The Setup Nobody Expected

NASA’s Ignition briefing in late March 2026 landed in the commercial station industry like a controlled demolition. After years of Commercial Low Earth Orbit Destination (CLD) Phase 1 funded agreements and the expectation of Phase 2 full-and-open competition awards, NASA officials told the industry directly: they no longer believe a commercial business case exists yet in low Earth orbit (LEO). The agency’s new proposal calls for NASA to procure a government-owned “core” module with two docking ports, attach it to the ISS, and require commercial station developers to dock their modules to that core rather than build standalone platforms.

Industry reaction was immediate and largely negative. CLD competitors had been building toward independent free-flying station architectures. The Ignition framework inverted that model, turning private station developers into module suppliers orbiting a government-owned hub. For Vast, Starlab, and Orbital Reef, that is a fundamental architecture mismatch. For Axiom, it is not. It is Tuesday.

The Contract That Already Existed

Axiom Space holds a firm-fixed-price, indefinite-delivery, indefinite-quantity (IDIQ) contract awarded by NASA on February 28, 2020, with a maximum potential value of $140 million, specifically to attach a habitable commercial module to the ISS. No other CLD competitor holds an equivalent ISS-attachment contract. The program has since evolved: NASA requested that Axiom install its Payload, Power, and Thermal Module (PPTM) first in early 2027, with Hab One following approximately nine months later, at which point the two-module combination would undock and become a free-flying station by early 2028.

That trajectory, ISS-attached first, free-flying second, is precisely the Ignition model. The NASA core module procured under Ignition would host commercial modules docked to it; those modules would eventually detach and operate independently. Axiom’s contract with NASA already funds a version of that transition. Where Ignition is a new acquisition framework for Vast, Starlab, and Orbital Reef, it is a ratification of Axiom’s existing program structure.

The competitor response has been frustration. The Commercial Space Federation, representing multiple CLD developers, signaled that companies want program stability, not a new plan. That is a legitimate grievance for operators who built capital structures around Phase 2 assumptions. It is not Axiom’s problem. Axiom’s capital structure was built around ISS-adjacent integration from the start.

Read more

There is a useful distinction, one that has not gotten nearly enough attention since March 24, 2026, between being stranded and being exposed.

Stranded means your capital is gone and your program is cancelled. Exposed means your capital is at risk and your program has changed shape. The second condition is more common right now, and more recoverable, but only if you understand precisely where you sit in the new architecture before the next round of procurement decisions.

NASA’s Ignition announcement restructured the Commercial LEO Destinations program in ways that every tier of the supply chain felt, but not in the same way. The agency formally shifted from certifying commercial operators to run complete station services, the original Phase 2 (full commercial station certification) ambition, toward a model in which NASA owns and operates a core module initially docked to the International Space Station (ISS), with commercial firms building add-on modules that attach to that core. The ISS is currently targeted for decommission in 2030, though NASA has historically extended ISS operational life under pressure. That clock gives this architecture both urgency and brittleness.

For the four active Commercial LEO Destination developers, Vast, Axiom Space, Starlab (the Airbus-Voyager joint venture), and Orbital Reef (Blue Origin and Sierra Space), the Ignition framework represents a forced strategy reset. But the downstream consequence that has received far less coverage is what it means for the supplier tiers below them.

The Two-Class Supplier Problem

The April 15, 2026 piece on this publication mapped the category of exposed suppliers. This analysis goes further: it names the structural mechanism that separates the most exposed firms from the merely disappointed ones.

The central dividing line is the contract vehicle. Firms that hold executed Funded Space Act Agreements for Phase 2 scope have a degree of financial protection, NASA has a documented obligation, and renegotiation at minimum creates a negotiating floor. Firms that accepted letters of intent, teaming agreement invitations, or verbal commitments from Phase 1 prime developers have no such floor. NASA has no binding obligation to those suppliers, and primes who issued teaming invitations ahead of Phase 2 awards have, in most cases, not executed binding subcontracts.

This creates two supplier classes with fundamentally different risk profiles operating under the same program umbrella. Both classes describe themselves as “CLD suppliers.” Only one has a legal instrument that matters.

Read more

The Signal: Precision, Not Brute Force

The FOE’s “Dark Horizons” scenario describes a 2040 environment defined by continuous, hard-to-attribute contestation below the level of declared war, with the space domain “changed through consistent attrition rather than decisive campaigns.” Within that environment, China’s gray-zone toolkit explicitly includes “sub-threshold jamming that mimics natural interference, precise spoofing masked as routine errors, and targeted micro-supply chain disruptions framed as product recalls” that “gradually weaken adversary capability and will.” The core of the tactic is the disguise: a supply disruption that arrives in the inbox as a quality issue, a safety notice, or a routine recall, and routes through commercial processes rather than military channels.

The FOE emphasizes that the cumulative effect of these measures is operational tempo erosion: launch schedules slip, satellite delivery milestones move to the right, and commanders lose confidence in “target custody and node status” amid ambiguous outages and delays. None of these individual events, on its own, looks like an act of war. Collectively, they shift the tempo of a campaign in favor of the actor orchestrating the disruptions, especially when combined with Unrestricted Spectrum Warfare (USW) across the electromagnetic spectrum. The question for any program manager is not whether supply chain interference is a threat vector; it is whether your particular program looks like an attractive target for a “product recall that isn’t.”

The Supply Chain Map: Where the Exposure Lives

The FOE describes China’s broader strategy as a fusion of joint space-military innovation and economic coercion, enabled by tight Party control over ostensibly commercial actors. Military-civil fusion, personally chaired by Xi Jinping, uses Party committees, state ownership, and guidance funds to integrate firms like CGSTL (Jilin-1), China Siwei, and Spacety into state demand for high-tempo Electro-Optical (EO) and Synthetic Aperture Radar (SAR) systems. That same fusion architecture gives Beijing structural visibility into global component flows in which Chinese manufacturers participate, including sectors critical to Western space programs.

Three categories stand out across U.S. defense and space systems.

Read more

The Decisions

The FCC made two distinct moves in May 2026. Neither was routine.

The first was FCC 26-26, released May 1. The order eliminated mandatory Equivalent Power Flux Density (EPFD) limits that had constrained non-geostationary orbit (NGSO) satellite operators since the late 1990s. In their place: voluntary coordination agreements between NGSO and geostationary orbit (GSO) operators. The FCC estimated the shift delivers up to seven times more usable spectrum capacity and over $2 billion in economic benefits -- figures drawn from the Commission’s own economic analysis accompanying the order -- without requiring a single additional launch.

The second move came twelve days later. The FCC approved SpaceX’s acquisition of approximately 65 megahertz of mid-band spectrum from EchoStar Corporation -- AWS-4, AWS H-Block, and unpaired AWS-3 licenses -- in a $17 billion transaction. That spectrum is the foundational layer for SpaceX’s next-generation Direct-to-Device (D2D) network. The transaction is expected to reach final consummation by November 30, 2027, with interim buildout milestones spanning nine years.

The FCC also granted AST SpaceMobile commercial authorization, in April 2026, to deploy 248 satellites providing Supplemental Coverage from Space (SCS), using low-band 700 and 800 MHz spectrum in coordination with Verizon, AT&T, and FirstNet. Half that constellation -- 124 satellites -- must be in orbit by August 2030.

AT&T, T-Mobile, and Verizon then announced plans for a joint venture to build a technology-neutral satellite D2D platform targeting unserved and underserved areas. The three carriers are not simply customers of the new D2D network. They are setting the technical interface standards that every hardware supplier will need to meet.

The Certification Clock

D2D services at commercial scale require hardware that does not exist today in sufficient volume. Consumer handsets authenticating via existing SIM identities to satellite constellations require chipsets, modems, and radio frequency (RF) front-end components that differ materially from current terrestrial 5G hardware.

The FCC has issued a one-year waiver allowing a broader range of end-user devices to connect to SCS services while the Commission finalizes equipment certification requirements. That waiver has a clock. When it expires, every device connecting to D2D networks needs a certified component stack. The FCC has a documented history of extending technical waivers in complex transitions -- that risk is real -- but procurement programs that wait for an extension rather than plan to the stated deadline are accepting schedule exposure they do not need to carry.

Interim D2D services -- messaging and emergency alerts -- could begin as early as late 2026 under SpaceX’s existing T-Mobile partnership, which operates under separate authorization. That capability is distinct from the full commercial D2D network authorized under the EchoStar spectrum acquisition, which cannot launch until after the transaction reaches final consummation on or before November 30, 2027. Component demand for the full network does not arrive in a single wave. It arrives in installments, each tied to a regulatory milestone.

Read more

From Gateway to Ground: What Actually Changed on March 24

The announcement that came out of NASA’s “Ignition” event on March 24, 2026, was framed as a strategic leap forward. Administrator Jared Isaacman told the crowd the agency was redirecting approximately $20 billion over seven years toward a permanent lunar surface base, structured in three deliberate phases, with Gateway “paused in its current form” (NASA, “Moon Base”). What the press release framing obscured is that “paused in its current form” is not a legal category. The Federal Acquisition Regulation (FAR) recognizes termination for convenience, suspension of work, and stop-work orders. It does not recognize “paused,” and that ambiguity is not semantic — it is financial (Ex Terra Media, “’Paused’ Is Not a Contractual Term”).^[1][2]

The Gateway supply chain breaks into three groups with meaningfully different exposure profiles. Hardware that has been delivered and accepted carries limited financial exposure but real strategic exposure, since its repurposing architecture is undefined. Hardware in active integration with no defined repurposing path is accumulating standby costs with no settlement trigger. And international module work under European Space Agency (ESA), Japan Aerospace Exploration Agency (JAXA), and Canadian Space Agency (CSA) Memoranda of Understanding (MOUs) carries no FAR termination provisions at all — renegotiation runs through diplomatic channels, not contract officers. Companies that categorized themselves correctly within thirty days of the March 24 announcement are in a defensible posture. Companies still awaiting guidance from their prime are not.

What NASA Is Actually Buying: The Three-Phase Architecture

The Moon Base program is not one program. It is three procurement architectures operating in sequence, and conflating them is the most expensive mistake a BD team can make (Ex Terra Media, “The $20B Moon Base Procurement Architecture”).

Phase One (now through 2029): Experiment and Learn is the only phase with live procurement instruments as of May 2026. NASA has committed approximately $10 billion and is riding existing contract vehicles: two draft Requests for Proposals (RFPs) live against the Commercial Lunar Payload Services (CLPS) 1.0 Indefinite Delivery, Indefinite Quantity (IDIQ) vehicle, and a Lunar Terrain Vehicle (LTV) Services down-select expected in the second half of 2026. Carlos García-Galán, NASA Moon Base Program Executive, confirmed at the May 27, 2026, briefing that Phase One encompasses 25 launches and 21 landings, targeting delivery of approximately four metric tons of cargo to the lunar south pole surface before 2029 (García-Galán, qtd. in Mack). The May 27, 2026, contract announcement totaling nearly $1 billion is Phase One’s first major visible milestone: $220 million each to Astrolab and Lunar Outpost for LTV design and delivery, a $234 million per-unit delivery contract to Blue Origin using the Blue Moon Mark 1 lander, and a $75 million Firefly Aerospace subcontract awarded through the Jet Propulsion Laboratory (JPL) for lunar drone deployment on the MoonFall mission (Strickland).

Phase Two (2029 to 2032): Establish Early Infrastructure carries another $10 billion and is the phase where international hardware becomes load-bearing. JAXA’s pressurized rover is a confirmed contribution. The acquisition vehicles for Phase Two are, as of late May 2026, blank paper. Two shaping Requests for Information (RFIs) published March 24 and 26 are the only active instruments: SAM.gov Notice 26-01-PS70, titled “Enabling Commercial Lunar Transportation to Support a Sustained Lunar Base,” and a second notice titled “Capability Demonstrations and Supply Chain Challenges for NASA Moon Base Development.” For companies that want to compete in Phase Two, responding to those RFIs is the only lever currently available. Note: all sub-tier supplier analysis for Phase Two constitutes structural inference from Phase One program architecture and the 2023 OIG report, as no acquisition instruments have been released.^[2]

Phase Three (2032 onward): Sustained Human Presence includes the Italian Space Agency’s (ASI) Multi-purpose Habitats, CSA’s Lunar Utility Vehicle, and cargo-capable human landing systems. No U.S. acquisition instruments are attached to Phase Three yet. The technical requirements entering Phase Three will be shaped by whoever responded to Phase One and Two RFIs, making early shaping participation a strategic asset, not an administrative burden (NASA, “Moon Base”). Note: Phase Three sub-tier analysis is structural inference only.

The Supply Chain Map: Who Is Actually Building This

As of May 2026, NASA’s prime contractor network contributing to the broader Moon Base program spans more than 2,700 suppliers across 47 states (Supply Chain Digital; Procurement Magazine). The prime tier includes Aerojet Rocketdyne, Axiom Space, Bechtel, Blue Origin, Boeing, Amentum, Jacobs, Lockheed Martin, Maxar Space Systems, Northrop Grumman, and SpaceX. Lockheed Martin holds the lead contract for the Orion spacecraft. Northrop Grumman built the solid rocket motor boosters for the Space Launch System (SLS). Boeing led SLS core stage production (NASA, “Artemis Partners”).

The Phase One roster introduced a second tier of emerging prime-adjacents. Astrolab and Lunar Outpost are now funded LTV developers with $220 million commitments each. Blue Origin secures its position as the Phase One delivery prime with the Blue Moon Mark 1 lander serving all three initial LTV delivery missions. Three missions originally designated under CLPS have been redesignated as Moon Base Missions 1 through 3: Blue Origin’s Blue Moon Mk.1 targeted for fall 2026, Astrobotic’s Griffin-1 for late 2026, and Intuitive Machines’ IM-3 for late 2026. Firefly Aerospace enters Phase One through the Elytra Dark drone deployment mission, which will test scouting capabilities in 2028 (Strickland; PBS NewsHour).

The sub-tier is where the supply chain map thins rapidly, and where the Office of Inspector General (OIG) data becomes most relevant. The OIG’s 2023 report on NASA’s management of the Artemis supply chain found that “most importantly, the Agency lacks visibility into its critical suppliers with many Artemis programs and projects not tracking their prime contractors’ supply chain impacts” (NASA OIG, IG-24-003 4). Even when subcontractor performance issues were identified, the OIG found they were not shared across Artemis teams to enable effective procurement decisions. NASA’s Logistics Management Division (LMD) was not utilized by Artemis programs or projects at the time of the report, and NASA’s initial Holistic Agency Study found that supply chain issues “are usually a surprise to the agency, with mitigation therefore reactive rather than proactive” (NASA OIG, IG-24-003 7; NASAWatch). The practical meaning for Phase One suppliers: the agency’s institutional supply chain visibility infrastructure was not adequate for Artemis, and Moon Base Phase One is launching before that infrastructure has been measurably rebuilt.

Read more

What This Means

The United States launched the Pax Silica Initiative in December 2025, aligning twelve nations around a shared trusted-vendor framework for semiconductors, artificial intelligence (AI) hardware, critical mineral refining, and advanced manufacturing — the same input categories running through virtually every satellite, launch vehicle, and space-services platform in commercial production today. Under Secretary of State Jacob Helberg described the coalition as equivalent to “what the G7 was to the industrial age,” and directed U.S. diplomats globally to use Pax Silica to identify infrastructure projects and coordinate economic security practices. Space hardware suppliers with China-anchored sourcing are not facing a future regulatory risk. They are accumulating qualification exposure today, against a Q3 2026 timeline that is already in motion.

The Risk

On December 12, 2025, the United States inaugurated the Pax Silica Initiative at a summit in Washington, D.C., bringing together Japan, South Korea, the Netherlands, Israel, Australia, Singapore, the UAE, and the United Kingdom under a shared declaration to build a “secure, prosperous, and innovation-driven silicon supply chain.” The initiative explicitly targets semiconductors, AI hardware, critical mineral refining, advanced manufacturing, and logistics. India joined the coalition in early 2026, and Qatar signed the declaration in January 2026, bringing named membership to eleven nations.

Helberg, speaking at the December 2025 summit, was direct about scope: “This grouping of countries will be to the AI age what the G7 was to the industrial age.” The Trump administration has been equally explicit: export controls, foreign investment screening, and anti-dumping enforcement will be coordinated across member nations. Suppliers outside the coalition’s trusted-vendor framework face increasing friction accessing member-country markets and contracts.

The twelve-month horizon is what matters. Coalition members are currently identifying choke points in the global supply chain and designing joint ventures and co-investment structures to displace them. Suppliers whose inputs or affiliates remain anchored to China without a documented diversification plan are accumulating exposure today that will materialize as qualification risk by end-of-year 2026.

Who Is Exposed

The companies with the most immediate exposure combine two characteristics: significant revenue from U.S. government or allied-nation primes, and a production or sourcing base concentrated in China.

In the commercial space sector, this describes a meaningful slice of the hardware ecosystem. Firms supplying satellite components, particularly radio-frequency (RF) systems, printed circuit board assemblies, power electronics, and electro-optical sensors, frequently source sub-tier inputs from Chinese manufacturers or maintain Chinese affiliates for cost reasons. The same applies to launch vehicle suppliers sourcing carbon fiber composites, precision-machined aluminum structures, and rare-earth permanent magnets used in actuators and reaction wheels.

Companies in the Terran Orbital supply network are a concrete reference point. Terran Orbital, L3Harris, Axiom Space, Safran, and Sierra Space collectively operate hardware procurement programs touching dozens of sub-tier suppliers. Any supplier to these primes that cannot demonstrate a “China plus one” sourcing posture, or that lacks a documented Association of Southeast Asian Nations (ASEAN) diversification plan, faces increasing risk of being downlisted from new program awards as Pax Silica compliance becomes a de facto qualification criterion for U.S.-aligned contracts.

Read more

The Regulatory Door Just Opened

For the better part of a decade, on-orbit refueling occupied a regulatory no-man’s land. Operators could build the hardware. They could plan the missions. What they could not reliably obtain was a clear authorization pathway — a single agency with the authority to say yes to a mission that involved proximity operations, propellant transfer, and docking with another operator’s asset.

That changed on March 24, 2026, when the Office of Space Commerce (OSC) released its proposal for a voluntary Space Commerce Certification (SCC) process. The framework creates a consolidated, interagency review pathway for commercial space activities not clearly governed by existing regulatory regimes — explicitly including satellite servicing, on-orbit refueling, proximity operations, and orbital computing. The proposal is voluntary and opt-in, which means operators can pursue it without being forced to abandon existing licensing relationships, but it provides something more valuable than a mandate: a predictable approval process that institutional customers — including the U.S. Department of Defense (DoD) — can point to when justifying commercial service contracts.

The practical implication is not subtle. If you are a program procurement manager or a BD team building a servicing business, the SCC framework is the bridge between demonstration missions and recurring commercial contracts. The market window is real. The question is who is actually ready to walk through it.

The Four-Mission Crucible: 2026 as the Proving Year

Before any commercial refueling market exists at scale, the technology has to work on orbit. Four U.S. government missions in 2026 are the most concentrated demonstration push the industry has ever seen, and each one will generate supply chain, performance, and pricing data that shapes the first commercial contract cycle.

Astroscale U.S. is the highest-profile mission. Funded by the U.S. Space Force’s Space Systems Command (SSC), the Astroscale U.S. Refueler is manifested for summer 2026 launch and will attempt to conduct the first-ever hydrazine refueling operations above GEO, targeting U.S. Space Force Tetra-5 satellites. The 300-kilogram spacecraft carries a refillable hydrazine tank designed to support two planned refueling operations. This is not a laboratory demonstration — it is a live operational mission supporting the warfighter, and Astroscale has publicly committed to it as the foundation for scalable commercial refueling services.

Orbit Fab is taking a parallel track, pursuing both government and commercial fuel-sale agreements simultaneously. The company’s Rapid Attachable Fluid Transfer Interface (RAFTI) refueling port was approved by the Defense Innovation Unit (DIU) as an accepted standard in 2024 and is already embedded in government procurement language. The DIU has contracted with Orbit Fab for the first in-space fuel sale in GEO orbit under its Refueling and Fuel Depot Initiative (RAPIDS) program. Orbit Fab has also signed the first government and commercial fuel delivery agreements in GEO and secured a European Space Agency (ESA) contract worth approximately $830,000 (≈€750,000) to work with telecom primes on integrating xenon refueling technology into GEO commercial satellites.

The standardization story matters here more than any single contract value. RAFTI is positioning itself as the docking standard the way USB-C positioned itself in consumer electronics — a single interface that commoditizes compatibility and makes every future satellite a potential

refueling customer. Every GEO satellite launched without RAFTI compatibility is a satellite that cannot be refueled by Orbit Fab’s system. That is a supply chain dependency that satellite manufacturers, operators, and investors need to be tracking now, not when the first commercial refueling contract is announced.

Read more

The Signal Nobody Wants to Say Out Loud

Dave Cavossa, president of the Commercial Space Federation (CSF), did not bury the lead at a February 2026 Aerospace Industries Association conference. “I think demand is about to go through the roof — for launch vehicles and space-qualified parts and just satellite equipment in general — in the next two or three years,” he said. “I’m concerned that the industrial base in the United States isn’t ready to support it.” That was February. The situation has not eased.

The 2026 State of the Space Industrial Base report — the seventh annual edition from a consortium of defense and intelligence community advisors — concluded that “technical capability is no longer the primary bottleneck for the U.S. space industrial base.” The bottleneck is alignment: policy, regulation, acquisition, infrastructure, and workforce moving fast enough together to meet what demand is already requiring. When the Pentagon’s own technical advisors say the problem is not technical, that is the signal.

The 2027 Crunch: Why This Date Is Not Arbitrary

Three converging demand signals are compressing against the same constrained supplier base, and they all peak in roughly the same 18-month window.

First: Jeff Schrader, Lockheed Martin’s VP of strategy and business development, stated publicly that the company is “seeing a 632% increase in satellite and space vehicle deliveries” over its long-range plan — a supplier network that spans approximately 13,200 vendors across 52 countries. That is not a rounding error on a slide deck. That is a prime contractor publicly warning that its sub-tier base cannot absorb what its own contracts require. Lockheed is forecasting Space segment sales of $13.5 to $13.8 billion in 2026, with projected growth of approximately 5% year-over-year at the midpoint driven by SDA Transport and Tracking Layer programs.

Second: The Space Development Agency (SDA) is already documenting the practical consequence. Tranche 1 of the Proliferated Warfighter Space Architecture (PWSA) is now a full year behind its original schedule, with Acting SDA Director Dr. Gurpartap Sandhoo attributing the delays explicitly to supply chain bottlenecks for optical communication terminals and encryption devices. The agency has launched 42 Transport Layer satellites but has not yet achieved full optical mesh network connectivity between them, and initiated a strategic pause in additional launches to address on-orbit technical issues.

Third: Space Force’s Space Systems Command (SSC) awarded 20 Other Transaction Authority (OTA) agreements worth up to $3.2 billion to 12 companies in late 2025 and early 2026, announced publicly on April 24, 2026, for Golden Dome SBI prototype development. The demonstration-by-2028 requirement is contractual, not aspirational — and it lands directly on the same radiation-hardened electronics, space-qualified solar panels, and propulsion suppliers already stretched by commercial constellation demand.

The Supply Chain Map: What Specifically Is Breaking

The failure modes are not abstract. Lockheed’s own supply chain leadership named them at the February 2026 AFA conference: on-board processors, mission computers, solar panels, propulsion systems, and optical inter-satellite links. These are not niche components. They are load-bearing elements of nearly every satellite program in the current pipeline.

The OISL terminal market deserves specific attention because it is the single most concentrated sub-tier risk in the current SDA architecture. As of April 2026, the entire SDA-qualified OISL supply base runs through four named providers: Mynaric (now absorbed into Rocket Lab following a $155.3 million acquisition closed April 14, 2026), Tesat-Spacecom US, Skyloom, and CACI. Rocket Lab’s acquisition reduced the count of independent OISL suppliers by one, making Rocket Lab the only listed company that vertically integrates launch, spacecraft, and OISL terminal production. An April 2026 delivery event documented that CACI supplied 21 optical communication terminals to Tesat’s 42 — a 2:1 production gap on the same launch vehicle. Program managers who have not mapped their terminal dependencies since the acquisition closed are operating from an outdated supply chain picture.

The Aerospace Industries Association (AIA) named its own set of structural vulnerabilities in its 2026 Space Priorities release: domestic manufacturing capacity, small business innovation gaps, workforce depth, and supply chain resilience against geopolitical disruption. China retains approximately 90% of global rare earth refining capacity and approximately 70% of natural graphite processing capacity, per the International Energy Agency’s 2024 Critical Minerals Outlook — and China’s April 2025 export restrictions on heavy rare earths and permanent magnets have already triggered documented disruptions across allied defense supply chains. The Trump administration’s tariff posture has added cost volatility on top of an already fragile supplier base, compressing investment decisions by sub-tier manufacturers.

A March 2026 Pentagon industrial base report found that DoD’s efforts to scale production and bring in nontraditional vendors have not yet “moved the needle.” Despite incremental production increases and new investment in critical minerals, the manufacturing base is not assessed as prepared for a potential high-intensity conflict scenario. The Pentagon’s top industrial base official, speaking at Xponential 2026 in May, stated: “We’ve outsourced our manufacturing to our greatest strategic competitor. We’ve become dependent upon multiple countries and companies for things we need vitally. We’ve forgotten how to mine. We’ve forgotten how to process.”

Read more

The Signal That Started This

On February 23, 2026, Starlab Space LLC announced it had completed its Commercial Critical Design Review (CCDR) with NASA in attendance — the 28th milestone on the NASA Space Act Agreement, and the program’s definitive pivot from drawing board to production floor. The CCDR was not a press release milestone. It was a contractual gate: completing it positions Starlab to qualify for a milestone payment from NASA under the CLD program and formally transitions the program from design to manufacturing and systems integration.

That transition has teeth. Vivace Corporation, a New Orleans-based manufacturer, was already under contract before the CCDR closed. It had been tapped in September 2025 to build Starlab’s primary aluminum structure, one of the largest single spaceflight structures developed for launch in the commercial station era. By February 2026, Vivace had begun active production at NASA’s Michoud Assembly Facility (MAF), the same Louisiana facility that built the core stage for the Space Launch System’s Artemis II mission. The hardware clock is running.

The Supply Chain Map

Starlab’s architecture is a deliberate “best-of-class” assembly of proven aerospace integrators. That strategy creates a specific risk profile worth naming by tier.

Primary Structure — Vivace Corporation (Single Source)

Vivace holds the contract to manufacture Starlab’s primary aluminum structure at Michoud. There is no publicly disclosed backup manufacturer for this component. Based on publicly available disclosures, no alternative U.S. facility has been identified with equivalent capability and active clearance for a structure of this scale — though the existence of non-public contingency planning cannot be ruled out. Vivace’s production timeline is the critical path for every downstream integration milestone.

Assembly, Integration, and Testing — Leidos (Single Source, Alabama)

In November 2025, Starlab named Leidos as its primary AI&T provider. Leidos will assemble and integrate all station components, conduct environmental and performance testing, and provide safety and mission assurance — all at a single Alabama facility. There is no second AI&T source disclosed. Leidos also brings Dynetics, its defense-and-space subsidiary, into the integration architecture. The concentration of assembly authority in a single contractor at a single site is a program-execution dependency that carries no disclosed redundancy.

European Design and Engineering — Airbus Defense and Space (Exclusive Partner)

Airbus is not a minority stakeholder in Starlab’s design. It is the exclusive provider of technical design and engineering services for the station and co-owner of Starlab Space GmbH, the European subsidiary based in Bremen, Germany. The Bremen co-location places Starlab adjacent to Airbus’s International Space Station (ISS) Columbus Module infrastructure and the European Service Module team for NASA’s Orion spacecraft. For European Space Agency customers — and for any program that depends on European technical inputs — Airbus is the sole gateway.

Read more

The Slip That Changed the Math

When Vast announced in January 2026 that Haven-1 would move from a May 2026 launch to no earlier than Q1 2027, the company framed it as a natural product of integration discipline, more data, greater certainty, progressively more precise timelines. That is accurate as far as it goes. But the supply chain implications of that framing deserve more scrutiny than they’ve received.

A slip of roughly nine months in a program that entered integration in late 2025 creates a specific kind of problem: the hardware that feeds Haven-1’s final assembly, solar arrays, propulsion units, avionics, docking mechanisms, now has to remain available and unallocated in a procurement environment that is pulling in the opposite direction. The broader commercial and defense space market is not standing still while Vast completes environmental tests. Every month the launch date extends is a month in which competing programs with deeper pockets and longer contract vehicles are absorbing the same constrained component supply.

Vast has publicly confirmed that nearly all of Haven-1’s primary structure is manufactured in-house at its Long Beach, California facility. The company has been explicit that solar arrays and thrusters are outsourced, two of the most capacity-constrained component categories in the current market. What has received less attention is the third layer: docking hardware, avionics, and mission computers. These are not incidental to launch readiness; they determine it.

What’s Outsourced, What’s at Risk

Vast’s vertically integrated build model is one of the company’s genuine differentiators. Founded in 2021, the company scaled to approximately 800 employees and completed primary structure welding in less than two years, a pace that most legacy station developers would consider implausible. The in-house manufacturing posture reduces Vast’s exposure to some of the sub-tier volatility that is currently affecting SDA contractors. But it does not eliminate external dependencies. It concentrates them.

Solar Arrays. Vast has confirmed that solar panels are externally sourced. This is the single most congested category in the current space hardware market. Space Systems Command (SSC) commander Lt. Gen. Philip Garrant identified solar panels explicitly as a cross-cutting supply chain challenge in February 2026, noting that the problem affects Space Rapid Capabilities Office (RCO), SDA, and SSC programs simultaneously. Lockheed Martin’s supply chain leadership echoed the concern, citing availability issues in solar panels alongside on-board processors, propulsion systems, and optical intersatellite links. For Haven-1, the question is not whether solar array supply is tight, it is. The question is whether Vast’s procurement position, secured before or after the January 2026 schedule slip, protected its delivery window or left it exposed.

Read more

There is a quiet paradox running through the modern space economy. It is not technological. It is not financial. It is perceptual.

Many of the most critical companies in the space supply chain are structurally indispensable, yet commercially invisible. They build the components that make missions possible, the subsystems that determine reliability, the materials that define survivability, and the manufacturing processes that quietly set the ceiling for performance. Yet downstream customers often behave as though these suppliers are interchangeable or, more precisely, they behave as though the dependency does not exist at all.

This is the hidden dependency problem. It is not a failure of engineering. It is a failure of brand cognition inside the space value chain.

In traditional terrestrial industries, dependency is often visible because the product itself carries the identity of its upstream inputs. Automotive manufacturers understand the value of Bosch or Denso systems. Aerospace primes recognize the gravity of reliance on specific turbine blade technologies or avionics architectures. In these systems, dependency is documented, discussed, and priced into strategic decisions.

In the space supply chain, however, dependency is frequently diffused across layers of integration, obscured by contracting structures, and absorbed into the identity of the final system integrator. The downstream customer interacts with the prime contractor, not the enabling ecosystem. As a result, the companies that create the highest functional leverage often remain unrecognized as strategic assets.

This creates a structural distortion in brand value.

When a downstream customer does not consciously recognize dependency, they do not assign commensurate strategic weight to the upstream supplier. This affects procurement behavior, pricing power, long-term contracting stability, and the supplier’s ability to shape roadmap decisions. In effect, invisibility becomes a tax on value realization.

The cost of this condition is not abstract. It manifests in four predictable ways across the space supply chain.

First, pricing compression emerges even in high-uniqueness components. Suppliers that deliver mission-critical performance improvements are still evaluated through cost comparison frameworks that assume substitutability. The absence of perceived dependency flattens differentiation.

Second, strategic exclusion occurs during early-stage design conversations. If a downstream customer does not cognitively anchor a supplier as essential, that supplier is not invited into architecture definition discussions where long-term value is created. Instead, they are introduced later as implementers rather than co-definers of capability.

Third, innovation feedback loops weaken. When dependency is not recognized, suppliers receive less upstream intelligence about future mission requirements. This limits their ability to evolve from component providers into system-shaping partners.

Fourth, and most critically, vulnerability is mispriced. Downstream organizations often fail to hedge against supply risk because they do not fully understand where their real dependencies exist. This leads to fragile architectures that appear robust on paper but are concentrated in hidden single points of failure.

The irony is that many upstream companies assume their importance is self-evident. In engineering terms, it often is. In market terms, it rarely is.

The space industry has historically rewarded technical excellence more than narrative visibility. But as the ecosystem matures into a multi-layered industrial supply chain, technical excellence alone is no longer sufficient to secure strategic recognition. Dependency must be made legible.

This does not mean exaggerating importance. It means revealing structural truth.

A space supply chain brand that fails to articulate its position in the dependency graph of a mission is effectively asking the market to infer its value without context. In a complex system, inference is unreliable. Integration obscures causality. And causality is the foundation of perceived indispensability.

The companies that solve the hidden dependency problem do three things differently.

They map their position in the mission architecture, not just the product catalog. They communicate downstream impact in terms of mission outcomes rather than component specifications. And they consistently educate the ecosystem on failure modes that are invisible until they occur.

This is not marketing in the traditional sense. It is system clarification. It is the discipline of making dependency visible before disruption makes it obvious.

There is a deeper strategic implication here for the space supply chain as a whole.

As space commercialization accelerates, the industry is shifting from a procurement-driven model to a dependency-managed model. The most valuable suppliers will not simply be those who produce the best components. They will be those who are recognized as irreplaceable nodes within mission-critical networks.

In that environment, invisibility is not neutrality. It is strategic erosion.

The hidden dependency problem does not mean downstream customers are unaware in an absolute sense. It means they are unaware in a way that influences behavior, allocation, and priority. Awareness without integration into decision frameworks is functionally equivalent to absence.

The task for space supply chain brands, therefore, is not to demand recognition. It is to engineer clarity. It is to make dependency explicit, structured, and impossible to ignore at the architecture level of mission design.

Because in the end, the most powerful position in any supply chain is not simply being essential.

It is being understood as essential before failure forces the realization.

About the Author

Michael Daily is the President of NewSpace Brand Builders, a strategic consultancy dedicated to advancing the branding, marketing, and communications excellence of the global space industry. With an extensive background in brand strategy, public affairs, and community strategy development, Daily established NewSpace Brand Builders to help organizations define their identity, strengthen their market position, and contribute to a sustainable and innovative space ecosystem. You can reach Mike at mike.daily@newspacebb.com or visit https://newspacebrandbuilders.com/

Subscribe now

The Peer-Level Declaration Changes the Stakes

When Isaacman sat before the Senate Commerce Committee in December 2025, he used language that previous NASA administrators had carefully avoided. China, he said, is “our great rival with the will and the means to challenge American exceptionalism across multiple domains, including the high ground of space.” By the April 2026 FY2027 budget hearing, the framing had sharpened further — NASA is prioritizing a lunar base by 2030 specifically because China is targeting the same location, and Isaacman told Congress directly that delay is not an option.

That political urgency creates a supply chain problem that budget lines cannot easily fix. A $624 million lunar base allocation and a 23% overall NASA budget cut happening simultaneously mean the industrial base is being asked to accelerate delivery under resource constraint. When demand compresses timelines, single-source dependencies stop being acceptable long-term risks and become immediate program threats. The space industry is entering that zone now, and it is not evenly prepared.

At a February 2026 industry conference, Lockheed Martin’s Jeff Schrader described a company projecting substantial growth in planned satellite and space vehicle deliveries, managed through a supplier network of approximately 13,200 vendors across 52 countries, as reported by Breaking Defense. That scale of delivery growth is significant. What it conceals is that across those 52 countries, four specific material and component categories carry disproportionate exposure to Chinese-controlled supply and not all of them are being addressed at equivalent speed.

Category 1: Rare Earth Elements — The Processing Chokepoint

The conversation about rare earth dependency has matured significantly in the past twelve months, partly because China forced the issue. In April 2025, China imposed export controls on seven heavy rare earth elements and high-performance permanent magnets, triggering what the Center for Strategic and International Studies (CSIS) described as “rapid disruptions across allied defense and industrial supply chains.” The export controls were later partially suspended through November 2026 as part of a trade negotiation posture, but the underlying structural dependency did not disappear with the suspension.

The dependency is not at the mining level. The United States has rare earth ore. Mountain Pass, California, operated by MP Materials, produces approximately 11.5% of global rare earth oxide output. Lynas Rare Earths, the Australian company building a heavy rare earth processing facility at Seadrift, Texas, signed a $96–137 million supply agreement with the Pentagon in March 2026. USA Rare Earth holds the Round Top deposit in Texas with processing infrastructure in development. These are meaningful developments.

The chokepoint is processing, specifically, the conversion of raw ore into separated oxides, alloys, and high-performance magnets. China commands roughly 90–95% of global rare earth processing capacity. Not mining. Processing. West Point’s Modern War Institute put the magnet production figure at nearly 90%. MP Materials’ July 2025 public-private partnership with the Department of Defense (DoD) a $400 million preferred stock investment, $150 million direct loan, 10-year price floor at $110/kg neodymium-praseodymium (NdPr), and magnet output commitments — represents the most serious U.S. government commitment to closing that gap to date. MP’s 10X magnetics manufacturing campus in Northlake, Texas is expected to break ground under a $200 million incentive package.

For space programs specifically, rare earth elements flow into three critical hardware families: reaction control system actuators and valves, electric propulsion systems using samarium-cobalt and neodymium-iron-boron magnets, and precision pointing mechanisms in satellite attitude control.

The policy response for rare earths is now the most developed of the four categories. MP Materials, Lynas, and the DoD investment structure provide a model, government-anchored offtake agreements, domestic processing investment, and supply contract structure, that the other three categories have not yet replicated. The rare earth response was triggered by a supply disruption. The other three categories are waiting for theirs.

Category 2: Solar Arrays and the Germanium Wafer Problem

Space-grade solar cells are not made from silicon. The highest-efficiency cells used in satellite power systems, triple-junction cells capable of converting 28–32% of incident solar energy, are built on germanium wafer substrates. Germanium wafers give the cell its structural base and bottom-junction, and they are foundational to the products from Spectrolab (a Boeing subsidiary), SolAero Technologies (a Rocket Lab subsidiary since 2022), and Azur Space (German) that supply most NASA, DoD, and commercial satellite programs.

China’s control over germanium is not incidental. In August 2023, China introduced export licensing requirements for gallium and germanium. In December 2024, Beijing banned gallium exports to the United States outright. The ban was suspended in November 2025 until November 27, 2026, but this suspension is explicitly conditional, a negotiating instrument, not a structural resolution. Readers should verify the current status of this suspension, as it is subject to change. A Chinese state-affiliated company, Yunnan Chihong Zinc and Germanium, launched a dedicated Space Solar Cell Germanium Wafer Construction Project in March 2025 with an interim production target of 1.25 million wafers by end of 2025, expanding Chinese domestic capacity even as it controls exports to competitors.

Each standard satellite requires approximately 6,000 to 15,000 germanium wafers for its solar array, with larger satellites requiring tens of thousands, according to Fastmarkets industry demand reporting. Spectrolab and SolAero are U.S.-based cell manufacturers with established DoD program relationships, but the substrate materials feeding their production are sourced at a tier that carries exposure to Chinese-controlled germanium supply.

The qualification pathway for alternative substrate suppliers is not short. U.S.-based germanium processing at space-wafer grade exists in limited form, Umicore operates germanium refining capacity in North America, and 5N Plus (Canadian) holds space-grade germanium wafer capability, but neither has publicly confirmed production volume equivalence to Chinese output for space-grade applications. The U.S. Department of Energy’s Critical Materials Institute has identified germanium wafer domestication as a gap, but no program-of-record commitment at production scale has been confirmed through public filings as of May 2026. The qualification timeline for a new substrate supplier entering an established cell manufacturer’s process is estimated at three to five years based on standard space component qualification precedent.

Program managers sourcing solar arrays for lunar surface systems, where array panels face extended operating cycles and higher radiation environments than low-Earth orbit (LEO) missions, face compounded risk. The material dependency is greatest precisely where the mission environment demands the most from the hardware.

Category 3: RF and Microwave Components — Foundry-Level Exposure

Radio frequency (RF) and microwave components sit at the intersection of space and defense in ways that make supply chain exposure politically sensitive and operationally critical. Satellites relay communications and sensing data through RF chains. Ground terminals communicate through microwave amplifiers. Phased array antennas on next-generation reconnaissance and communications satellites demand monolithic microwave integrated circuits (MMICs) at scale. Mercury Systems, Crane Aerospace and Electronics, API Technologies, and Teledyne Technologies are the primary U.S. suppliers in this tier.

The dependency is not at the MMIC design level, the United States has credible MMIC design capability. The dependency is at the substrate fabrication level. Gallium nitride (GaN) and gallium arsenide (GaAs) are the two dominant substrate materials for high-performance RF MMICs used in space and defense applications. Gallium is the key input. China imposed an outright ban on gallium exports to the United States in December 2024. The ban was suspended through November 2026, but gallium prices responded sharply to the initial announcement and have not returned to pre-2023 baselines. Again, readers should verify the current suspension status before making procurement decisions.

A 2021 Microwave Journal survey of RF GaN fabrication facilities identified eight global providers with 36 process variants. Non-Chinese alternatives exist, primarily through U.S. DARPA-funded foundries (including those supported under the Microelectronics Commons program) and European producers such as United Monolithic Semiconductors (France) and Ferdinand-Braun-Institut (Germany), but qualifying a new foundry for a flight program is a three-to-five year process that program offices rarely initiate without a specific contract trigger. DARPA’s ongoing investment in domestic GaN foundry capacity through programs such as the Technologies for Mixed-Mode Ultra Scaled Integrated Circuits (T-MUSIC) initiative represents the most credible domestic alternative development path, though no production-scale domestic qualification for flight hardware has been publicly confirmed through program-of-record filings as of May 2026.

The risk pattern is a gap between structural awareness that foundry concentration is a problem and the program-level action to begin alternative qualification before a supply disruption forces the issue. Programs that have not initiated that process should assess whether their current gallium inventory and foundry relationships would sustain production through a reimposition of Chinese export controls beyond November 2026.

Read more

The satellite forecast is striking: 16,900 small satellites projected for launch between 2026 and 2035, according to Novaspace’s 11th-edition market report, averaging roughly 640 kilograms of hardware delivered to orbit every single day. That figure alone would stress any specialty components supply chain. But smallsats are only part of the picture, and focusing on the headline number understates the actual demand problem.

In April 2026, Space Systems Command (SSC) announced that the U.S. Space Force had awarded 20 Other Transaction Authority (OTA) contracts worth up to $3.2 billion to 12 companies tasked with prototyping orbital Space-Based Interceptors (SBIs) capable of boost, midcourse, and glide-phase missile engagements by 2028. The program’s goal is to demonstrate initial SBI capability within two years and ultimately deploy a proliferated Low Earth Orbit (pLEO) constellation of interceptors as part of the broader Golden Dome architecture. The 12 awardees include Anduril Industries, Booz Allen Hamilton, General Dynamics Mission Systems, GITAI USA, Lockheed Martin, Northrop Grumman, Quindar, Raytheon, Sci-Tec, SpaceX, True Anomaly, and Turion Space. Each prototype carries power requirements that must be met by space-qualified solar cells from a supplier base already serving the commercial and civil markets — and none of that demand appears in the commercial procurement queues.

Simultaneously, SpaceX’s next-generation Starlink build cycle has evolved beyond a simple satellite refresh. Elon Musk publicly described “Space AI Data Centers” as the optimal solution to Earth’s power and heat dissipation bottlenecks for artificial intelligence computing, identifying two binding constraints for orbital AI infrastructure: power generation and heat rejection. The next-generation Starlink architecture is widely expected to require substantially larger solar arrays per satellite than the current constellation, driven by the power requirements of on-orbit AI processing workloads. Musk’s stated technical requirements, high efficiency, scalability, and mass deployability, are driving SpaceX toward a fundamental restructuring of its solar supply chain, not merely an incremental order increase.

Sovereign constellation programs from Europe, India, the UAE, Japan, and others are adding additional demand layers that were not in most supply-chain forecasts eighteen months ago. The result is a demand curve that is steeper, broader, and less transparent than the headline satellite count suggests.

Who Actually Makes These Cells

Two names dominate every conversation about space-grade solar cells: Spectrolab, a Boeing subsidiary based in Sylmar, California, and AZUR SPACE Solar Power GmbH, a 5N Plus subsidiary headquartered in Heilbronn, Germany. Spectrolab has delivered more than 6.5 million multijunction space solar cells over its lifetime and powers everything from GPS satellites to Mars rovers with triple-junction gallium arsenide (GaAs) cells reaching 33% efficiency with its latest XTE+ product line. AZUR holds a comparable position in Europe and serves many of the same prime contractors, with a product line spanning 3G28, 3G30, and advanced variants used across commercial, civil, and defense constellations.

The rest of the qualified supplier base is thin in ways that matter. SolAero Technologies, acquired by Rocket Lab in January 2022, was characterized at acquisition as one of only two companies in the United States producing high-efficiency, space-grade solar cells — a description from Rocket Lab’s own acquisition announcement that is the most precise public statement available about the structural thinness of the domestic GaAs supply base. SolAero now operates as a Rocket Lab subsidiary, meaning its capacity is not fully available to the open commercial market on equal terms with independent suppliers. CESI in Italy rounds out the European-adjacent supplier base for GaAs cells.

MicroLink Devices, based in Niles, Illinois, occupies a distinct position in the qualified supplier ecosystem. MicroLink produces high-efficiency epitaxially lifted off (ELO) triple-junction cells with AM0 efficiency exceeding 30% and specific power exceeding 2,000 watts per kilogram specifications that make them attractive for mass-sensitive applications where power-to-weight ratio matters more than absolute panel area. MicroLink has an expansion program underway toward a roughly one-megawatt annual facility. To contextualize that ceiling: a single large GEO satellite can require 15 to 30 kilowatts of power, and orbital AI data center satellites, each requiring approximately 100 kilowatts of power based on published estimates of Musk’s stated requirements, would each consume the equivalent of 10% or more of MicroLink’s entire planned annual output. The math on secondary-supplier scale against next-generation demand is not a future problem. It is a present one.

SpaceX has famously departed from the GaAs consensus for Starlink, sourcing silicon-based solar cells from Taiwan Solar Energy Corp (TSEC), Taiwan’s largest photovoltaic manufacturer and a company that simultaneously supplies Tesla’s solar roof tile program, accounting for approximately 30% of TSEC’s annual capacity. TSEC Chairman Liao Kuo-jung has publicly described TSEC as a direct representative of the SpaceX supply chain, and TSEC cells are confirmed to be powering current LEO Starlink satellites. The rationale for the silicon approach was cost: GaAs cells were prohibitively expensive at Starlink’s volume, and SpaceX compensated for silicon’s lower efficiency by deploying significantly larger panels. The consequences of that tradeoff are now reverberating through the supply chain in ways no one fully mapped when the decision was made.

The Capacity Expansion Picture: Real Progress, Insufficient Scale

AZUR has been the most aggressive expander in the qualified GaAs supplier base. After a 35% capacity increase in 2024 and a 30% increase in 2025, the company announced in February 2026 an additional 25% expansion targeting second-half 2026 activation, covering back-end and front-end operations through process optimization and additional automation. The investment specifically included additional metal-organic vapor phase epitaxy (MOVPE) reactors, a detail that reveals where the actual binding constraint sits. MOVPE reactor capacity, not cell assembly throughput, is what ultimately limits how many triple-junction cells AZUR can produce in a given period. Over three years, AZUR will have grown capacity by roughly 118% on a compounding basis. That is a credible operational achievement for a specialty semiconductor manufacturer operating in a capital-intensive, long-lead-time fabrication environment.

The problem is baseline. AZUR’s expansion is being measured in percentage points atop a production base that was already considered a bottleneck before the 2026 demand acceleration. The germanium wafer substrate market, the foundational input for all GaAs triple-junction cells, had a total global value of approximately $125 million in 2024, projected to grow to $138 million in 2025 and $253 million by 2032 at a 10.5% compound annual growth rate. That is a small market for a material that must underpin 16,900 satellites, a proliferated SBI constellation, and multiple sovereign megaconstellations. The entire germanium wafer substrate market for space solar cells is smaller than the annual revenues of a mid-sized defense subcontractor.

Spectrolab has not announced comparable expansion programs at the scale the new demand curve requires, and its revenue base reflects a mature, government-relationship-anchored business model with limited structural incentive to make speculative capital-intensive investments that commercial megaconstellation volumes would demand without long-term take-or-pay commitments in hand. Spectrolab does not publish production capacity figures; the absence of public expansion announcements, against a backdrop of documented demand acceleration, is itself a signal worth building into procurement planning assumptions.

The availability gap is already showing up downstream. Satellite integrators and mission designers have reported significant delays in solar cell procurement, with lead times lengthening and costs rising materially. For small satellite programs operating on fixed-price contracts, this is not a theoretical risk it is an active schedule and margin threat that is compressing the contingency buffers programs depend on when hardware deliveries slip.

The Sub-Tier Nobody Mapped: Germanium, Umicore, and China Germanium

The supply chain risk in space-qualified solar cells does not begin at Spectrolab or AZUR. It begins one level deeper, at the germanium wafer substrate, the semiconductor foundation on which every GaAs triple-junction cell is built.

Germanium’s properties make it nearly irreplaceable for this application: high electron mobility, excellent radiation resistance, and thermal stability in the extreme orbital environment. There is no commercially qualified alternative substrate for space-grade triple-junction GaAs cell production at scale. The germanium wafer must be produced through a crystal-pulling process that requires specialized equipment, tight process controls, and long qualification timelines. New entrants into this sub-tier do not arrive quickly.

The global market for germanium wafer substrates for space solar cells is dominated by two players: Umicore, a Belgian materials technology company with its germanium crystal-pulling facility in Olen, Belgium, and China Germanium, a state-linked Chinese enterprise. Umicore is one of the few facilities in the world capable of pulling dislocation-free germanium ingots at the quality levels required for space-grade cell production, and the company has publicly confirmed its role powering satellites and Mars missions. The Umicore-AZUR relationship is not inferred, it is documented through joint ESA technical studies and a publicly disclosed germanium recycling collaboration in which Umicore brings its recycling process directly to AZUR SPACE installations, with AZUR’s backgrinding waste stream feeding back into Umicore’s germanium supply chain. AZUR’s February 2026 expansion investment in additional MOVPE reactors directly references the epitaxy-level constraint that this sub-tier bottleneck creates.

The China Germanium presence in this sub-tier deserves direct analytical attention. China is the world’s dominant producer of raw germanium, accounting for roughly 60% of global refined output, and China Germanium, as a state-linked enterprise, operates within a supply chain that the U.S. government has explicitly identified as a strategic vulnerability. In a scenario where U.S.-China trade tensions escalate or export controls tighten, a germanium supply disruption would propagate not just through AZUR but through the entire GaAs cell production base. Spectrolab’s germanium sourcing strategy is not publicly disclosed, and that opacity is a risk in itself for any program that has not independently mapped its sub-tier dependencies to this level.

The practical implication for supply-chain leaders is significant: even a fully funded, contract-in-hand procurement from AZUR or Spectrolab does not immunize a program from sub-tier germanium risk. Multi-year allocation agreements for cells are necessary but not sufficient. The actual binding constraint may sit one level below where most procurement teams are looking.

Read more

What This Means

The satellite propulsion market is growing at 16.6% annually — a number that sounds reassuring until you map the actual supplier base. Fewer than a dozen manufacturers can deliver radiation-tested, LEO-qualified thrusters at the scale a 16,900-satellite pipeline demands, and the open-market supplier pool is actively shrinking as constellation operators acquire propulsion companies for internal use. Supply-chain leaders who have not validated secondary sourcing options for every thruster category in their program are carrying a single-source risk they probably haven’t named on any risk register — and the 2028 Golden Dome demonstration deadline is about to make that problem measurably worse.

Who Makes the Thrusters for 16,900 Satellites

Sixteen thousand nine hundred satellites. That’s the Novaspace forecast for small satellite launches through 2035 — a number The Journal of Space Commerce reported on April 29, confirmed by Analysys Mason’s March 2026 projection of 36,528 constellation satellites launching between now and 2034. Pick whichever figure you prefer. Both require a propulsion supply chain that does not currently exist at the scale they imply.

The satellite propulsion market is growing, no question. Research and Markets pegged it at $5.93 billion in 2025, projecting a jump to $6.92 billion in 2026 on a 16.6% compound annual growth rate (CAGR), with a path to $12.22 billion by 2030. That’s a market that looks, by the numbers, like it’s scaling to meet demand. The problem with aggregate market figures is that they tell you nothing about the supplier concentration at the specific sub-tiers that matter — the handful of manufacturers capable of delivering radiation-tested, flight-heritage thrusters qualified for low Earth orbit (LEO) radiation environments, on the production schedules that constellation integrators are now locked into.

The supply side of this equation is thinner than the market growth charts suggest. And it’s getting thinner.

The Signal: Two Numbers Moving in the Wrong Direction

In April 2026, Muon Space acquired Starlight Engines. The Journal of Space Commerce covered the move as a supply chain consolidation signal: Muon Space’s acquisition of Starlight Engines removes an independent propulsion supplier from the open market at the same moment that constellation-scale production demand is reaching program-of-record levels.

This is the pattern to watch. When constellation operators acquire propulsion suppliers rather than sourcing from them, they are making a rational institutional decision — vertical integration hedges their own supply risk. But for every other constellation integrator in the market, each acquisition narrows the open-market supplier pool. The buyer solved their problem. Everyone else’s problem got worse.

Read more

What This Means

Rocket Lab’s $155.3 million acquisition of Mynaric AG, closed April 14, 2026, didn’t just consolidate a supplier — it restructured the competitive map for optical inter-satellite link (OISL) terminals at the exact moment Space Development Agency (SDA) demand is accelerating toward Tranche 2 and Tranche 3 deployments. Four named suppliers constitute the SDA-qualified supply base: Mynaric (now Rocket Lab), Tesat-Spacecom US, Skyloom, and CACI. Program managers and supply-chain leaders who haven’t mapped their terminal dependencies since the acquisition closed are working from an outdated picture.

When Rocket Lab completed its $155.3 million acquisition of Mynaric AG on April 14, 2026, adding the Munich-based firm’s CONDOR Mk3 and HAWK terminal lines to its space systems portfolio, most coverage focused on the deal mechanics. That framing missed the more consequential story: the OISL terminal market — already narrow before the deal — now has a significantly different structure, and the implications run directly into the SDA’s Proliferated Warfighter Space Architecture (PWSA) procurement timeline.

Mynaric began volume production of the CONDOR Mk3 in Q1 2024, but the ramp was not clean. Production delays through mid-2024 were caused by lower-than-expected yields and supplier shortages of key components. Rocket Lab has stated publicly that it plans to expand Mynaric’s production capacity to scale supply for both commercial and government customers. What that expansion looks like in practice — timeline, capital allocation, facility scope — has not been publicly detailed as of publication.

The Four Named Suppliers and What They Carry

The SDA’s terminal supply base, as publicly documented, runs through four named providers: Mynaric (now under Rocket Lab), Tesat-Spacecom US, Skyloom, and CACI. Each carries different risk and capacity profiles.

Read more

When the Arizona Space Commission made its formal case at The Broadmoor in Colorado Springs last week, alongside state delegations from Alaska and Colorado, the optics were easy to read as enthusiasm. State governments showing up at the space industry’s annual gathering to wave their flags, hand out brochures, and pitch their regional workforce to anyone who’ll listen — it’s a familiar pattern at conferences in any industry. The decision-relevant question is whether this time is different. The answer, based on what these three delegations actually argued and what their industrial bases actually contain, is yes. Not dramatically different. But different enough to warrant a sourcing conversation that most prime contractor supply chain teams haven’t had.

The stakes aren’t abstract. Breaking Defense documented in February 2026 a 632% increase in satellite and space vehicle deliveries over Lockheed Martin’s long-range plan, with the company’s multi-tier supplier network already showing documented strain across components including on-board processors, solar panels, propulsion systems, and optical intersatellite links. Congress held hearings on supply chain bottlenecks in early 2026. The Aerospace Industries Association (AIA) released its 2026 Space Priorities calling for supply chain resilience as a named national priority. When the top tier of the industrial base is documented as stretched, the logical response is to formally map what lives in the second and third tiers. That’s where state-level clusters operate.

Read more

The Clock Is Running

The reactor hasn’t been built. The fuel exists in limited supply from a small number of qualified producers. The regulatory clock is already running. And the vendor selection window for America’s first orbital nuclear power system opened in Colorado Springs on April 14, 2026.

When Michael Kratsios, Director of the White House Office of Science and Technology Policy (OSTP), stepped to the podium at the 41st Space Symposium, the space industry braced for the kind of policy address that sounds important and commits to nothing. Instead, Kratsios announced the issuance of National Security and Technology Memorandum-3 (NSTM-3), formally launching the National Initiative for American Space Nuclear Power, with two hard public deadlines: space reactors in orbit by 2028 and a fission surface power system on the lunar surface by 2030.

Those dates are not aspirational. The memorandum implements Executive Order 14369, “Ensuring American Space Superiority,” and instructs the National Aeronautics and Space Administration (NASA), the Department of Energy (DOE), and the Department of Defense (DoD) to coordinate across design, development, testing, and training requirements. It directs all three agencies to leverage private industry resources where possible. For companies positioned in the right supply chain tiers, that instruction is the most commercially consequential sentence in the document.

What the Memo Actually Is

The NSTM-3 architecture tells you what kind of procurement this will be. Within 30 days of issuance, meaning by mid-May 2026, NASA must formally initiate a program to develop a mid-power space reactor, with a lunar Fission Surface Power (FSP) variant ready for launch by 2030 and an option for a nuclear electric propulsion (NEP) demonstration to follow. That 30-day instruction is not a planning milestone. It is a program start trigger. The RFP activity, the teaming conversations, and the regulatory pre-positioning that follows have already begun whether or not they have been publicly announced.

The three-agency structure creates a specific procurement dynamic that narrows the competitive field more sharply than most program offices would prefer to acknowledge publicly. A vendor seeking a prime or critical sub-tier role must satisfy NASA’s performance and schedule requirements, DOE’s nuclear material compliance framework, and DoD’s security requirements simultaneously. That integration requirement eliminates most of the commercial small modular reactor developers now generating investor attention in the terrestrial energy market and concentrates the near-term opportunity among firms with prior space heritage or classified program experience.

The language around private industry leverage is also worth reading in context. NASA and DOE have been developing nuclear space power concepts in-house and through national laboratories since the Kilopower program began in earnest after 2015. NSTM-3 does not dismantle that government-led development track. It adds a commercial lane alongside it. That is structurally similar to how NASA’s Commercial Low Earth Orbit Destinations program sits alongside the International Space Station: the government defines the requirement, sets the safety framework, and acts as anchor customer, while industry owns the infrastructure. For reactor developers, the question is whether they want to compete for a government development contract, a commercial service contract, or both.

Read more

Walk the floor of The Broadmoor during the 41st Space Symposium, and the commercial instinct is to follow the crowds. The commercial space station booths drew the biggest lines. The artificial intelligence and autonomous systems track filled every seat. Gen. Stephen Whiting’s keynote on the United States Space Command’s (USSPACECOM) Year of Integration drew standing-room overflow.

Two exhibits near the back of the hall drew almost none of that attention. A set of elastic-wheel lunar rover tire prototypes from Bridgestone sat beside advanced materials displays, unaccompanied by any announcement or press event. Nearby, the Naval Research Laboratory (NRL) showcased a debris-detection concept called LARADO and a detailed model of its 45-by-100-foot Proximity Operations Laboratory — a simulation facility covering rendezvous, docking, and robotic satellite grappling — to an audience that was largely on its way to the next panel.

Those two booths contained more commercially actionable supply-chain intelligence than most of what appeared in post-Symposium coverage. The reason is structural: the space industry covers missions. Supply chains are harder to frame as a press moment, harder to render as a visual, and populated by companies — Bridgestone’s industrial tire division, defense electronics suppliers, robotics manufacturers — that don’t speak fluent “new space.” So they go uncovered. And the commercial window they represent stays open a little longer for the organizations paying attention.

Three supply chain categories surfaced at the 41st Symposium deserve immediate attention from C-suite executives, supply-chain leaders, and investors tracking the lunar economy. Together, they map the industrial base required to sustain — not just reach — a permanent human and robotic presence on and around the Moon.

Read more

Somewhere in the ground segment of a commercial satellite operator, there is a modem management platform that last received a security patch fourteen months ago. The operator knows about it. Their cybersecurity vendor knows about it. Their insurance underwriter has a note about it somewhere in a risk model.

What is new — and what the 41st Space Symposium made unmistakably clear — is that the U.S. government knows about it too, and is now deciding which commercial operators are trustworthy enough to be woven into national security space architecture.

That is not a future concern. It is the operational context for 2026. USSPACECOM’s self-declared “Year of Integration” is the year the government moves from identifying commercial capabilities to operationalizing them. The entry fee, as Gen. Whiting made explicit in Colorado Springs, includes a cybersecurity posture that meets the Pentagon’s standards — not eventually, but before the next contract cycle. Commercial operators who have treated cybersecurity as an information technology cost center are about to discover it is a business development prerequisite.

The Signal: February 24, 2022, and Why It Still Matters in 2026

At approximately 3:02 a.m. Coordinated Universal Time (UTC) on February 24, 2022 — the same hour Russian forces crossed into Ukraine — a wave of focused, malicious traffic began disabling modems on Viasat’s KA-SAT satellite network. Within hours, tens of thousands of broadband customers across Ukraine and Europe lost service. Ukrainian military communications were disrupted. Wind turbine operators across Germany and Central Europe lost remote monitoring access to roughly 5,800 turbines.

Researchers at SentinelLabs identified the attack vector as “AcidRain,” a purpose-built wiper malware designed to remotely erase vulnerable modems and routers, and attributed it to the ground segment rather than the satellite itself. Viasat’s own incident report independently confirmed that attackers gained access to the satellite management network through a misconfigured virtual private network (VPN) appliance and deployed a destructive payload across the service footprint. It is important to note that while Viasat confirmed the VPN attack vector and the mechanism of modem erasure, Viasat disputed elements of SentinelLabs’ supply-chain framing — these represent two separate confirmations from two separate investigative tracks, and should be read as complementary rather than mutually validating.

Four years later, Gen. Whiting referenced the Viasat attack at the 41st Space Symposium not as cautionary history but as a live operational lesson. USSPACECOM keeps returning to this attack because the structural vulnerabilities it exposed have not been uniformly addressed across the commercial satellite sector.

The Supply Chain Map: Where the Exposure Lives

The Viasat attack is instructive precisely because it targeted a layer most commercial operators treat as secondary: ground segment management infrastructure. Understanding why requires mapping how the commercial satellite supply chain is actually structured.

A commercial satellite operator’s supply chain has at minimum three distinct technical layers, each carrying its own cyber exposure profile. The space segment — the satellite itself — is expensive, visible, and relatively well-hardened. The link segment — the radio frequency (RF) communications path — is regulated and monitored. The ground segment — network operations centers, modem management platforms, gateway infrastructure, and customer-premise equipment — is where cost pressure, vendor fragmentation, and legacy systems concentrate.

The ground segment is also where, as the Viasat attack demonstrated, a state-level adversary can achieve strategic effect without ever touching the satellite. The AcidRain malware required no space-domain expertise. It required access to a misconfigured VPN.

The USSF’s April 2024 Commercial Space Strategy is explicit about this layered exposure, stating that “cybersecurity is a foundational requirement for any commercial provider to be considered for USSF integration” and that providers will be evaluated against NSA, NIST, and DISA standards “across all segments — ground, link, and space.” That three-segment framing matters: the government is not satisfied by satellite hardening alone. Ground infrastructure is in scope.

For operators with fragmented vendor stacks — common among small and mid-tier commercial satellite service providers who assembled ground systems from multiple contractors across multiple years — the compliance challenge is not a single audit. It is a supply chain-level exercise in mapping every point of access, every legacy integration, and every third-party dependency. Note: specific sub-tier vendor cybersecurity posture claims beyond what the USSF Commercial Space Strategy explicitly covers are inferred from program structure and have not been independently verified through primary sources; all such characterizations should be read as structural inference, not confirmed operator-level assessment.

That exercise has not been uniformly conducted. The Apollo Insight wargame, run by USSPACECOM in March 2026 under the direction of Cmdr. Heather Thomas, USSPACECOM’s Commercial Integration Branch Chief, gathered senior leaders from more than 60 commercial companies to simulate a nuclear anti-satellite weapon deployment in orbit and map which commercial technologies could contribute to a response. The exercise surfaced not only technical gaps but decision-making latency issues in government-commercial coordination — and the cyber layer was identified as a concentrated point of vulnerability in any integrated response architecture. These details are reported from the 41st Space Symposium comprehensive report; independent USSPACECOM press release confirmation of specific wargame findings was not available as of publication.

A second tabletop exercise, “Campaigning with Commercial Partners,” is scheduled for June 24, 2026, at The Aerospace Corporation in Colorado Springs, Colorado. Operators and prime contractors who participate will be shaping the government’s expectations for what a cybersecurity-ready commercial partner looks like — before those expectations are codified into contract requirements.

Read more

The Business Case, Stated Plainly

There is a particular kind of credibility that comes only from letting people watch you build a thing. At the 41st Space Symposium in Colorado Springs, Vast set up a live video feed of Haven-1 on the exhibit floor — not an animation, not a scale model, but a real-time stream of a full-scale station module being assembled at their Long Beach, California facility. For three years, the commercial space station market has been the topic of sometimes heated debate in congressional budget hearings, conference rooms, and investor decks. Vast answered by pointing a camera at the hardware.

That hardware represents hundreds of millions in deployed capital and the furthest any commercial entity has advanced toward launching and operating a crewed orbital platform. Haven-1 is targeting a Q1 2027 launch — contingent on clearing environmental testing at the National Aeronautics and Space Administration (NASA)’s Neil Armstrong Test Facility in Sandusky, Ohio, scheduled for later in 2026. How it performs there, and what happens to NASA’s CLD funding structure in the months surrounding that test campaign, will determine whether Haven-1 opens a new chapter in commercial space or becomes a cautionary note in it.

Read more