The Quick Summary
Standard Thermal is the company I've been leading for the last ~2 years.
The purpose of Standard Thermal is to make energy from solar PV available 24/7/365 at a price that is competitive with US natural gas.
Our technology works by storing energy as heat in the least expensive storage material available - large piles of dirt. Co-located solar PV arrays provide energy (as electricity) and are simpler and cheaper than grid-connected solar farms. Electric heating elements embedded in the dirt piles convert electricity to heat. Pipes run through the pile, and fluid flowing through them removes heat to supply the customer. The capital cost, not including the solar PV, is comparable to natural gas storage at less than $0.10/kilowatt-hour thermal and 1000x cheaper than batteries.
These are the customer archetypes we can save the most money for right now:
-
Solar developers with oversized arrays greater than 300 kilowatts and heat demand at the location. Our system can store the summer excess production for winter thermal demand.
-
Isolated energy users forced to use propane or fuel oil, typically using more than 50,000 gallons of propane per year.
The storage system can provide hundreds of megawatts of thermal demand as long as land is available (you can check land usage using our estimator). The technology requires some scale to be cost-effective, hence the minimums.
Our medium-term goal is to repower coal power plants to supply electricity during seasonal lows. Repowering means the stored heat creates steam on demand for the turbines instead of burning coal.
We have been continuously prototyping designs at our 100-kilowatt test site in Oklahoma since last fall and are nearing our modular commercial form factor. It will be built and tested in the next several months, then it will be ready for copying and pasting at customer sites.
Orca Sciences is incubating our company. We plan to spin out and close on fundraising early in 2026.
Read on if you'd like to know more!
Problem Statement
Raw solar PV direct current output has become cheaper than fossil fuels in many places in the last few years. But ultra-cheap energy from solar PV only means so much if it's not in the correct form or available when its needed. The shift in costs makes it the correct time to develop technologies that were irrelevant when fossil fuels were the cheapest source of energy everywhere.
Batteries will handle much of the daily electricity storage and firming. But several other technologies are required for solar PV to reach its full potential. One of those is a very inexpensive, reasonably efficient storage system that can hold a few months' worth of energy to flatten seasonal differences in solar production (a niche usually served by fuels). We believe thermal storage is the way to accomplish that.
I wrote a companion post called "Expanding the Marginal Energy Source," which further details how solar PV needs can become the dominant energy source.
The Standard Thermal Idea
Standard Thermal is about abundance. Eli Dourado and I published a paper titled "Energy Superabundance: How Cheap, Abundant Energy Will Shape Our Future" in 2022, which details the benefits humanity would enjoy with more access to cheap energy. Standard Thermal is one part of making that a reality.
In the Standard Thermal system, solar PV provides the energy. It is already the cheapest source of energy in much of the world. Our thermal storage is affordable and scalable enough to allow 24/7/365 supply without overbuilding the solar array.
A typical site is a factory, power plant, or town with a large earthen mound at the edge. The mound might be the size of a house for a smaller factory, and up to many football fields for a large power plant. Surrounding the earthen mound will be high-density, low-profile solar arrays.
Electricity from the solar arrays flows to heating elements in the earthen mound, building up heat. The storage temperature is 600 °C or higher. The outer mass of the mound, plus a favorable volume-to-area ratio, insulates and minimizes heat loss. Pipes embedded in the mound carry fluid that delivers heat to users.
Both heat and electricity are expensive to move. The heat storage must be near the customer, and the solar array must be near the storage to achieve real cost reductions. The design is also modular. Individual blocks of solar PV can be copied and pasted, and dozers can push more dirt. That repeatability reduces development costs and risk.
The Standard Thermal Storage System
A significant portion of human energy demand is for heat, whether that is for factories or heating buildings. Thermal storage is very efficient at delivering heat to these users. There is an efficiency penalty converting back to electricity; round-trip efficiency is 40%-45%, but sometimes the steady supply of electricity is worth it.
Harvesting sunlight with solar PV panels and storing it in a widely available material enables low-cost energy virtually everywhere on Earth.
Markets We Want to Serve
Isolated Energy Users
Some rural heat users are away from the existing energy infrastructure, and rural delivery systems can't meet their demand. They must migrate to more expensive fuels that are easier to transport, like propane or fuel oil.
Examples in this category are mining and agriculture. These customers have open land nearby and suffer from high energy costs. A solar PV array and Standard Thermal storage system can be cheaper today. Roughly 10 to 40 gigawatts of solar PV capacity might satisfy US demand, requiring more than $10 billion in investment.
From a more global perspective, most humans and their endeavors are isolated energy users. Co-located solar and cheap thermal storage can provide costs comparable to today's cheap energy. Typical steam process heat categories without access to affordable gas or coal require a few terawatts of solar to meet demand today and more in the future as the world becomes richer.
High Latitude Solar Developers
Solar arrays might produce 5x-6x more electricity in summer than in winter in high latitude regions. Batteries are too expensive to move excess electricity between seasons. But, solar panels are inexpensive, and it is common to oversize a solar array to meet basic electricity needs (lights, etc.) during non-summer months. The panels still produce electricity when it is cloudy or the sun is lower in the sky, but just not as much as in summer. The extra solar capacity covers many more days.
An obvious trade is to store the excess summer electricity from oversized arrays as heat to provide winter heating. The economics look very attractive because the electricity is “free” to utilize. And winter fuel is often expensive in these places, especially in Europe and Asia.
The concept could provide a generalized solution for seasonal solar PV variation. A town could have a solar array on the edges of town to provide electricity nearly year-round. It would fill storage with the summer excess to power its district heating system in winter. The energy demand for existing district heating systems would require 3 to 5 terawatts of solar to satisfy it (several times cumulative global solar installation).
Repowering Thermal Power Plants
Seasonal electricity storage is the sexiest (and largest) application. Powering turbines from existing coal power plants by producing steam from thermal storage is the most economical path to electricity storage. Power plants are expensive, and there are plenty of existing ones. It could take up to ~25 terawatts of solar to convert them all.
Thermal storage can make these power plants operate more efficiently. Exhaust air scrubbers with high fixed and variable operating costs aren't necessary. Thermal efficiency improves because heat isn't lost up the stack, in solids, or running scrubbers. Facilities can start faster, making them more relevant in electricity markets. The cost and operating profile emulates a natural gas power plant.
The ideal number of plant conversions is a trickier question because of the difficulty in predicting how much other generation sources, like solar PV linked with batteries, will supply.
Some countries with growing electricity demand might want facilities to run in a more "baseload" configuration in the short term. The math is simple: Is solar PV heat cheaper than coal plus coal's inefficiencies? The numbers might favor solar over coal imports, but mine-mouth power plants will take longer to cross over.
There are many ways to model the long-term question, and a recent paper I co-authored with several Stanford researchers is one entry. Most results converge around a technology like inexpensive thermal storage providing steady power on days when solar output is poor and electricity prices are higher. Thermal storage might supply around 10% of the total electricity supply in this configuration. Beating coal becomes much easier in this scenario because lower coal plant capacity factors hurt the economics.
Coal supplies around one-third of global electricity today, and electricity demand should keep growing. A substantial share of plants could undergo conversion in the most optimistic scenarios.
Why is Thermal Storage a Good Option?
Available Storage Materials
Requirements always determine what engineering choices are available. A constraining goal narrows the options.
Matching natural gas storage costs is a good constraining goal. Storing natural gas in depleted reservoirs is absurdly cheap, with a capital cost of $0.05-$0.10/kWh. For reference, batteries getting under $100/kWh is exciting, a 1000x higher cost per unit of storage. There are only a few materials that are even viable to match that cost. They are energetic chemical reactions, water, air, and dirt/rock.
Chemical storage works very well; that is what natural gas and other hydrocarbons are. They will always be favorable in some applications. But, extra steps and conversions raise the cost of producing or mining them. Cheaper options, even if they can't do everything a fuel can, would be nice.
Chemical fuels are so hard to beat because they carry so much energy per unit of mass, and the cost of that mass is very low compared to most products humans produce. By default, storing energy using less mass-efficient methods requires nearly free mass to be competitive.
Water and air are nearly free, so they are worth consideration. Unfortunately, both are forms of gravity/pressure storage, which have very low density, roughly 100x worse than thermal storage. Water and air are cheap, but managing their containers and movement is not. Air has the added issue of being compressible, so pressurizing it and recovering energy are more expensive and less efficient than water.
Thermal storage could also be attractive. A significant portion of human energy usage is in the form of heat. Conversion losses for these applications can be small. Conversion back into electricity is 40%-45%, which is not perfect, but better than many alternatives. The surface footprint is also very good, roughly an order of magnitude better than batteries because dirt piles a lot higher and closer together than shipping containers of batteries. A pile that could run a large power plant for months would look small in comparison to the cooling water reservoir.
Simple Thermal Storage Cost Analysis
There are two adjustable cost metrics: the cost of the storage material and the heat stored per unit. Most thermal storage is sensible storage, meaning heat changing the temperature of the material stores or releases heat rather than phase change or reaction. The material needs to be very inexpensive and subjected to high temperature swings to achieve acceptable $/kWh figures.
Dirt and rocks are practically free where they are, but moving, sorting, and processing the dirt adds cost very quickly. Giant mounds or embankments using nearby material cost ~$1/ton. Trucking sand or gravel costs $20-$50/ton. Refractory bricks made from purified dirt are ~$500/ton.
Dirt's thermal properties are similar to refractory bricks. It is composed of similar materials, like silicon dioxide and aluminum oxide. These oxides can handle very high temperatures exposed to air. Bricks have a more uniform structure than dirt, but piled-up dirt doesn't need fracture toughness or consistent thermal expansion. Dirt is an excellent value at one hundredth the cost per unit of storage.
Dirt stores around 20 kilowatt-hours thermal per ton for each 100 degrees Celsius of temperature change. A 400-degree Celsius swing would store 80 kilowatt-hours per ton and have a capital cost $0.013/kilowatt-hour, assuming $1/ton dirt moving costs. That is very competitive with natural gas storage, but selecting an inexpensive storage material is a tiny part of the battle. The heat addition and removal systems within the dirt mound can't cost more than $0.05-$0.08 per kilowatt-hour thermal. Standard Thermal has had to put in a lot of effort to reach costs in this range. Thermal storage systems are much more challenging to build than the material costs let on.
Engineering Cheap Storage
Beating natural gas storage costs requires several paradigm changes and adjustments to keep costs from ballooning to irrelevance.
The Challenges of Rapid Charge and Discharge
There is an anchoring to the idea that alternative energy storage technologies need to charge and discharge quickly. Rapid cycling is an incredible disadvantage for thermal storage for two reasons:
-
Heat Transfer is Slow
Materials that are inexpensive and tolerate high temperatures while exposed to air tend to be oxides that are insulators (dirt, sand, rocks, bricks, etc.). The thermal conductivity is poor, and it is difficult to withdraw heat from more than a few centimeters within minutes or hours.
A block of hot mass with dimensions in tens of meters can take years to fully cool. Dry rock geothermal electricity generation calls for giant fracture networks between two wells to overcome the heat transfer limitation.
Regenerative heat exchangers at steel mills are one of the few scaled thermal storage applications. They use special "honeycomb" bricks that maintain high surface area-to-volume ratios to handle ~25 cycles per day.
Source Hot Blast Stove Checker Bricks
Attempts to get around the heat transfer deficiency tend to be complicated. Pumping molten salts was a popular idea for a time. The Department of Energy has a concept that conveys sand through heaters and stores it in a silo until it drops through a heat exchanger to export heat. Other designs pump oil or molten salt through gravel. Several startups use graphite in an inert container to get higher thermal conductivity while preventing oxidation.
Any heat storage design must carefully consider heat transfer.
-
Temperature Cycling is Damaging
Heated materials undergo transformations. They tend to expand as they heat and contract as they cool, but there are also break points where the basic structure can change. Highly engineered or ordered materials become disordered and fall apart when exposed to thermal cycling.
Most heat storage concepts require engineered systems to achieve charge/discharge rate goals, making for a very severe techno-economic challenge.
Thankfully, there is little market reasoning for rapid cycling. We don't have to fight physics.
The Case for Slow Discharge
Charge and discharge rates on the order of months are fine.
-
Many Customers Use Energy Steadily
More than 10% of world energy consumption is industrial process heat. Almost all these facilities run 24/7 or in defined daily cycles. The energy usage in any one day compared to the cumulative storage needed to shift summer excess to winter usage is tiny. Models project 1500 megawatt-hours of storage for every 1 megawatt of output capacity in a typical US location. Charging with solar requires higher rates than satisfying demand because it delivers the same amount of energy at a lower capacity factor. The charging pace is still leisurely and benefits from electricity's ability to provide high charging temperatures.
-
Existing Winter Seasonal Storage is Slow
Winter demand is an application that already uses seasonal storage. The clearest example is natural gas. US consumption is less than supply in summer and shoulder seasons, but greater in winter. The industry stores large amounts of gas near customers for winter consumption. Many of these storage sites are depleted gas fields. They can't fill or empty faster than on a timescale of months. Adding more extraction and injection wells is expensive, and there is no need for it.
The natural gas storage seasonal cycle. Source: EIA
Thankfully, people don't use their entire winter allotment of heating fuel over 1-2 days. They use energy much more regularly, and daily usage is small compared to the total stored.
-
Batteries Solve Fast Cycling In Electricity Markets
I'm well into the 99th percentile of battery optimists, which is unusual for someone working on alternative storage technologies. The improvement in battery technology over the last twenty years has been incredible. They are already the least expensive way to add grid peaking capacity, and they will move a growing share of daily electricity generation.
People underestimate batteries. There is still significant room to drop costs in cells, packs, and construction/integration. Sub-$100/kWh installation costs are already present in much of the world and should come even to the US in time (tariffs, permitting, and the electrical code are the obstacles). These costs are transformative even in simplistic models, but the various revenue streams for batteries are also underappreciated. They can arbitrage prices across time, provide ancillary services, and arbitrage transmission constraints. The lower installation costs and multiple revenue streams mean that the daily spread between the high and low electricity prices will be much tighter, more like a few tens of dollars per megawatt-hour. That tight spread disadvantages technologies that take advantage of daily volatility with rapid charging or discharging.
Instead, markets will send signals to generation sources with low fixed costs that can provide a steady supply over days when solar, wind, or gas are constrained. Batteries will still be too expensive to arbitrage these seasonal differences effectively. Thermal electricity storage is orders of magnitude cheaper for storage, but worse at daily cycling. It can act as a complement to batteries and fill the market niche.
If we accept slower charge and discharge rates, then engineering becomes much simpler. Insulation and other factors benefit from the low thermal conductivity of insulating oxides. The thermal mass doesn't need to be highly engineered; it can be regular dirt piled up with some extraction pipes thrown in.
Spending more on electric heaters or heat transfer surface area to improve charge/discharge isn't necessary for the vast majority of applications. Dirt-cheap mass is the main advantage of thermal storage, and it is important to lean into that and take advantage of truly "seasonal" energy shifting.
Energy From the Sun
Even free storage isn't helpful if the cost of energy from the solar arrays is more than natural gas. Lazard's yearly report is the industry benchmark for solar's cost per megawatt-hour. The 2025 edition pegs the best-case US cost at $38/megawatt-hour, equivalent to more than $11/MCF of natural gas. The trading hub price for natural gas has been $3-$4/MCF in 2025 (and many years before that), exposing a significant problem for our thesis.
A closer look reveals the issue. The operating cost of a solar farm is very low at $4/megawatt-hour, but the capital cost comes in at $1150/kilowatt in the US, contributing $34/megawatt-hour. Commodity solar panels cost $80/kilowatt globally and $200-$250/kilowatt in the US, meaning there is a lot of waste (and opportunity) in today's solar farm capital cost.
The first slash at these costs comes from co-locating the solar array with the storage system. More than $300/kilowatt of cost comes from preparing electricity for export or connecting to the grid. These items include inverters, medium-voltage transformers, switchgear, substations, high-voltage transformers, power lines to the grid, and all the project management overhead to build these systems.
The next phase is to simplify the arrays themselves. Almost all US utility-scale systems use trackers that keep panels faced towards the sun throughout the day. These are expensive, add a lot of mechanical complexity, and cover more land to keep rows from shading each other. Instead, the array can use fixed racking where the panels lie nearly flat. These structures are very lightweight and simple because wind loads are lower. The total area is 2x-3x smaller than the tracking arrays because the panel spacing is tight. Operating costs even fall because there is less to break and less land to mow or pay rent on.
Our Test Site Solar Array
The financial implications of the racking change are profound. Install labor and direct racking costs fall, operating costs nearly halve. The electrical wiring decreases 50%, making wiring runs without increasing voltage much more practical. The reduced area opens up land constrained sites. The supply chain shortens dramatically because there are fewer unique parts and longer lead time items. The arrays are simple enough that an outside EPC firm isn't required, knocking off another chunk of cost. Each panel produces less electricity without trackers, but panels are relatively inexpensive, and the other benefits are well worth the penalty.
The result of these changes is that the installation cost for the solar array is the panel cost plus $100-$150/kW. A US array would be $300-$350/kWh, and the levelized cost falls to as low as $16/megawatt-hour, or $4.70/MCF of gas. An array with free market module prices would be equivalent to $3/MCF.
These costs put us within the range of US natural gas prices, and well in the money in most international markets. The consideration is what the end customer actually pays rather than the hub price. There are some large gas consumers, think of a giant ammonia plant, located near gas fields that might pay less than the trading hub price. These will take some time for economic crossover. In most of the US, customers pay 1.5x to 3x the trading hub price to cover transportation, storage, and distribution. Or power plants might be paying 100x the normal price for gas during power crunches. A fixed energy supply cost via co-located solar would be a big advantage for a converted power plant. Even with high US module prices, there is an opportunity to be competitive in many regions.
Progress and Commercialization
There was little doubt that a technology like this was technically possible. All the risk is in achieving the incredulously low costs. One of the most critical questions was how engineered the system needs to be. Pure and engineered materials can help make things predictable and reliable, but they are expensive. Ultra-low cost requires using them sparingly. Another concern is unknown costs. Typically, projects miss cost estimates not because a category is poorly estimated but because entire line items are left out. Constructing an end-to-end system became one of the overriding priorities to test these hypotheses.
So far we've:
-
Built a 100-kilowatt solar array to power testing.
We completed the array within 6 months of saying "Go," including permitting, order lead times, and construction (which was a few weeks at the end). Part of that speed was the simplicity and lack of interconnection, and part was clever (and lucky) planning. We completed all the engineering, procurement, and construction ourselves to prove its feasibility.
-
10 Months of Destructive Heater Testing
The biggest cost miracle is the heating elements. Ordering package systems from existing vendors can easily cost hundreds of dollars per kilowatt, even though the resistor material itself is $1-$2/kilowatt. Structure, insulation, controls, and assembly have astronomical costs. The extras are there because heating elements need to last years, but can literally die in one second in conditions outside their boundary. There are only a few materials that are even acceptable as resistors.
Our cost goal for the heaters is $10/kW, close to the cost of the resistor material. A goal so ambitious obviously requires designing and building our own heaters. We have large piles of dead heaters, and now know about every way one can die. Our latest testing suggests we will achieve our cost and reliability goals, and the effort to get there has been incredible.
-
Rearchitecting Solar PV Electrical Integration and Safety
Another unpleasant surprise was that the entire solar industry relies on one narrow regulatory pathway that involves inverters and AC interconnection. The equipment needed to satisfy the electrical code for an off-grid DC solar array doesn't exist at any scale without going to grid-forming inverters (we had to buy some discontinued equipment on eBay to complete the test site). Companies usually embed these systems in a grid-tied inverter, which we don't have.
Solar array safety is challenging compared to wiring in a house or factory due to the direct current and constant current nature of solar modules. We've had to design our own controllers and safety systems to have a chance at deploying a commercial project. These are usually "outsourced" to very specialized inverter manufacturers.
-
Moving Dirt
Building an inexpensive dirt pile is one thing, but embedding pipes and heaters while keeping it cheap is another. We've gone through at least a half dozen configurations before finally settling on a viable construction method.
One version of the mound.
-
Heating Dirt
Lab-scale tests of heating dirt are easy. But, heat flow in a large mound can be more complex with much longer diffusion paths, especially during the early phases (heating the first time is like making bricks in situ). Heating in a test mound has been critical for developing models that are reliable at scale.
The Team
We are now an army of three permanent employees. Jimmy Williams joined in the summer of 2024 to focus on solving the heater cost and reliability challenge. Brian Pal joined more recently to tackle final systems engineering hurdles and refine our prototypes into more robust and replicable commercial designs.
We also have a horde of capable contractors who help with construction or provide specialized engineering and regulatory services.
Working with this team has been a wonderful experience, and their abilities and extremely specific experience has shaved years and tens of millions (or hundreds of millions?) of dollars off the development process.
The Constant Pressure of Cost and First Customer Systems
Cost tries to balloon at every turn. Almost every system component has undergone at least one cost crisis, if not many, where that single component could blow the entire project budget. Having a component that costs anything to buy and install is a problem when everything has to be practically free.
Balance-of-system and soft costs are the most challenging, like the heater example. Theoretically, many things can be very cheap, but only under extremely narrow conditions. Any perturbation quickly requires a slew of mitigations that increase the cost and complexity. Rarely do these narrow windows for different subsystems align.
Working out the details has been the most extreme global optimization problem I've worked on. I've had to develop a detailed understanding of solar cells, the electrical code, unsteady state heat transfer, and many other topics in between (My blog output decreased because I had to study at night instead of writing blog posts). We now know what we need to build and will be finalizing the commercial design over the next few months.
The goal is to build the first customer systems in 2026. We are working with several potential customers to ensure our system can successfully integrate with their energy usage (designing a universal, copy-and-paste system that can integrate easily with different customer facilities is another tricky part of the global optimization). Our first system can offer savings for fuel oil, propane, and higher-priced natural gas users. Follow-on systems, especially those larger than a few megawatts, can match the sub-$0.10/kWh natural gas storage cost. Matching energy cost depends on the location and the solar panel cost.
If signing up for a system like this interests you, please contact me at austin@standardthermal.com.
Maintaining a Competitive Edge
Differentiation from Existing Thermal Storage Startups
I would not be working on this problem if I agreed with the development path of other firms in the space.
Every other serious effort is a highly engineered system that has a day to a few days' worth of heat storage. That means competing head-on with batteries, whose cost is already below that of many thermal storage hopefuls.
The Standard Thermal system has months of storage capacity, a cost per kilowatt-hour several orders of magnitude lower than highly engineered systems, and is scalable with a fraction of capital and human resources.
The Strategy
"What is the moat with a dirt pile?" is a common question. Most sites will have drainage moats, but that usually isn't the answer people are looking for.
The main differences between energy projects like this and software are the difficulty of scaling and customer stickiness. The technology itself is far from enough to achieve scale due to thickets of regulations, permits, certifications, cost pressures, and financing hurdles that add friction to growth. However, the on-site nature of the projects ensures a long-term relationship with the customer. The project size unlocks low customer acquisition costs.
There is a well-worn path for scaling infrastructure technologies. Expensive capital builds the first examples. Then there might be debt financing available to borrow against an operating project. Then the debt for new construction. And finally, someone else pays for new projects.
The reasoning behind this path is scale and capital efficiency. Developing projects is demanding, and one firm can only take on so many before the overhead and risk become crippling. Each rung up the strategy ladder reduces capital requirements and company risk.
There might be dozens of renewable energy developers with a market cap of more than a billion dollars. No single player can dominate due to the friction and barriers in each project. Each project having a pseudo-monopoly over that particular location also maintains the value of existing systems. The size of the company is related to the size of its projects - big companies have big projects. The same pattern holds in other industries, like oil and gas, where the leading companies operate the most productive fields and focus on megaprojects. Even within the shale industry, the most successful players found the most extensive high-return plays. The company with an all-encompassing vision of being in every play, Chesapeake, declared bankruptcy while its founder died an untimely death.
Another unusual feature of ultra-cheap storage is that the actual storage is a relatively small portion of the cost. But project revenue and viability are solely reliant on it. Lenders must evaluate the risk based on the storage system performance, not low-risk solar arrays or customers with good credit.
A multi-phase strategy emerges from these realities:
-
Build Initial Projects
The first projects need to validate the technology, provide some revenue, and create a track record for future financing. There is no "chasm of death" as both the off-grid solar and storage systems can scale down to a few megawatts without massive penalties. The diversity of energy users is enough that potential customers with expensive energy exist in this range.
-
Scale Up Size and Financing
Projects in the tens of megawatts have cost benefits and conserve corporate overhead, allowing faster growth. The track record of the initial projects should open up financing.
-
Push Lending Rates and Cost Down
Several trends work to make the technology viable in more locations. Solar panels are a significant portion of CAPEX, and continued cost decreases have a disproportionate impact on the levelized cost of heat delivered. Improving solar panel efficiency reduces the land required and saves money on racking and labor. Other custom components, like electrical safety systems and controllers, decrease in cost with scale. Interest rates for borrowing against projects decrease, which has an outsized impact on project breakeven.
-
Provide Value No One Else Can
The technology can only continue scaling by enabling many project development firms to deploy it. Becoming some version of an engineering, procurement, and construction (EPC) firm is a necessity. The bankability and turn-key system can create an excellent buyer surplus by decreasing the overhead and levelized cost of energy for the entire project. Offering a fair price plus friction from patents leaves little opportunity for new entrants and preserves the long-term maintenance and operations relationship.
Many of the necessary components to replicate the system require time (testing, certifications, etc.). Our system costs will continue to fall and become more bankable. That increases the free cash flow hole and time-to-market a new entrant has to weather. It will be much easier to make money by buying the thermal storage system and developing projects rather than trying to copy it.
-
Further Upsize Projects and Repeat
Larger projects enable the firm to grow without choking on overhead. These projects will almost certainly be power plant conversions or entire city district heating, which are often 10x-100x larger than typical process heat applications. The same follow-on move to EPC unlocks global scale.
Implementation of the plan requires excellent execution, but this is always the case with startups. Each rung on the ladder unlocks durable value. Falling short could easily mean being a unicorn project developer instead of being zeroed out.
Conclusion
Handling high seasonal solar variation, dealing with the last 10% of electricity generation in renewables-dominated grids without fossil fuels, and reducing the burden of energy costs in fuel-poor regions have been classified as near-impossible problems at various times.
It is rare to work on something that has the potential to be so impactful and also benefits from my passions for energy and engineering minutiae or low-level global optimization. I'm grateful for that. There is an enormous amount of work ahead to make the impossible possible, and I'm more motivated than ever to see it through and scale the technology.
Building Ultra Cheap Energy Storage for Solar PV
2025 August 18 Twitter Substack See all postsSolar still needs storage cheap enough to move energy between seasons. I have been working on a startup to do this for the last two years.
The Quick Summary
Standard Thermal is the company I've been leading for the last ~2 years.
The purpose of Standard Thermal is to make energy from solar PV available 24/7/365 at a price that is competitive with US natural gas.
Our technology works by storing energy as heat in the least expensive storage material available - large piles of dirt. Co-located solar PV arrays provide energy (as electricity) and are simpler and cheaper than grid-connected solar farms. Electric heating elements embedded in the dirt piles convert electricity to heat. Pipes run through the pile, and fluid flowing through them removes heat to supply the customer. The capital cost, not including the solar PV, is comparable to natural gas storage at less than $0.10/kilowatt-hour thermal and 1000x cheaper than batteries.
These are the customer archetypes we can save the most money for right now:
Solar developers with oversized arrays greater than 300 kilowatts and heat demand at the location. Our system can store the summer excess production for winter thermal demand.
Isolated energy users forced to use propane or fuel oil, typically using more than 50,000 gallons of propane per year.
The storage system can provide hundreds of megawatts of thermal demand as long as land is available (you can check land usage using our estimator). The technology requires some scale to be cost-effective, hence the minimums.
Our medium-term goal is to repower coal power plants to supply electricity during seasonal lows. Repowering means the stored heat creates steam on demand for the turbines instead of burning coal.
We have been continuously prototyping designs at our 100-kilowatt test site in Oklahoma since last fall and are nearing our modular commercial form factor. It will be built and tested in the next several months, then it will be ready for copying and pasting at customer sites.
Orca Sciences is incubating our company. We plan to spin out and close on fundraising early in 2026.
Read on if you'd like to know more!
Problem Statement
Raw solar PV direct current output has become cheaper than fossil fuels in many places in the last few years. But ultra-cheap energy from solar PV only means so much if it's not in the correct form or available when its needed. The shift in costs makes it the correct time to develop technologies that were irrelevant when fossil fuels were the cheapest source of energy everywhere.
Batteries will handle much of the daily electricity storage and firming. But several other technologies are required for solar PV to reach its full potential. One of those is a very inexpensive, reasonably efficient storage system that can hold a few months' worth of energy to flatten seasonal differences in solar production (a niche usually served by fuels). We believe thermal storage is the way to accomplish that.
I wrote a companion post called "Expanding the Marginal Energy Source," which further details how solar PV needs can become the dominant energy source.
The Standard Thermal Idea
Standard Thermal is about abundance. Eli Dourado and I published a paper titled "Energy Superabundance: How Cheap, Abundant Energy Will Shape Our Future" in 2022, which details the benefits humanity would enjoy with more access to cheap energy. Standard Thermal is one part of making that a reality.
In the Standard Thermal system, solar PV provides the energy. It is already the cheapest source of energy in much of the world. Our thermal storage is affordable and scalable enough to allow 24/7/365 supply without overbuilding the solar array.
A typical site is a factory, power plant, or town with a large earthen mound at the edge. The mound might be the size of a house for a smaller factory, and up to many football fields for a large power plant. Surrounding the earthen mound will be high-density, low-profile solar arrays.
Electricity from the solar arrays flows to heating elements in the earthen mound, building up heat. The storage temperature is 600 °C or higher. The outer mass of the mound, plus a favorable volume-to-area ratio, insulates and minimizes heat loss. Pipes embedded in the mound carry fluid that delivers heat to users.
Both heat and electricity are expensive to move. The heat storage must be near the customer, and the solar array must be near the storage to achieve real cost reductions. The design is also modular. Individual blocks of solar PV can be copied and pasted, and dozers can push more dirt. That repeatability reduces development costs and risk.
The Standard Thermal Storage System
A significant portion of human energy demand is for heat, whether that is for factories or heating buildings. Thermal storage is very efficient at delivering heat to these users. There is an efficiency penalty converting back to electricity; round-trip efficiency is 40%-45%, but sometimes the steady supply of electricity is worth it.
Harvesting sunlight with solar PV panels and storing it in a widely available material enables low-cost energy virtually everywhere on Earth.
Markets We Want to Serve
Isolated Energy Users
Some rural heat users are away from the existing energy infrastructure, and rural delivery systems can't meet their demand. They must migrate to more expensive fuels that are easier to transport, like propane or fuel oil.
Examples in this category are mining and agriculture. These customers have open land nearby and suffer from high energy costs. A solar PV array and Standard Thermal storage system can be cheaper today. Roughly 10 to 40 gigawatts of solar PV capacity might satisfy US demand, requiring more than $10 billion in investment.
From a more global perspective, most humans and their endeavors are isolated energy users. Co-located solar and cheap thermal storage can provide costs comparable to today's cheap energy. Typical steam process heat categories without access to affordable gas or coal require a few terawatts of solar to meet demand today and more in the future as the world becomes richer.
High Latitude Solar Developers
Solar arrays might produce 5x-6x more electricity in summer than in winter in high latitude regions. Batteries are too expensive to move excess electricity between seasons. But, solar panels are inexpensive, and it is common to oversize a solar array to meet basic electricity needs (lights, etc.) during non-summer months. The panels still produce electricity when it is cloudy or the sun is lower in the sky, but just not as much as in summer. The extra solar capacity covers many more days.
An obvious trade is to store the excess summer electricity from oversized arrays as heat to provide winter heating. The economics look very attractive because the electricity is “free” to utilize. And winter fuel is often expensive in these places, especially in Europe and Asia.
The concept could provide a generalized solution for seasonal solar PV variation. A town could have a solar array on the edges of town to provide electricity nearly year-round. It would fill storage with the summer excess to power its district heating system in winter. The energy demand for existing district heating systems would require 3 to 5 terawatts of solar to satisfy it (several times cumulative global solar installation).
Repowering Thermal Power Plants
Seasonal electricity storage is the sexiest (and largest) application. Powering turbines from existing coal power plants by producing steam from thermal storage is the most economical path to electricity storage. Power plants are expensive, and there are plenty of existing ones. It could take up to ~25 terawatts of solar to convert them all.
Thermal storage can make these power plants operate more efficiently. Exhaust air scrubbers with high fixed and variable operating costs aren't necessary. Thermal efficiency improves because heat isn't lost up the stack, in solids, or running scrubbers. Facilities can start faster, making them more relevant in electricity markets. The cost and operating profile emulates a natural gas power plant.
The ideal number of plant conversions is a trickier question because of the difficulty in predicting how much other generation sources, like solar PV linked with batteries, will supply.
Some countries with growing electricity demand might want facilities to run in a more "baseload" configuration in the short term. The math is simple: Is solar PV heat cheaper than coal plus coal's inefficiencies? The numbers might favor solar over coal imports, but mine-mouth power plants will take longer to cross over.
There are many ways to model the long-term question, and a recent paper I co-authored with several Stanford researchers is one entry. Most results converge around a technology like inexpensive thermal storage providing steady power on days when solar output is poor and electricity prices are higher. Thermal storage might supply around 10% of the total electricity supply in this configuration. Beating coal becomes much easier in this scenario because lower coal plant capacity factors hurt the economics.
Coal supplies around one-third of global electricity today, and electricity demand should keep growing. A substantial share of plants could undergo conversion in the most optimistic scenarios.
Why is Thermal Storage a Good Option?
Available Storage Materials
Requirements always determine what engineering choices are available. A constraining goal narrows the options.
Matching natural gas storage costs is a good constraining goal. Storing natural gas in depleted reservoirs is absurdly cheap, with a capital cost of $0.05-$0.10/kWh. For reference, batteries getting under $100/kWh is exciting, a 1000x higher cost per unit of storage. There are only a few materials that are even viable to match that cost. They are energetic chemical reactions, water, air, and dirt/rock.
Chemical storage works very well; that is what natural gas and other hydrocarbons are. They will always be favorable in some applications. But, extra steps and conversions raise the cost of producing or mining them. Cheaper options, even if they can't do everything a fuel can, would be nice.
Chemical fuels are so hard to beat because they carry so much energy per unit of mass, and the cost of that mass is very low compared to most products humans produce. By default, storing energy using less mass-efficient methods requires nearly free mass to be competitive.
Water and air are nearly free, so they are worth consideration. Unfortunately, both are forms of gravity/pressure storage, which have very low density, roughly 100x worse than thermal storage. Water and air are cheap, but managing their containers and movement is not. Air has the added issue of being compressible, so pressurizing it and recovering energy are more expensive and less efficient than water.
Thermal storage could also be attractive. A significant portion of human energy usage is in the form of heat. Conversion losses for these applications can be small. Conversion back into electricity is 40%-45%, which is not perfect, but better than many alternatives. The surface footprint is also very good, roughly an order of magnitude better than batteries because dirt piles a lot higher and closer together than shipping containers of batteries. A pile that could run a large power plant for months would look small in comparison to the cooling water reservoir.
Simple Thermal Storage Cost Analysis
There are two adjustable cost metrics: the cost of the storage material and the heat stored per unit. Most thermal storage is sensible storage, meaning heat changing the temperature of the material stores or releases heat rather than phase change or reaction. The material needs to be very inexpensive and subjected to high temperature swings to achieve acceptable $/kWh figures.
Dirt and rocks are practically free where they are, but moving, sorting, and processing the dirt adds cost very quickly. Giant mounds or embankments using nearby material cost ~$1/ton. Trucking sand or gravel costs $20-$50/ton. Refractory bricks made from purified dirt are ~$500/ton.
Dirt's thermal properties are similar to refractory bricks. It is composed of similar materials, like silicon dioxide and aluminum oxide. These oxides can handle very high temperatures exposed to air. Bricks have a more uniform structure than dirt, but piled-up dirt doesn't need fracture toughness or consistent thermal expansion. Dirt is an excellent value at one hundredth the cost per unit of storage.
Dirt stores around 20 kilowatt-hours thermal per ton for each 100 degrees Celsius of temperature change. A 400-degree Celsius swing would store 80 kilowatt-hours per ton and have a capital cost $0.013/kilowatt-hour, assuming $1/ton dirt moving costs. That is very competitive with natural gas storage, but selecting an inexpensive storage material is a tiny part of the battle. The heat addition and removal systems within the dirt mound can't cost more than $0.05-$0.08 per kilowatt-hour thermal. Standard Thermal has had to put in a lot of effort to reach costs in this range. Thermal storage systems are much more challenging to build than the material costs let on.
Engineering Cheap Storage
Beating natural gas storage costs requires several paradigm changes and adjustments to keep costs from ballooning to irrelevance.
The Challenges of Rapid Charge and Discharge
There is an anchoring to the idea that alternative energy storage technologies need to charge and discharge quickly. Rapid cycling is an incredible disadvantage for thermal storage for two reasons:
Heat Transfer is Slow
Materials that are inexpensive and tolerate high temperatures while exposed to air tend to be oxides that are insulators (dirt, sand, rocks, bricks, etc.). The thermal conductivity is poor, and it is difficult to withdraw heat from more than a few centimeters within minutes or hours.
A block of hot mass with dimensions in tens of meters can take years to fully cool. Dry rock geothermal electricity generation calls for giant fracture networks between two wells to overcome the heat transfer limitation.
Regenerative heat exchangers at steel mills are one of the few scaled thermal storage applications. They use special "honeycomb" bricks that maintain high surface area-to-volume ratios to handle ~25 cycles per day.
Source Hot Blast Stove Checker Bricks
Attempts to get around the heat transfer deficiency tend to be complicated. Pumping molten salts was a popular idea for a time. The Department of Energy has a concept that conveys sand through heaters and stores it in a silo until it drops through a heat exchanger to export heat. Other designs pump oil or molten salt through gravel. Several startups use graphite in an inert container to get higher thermal conductivity while preventing oxidation.
Any heat storage design must carefully consider heat transfer.
Temperature Cycling is Damaging
Heated materials undergo transformations. They tend to expand as they heat and contract as they cool, but there are also break points where the basic structure can change. Highly engineered or ordered materials become disordered and fall apart when exposed to thermal cycling.
Most heat storage concepts require engineered systems to achieve charge/discharge rate goals, making for a very severe techno-economic challenge.
Thankfully, there is little market reasoning for rapid cycling. We don't have to fight physics.
The Case for Slow Discharge
Charge and discharge rates on the order of months are fine.
Many Customers Use Energy Steadily
More than 10% of world energy consumption is industrial process heat. Almost all these facilities run 24/7 or in defined daily cycles. The energy usage in any one day compared to the cumulative storage needed to shift summer excess to winter usage is tiny. Models project 1500 megawatt-hours of storage for every 1 megawatt of output capacity in a typical US location. Charging with solar requires higher rates than satisfying demand because it delivers the same amount of energy at a lower capacity factor. The charging pace is still leisurely and benefits from electricity's ability to provide high charging temperatures.
Existing Winter Seasonal Storage is Slow
Winter demand is an application that already uses seasonal storage. The clearest example is natural gas. US consumption is less than supply in summer and shoulder seasons, but greater in winter. The industry stores large amounts of gas near customers for winter consumption. Many of these storage sites are depleted gas fields. They can't fill or empty faster than on a timescale of months. Adding more extraction and injection wells is expensive, and there is no need for it.
The natural gas storage seasonal cycle. Source: EIA
Thankfully, people don't use their entire winter allotment of heating fuel over 1-2 days. They use energy much more regularly, and daily usage is small compared to the total stored.
Batteries Solve Fast Cycling In Electricity Markets
I'm well into the 99th percentile of battery optimists, which is unusual for someone working on alternative storage technologies. The improvement in battery technology over the last twenty years has been incredible. They are already the least expensive way to add grid peaking capacity, and they will move a growing share of daily electricity generation.
People underestimate batteries. There is still significant room to drop costs in cells, packs, and construction/integration. Sub-$100/kWh installation costs are already present in much of the world and should come even to the US in time (tariffs, permitting, and the electrical code are the obstacles). These costs are transformative even in simplistic models, but the various revenue streams for batteries are also underappreciated. They can arbitrage prices across time, provide ancillary services, and arbitrage transmission constraints. The lower installation costs and multiple revenue streams mean that the daily spread between the high and low electricity prices will be much tighter, more like a few tens of dollars per megawatt-hour. That tight spread disadvantages technologies that take advantage of daily volatility with rapid charging or discharging.
Instead, markets will send signals to generation sources with low fixed costs that can provide a steady supply over days when solar, wind, or gas are constrained. Batteries will still be too expensive to arbitrage these seasonal differences effectively. Thermal electricity storage is orders of magnitude cheaper for storage, but worse at daily cycling. It can act as a complement to batteries and fill the market niche.
If we accept slower charge and discharge rates, then engineering becomes much simpler. Insulation and other factors benefit from the low thermal conductivity of insulating oxides. The thermal mass doesn't need to be highly engineered; it can be regular dirt piled up with some extraction pipes thrown in.
Spending more on electric heaters or heat transfer surface area to improve charge/discharge isn't necessary for the vast majority of applications. Dirt-cheap mass is the main advantage of thermal storage, and it is important to lean into that and take advantage of truly "seasonal" energy shifting.
Energy From the Sun
Even free storage isn't helpful if the cost of energy from the solar arrays is more than natural gas. Lazard's yearly report is the industry benchmark for solar's cost per megawatt-hour. The 2025 edition pegs the best-case US cost at $38/megawatt-hour, equivalent to more than $11/MCF of natural gas. The trading hub price for natural gas has been $3-$4/MCF in 2025 (and many years before that), exposing a significant problem for our thesis.
A closer look reveals the issue. The operating cost of a solar farm is very low at $4/megawatt-hour, but the capital cost comes in at $1150/kilowatt in the US, contributing $34/megawatt-hour. Commodity solar panels cost $80/kilowatt globally and $200-$250/kilowatt in the US, meaning there is a lot of waste (and opportunity) in today's solar farm capital cost.
The first slash at these costs comes from co-locating the solar array with the storage system. More than $300/kilowatt of cost comes from preparing electricity for export or connecting to the grid. These items include inverters, medium-voltage transformers, switchgear, substations, high-voltage transformers, power lines to the grid, and all the project management overhead to build these systems.
The next phase is to simplify the arrays themselves. Almost all US utility-scale systems use trackers that keep panels faced towards the sun throughout the day. These are expensive, add a lot of mechanical complexity, and cover more land to keep rows from shading each other. Instead, the array can use fixed racking where the panels lie nearly flat. These structures are very lightweight and simple because wind loads are lower. The total area is 2x-3x smaller than the tracking arrays because the panel spacing is tight. Operating costs even fall because there is less to break and less land to mow or pay rent on.
Our Test Site Solar Array
The financial implications of the racking change are profound. Install labor and direct racking costs fall, operating costs nearly halve. The electrical wiring decreases 50%, making wiring runs without increasing voltage much more practical. The reduced area opens up land constrained sites. The supply chain shortens dramatically because there are fewer unique parts and longer lead time items. The arrays are simple enough that an outside EPC firm isn't required, knocking off another chunk of cost. Each panel produces less electricity without trackers, but panels are relatively inexpensive, and the other benefits are well worth the penalty.
The result of these changes is that the installation cost for the solar array is the panel cost plus $100-$150/kW. A US array would be $300-$350/kWh, and the levelized cost falls to as low as $16/megawatt-hour, or $4.70/MCF of gas. An array with free market module prices would be equivalent to $3/MCF.
These costs put us within the range of US natural gas prices, and well in the money in most international markets. The consideration is what the end customer actually pays rather than the hub price. There are some large gas consumers, think of a giant ammonia plant, located near gas fields that might pay less than the trading hub price. These will take some time for economic crossover. In most of the US, customers pay 1.5x to 3x the trading hub price to cover transportation, storage, and distribution. Or power plants might be paying 100x the normal price for gas during power crunches. A fixed energy supply cost via co-located solar would be a big advantage for a converted power plant. Even with high US module prices, there is an opportunity to be competitive in many regions.
Progress and Commercialization
There was little doubt that a technology like this was technically possible. All the risk is in achieving the incredulously low costs. One of the most critical questions was how engineered the system needs to be. Pure and engineered materials can help make things predictable and reliable, but they are expensive. Ultra-low cost requires using them sparingly. Another concern is unknown costs. Typically, projects miss cost estimates not because a category is poorly estimated but because entire line items are left out. Constructing an end-to-end system became one of the overriding priorities to test these hypotheses.
So far we've:
Built a 100-kilowatt solar array to power testing.
We completed the array within 6 months of saying "Go," including permitting, order lead times, and construction (which was a few weeks at the end). Part of that speed was the simplicity and lack of interconnection, and part was clever (and lucky) planning. We completed all the engineering, procurement, and construction ourselves to prove its feasibility.
10 Months of Destructive Heater Testing
The biggest cost miracle is the heating elements. Ordering package systems from existing vendors can easily cost hundreds of dollars per kilowatt, even though the resistor material itself is $1-$2/kilowatt. Structure, insulation, controls, and assembly have astronomical costs. The extras are there because heating elements need to last years, but can literally die in one second in conditions outside their boundary. There are only a few materials that are even acceptable as resistors.
Our cost goal for the heaters is $10/kW, close to the cost of the resistor material. A goal so ambitious obviously requires designing and building our own heaters. We have large piles of dead heaters, and now know about every way one can die. Our latest testing suggests we will achieve our cost and reliability goals, and the effort to get there has been incredible.
Rearchitecting Solar PV Electrical Integration and Safety
Another unpleasant surprise was that the entire solar industry relies on one narrow regulatory pathway that involves inverters and AC interconnection. The equipment needed to satisfy the electrical code for an off-grid DC solar array doesn't exist at any scale without going to grid-forming inverters (we had to buy some discontinued equipment on eBay to complete the test site). Companies usually embed these systems in a grid-tied inverter, which we don't have.
Solar array safety is challenging compared to wiring in a house or factory due to the direct current and constant current nature of solar modules. We've had to design our own controllers and safety systems to have a chance at deploying a commercial project. These are usually "outsourced" to very specialized inverter manufacturers.
Moving Dirt
Building an inexpensive dirt pile is one thing, but embedding pipes and heaters while keeping it cheap is another. We've gone through at least a half dozen configurations before finally settling on a viable construction method.
One version of the mound.
Heating Dirt
Lab-scale tests of heating dirt are easy. But, heat flow in a large mound can be more complex with much longer diffusion paths, especially during the early phases (heating the first time is like making bricks in situ). Heating in a test mound has been critical for developing models that are reliable at scale.
The Team
We are now an army of three permanent employees. Jimmy Williams joined in the summer of 2024 to focus on solving the heater cost and reliability challenge. Brian Pal joined more recently to tackle final systems engineering hurdles and refine our prototypes into more robust and replicable commercial designs.
We also have a horde of capable contractors who help with construction or provide specialized engineering and regulatory services.
Working with this team has been a wonderful experience, and their abilities and extremely specific experience has shaved years and tens of millions (or hundreds of millions?) of dollars off the development process.
The Constant Pressure of Cost and First Customer Systems
Cost tries to balloon at every turn. Almost every system component has undergone at least one cost crisis, if not many, where that single component could blow the entire project budget. Having a component that costs anything to buy and install is a problem when everything has to be practically free.
Balance-of-system and soft costs are the most challenging, like the heater example. Theoretically, many things can be very cheap, but only under extremely narrow conditions. Any perturbation quickly requires a slew of mitigations that increase the cost and complexity. Rarely do these narrow windows for different subsystems align.
Working out the details has been the most extreme global optimization problem I've worked on. I've had to develop a detailed understanding of solar cells, the electrical code, unsteady state heat transfer, and many other topics in between (My blog output decreased because I had to study at night instead of writing blog posts). We now know what we need to build and will be finalizing the commercial design over the next few months.
The goal is to build the first customer systems in 2026. We are working with several potential customers to ensure our system can successfully integrate with their energy usage (designing a universal, copy-and-paste system that can integrate easily with different customer facilities is another tricky part of the global optimization). Our first system can offer savings for fuel oil, propane, and higher-priced natural gas users. Follow-on systems, especially those larger than a few megawatts, can match the sub-$0.10/kWh natural gas storage cost. Matching energy cost depends on the location and the solar panel cost.
If signing up for a system like this interests you, please contact me at austin@standardthermal.com.
Maintaining a Competitive Edge
Differentiation from Existing Thermal Storage Startups
I would not be working on this problem if I agreed with the development path of other firms in the space.
Every other serious effort is a highly engineered system that has a day to a few days' worth of heat storage. That means competing head-on with batteries, whose cost is already below that of many thermal storage hopefuls.
The Standard Thermal system has months of storage capacity, a cost per kilowatt-hour several orders of magnitude lower than highly engineered systems, and is scalable with a fraction of capital and human resources.
The Strategy
"What is the moat with a dirt pile?" is a common question. Most sites will have drainage moats, but that usually isn't the answer people are looking for.
The main differences between energy projects like this and software are the difficulty of scaling and customer stickiness. The technology itself is far from enough to achieve scale due to thickets of regulations, permits, certifications, cost pressures, and financing hurdles that add friction to growth. However, the on-site nature of the projects ensures a long-term relationship with the customer. The project size unlocks low customer acquisition costs.
There is a well-worn path for scaling infrastructure technologies. Expensive capital builds the first examples. Then there might be debt financing available to borrow against an operating project. Then the debt for new construction. And finally, someone else pays for new projects.
The reasoning behind this path is scale and capital efficiency. Developing projects is demanding, and one firm can only take on so many before the overhead and risk become crippling. Each rung up the strategy ladder reduces capital requirements and company risk.
There might be dozens of renewable energy developers with a market cap of more than a billion dollars. No single player can dominate due to the friction and barriers in each project. Each project having a pseudo-monopoly over that particular location also maintains the value of existing systems. The size of the company is related to the size of its projects - big companies have big projects. The same pattern holds in other industries, like oil and gas, where the leading companies operate the most productive fields and focus on megaprojects. Even within the shale industry, the most successful players found the most extensive high-return plays. The company with an all-encompassing vision of being in every play, Chesapeake, declared bankruptcy while its founder died an untimely death.
Another unusual feature of ultra-cheap storage is that the actual storage is a relatively small portion of the cost. But project revenue and viability are solely reliant on it. Lenders must evaluate the risk based on the storage system performance, not low-risk solar arrays or customers with good credit.
A multi-phase strategy emerges from these realities:
Build Initial Projects
The first projects need to validate the technology, provide some revenue, and create a track record for future financing. There is no "chasm of death" as both the off-grid solar and storage systems can scale down to a few megawatts without massive penalties. The diversity of energy users is enough that potential customers with expensive energy exist in this range.
Scale Up Size and Financing
Projects in the tens of megawatts have cost benefits and conserve corporate overhead, allowing faster growth. The track record of the initial projects should open up financing.
Push Lending Rates and Cost Down
Several trends work to make the technology viable in more locations. Solar panels are a significant portion of CAPEX, and continued cost decreases have a disproportionate impact on the levelized cost of heat delivered. Improving solar panel efficiency reduces the land required and saves money on racking and labor. Other custom components, like electrical safety systems and controllers, decrease in cost with scale. Interest rates for borrowing against projects decrease, which has an outsized impact on project breakeven.
Provide Value No One Else Can
The technology can only continue scaling by enabling many project development firms to deploy it. Becoming some version of an engineering, procurement, and construction (EPC) firm is a necessity. The bankability and turn-key system can create an excellent buyer surplus by decreasing the overhead and levelized cost of energy for the entire project. Offering a fair price plus friction from patents leaves little opportunity for new entrants and preserves the long-term maintenance and operations relationship.
Many of the necessary components to replicate the system require time (testing, certifications, etc.). Our system costs will continue to fall and become more bankable. That increases the free cash flow hole and time-to-market a new entrant has to weather. It will be much easier to make money by buying the thermal storage system and developing projects rather than trying to copy it.
Further Upsize Projects and Repeat
Larger projects enable the firm to grow without choking on overhead. These projects will almost certainly be power plant conversions or entire city district heating, which are often 10x-100x larger than typical process heat applications. The same follow-on move to EPC unlocks global scale.
Implementation of the plan requires excellent execution, but this is always the case with startups. Each rung on the ladder unlocks durable value. Falling short could easily mean being a unicorn project developer instead of being zeroed out.
Conclusion
Handling high seasonal solar variation, dealing with the last 10% of electricity generation in renewables-dominated grids without fossil fuels, and reducing the burden of energy costs in fuel-poor regions have been classified as near-impossible problems at various times.
It is rare to work on something that has the potential to be so impactful and also benefits from my passions for energy and engineering minutiae or low-level global optimization. I'm grateful for that. There is an enormous amount of work ahead to make the impossible possible, and I'm more motivated than ever to see it through and scale the technology.