage as the Swiss Army knife for the grid. And I think it's absolutely true. What's interesting about that, though, is to extend the analogy-- and hopefully this isn't-- I'm doing this on the fly so it's not too much of a stretch.
But with a lithium ion battery, or with any battery that degrades every time you cycle it, think about, that's like using the knife on your Swiss Army knife and having every other tool inside that device dull, as you're using it. The way that-- one of the best things about having a battery storage system that doesn't degrade with cycle life, is that you can use it for multiple applications in parallel. And this is a dispatch profile curve from one of the projects that we built recently in South Australia. And essentially, what we're doing here is, not only are we taking solar generation in the middle of the day and dispatching it into the evening when it's most valuable, but we're also dispatching into the morning period based on energy absorbed overnight.
And we're also, 24 hours a day, addressing the frequency and voltage fluctuations on the electric grid. So really stacking those three different applications on top of one another. Using different tools out of that Swiss Army knife all at the same time, all in one project. And that, in our view, is how you get the most value and most benefit out of one of these batteries, over its lifetime.
Just a quick look inside one of our batteries. We talk about this as a flow battery. That's literally because we are flowing a liquid electrolyte that is stored in two big electrolyte tanks. We're flowing that through what we call our cell stacks, which is where that charge and discharge reaction actually takes place. That separation of energy storage in the electrolyte tanks, and power generation in the cell stacks, allows us to really manage that charge discharge reaction very effectively. It means that we can expand those tanks at very, very low incremental cost.
If we want to go from 4, to 8, to 12 hours of storage duration, as an example. It allows us to manage the thermal characteristics of managing that charge and discharge reaction, again, to promote that very, very long, very durable lifecycle of one of these devices. People often ask us, what the heck is vanadium? Where you've probably seen it before is in, if you've got a box of wrenches in your garage somewhere, Chrome vanadium is one of the materials that's used to strengthen hand tools. It is one of the most abundant elements on Earth. It's the 13th most common element in the Earth's crust, more available than copper, typically used in steel strengthening. And so, although it's available all around the world, where it's generally produced is in close proximity to steel production.
So that's the technology in general. Just to talk, I'll give you guys a really quick overview about Infinity. Our view is that the renewable shift is going to stall without energy storage, for all the reasons we talked about. Lithium-ion batteries are not going to meet all of the future needs, some but not all.
And that we deliver the alternative to that solution. You know we were established last year through a merger of two companies, redT Energy and Avalon Battery. Avalon was the company that I was the founder of, back in 2013. And we've got all of the trappings that you would expect of a technology company like ours, that are really operating all around the world to, deliver these projects to our customers today.
And I think I will leave it there. SARAH SIMON: That's great. Thank you very much, Matt. That's super.
Good introduction, and we've got a bunch of questions that we will get to after we have our second speaker. So at this point, I'd like to introduce Dharik Mallapragada, and have him talk to us about some of the studies and things going on at the MIT Energy Initiative. DHARIK MALLAPRAGADA: Thank you, Sarah. Let me just get my screen up. Let me know if you can see my screen. SARAH SIMON: You're good.
DHARIK MALLAPRAGADA: So my name is Dharik Mallapragada. I'm a Research Scientist at the MIT Energy initiative. And following that great introduction from Matt, I'd like to tell you a little bit about the research we're doing at MITEI. To think about the long run value of energy storage as we think about deep decarbonization of, not just electricity systems, but more broadly energy systems.
Everybody has their own favorite analogy. This is one of my favorite charts, from my colleague Patrick Brown, who looked at the variability of renewables availability and correlate that with the load variability that you would see. And you can do this for any region.
But he looked at this for Texas in 2015. And this, I think, illustrates the case for energy storage. Which is, I think, very well made by Matt in his earlier remarks. Which is that you see, as you have systems where you are trying to meet just-in-time delivery of electricity to manage supply and demand variations, as you're adding more variability on the supply side, you end up with variations that occur over multiple timescales.
And so with solar, for example, you see variations that occur through the course of a day, where solar peaks. And then you have declining solar towards sunset and sunrise. With wind, when you start to see these sort of long periods of variability, in this case in January, and you also get periods of cloudy days that may be correlated with that, that could lead to even longer periods of variability. So this kind of makes the case for why energy storage is particularly interesting. And we are trying to think about the role for energy storage from this multidimensional challenge.
That we see here, in some ways, an opportunity for a variety of energy storage technologies to be deployed. So going with the Swiss Army knife analogy, there's multiple different use cases for energy storage. And this is a report from the EIA from about two years back, looking at the capacity of energy storage deployments in the US. And you can see that the curve is very much on an exponential growth trend. And then they're broken down here by regions, as obviously you see a lot of activity in California. But I'd like to focus on these applications here, where you're starting to see many different applications exist for energy storage.
But as the cost of batteries come down, particularly lithium-ion batteries but also other technologies, you're increasingly shifting, in terms of the timescale, over which energy storage is becoming relevant. So frequency regulations is in some ways the shortest timescale, in the power system context, over which energy storage would be relevant. But now as you have cheaper costs of just storing energy, resulting from cheaper battery storage costs, you're able to think about applications that are increasingly moving into the space of providing energy, rather than just providing power.
And so some examples that I've listed out here, again this is a-- some of these news stories are a bit dated but they kind of give you an indication of the scale of deployments that are being contemplated. So one point I wanted to highlight was, energy storage comes in many forms and introduces complexity in grid operations through a couple of different ways. And so let me just pull through all of the animation on the slide.
Somehow it's-- Yep. So there's first the aspect that all energy storage does not look the same. So Matt kind of described very well the differences between traditional lithium-ion batteries, which are closed systems where you have limited degrees of freedom with respect to energy capacity and power capacity. Versus flow systems that have different degrees of freedom here.
Versus, you could think about systems that look more like hydrogen storage, where you have potentially the option to independently size the electrolyzer, the hydrogen storage system, as well as the fuel cell, if that turns out to be economical. Versus other systems that are more geographically constrained, which would include compressed air energy storage or pumped hydro storage, which is the dominant form of energy storage today. So there's a diversity here. And this diversity is coupled with the fact that many of these technologies, not all, can also be deployed at different points in the electricity supply chain. Starting from the grid scale storage applications, which is where most of my presentation will focus on.
But also end user applications closer to customers, both residential and commercial customers, where there are a number of considerations by distributed energy storage might be very valuable. Finally the key element of energy storage, that introduces complexity into the grid, is the fact that it couples operations from one hour to the other. And this is illustrated in this example chart of a dispatch for a system, for the power system in India, where if you have energy storage charging up during the middle of the day, when you have high solar, and discharging late at night, you're inherently linking the dispatch of the system across the entire day. And this linking creates challenges when you have uncertainties with respect to load, as well as the availability of resources. And so this complexity needs to be thought through, as we think about the assessment of energy storage technologies and their value, and where they would be most beneficial. So in our group we have been trying to think about these questions from the perspective of trying to model what future grids would look like and what are the cost optimal pathways for decarbonizing the energy of the electricity system, and more broadly the energy system as a result.
So as I mentioned earlier, variability is kind of critical to incorporate in these types of assessments. And so our modeling toolkit builds on our understanding of variability and resources, primarily renewables, but also variability in demand, coupled with all of the complexity that one needs to account for with respect to the operational detail, to capture the value of resources like energy storage. And then electricity markets come also in many different forms and have different regulations around how costs are covered.
And so we can represent many of those features in our model as well. So in today's presentation, I want to give you two quick case studies that highlight the approaches that these types of system modeling tools can be used for, to understand the value of energy storage. So one of this is a paper that came out last year, looking at the system value of battery storage in the bulk power system, where we are particularly focused on storage technologies represented by lithium-ion type storage.
And we wanted to look at the various services that energy storage could provide, going from energy network referral, which is operating reserves, as well as capacity deferral. And looking at the stacked value that energy storage could recover from participating in all of these markets, recognizing that some of these markets might be coincident, and may not be available at the same time. And so understanding the value side of the equation will help us then understand where the costs are today, and where the costs are likely to be in the future, and how much value that energy storage can provide for the system. So we ran our case studies for two different types of regions, one looking at conditions that are similar to the Northeast and other looking at conditions similar to the Texas region. And what I'm showing you here in this chart is looking at the incremental value of energy storage for different levels of renewables penetration in the system, going from 40%, 50%, 60% of annual generation, for different levels of storage deployment as a percentage of peak demand. So there are a couple of interesting results that we formed from this study, which was related to the fact that increasing energy storage reduces the incremental marginal value of these resources.
So they sort of compete with one another. But at the same time increasing renewables penetration is strongly correlated with increasing value for energy storage. And then thinking about these valuations, one can then compare those valuations against what one thinks are the costs of these resources, to get a sense of what's the cost optimal deployment of energy storage in these types of systems. Another interesting piece of analysis was that we were able to decompose this value into the various contributors. And when we break down these value contributors, what we see is that network deferrals, as well as capacity deferral, namely in reducing the amount of renewables that you would need to deploy to get to a particular renewable generation goal, as well as deferral of thermal capacity by having the availability of storage to provide generation during times of low renewables availability, those three components are among the major drivers for value.
And so the arbitrage value, implicitly, is tied to the capacity deferral value. And so how this value can be monetized in electricity markets today is an open question, in terms of how capacity markets quantify the value of energy storage. And perhaps there are ways in quantifying that. But this analysis points to the capacity substitution value as being the dominant driver for energy storage.
So sort of building on that team, one of the questions that has become a hotly debated issue, is around the ability of storage to displace gas generation. And so I like to start out by saying that storage and gas generation are very different entities on the power system. One is a generation resource. One is an energy storage resource.
But in the context of these experiments, one can start to look at how the storage configurations, and the system configurations, influence the displacement of natural gas generation in the system. So what you see is that systems with storage systems which have longer durations tend to get larger natural gas displacement. All else remaining equal, and you start to see similar effects that are available irrespective of the resource variability.
You also start to see some marginal decline in substitution value, as you start to increase renewables penetration, which is indicating that you need longer and longer durations to get rid of the last units of gas in the system, presumably. So one of the other nuggets of this analysis, which is now also being considered as part of the future of storage study that the MIT Energy Initiative is undertaking as part of a broader look at energy storage, is to look at the role for demand flexibility, which is often cited as a mechanism for balancing renewables penetration. And so demand flexibility, which could be things like incentivizing charging of electric cars at certain times of the day, can affect the value for energy storage in these systems. But they're mostly focusing on short duration storage applications, where energy storage is basically being used for intraday energy shifting. So the question really becomes-- so we have talked about opportunities for energy storage with an intraday shifting and opportunities for energy storage that are looking at durations in the range of under 12 hours. But as we start to think about higher and higher renewables penetration, the question really becomes, what are the technology space.
And what are the opportunities for longer duration storage technologies. So I'd like to just point your attention to this particular chart from one of the ARPA-E funded programs focused on long duration storage where you can look at, even with most optimistic lithium-ion storage cost assumptions, the duration at rated power is occurring at a very high cost and plateaus at a relatively high cost. So one really needs to be thinking about much cheaper energy capital cost technologies, to be able to get the desired duration, to be able to be much beyond the daily cycling element.
So in our group we have been thinking about, what are the value for some of these long duration technologies in deeply decarbonized electricity systems and also understanding, what are the most important design attributes. As I mentioned earlier, there are a number of different design attributes for each of these technologies. And systems modeling at an early stage can help to prioritize what are the most important design attributes. So we recently published a paper, in collaboration with Professor Jesse Jenkins group at Princeton University, where we looked at the design space for long duration storage and energy storage in decarbonized power systems. And when we define the design space, we define the design space based on five broad parameters, looking at the capital costs of the various components, as well as the charging and discharging efficiencies.
You can take the five-dimensional view of energy storage technologies. We combed the entire literature that's available on this topic and then came up with these broad, I would say broad-strokes regional classifications on where different energy storage technologies lie. And I will note that this is an evolving target.
So this is not meant to be definitive. But it's meant to be illustrative of thinking about these storage technologies on the axis of efficiency, energy capital costs, and power costs. So once we have this mapping, we can then look at this mapping and its relative value, in the context of a decarbonized power system.
And so the way we define a decarbonized power system is to look at systems that have varying types of demand, and renewables availability and costs, as well as competing technologies that are not storage based but also meet low-carbon generation requirements. So in the next few slides I just want to quickly give you a sense for some of the insights that we got from this work. So this is a chart that's kind of demonstrating, as a function of roundtrip efficiency, the power costs on the x-axis and the energy capital costs in each panel, the system cost reduction achieved by having long-duration storage deployment, based on those five attributes that I mentioned.
And so the heat map here basically points to, as you go from right to left on this chart, as you're decreasing the energy capital costs, you see that the system cost reductions that are achievable by long-duration storage technologies tend to become greater and greater, approaching 45% to 50%, in some cases. And so this is kind of highlighting how critical energy capital costs are for energy storage technologies. But also some guidance on what the key thresholds might be, such as $20 per kilowatt hour seems like a key inflection point for these types of applications. Another important parameter becomes discharge efficiency.
So as you look at-- here we're throwing roundtrip efficiency. But if you think about just the discharge efficiency component, increasing the roundtrip efficiency as a result of increasing the discharge efficiency, also increases the value of these resources significantly. And the lines that I'm showing you here, basically indicate the overlapping between our analysis, relative to where we think technologies lie today. Another aspect that always has been brought up with respect to energy storage technologies is, we are thinking about decarbonization of electricity. While at the same time, trying to expand the use of electricity for a number of different uses.
So these include transportation. These include residential commercial heating. And so one of the questions we had, was to try and understand, how does electrification, particularly of energy uses in cold climates where potentially electrification of heating could be significant, added to total demand, impact the overall design space long duration storage technologies? And so what we find here is that because electrification of heating systems results in winter peaking electricity demand, the design space shifts towards lower energy capital costs, in terms of the threshold energy capital cost requirements. So compared to the previous you see, generally speaking, for the same design point, you would get lower value, based simply on the fact that the electricity load profile is more spaced out and includes variations from winter peaking heating systems. Finally, the very last slide talks about, basically, thinking about duration as a metric for energy storage technology. So I'd like to clarify how we define duration here because duration tends to be defined in very different ways.
So in our assessment, we are defining duration as the deliverable energy, accounting for the discharge efficiency, divided by the discharge power capacity. So it is the duration at the rated power of the long-duration storage device. And so what you see is that, for the regions where it's most valuable, which was $20 kilowatt hour and below, perhaps above 40-50% roundtrip efficiency, you start to get a sense of what other kinds of durations where energy storage would add the maximum value to the system. So there are three broad regimes I'd point out. So at the low end, if you have higher energy capital costs, you're talking about hourly durations, so durations on the order of hours.
If you move into the lower energy capital cost territories, then durations are more in the daily. And then if you move to the very low energy capital cost scenarios, then durations end up being more on the order of weeks. So this gives you a sense of using these kind of design space approaches to understand and evaluate the role long duration storage technologies.
So with that I'll stop. There's a bunch of insights here that I'm sure you can read later on. But I'm happy to turn it back to Sarah and Ramon for questions. SARAH SIMON: Thank you very much Dharik. Yes we are going to be posting these slides, also a recording of the session today.
But that'll take a little longer. We have had a lot of really good questions because we're kind of tickling the tech questions that MIT alums like to ask. There are a lot of co-benefits and co-costs of the different ways we make electricity. And I would like to-- we have a lot of questions about the material inputs, which we will get to. I would like to ask first, though, about the safety of the different kinds of systems, especially the flow battery ones. Either of you could answer.
Certainly it's directed at the Invinity system. MATT HARPER: Sure, yeah. Happy to talk about that. We like to quip that our battery is more likely to put out a fire, than start one because of its liquid nature.
We do all sorts of testing, both in our lab and with some of the National labs, to make sure that we comply with standards, like UL 9540A, which is the quintessential standard for non flammability and battery systems. For us that makes a huge difference. Excuse me Not only does it keep ancillary costs low, we don't need fire suppression.
We don't need active fire monitoring on our systems. But it also means that in installations at close proximity to critical infrastructure, close proximity to things like schools, the siting and install requirements for the batteries-- or the vanadium flow batteries, are a lot easier. RAMON BUENO: Yeah, I mean, some of the other questions that were asked had to do with-- aside from vanadium-- what other elements are needed and how scarce they are, or what countries are available, environmental impacts, and that sort of thing.
So I think that's sort of a general question that I think applies to all storage technologies. Whatever you can offer. SARAH SIMON: So let's start with the very specific ones about hydrogen. People are-- Jim Papadopoulos and Claude Gerstle have both asked about how flow batteries might relate to hydrogen storage, because we're trying to talk about storing a lot of energy.
A cell phone has, what, a 10 to the minus 3 kilowatt hour storage. A car has 10 to the first kilowatt hour storage in the battery. But here we're talking about two megawatts of storage. And is that something which hydrogen fuel holds as much energy. How do they compare on the hydrogen level? Are there benefits, big benefits? MATT HARPER: I'm happy to start it in the dark. Maybe I'll pass this to-- RAMON BUENO: Sure.
MATT HARPER: Look, I spent the first 10 years of my career in hydrogen and fuel cells, so can speak a little bit to it. The truth is that if we were to take that storage landscape chart that I showed and extend it all the way up to the right, there are a number of emerging situations where very, very long duration storage makes a ton of sense. When you are trying to take a tremendous amount of wind energy generated in the Midwest in the fall and use that to heat homes in the Northeast through the winter, that's a kind of storage that electrochemical systems are not particularly good at serving.
You can think about the relative cost of the power conversion and the relative cost of the storage medium. And when you're generating hydrogen, and then using that hydrogen as an energy carrier, the storage medium itself is effectively-- can be extremely inexpensive. Those are the kind of characteristics that you would look for in a technology that could serve those, not just hours of storage, but weeks to months of storage. And I think there's a lot of work to be done in that space. But hydrogen certainly has the potential to step in to some of those kinds of applications. DHARIK MALLAPRAGADA: That, I think, covers most of the stuff with respect to hydrogen's use as a storage resource for the grid.
The point I'd also mention is hydrogen is also a fuel and energy carrier. And it has values outside electricity that will impact its value, very likely, in the electricity system. So if you think about all of the difficult to electrify sectors that we have in the economy today, there's growing interest to think about the role for hydrogen. And in some ways
the volumes that might be used for hydrogen in that space might be much, much higher than the volumes of hydrogen that are used for the grid. And might just come along for the ride in some ways, from the same set of assets. So our assessment, we think about hydrogen storage in the framework that I shared.
You can think about it just as a storage technology, with all of those different components and the attributes that Matt described. But I think that would only be a very limited view of hydrogen. What you really need to do is expand the envelope and say, where else can hydrogen be used? And what does that do for the economics of hydrogen for the grid? SARAH SIMON: Let me, let me-- I'm sorry. MATT HARPER: Sorry, go ahead, Sarah.
SARAH SIMON: I was just going to ask the very specific questions, just to make sure we've covered people's things. Mr Papadopoulos, or professor Papadopoulos, said he was seeking an opinion on H2 cavern storage, how it drives fuel cells. What kind of CAPEX dollars per kilowatt hour would it take to make a big difference? And then we also had a question from Claude Gerstle. Which I can't find again.
But it was about, how are the costs of hydrogen compared with this battery storage? And again, I think he was thinking strictly of electricity. Is it something that is interchangeable or the batteries a clear, better choice? MATT HARPER: So I haven't run the absolute numbers on what does and doesn't make sense in hydrogen storage. But what I will say is that, relatively speaking, compared with electrochemical storage, the cost of energy conversion-- going from chemical or some other form of energy to electricity-- is comparatively high. The cost, especially if you're using underground caverns, of storing that storage medium is comparatively low. And so what that argues for is very, very long duration storage.
Because again, you're in effect taking advantage of the relative advantages of that energy and storage side of the equation. DHARIK MALLAPRAGADA: And just to add to that, the discharge efficiency, as I mentioned in some of our findings, is a key parameter. So you lose a lot going from electricity to hydrogen, and then back to electricity, relative to some of the electrochemical technologies. And so there is a, I think-- there's these multiple timescales, that when you have to think about the electricity system and different storage technologies, may have a more dominant role.
And so hydrogen, as Matt highlighted, has more on the long duration, the really long duration storage realm, depending on how you define it. RAMON BUENO: There's a question about, have you looked at flow-battery applications in railroad locomotives. And if so, what point has this work reached. MATT HARPER: So we've not.
We look primarily at stationary applications. One of the downside-- could be a downside or it can be a positive-- but one of the differences between our technology and lithium-ion is that we're comparatively heavier. You're not going to see a vanadium flow battery in a car anytime soon.
Now there are people looking at VFBs for inland shipping applications, especially in Europe where they have very rigid controls on emissions from transportation means. I can see possibly where trains could be an application, as well. But it's not been an active conversation with us, to date. SARAH SIMON: OK we have several questions on the issue of vanadium and other inputs.
You had mentioned that vanadium is common everywhere around the world. Of course it is a very toxic material. It used to come out of oil power plants that we got oil from Argentina. It's really all over the place. What other kinds of inputs and life cycle material issues about sustainability for these flow batteries, and then in the whole system-- are batteries-- we've heard a lot about lithium batteries being limited sources of extraction and we can't recycle them well. And so what are some of the life cycle material questions about flow batteries and then batteries in general.
MATT HARPER: Sure. So, first of all, I'll push back a little bit on the vanadium toxicity point. You're absolutely right that the fly ash and petcoke that comes out of heavy oil refining, heavy oil burning, is a nasty material.
It's not necessarily the vanadium itself that is problematic. It's the overall waste that contains a lot of vanadium. In our view, that's actually a huge opportunity, right? A lot of the vanadium sources that we look at are taking that spent petroleum coke, or fly ash, and reprocessing it to extract the vanadium. And therefore, not only are you getting rid of a hazardous waste problem, but you're also gaining an effective and useful material from that. Part of the reason why the vanadium-- There's a lot of different flow battery chemistries out there. One of the reasons the vanadium system is so effective at maintaining that very, very long duration charge and discharge capability, without degrading, is because both the positive and the negative couple inside the battery use vanadium ions, right? Both the positive and negative electrolyte are, in effect, exactly the same material.
It's just that, on the negative side you're switching between a plus two and a plus three ionic state. And on the positive, it's between a plus four and a plus five ionic state. What that means is, any crossover between those two doesn't degrade the battery in the long run.
That's how we get 25 to 30 years of service. But it also means that it is exclusively that vanadium material that is the working element inside the system. Aside from that, vanadium is suspended in sulfuric acid, in essentially battery acid. The same as what would be in your car battery. The rest of the product is like-- I think I said earlier-- but 95% by weight is polyethylene, steel, aluminum, copper, really just common, common materials that can be completely recycled inside existing municipal or industrial recycling programs. RAMON BUENO: One question, Dharik, in your work, you mentioned how you're looking at these complex combinations of scenarios to try to identify, within the whole range of designs, possibilities where innovation is best, or efforts at innovation are best targeted, or funding whatever.
I wonder if both of you can comment on, from the-- not just the technology side, but from the other side, the regulation and the other things that affect what actually gets recognized as value in a tariff setting, and all that kind of stuff. Where do you see storage needing greater recognition, because of this rapidly changing set of values? DHARIK MALLAPRAGADA: Yes I can start. And maybe Matt, since you're in the business world, you'll have a much more practical perspective on this. So from our perspective, what we see is that much of the value for storage in many of these applications, especially if you are thinking about high value applications that are also involving large quantities of energy storage deployment, they may not be monetizable in today's market constructs. So just to give you an example, if you think about storage is a deferral asset for transmission, in many jurisdictions in the US, that's not actually a viable alternative that's being contemplated. And storage, as in terms of being able to provide capacity, for example, is a moving target, in terms of how much capacity storage can provide to the system, as a function of the system condition.
And so the capacity market rules, basically, have to keep up, basically, with the evolving grid dynamics to be able to remunerate storage. And you're seeing movement in that front, in different parts of the country. So there is regulation that's needed, in some ways, to be able to renumerate storage for all the value that it can provide. And there is some movement in that frame. But it does not-- what we are modeling here is a very much a purist and idealistic scenario that captures the full value.
And in reality, there are still many markets that are not available for storage. MATT HARPER: Yeah look, I totally agree with you and reinforce some of those points. One of the things that we've seen-- we do a lot of our business in parallel with the solar industry. And what we've seen in solar in the last couple of years, especially in California around things like net energy metering, are situations where there's a tremendous battle between the regulators, and some of the solar companies, around what the right way of regulating these technologies and installations are. Because of that, I think that some of the regulators have taken a more proactive approach, to trying to understand what storage can do before they set the rules around it.
Which is, of course, frustrating for us. We want the rules here today so we can sell into it. But it's actually-- that measured approach is not necessarily a bad one. The thing that is beneficial to us, I think, is that we see we've seen over and over again, in specific projects, where-- I'll use the example of the project I showed you some data from in South Australia. We approached the regulator and said, look, we want to install solar plus storage.
And they said, well, there's no rules for that. We don't know how to permit you. Go back to the drawing board. And what we were able to go to them with was a series of arguments around, look what we're doing is we're taking solar energy-- you already have an excess of that energy-- we're going to move it to a more beneficial portion of the day.
At the same time, we're going to be doing regulation services. At the same time, we're going to be providing black start in that particular region of the grid. That stacking of values on top of one another-- again, different pieces of the Swiss Army knife-- is a great argument to be able to go to the regulator and say, look, this benefits a number of different portions of what you're trying to do. Let's figure out a way to make it happen. SARAH SIMON: I'd like to mention that in Massachusetts, the Clean Energy Council-- CEC, MassCEC-- is just about finishing up a round of projects to look at storage as pilot and how to develop the market. And I believe that we're kind of at the front of the pack in trying to figure out the regulatory questions, the permitting questions.
But the MassCEC has quarterly reports on these different pilot projects that they're doing to try and see how it works. Now it's not-- they're not big storage. They're mostly-- the pilots that people proposed were mostly based on lithium batteries, and power walls, and things. But it is another source for information on that. There was a very specific question here about how much storage capacity-- Luke Frechette asked, how much storage capacity would be required to match the renewable generation in 2030 or 2050? And what technologies are the best at this point for the scale of what's needed? DHARIK MALLAPRAGADA: So I love these questions because this is exactly what I would like to answer myself.
And we're working on it. But I think the one thing we know is it's very much region dependent. And as I showed in some of the examples that we looked at, the value of storage is impacted, not just by what storage is and what the technology definition is, but also by everything else that's happening on the system. So I'd be very brave if I'm going to be making some guesses. RAMON BUENO: Matt, you used the phrase black start.
One of the questions from Robert Poor was, how would a black restart work with flow batteries? MATT HARPER: Yeah, and again, it's one of-- I mean we have our batteries today operating in microgrid applications and backup power. So certainly at a commercial and industrial level, or a microgrid level, we're able to execute that today. Black start of the grid is something that is a scale that is beyond what we currently do. But certainly, fundamentally, the technology is capable of doing those kind of black start applications. SARAH SIMON: There's a question from Peter Dreher.
How much would time of day residential electricity prices curb demand during peaks and make storage unnecessary? DHARIK MALLAPRAGADA: Yeah, so I think the example that I highlighted about flexible demand is kind of alluding to that point. So you see that if you do have demand flexibility that could be enabled by just tariffs, that could be enabled by behind the meter resources, a combination of factors, you can effect the amount of grid scale storage that you would need but primarily in the intraday time frame rate. So you're not really going to be dealing with weekly peaks, or weekly needs for storage, or even multi-day needs for storage by not charging your car. But you can certainly manage the sunrise and sunset peaks that you create on the system by just deferring when you charge your car. So it can make a pretty substantial difference in that setting. SARAH SIMON: That's good.
That's great. Trevor asked whether the combination of lithium-ion plus flow batteries might work well because of the duration and the discharge characteristics. Is that possible? MATT HARPER: Absolutely, in fact, the biggest standalone storage project we're building right now is at the Energy Superhub Oxford over in the UK.
And that's a hybrid system of our flow batteries with a large lithium array. Essentially, what we do is we provide the tip of the spear, right, the very first response when that plant is called to provide power. And then only when that response exceeds a certain threshold or exceeds a certain duration, the lithium ion batteries are called into play. That means that they're cycling less frequently. and therefore-- SARAH SIMON: They're your peakers.
MATT HARPER: That's right. We're the peaking plant within the plant. SARAH SIMON: And the lithium is the peak for your vanadium flow. We have about one minute left for questions because I wanted to wrap this up. But David has asked about vanadium flow battery roundtrip efficiency and also-- Yeah let's start with that one. MATT HARPER: Sure yeah, at the cell level we're about 85%, which is slightly lower than lithium, but not materially so.
The big advantage though, is that we don't see degradation of roundtrip efficiency over time. So whereas the lithium cell might lose 20% RTE over its life, we stay roughly at that same number throughout. SARAH SIMON: This is great. Wow, well this has certainly been good discussion. I want to emphasize that we are trying, with our network, to provide a lot of platform for alums to get together.
You have the people's names. If you are an alum you can get into our directory, and write to these people directly or write back to EESN. Visit our website EESN.AlumGroup.MIT.edu. And it's been a pleasure having you folks today. This was a very interesting and lively discussion that we had. It's a very important issue coming up.
And if people want to join EESN and help us with the monthly webinars like this, or if you have an event going on in your club, your MIT club or your area that you happen to participate in, we'd be more than happy to help promote it and to connect you with other people doing the same. We have a group of ambassadors from the different clubs and regions, which you could join and get some of this information flowing about what different groups of alums are doing around the world on energy environment and sustainability. And today we often call that climate action. So I hope you'll join us again next month. RAMON BUENO: We will try to get answers to-- if the speakers are amenable to it-- to some of the questions that we never got to. And thank you for the audience.
We know that some of you repeat your attendance. What I think we're always finding very helpful is if you have suggestions for topics that you would like to-- or speakers that you would like to have us consider in our programming. SARAH SIMON: So goodbye for this month. Have a great spring. Enjoy your changing situation for our lockdowns. And thank you very much, Matt and Dharik.
Bye Now. RAMON BUENO: Thank you both.