Making qubits is hard and manufacturing them even harder

June 15, 2026

The upcoming challenges in qubit manufacturing.

The quantum computing race used to be framed as a physics problem: more qubits, higher fidelity … the framing is clear. But the battleground has moved to the fab: can we make thousands of qubits, reproducibly, on a wafer (or elsewhere), with a supply chain we control?

Written by Eric Mounier, PhD, Chief Analyst, Photonics at Yole Group, this is a “standard” manufacturing story for an emerging tech: front-end, packaging, assembly and test, and an important part of Yole Group’s upcoming report: Quantum Technologies 2026.

Front-end: each modality, its own process zoo!

Quantum is not like standard CMOS, and there is no single “quantum process flow”. Every qubit modality has its own challenges:

Superconducting qubits live or die by the Josephson junction: a tunnel barrier a few atoms thick, historically made by double-angle shadow evaporation and lift-off (more a lab technique than a fab technique). Junction uniformity sets qubit frequency reproducibility, and a few percent of spread ruins a multi-qubit processor. Also, two-level-system (TLS) defects in oxides and interfaces drain coherence. Process mastery here means ultra-clean superconducting metals, controlled oxidation, and lift-off-free junction processes that transfer to industrial toolsets.
Silicon spin qubits are the closest cousins to CMOS and run on 300 mm lines (imec, CEA-Leti, Intel). But “close” is not CMOS: they need isotopically enriched silicon-28 and uniformity far beyond transistor specs. One example is Quobly, which partners with STMicroelectronics at Crolles, with SOITEC supplying wafers, or Equal1 (Ireland) with TNO for SiGe qubits.
Photonic qubits leverage maturing Si photonics most directly, e.g., PsiQuantum and Xanadu with GlobalFoundries on 300 mm. Yet the standard PDK is not enough: ultra-low-loss waveguides, new switch materials like barium titanate, and superconducting single-photon detectors (SNSPDs) that no standard photonics flow incorporates. BTO also requires specific toolsets that need extensive development.
Trapped-ion chips and neutral-atom platform scaling means integrating photonics and specific packaging. For these two modalities, photonic assembly poses critical challenges.

Why this is not “only” standard CMOS or Si photonics

There are at least three structural differences.

First are materials: superconductors, exotic oxides, indium are often unwelcome contaminants in a CMOS fab.
Second is metrology. A chip can pass every room-temperature inspection and still fail at 20 millikelvin, because the killer defects are invisible to in-line metrology.
Last is economics: quantum volumes are wafers per month, not per day. Lithography nodes are rarely the issue here, but rather interfaces, losses, very low yield, throughput, and reproducibility.

Finding the right foundry partner is a real CHALLENGE

Academic cleanrooms are excellent and offer flexibility but unfortunately are not designed for volume.

RTO pilot lines (imec, CEA-Leti, VTT, Fraunhofer for superconducting pilot lines) bridge part of the gap. Commercially, the list of foundries willing to run superconducting metals (or any other exotic material) at low volumes with quantum-grade process control is remarkably short, which is why GlobalFoundries’ new Quantum Technology Solutions business unit and SkyWater’s quantum-foundry positioning matter so much.

The choice of secure foundry is now as strategic as the choice of qubit modality.

Securing the supply chain, the strong IonQ–SkyWater signal

The clearest evidence that manufacturing is the new strategic high ground: IonQ announced a $1.8B acquisition of SkyWater Technology, internalizing a trusted U.S. foundry with advanced packaging to become a vertically integrated quantum platform company. So what is the optimal route: acquisition (IonQ–SkyWater), deep partnership (PsiQuantum–GlobalFoundries, Infineon fab at Villach-Oxford Ionics), or in-house fab (Rigetti, IQM)?

Every serious player must answer the same question: who controls my qubit supply chain?

Add sovereignty demands from government customers, and captive manufacturing becomes a competitive moat. This is where Yole Group adds value: our teams help players structure and secure these supply chains, and translate the complexity of quantum into clear, actionable insight for investors and industrials alike.

Packaging, the silent differentiator

Connecting qubits, assembling QPUs, connecting the electronic control chips …: all these approaches are packaging-related, a topic Yole Group has been involved in for more than a decade.

Superconducting QPUs require flip-chip packaging with superconducting indium bumps, through-silicon vias (TSV), and multi-chip stacks. Thousands of microwave lines into a dilution refrigerator demand cryo-flex cabling and, eventually, cryo-CMOS control.

Trapped-ion packages must hold sub-micron optical alignment through ultra-high-vacuum bakes. A new kind of player is emerging around exactly this challenge: QuantWare, for instance, has built its entire business as a dedicated quantum packaging foundry, and its VIO architecture, delivering control signals vertically through a stack of chiplets rather than crowding them across a 2D surface, is an original, packaging-first answer to the qubit I/O bottleneck.

Packaging and assembly into a final form factor that would obviate the need for the quantum computer to be the size of Central Park, NY would also decide quantum manufacturing leadership.

Testing, the hidden cost of cold

This is probably the most underestimated challenge and something we try to avoid discussing now…

Coherence and fidelity only exist at cryogenic temperatures, and a cooldown takes days. The answers are emerging today: room-temperature proxies (junction resistance mapping), cryogenic wafer probers (Bluefors with Afore probing 300 mm wafers down to 1 K, FormFactor’s 4 K systems cooling in hours instead of days), and characterization stacks from Zurich Instruments, Quantum Machines, Qblox, and Keysight. But a true “quantum ATE” industry does not exist yet. It will have to. Mapping this emerging test and metrology landscape is precisely where Yole Group’s expertise comes in, a topic we have followed across the semiconductor industry for decades, and it is a core focus of our upcoming quantum report.

The emerging quantum tool ecosystem

A specialized equipment base is forming anyway: Plassys-Bestek (UHV deposition for Josephson junctions), Oxford Instruments (plasma processes and cryogenics), Bluefors (dilution refrigerators), FormFactor and Afore (cryogenic probing), Delft Circuits (cryogenic I/O). Tracking how this equipment base consolidates is long-standing Yole Group territory, from CMOS to MEMS and photonics, and now quantum, and Yole Group’s report maps who is positioned to become the Applied Materials or ASML of the field.

But, although they have exceptional technologies, these companies are not yet to quantum what Applied Materials and ASML are to CMOS.

What comes next?

Qubit counts will keep making headlines. But the winners will also master oxidation chambers, indium bumps, cryogenic probe cards … and secure the fabs to run them.

In September 2026, we will release our new2026 Yole Group quantum report. As quantum is now in a strong engineering phase, we will examine challenges for front-end processes, the foundry and packaging landscape, supply chain strategies, testing, and the equipment ecosystem.

For almost three decades, Yole Group has decoded emerging technologies where it matters most: at the intersection of markets and manufacturing. Analysts have seen this inflection point before, in MEMS, in photonics, in advanced packaging. Now it’s quantum’s turn.

Stay tuned on www.yolegroup.com!

About the author

Eric Mounier, PhD, is Chief Analyst, Photonics at Yole Group.

With more than 30 years’ experience in the semiconductor industry, Eric provides daily in-depth insights into emerging semiconductor technologies, including quantum technologies, the Metaverse, terahertz, photonics, and sensing.

Based on relevant methodological expertise and a significant technological background, Eric works closely with all of Yole Group’s teams to highlight disruptive technologies and analyze business opportunities through technology & market reports and custom consulting projects.

Eric has spoken at numerous international conferences, presenting Yole Group’s vision of emerging semiconductor technologies, markets, and applications.

Previously, Eric held R&D and Marketing positions at CEA-Leti (France).

Eric Mounier has a PhD in Semiconductor Engineering and a degree in Optoelectronics from the National Polytechnic Institute of Grenoble (France).

EPIC Technology Meeting on Photonics for Quantum Technologies 2026

Yole Group will be attending the event to share insights on the evolving quantum and photonics landscape. Join us to discuss market trends, technology developments, and emerging opportunities shaping the future of quantum technologies. We look forward to connecting with industry leaders and exploring new collaborations.

Yole Group’s speaker: Eric Mounier, PhD Chief Analyst, Photonics at Yole Group