Jeff Thornburg helped turn a government research project into SpaceX’s most powerful rocket engine. Now, he’s trying to do the same thing at his startup Portal Space Systems, which is taking an idea set aside by NASA and turning it into high-powered propulsion for the next generation of spacecraft.

Portal, founded in 2021, announced a $50 million Series A funding round Thursday that values the company at $250 million. The round was led by Geodesic Capital and Mach33, alongside Booz Allen Ventures, ARK Invest, AlleyCorp, and FUSE.

The company is developing a technology called solar thermal propulsion. Today’s standard satellite engines either burn chemical fuel or convert the sun’s energy to electricity, using that to power efficient but low-powered thrusters. Portal’s engines would instead concentrate the heat of the sun, using that to heat propellant and move the spacecraft along at high speed.

The technology has been the subject of investigation in government research labs since the 1960s, most recently as a concept for sending a probe into interstellar space, but has yet to make it into orbit. Thornburg, along with co-founders Ian Vorbach and Prashaanth Ravindran, plans to change that in the next two years.

Thornburg began his career in the U.S. Air Force, where he worked on a program to develop an efficient, powerful next-generation kind of rocket engine that engineers call full-flow staged combustion. A decade later, he was wooed to SpaceX by Elon Musk to turn those concepts into the Raptor engine that now powers the company’s massive Starship.

After stints working at Stratolaunch and Amazon’s Project Kuiper, he turned to propulsion once again.

A new kind of rocket engine

Solar thermal power is, in Thornburg’s view, the next logical step in rocket tech. NASA had studied the technology extensively in the late nineties and concluded it provides better performance in many cases. It wasn’t developed further because there wasn’t enough demand for in-space mobility, according to a 2003 report commissioned by NASA.

With satellites and space probes flying much more rarely back then, it was simpler just to use a more powerful rocket than invest in in-space propulsion. Now, with thousands of new satellites flying every year and the U.S. military demanding spacecraft that can fly quickly between orbits to surveil or threaten rivals, that calculus has been turned on its head.

“It’s no longer acceptable to move slowly on orbit,” Thornburg told TechCrunch. “You know, China’s running circles around our spacecraft. We need equivalent capability.”

Portal has already received $45 million in strategic funding from the U.S. military, on top of $67.5 million in private capital, thanks to the potential for using its technology for orbital warfare, according to Booz Allen Ventures’ managing director Travis Bales.

And in a future where we may see millions of satellites in orbit around the Earth providing communications and computing services, satellite operators will need cheaper means of maneuvering spacecraft out of each other’s way, notes Aaron Burnett, the CEO of the aerospace-focused venture fund Mach33. Burnett sees Portal becoming a “space mobility prime,” providing propulsion to a variety of users.

The path to orbit

To get there, the company will need to get its technology working in orbit. Its flight electronics were launched on a shakedown cruise around the planet last week, and another prototype spacecraft is expected to launch in October. The company will demonstrate a working prototype of its engine with the launch of its first SuperNova spacecraft — a “fighter jet for orbit,” per Thornburg — expected in 2027.

Portal benefits from recent advancements in additive manufacturing and materials science, which have led to the development of the company’s combined solar concentrator and nozzle, the Hex thruster.

Rocket nerds think that nuclear-powered rockets are the next step to unlock transportation throughout the solar system, but the regulatory and legal challenges of building such a system make it beyond a startup’s pay grade.

But Portal’s engine also gives the company a head start on one version of an atomic rocket, a nuclear thermal propulsion system — essentially replacing the heat of the sun with that of a reactor. When the U.S. government is ready to build it, Thornburg’s team will have already proved out many of the moving parts in orbit.

“I’ll be able to help mature this technology much faster on orbit than we ever will by trying to build a $2 billion ground test facility that’s nuclear safe,” Thornburg said.

Collide Capital, founded by Brian Hollins and Aaron Samuels, announced Thursday the close of a $95 million Fund II. The firm, founded in 2021, backs early-stage companies in fintech, supply chain, and the future of work. 

It’s backed 75 companies to date after closing its inaugural $66 million Fund I in 2022. Hollins said it took them around 13 months to raise this latest fund, which they hope to deploy over the next 3.5 years. The landscape for many emerging fund managers is tough, but the Collide Capital team comes with a track record and strong pedigree (Hollins spent a decade at Goldman, Lightspeed, and Slow, while Samuels worked at Bain, Lightspeed, and co-founded AfroTech, one of the largest tech conferences in the world). 

Limited partners (LPs) in Fund II include the University of California Endowment (UC Regents) — which anchored their last fund — Accolade Partners, Fairview Capital, Goldman Sachs, and JPMorgan. The average check size will be between $1 million and $3 million, and the team hopes to back at least 30 companies, having already cut checks to five. (Other companies in the firm’s portfolio include Culina Health and Helios). 

“We’re most interested in platforms enabling automation, real‑time collaboration, and faster, data‑driven decision making,” Hollins said. 

Aside from technology, Hollins and Samuels said the firm is also excited to expand its Collide Campus program, launched in 2022 to help mentor the next generation of founders and venture capitalists. The expansion is separate from the fund raise. The program encompasses an undergraduate initiative that trains students in VC and entrepreneurship, and a graduate fellowship program that sees students working alongside the Collide team as investors and apprentices. 

The undergraduate campus program is on more than 20 campuses, including Harvard and John Hopkins, and Samuels said more than 50 students have passed through the program so far and have landed top jobs at places like General Catalysts and, of course, Collide. Overall, the program helps the firm source deals and talent. Samuels said they started the program as it was something they wished they had when they were students. 

“We’re connecting the best and the brightest with venture capital to match their grit and determination to build businesses the world needs,” he said. 

Rare earth elements might represent small slivers of the global metals market, but they’re big sticks on the geopolitical stage, where China wields its considerable heft as a bargaining chip during trade disputes.

Countries like the U.S. have started taking steps to reverse China’s dominance. New mines have opened, and manufacturers are popping up to make magnets and motors. But it has been a slow process. China built its position over several decades

“The unfortunate part is that in order to be able to support an entire industrial base, you’ve got to have that whole supply chain node-matched in terms of capacity,” Zach Detweiler, co-founder and CEO of Radify Metals, told TechCrunch. He thinks one node in particular has been overlooked, the part that turns metal oxides into pure metals. “That’s this missing middle we’ve identified,” he said.

Most metal refining uses either heat or water (in combination with other elements) to strip metal oxides of their oxygen, leaving pure metal that’s easier to incorporate into alloys that make, for example, stronger magnets or more efficient electronics. Both processes are effective, but also highly polluting.

Another option has been known for a while, but it was considered too expensive to use in commercial production. It involves stripping oxygen using plasma, which is basically a superheated soup of super-energetic particles. The process’s only waste is water vapor.

Now, Radify thinks it has cracked the plasma problem through a combination of more efficient power electronics and some clever engineering to handle the metallic powders. It recently gave TechCrunch an exclusive look at its technology. The startup has raised just under $3 million from investors including Overture, Founders Inc., Mana Ventures, and Acequia Capital.

Radify said its reactor can transform a wide range of metal oxides. Today, it’s focused on dysprosium and samarium, two rare earths that are key ingredients in magnets and electronics.

Inside the reactor, hydrogen is heated until it forms a plasma, then metal oxide powder is blown into the chamber, where it’s reduced — that is, the oxygen that’s bound to the metal is removed. On the other end, pure metal emerges. Need to make a different metal? Just tweak the operating parameters for the reactor.

The design allows the company to make smaller reactors compared with the massive equipment the industry uses today. That could reduce manufacturing costs while also adding flexibility.

“When China bottoms out the price of dysprosium and starts selling it for a dollar a kilogram because they want to kill a lot of the businesses out there, we don’t die because we can pivot to titanium or zirconium,” Detweiler said. “We become substantially more resilient to a lot of the volatility that’s endemic of the metals industry.”

Currently, the five-strong Radify team is refining the technology in its lab in Campbell, California. It hopes to produce several kilograms of pure metal per day by the end of the year. In the coming months, it plans to raise more money to build a pilot reactor capable of making up to 100 kilograms per day.

If Radify can reproduce the results it’s seeing today at a larger scale, it could give Chinese producers a run for their money. Today, rare earths outside of China sell for several times the Chinese price. Detweiler thinks that Radify can produce rare earths that cost just 50% more than China’s prices in the near term. “But I also expect Chinese spot prices to go up,” he added. Once the company starts scaling, it expects to reach parity and possibly lower.

Beyond rare earths, the company is exploring a range of other metals, including hafnium, uranium, scandium, and titanium, which are used in electronics, aerospace, and other fields. The plasma-based technology also works on more widely used metals like iron and aluminum, though currently its not efficient enough to challenge the incumbents. But if it eventually gets to that point, “we’ll have completely changed how humans make metal,” Detweiler said.