Who will pay for moving factories into space?

In February 2026, LambdaVision, an artificial retina R&D developer, reserved payload positions and commercial spaces on the yet-to-be-operational Starlab commercial space station. Four months later, it signed a memorandum of understanding with Vast to prepare for future research and manufacturing activities on the Haven series space stations.

Prior to that, the company had completed nine International Space Station missions. It aims to develop artificial retinas with more stable quality, while gravity on Earth has caused troubles for the thin-film manufacturing process involved.

Exploration efforts in China are also advancing. Recently, the development progress of the "Yixian-B Satellite" project, a cooperation between CAS Space and the Third Affiliated Hospital of Sun Yat-sen University, has drawn public attention. The project plans to carry payloads for stem cell culture and observation to conduct biological experiments under microgravity conditions.

From pharmaceuticals and protein thin films to semiconductor materials, a number of problems that are difficult to solve in ground factories are now seeking solutions in space. However, to turn high-quality samples from laboratories into long-term orders from customers, three barriers concerning quality, cost and delivery lead time still need to be overcome.

Space manufacturing has already found buyers willing to give it a try. Who will be the long-term purchasers next?

The first batch of buyers target hard-to-solve problems in ground factories

The artificial retina developed by LambdaVision is composed of multiple layers of protein thin films that need to be deposited layer by layer. The uniformity of the film coating directly affects the product performance.

According to information disclosed by the International Space Station U.S. National Laboratory in June 2026, LambdaVision cooperated with Space Tango to load the automated production system into the CubeLab device.

Space experiments show that compared with ground-produced samples, artificial retinas prepared under microgravity have improved in uniformity, optical performance and repeatability. The team is also continuously optimizing fault detection and quality control mechanisms.

What LambdaVision aims to improve is the uniformity of the thin films, as well as the performance gaps between different batches of products.

Microgravity can weaken the impact of sedimentation and buoyancy-driven convection on some processing procedures, but these advantages need to be verified in specific materials and processes. What customers ultimately purchase are products that pass all quality inspections. The space environment itself cannot replace formal quality standards.

UK-based Space Forge focuses on semiconductor materials, hoping to leverage the space environment to manufacture high-quality substrates — the basic materials required for chip production. ForgeStar-1 is tasked with the technology demonstration mission.

There are multiple processing steps from substrates to finished chips. Whether the material advantages obtained in space can withstand subsequent processing, adapt to the production lines of customers, and finally be reflected in the improvement of device performance or yield, is the fundamental basis for customers to make procurement decisions.

All these explorations are targeting the process links that are difficult to make breakthroughs in ground manufacturing. The degree of improvement that space can bring determines whether such a round trip is worthwhile.

Pharmaceutical companies that entered this field earlier have taken a different path.

A paper published by the Merck team in 2019 shows that researchers carried out crystallization experiments of pembrolizumab on the International Space Station and obtained crystal suspension with more uniform particle size distribution. Subsequently, they applied the insights gained from the experiments to ground production, and prepared uniform suspension with suitable viscosity and injectability through rotational mixing and temperature gradient control.

Researchers found the direction for improvement through space experiments, and the subsequent processes are completed on the ground. This does not mean that the related preparations have been commercialized. Customers may only need to send their experiments to space instead of moving the entire production process there.

Pharmaceutical companies are willing to pay for an experiment if it can solve valuable R&D problems. For orbital platforms, this revenue may terminate once the project is completed. To continue sharing the benefits of the achievements, they need to obtain corresponding rights and interests in intellectual property, process licensing or joint development activities.

Space manufacturing has thus formed different business models: experimental services are charged by project, process development can generate revenue through technology licensing, and material production relies on continuous product supply. The increase in the number of flights will bring different revenue levels, depending on what the customers are purchasing.

Long-term orders are still pending

In May 2026, Varda announced a cooperation with United Therapeutics to study the use of microgravity to improve the formulations of therapeutic drugs for rare lung diseases. The announcement only disclosed the research cooperation, and its abstract did not list the long-term procurement volume or the amount of production orders.

Pharmaceutical companies have entered this field with specific demands, but results are still awaited from joint research to continuous mass production.

Varda's process connects ground process development, gravity condition screening, microgravity processing and post-flight characterization. Before launching, it is necessary to screen the objects worthy of verification, and after returning, the processing effect needs to be analyzed. The orbit is only part of the entire R&D process.

LambdaVision has already started booking berths for future production and negotiating cooperation with other platform providers. However, these arrangements only solve the problem of where to produce, and whether the artificial retina can be sold and how many units can be sold still remain to be verified.

As of September 2026, the company is still in the pre-clinical stage. It announced in the same month that it plans to use Helogen's autonomous orbital platform for the first time in October to test the protein used in artificial retinas and conduct in-orbit characterization. The new mission will help it evaluate the protein properties in space and further improve quality control.

It needs to pass two barriers: first, to prove that the process can run stably, and second, to prove that the product performance meets customer requirements. Completing one flight can only answer one of the two questions.

The "Yixian-B Satellite" depends on the actual tasks it undertakes.

The October 2024 announcement from the Third Affiliated Hospital of Sun Yat-sen University shows that the project has completed the scheme design review and kicked off development, with key focuses including in-orbit intelligent experiments on lab-on-a-chip, as well as verification of cell growth and differentiation under microgravity. It was originally scheduled to be launched in the first half of 2025.

According to media reports in September 2026, the verification time is adjusted to the end of that year to the beginning of the next year, and the capabilities including stem cell culture, in-orbit observation, environmental control and data downlink are listed.

These capabilities first serve space biological experiments. Whether the culture system can operate normally, whether the observation results are valid, and whether the data can be transmitted stably back to the ground, determines whether subsequent research can be carried out.

Domestic scientific research layouts are also expanding.

In September 2026, the Hong Kong SAR Government announced that the Space Manufacturing Technology Innovation Center led by the Hong Kong Institute of Science & Innovation, Chinese Academy of Sciences, plans to carry out its first space mission in 2027.

These missions include in-orbit operable culture and testing equipment, controllable material processing devices, unattended operation systems, as well as control experiments and performance analysis after returning to the ground.

They may generate equipment and service revenue first. As for the large-scale product market, it still needs to wait for continuous customer verification.

For space factories, calculate the round-trip cost first

Even if customers are willing to purchase the products, the products need to be transported back to the ground smoothly.

For the "manufactured in space, used on Earth" business, launch is only the first leg of the journey. Ground screening, payload development, platform integration, in-orbit operation, return and recovery, as well as subsequent transportation and testing, all need to be included in the delivery cost.

The total payload mass is not equal to the product output. Reaction vessels, culture medium, environmental control equipment and structural parts take up a lot of space, and the materials that can actually be delivered only account for a small part. The drop in the launch price per kilogram of rockets, when transmitted to each batch of qualified products, also depends on equipment utilization rate and processing yield.

The return system has its own technical thresholds.

NASA disclosed that on January 29, 2026, Varda's W-5 return capsule landed in southern Australia, completing its return for the first time entirely using Varda's self-manufactured C-PICA thermal protection material. The technology is licensed by NASA, which also provided manufacturing and flight test support.

Commercial return now has an additional flight-verified thermal protection solution. From the landing of the return capsule to the delivery of products to customers, sample preservation, transportation and acceptance still need to be completed.

First-stage rocket recovery cannot replace this capability. First-stage recovery solves the reusability of launch vehicles, while orbital sample return needs to bring the produced payload back to the ground. The two correspond to different tasks and cost structures.

In addition, long-term procurement requires multiple scenarios to be included in the arrangements, including how to deal with delays, who will bear the losses caused by failures, and when alternative missions can be executed.

LambdaVision has been looking for future production sites in advance, and Varda is connecting orbital experiments into the R&D processes of pharmaceutical companies. Buyers are emerging, but it is still unclear who has established long-term procurement mechanisms for such ground-produced products.

This article is from the WeChat official account "Xingdong Wuji", the author is Julian, and it is published by 36Kr with authorization.