Down Arrow Button Icon

This article is part of Fortune‘s Blueprint for a climate breakthrough package, guest edited by Bill Gates.

In an effort to make renewable energy more reliable, one company wants to send trains carrying tons of rocks up and down a hill. Another wants to use cranes to stack and unstack heavy blocks.

Startups are racing to create new technologies—many seemingly far-out—to help in the shift to renewable energy. Their goal is to solve one of the solar and wind industry’s biggest problems: how to store energy over long periods of time in order to fill lulls created by clouds, darkness, or lack of a breeze.

“There’s been all kinds of stuff people have been messing with,” says Joe Osha, analyst at investment banking firm JMP Securities. “So far, no one has managed to scale successfully.”

Energy-storage upstarts see a huge opportunity in what is expected to be an aggressive shift to clean energy under President Joe Biden. By 2050, he wants the U.S. to be at net-zero emissions, the point at which the country eliminates as much pollution as it emits.

Practically speaking, that means increasing the country’s reliance on solar and wind energy. It also means creating technologies that are better at storing energy than lithium-ion batteries, which many utilities currently use but are too expensive and bulky to store large quantities of energy for days.

One company working on the problem is Malta Inc., a startup spun out of Google X, Google’s “moonshot” factory, in 2018. Malta is betting that using molten salt and coolant to store energy created from solar and wind farms will be the cheapest and most effective technology.

When needed, the energy from the salt, liquefied at 565 degrees Celsius, and coolant, stored at minus 60 degrees Celsius, would be converted back into electricity. Malta CEO Ramya Swaminathan says she believes the company has an advantage over its rivals because the materials used in its storage system have a proven track record in other contexts, reducing the risk of failure. “They’re industrial products that have been made for decades,” she says.

Climate Package 2021-Ramya-Portrait — Ramya Swaminathan, CEO of Malta.

Malta’s planned system, which would look somewhat like an oil tank farm, would cover at least 10 acres. It would provide 100 megawatts—enough for a large town—for at least 10 hours at a cost of less than $100 per kilowatt hour.

Malta’s biggest investor is Breakthrough Energy Ventures, which led a $26 million round of investment after the spinout from Google X. The venture capital firm boasts deep-pocketed investors including tech billionaires Bill Gates, Jeff Bezos, and Alibaba cofounder Jack Ma.

But Malta faces major challenges. For now, it has no customers, and it doesn’t expect its first project to debut until 2024 or 2025.

Malta also must prove that it’s different from its bankrupt predecessors like SolarReserve, which developed the Crescent Dunes solar plant near Las Vegas, which used molten salt to store energy before shutting down in 2019. Another struggling company, Aquion, created batteries that used salt water to store energy.

Malta must also cover the huge cost of building and maintaining its system. Then there’s the hurdle of the rules and regulations laid out by industry organizations, which don’t account for storage technologies like Malta’s.

All the while, Malta’s competitors are signing deals and landing new investment. For example, Swiss-based Energy Vault, the company that stores energy by stacking and unstacking blocks, landed $110 million in investment from SoftBank in 2019. The company uses energy from a solar farm, for example, to stack blocks to make a tower, and then, when power is needed, it unstacks them, turning the kinetic energy produced back into electricity.

Climate Package 2021-Malta Layout — A rendering of Malta’s system.

Meanwhile, Santa Barbara–based ARES North America, the company working on running rock-filled trains up and down hills, broke ground in October on its first project, outside Las Vegas. The company expects to use electricity from renewable energy to drive railcars up a hill. Then later, it would roll the cars back down the hill to convert the energy released back into electricity.

“It comes down to who can get to the market at the right time,” says Patrick Leslie, energy-storage consultant at LevelTen Energy, a software service for for energy buyers and sellers. “There’s a chance they may be too early, too late, not bankable, or the technology doesn’t work.”

The automotive industry has poured billions of dollars of investment into the innovation of lithium-ion batteries, helping drive down their cost to an average of $137 per kilowatt hour, according to research company BloombergNEF. The batteries also can provide power over longer periods of time than what they could before.

The rapid improvements in battery technology present a challenge for emerging energy-storage companies focused on serving utilities. Those storage companies hope their technologies will store energy for longer than batteries can, in a way that would let utilities easily tap the power as demand increases.

“If you’re a grid storage [company] trying to compete with lithium ion, it’s going to be impossible,” says Gene Berdichevsky, CEO of battery materials company Sila Nanotechnologies.

Although lithium-ion batteries will improve and become cheaper over time, they aren’t ideal for long-duration energy storage, Berdichevsky says. That’s because utilities would have to spend money buying more batteries every time they want to extend the amount of time they can store energy, whereas doing so with the alternatives may be more cost-effective. Malta, for example, says it would be able to extend the time it could provide power from its 10-acre facilities to 16 hours from 10 hours simply by adding more molten salt and coolant to its tanks.

Though the demand for long-duration energy storage is still limited, some states are already seeking to tap the emerging technology. Last year, eight public electricity agencies in California issued a request for proposals for a company to offer 500 megawatts of long-duration storage. And in May, a utility in Minnesota signed a deal to pilot an energy-storage project with Form Energy, a Malta competitor, that is to premiere sometime around 2025.

Form, which has raised more than $50 million for a low-cost, long-duration battery, is expected to provide 1 megawatt of electricity for 150 hours at one time. That would be enough for a city to make it through a polar vortex, says Form CEO Mateo Jaramillo, and to retire a polluting coal plant.

These emerging technologies could have a massive impact on the future of renewable energy, ultimately aiding in the fight against climate change. But JMP analyst Osha says he’ll believe it when he sees it. “Talk is cheap,” he said. “It’s all well and good to say I have something that works versus demonstrating that it works.”