Space mirrors, solar panels, fools, and their money

I wonder if fraudulent or foolish investment schemes are more likely to attract victims if the schemes involve high technology, saving the planet, or, especially, both. This came to mind when reading a baffling article published in the New York Times two days ago [1], about a company that wants to put mirrors in space to bounce sunlight Earthward to power solar panels on the ground at night. “‘We’re trying to build something that could replace fossil fuels and really power everything,’ Ben Nowack, Reflect Orbital’s chief executive, said in an interview.” It is, of course, true that solar panels as conventionally used are only productive during the daytime. But do space mirrors for nighttime solar power make any sense? Especially to “really power everything”? The article continues, “The company has raised more than $28 million from investors.”

There’s no shortage of holes to poke. Here’s one that I posed to my Physics of Energy and the Environment class (for non-science majors) yesterday, leading into the exercise by reminding them that one of the aims of the course is to produce scientifically literate citizens, capable of critically and quantitatively evaluating claims. I also emphasized that they possess all the tools they need. (It’s week 9 of the 10-week term.)

Let’s consider the best-case scenario of a space mirror perfectly focusing light onto terrestrial solar panels. (No, this isn’t possible, but we’ll be optimistic potential investors!) In our class, we already calculated how much solar power hits the Earth, a precursor to figuring out what sets the temperature of the planet and how the atmosphere modifies this. A one-meter by one-meter square in space, at the Earth-sun distance, intercepts about 1.4 kW of power in the form of sunlight. As a result, at noon on a sunny day, the power per square meter hitting the earth’s surface is about 1.0 kW. It’s less in the morning or evening or on a cloudy day, but again let’s take the best-case scenario and imagine 1 kW / m² as our solar power density, and the effective output of our space mirror. “The largest mirrors are planned to be nearly 180 feet wide”. That’s about 60m × 60m, or 3600 m². A good commercial solar panel has an efficiency of 20% for the conversion of sunlight to electricity [2]. Electricity presently costs about 10 cents ($0.10) per kilowatt-hour. Given this, what’s the value ($) of electricity generated in one hour by a solar panel powered by one space mirror? I asked the students to work on this. To summarize the setup:

• mirror area: 3600 m²
• power density: 1 kW / m²
• solar panel efficiency: 20%
• price of electricity: $0.1 / kilowatt-hour
• The question: What’s the value ($) of electricity generated in one hour by a solar panel powered by one space mirror?

You can try it yourself…

In the meantime, a pedagogical note. I asked everyone to work on this, and about a quarter of the class quickly came up with an answer. Walking around to chat with students and see how they were doing, it was clear that for many, the stumbling point was math: What is 20% of 3600, either expressed as 3600 x 0.2, or 3600 divided by 5? There are, in a college course at the “flagship” university of Oregon’s higher education system, a sizeable fraction of people who state that they are incapable of this. Obviously one could plug it into a calculator, but a basic fluency with simple arithmetic allows for intuition, quick assessments, and spending time on machine-assisted tasks that are actually difficult. Because I’m actually fond of this class, I yelled, “You are not idiots!” They all know what 35/5 is, which is close enough to 36/5 for physics (as we experience throughout the course), even if they freeze at the thought of a non-integer quotient. (I also pointed out that if they don’t know what 35/5 is, they shouldn’t tell me, because I couldn’t take it.) This seemed to help, and the majority of the class surmounted the hurdle. It’s a sad state of affairs, though, and the students have been badly served by an educational system, and a society, that has steadily eroded its expectations, often under the guise of compassion. (It is certainly not compassionate to leave students uneducated.)

The answer: $72. That’s the value of one hour of space-mirror-powered electricity, in the best-case scenario. How much will Reflect Orbital charge? $5000 !

Obviously, $5000 is far larger than $72. For $5000 (per hour!), one could buy a lot of batteries to store power generated in the daytime, or build a lot of electrical transmission infrastructure, or rent a lot of monkeys to crank generators by hand.

The students were stunned. The New York Times article has the usual flaw of wanting to appear fair, presenting “both sides” of a story even if one of the sides is absurd, but it does cite people who are critical of the concept. “But when you start crunching the numbers, and the numbers are pretty easy to crunch, then you find there’s a lot of serious issues with it” (astronomer Michael Brown).

The calculation above highlights just one of many fatal issues with the company’s scheme. To briefly note another: the light will spread out as it travels from the mirror to the earth, covering a large area with quite dim light. In the article, Brown estimates the area as 18 square miles! One would have to install panels over that entire span to capture the (best-case) kilowatt of power.

As noted, investors have given Reflect Orbital over $28 million. I told the class that one of our learning goals is that they’re never one of these people. It would be fascinating to find out who the investors are and interview them, asking how they assessed the company and came to their decision to part with their cash. (I have no idea whether information about investors is easy to gather. It would be a great task for talented New York Times reporters.)

What should one think of Reflect Orbital? There seem to be three possibilities: (1) Reflect Orbital is run by idiots. (2) Reflect Orbital is run by criminals, unethically pitching nonsensical, nonscientific schemes to a gullible audience. (3) The space-mirror solar power idea is just a hook to get money and attention for some other aim, to be announced later. The Reflect Orbital website doesn’t clarify anything, though we get more non-sequiturs meant to dazzle the ignorant, like:

None of these are cheery conclusions. I’d bet on (2), and the likelihood of this being a fraud bothers me more than most frauds because of its faux-ecological costume.

Today’s illustration…

The ‘Z’ Lounge Chair” by Poul Jensen, mid 20th century. I painted it based on this photo. Cost: $3000, or 36 minutes of space-mirror electricity.

— Raghuveer Parthasarathy, March 11, 2026

Notes:

[1] Chang K, Tabuchi H. “A Night Light in the Sky? Reflect Orbital Wants to Launch a Big Space Mirror.” The New York Times, 2026 Mar 9; Available from: https://www.nytimes.com/2026/03/09/climate/space-mirror-satellite-solar.html
[2] If you’re wondering: Couldn’t I get an extra factor of 5 with a 100% efficient solar panel? Physical laws constrain the efficiency to at most 34%; the 20% efficiency of current commercial panels is remarkable.