Hardware alone is not enough. Without algorithms, quantum computers risk remaining impressive without delivering value.
For generations, the development of new materials has followed a slow and costly rhythm of hypothesis, synthesis, failure, and repetition — a cycle that industries dependent on energy storage, medicine, and sustainable supply chains can no longer afford. Fraunhofer ISC, a German institute of chemical synthesis, and Algorithmiq, a quantum computing firm from Milan, have joined forces to interrupt that rhythm, applying hybrid quantum-classical algorithms to simulate molecular behavior before a single experiment is run. Their collaboration, grounded in Algorithmiq's $2 million Wellcome Leap prize-winning work in drug discovery, asks a question that is no longer merely theoretical: can computation replace years of laboratory trial-and-error with prediction? The answer, it seems, is beginning to take shape.
- Bringing a novel material to market can take a decade or more — an unacceptable timeline for industries racing to find alternatives to scarce resources and expensive medicines.
- Classical supercomputers buckle under the mathematics of simulating quantum molecular behavior, leaving vast regions of the materials landscape effectively unreachable.
- Algorithmiq's hybrid approach — routing the hardest quantum calculations to quantum processors while classical systems handle optimization — sidesteps the limitations of today's noisy, error-prone quantum hardware.
- The Fraunhofer ISC partnership pairs algorithmic sophistication with deep synthesis expertise, aiming to screen out failing candidates digitally before any lab work begins.
- The vision now extends beyond discovery: digital twins that predict how a material will synthesize, perform, and be recycled across its entire lifecycle — without exhaustive physical testing.
- The industry's fixation on qubit counts and processor speed is giving way to a harder, more consequential question: can quantum machines actually solve real problems faster than the tools they are meant to replace?
The problem is as old as chemistry itself: developing a genuinely new material can consume a decade of repeated synthesis, testing, and failure. For industries chasing affordable medicines, efficient batteries, or alternatives to geopolitically fragile raw materials, that pace has become untenable. The central question driving a new partnership between Fraunhofer ISC and Algorithmiq is whether computation can replace much of that trial-and-error with prediction.
Fraunhofer ISC brings decades of hands-on expertise in chemical synthesis and materials characterization. What it has long lacked is a way to eliminate unsuitable candidates before the laboratory work begins. Classical computers can simulate molecular behavior, but the mathematics of quantum effects — how electrons interact across a molecule — demands approximations that erode accuracy and limit reach. Quantum computers, in principle, simulate these effects directly, opening up regions of the materials landscape that classical methods would never explore in reasonable time.
Algorithmiq, based in Milan, has built its reputation not on waiting for perfect quantum hardware, but on making imperfect hardware useful. Its hybrid strategy assigns the genuinely quantum portions of a problem to quantum processors while classical systems handle optimization and analysis — a pragmatic acknowledgment that today's machines are noisy but not useless. In April, the company won a $2 million Wellcome Leap prize for demonstrating this workflow in drug discovery, establishing that the approach yields real scientific advantage now, not in some distant future.
The partnership with Fraunhofer ISC extends that success into materials science. Miriam Unterlass, director of Fraunhofer ISC, describes simulations capable of revealing 'white spots' — compounds no one was explicitly seeking, whose properties could prove invaluable. She envisions a future where digital twins predict synthesis outcomes and recycling potential across a material's entire lifecycle, making the laboratory a place of confirmation rather than exhaustive exploration.
Algorithmiq's CEO Sabrina Maniscalco is direct about where the real challenge lies: hardware alone is not enough. Without advances in algorithms and software, quantum machines risk remaining scientifically impressive while delivering no industrial benefit. The partnership is, at its core, an argument that the quantum computing industry has been asking the wrong question — and that the right one is simply whether these machines can solve real problems faster than anything that came before.
The problem is old and urgent: developing a new material takes years. A chemist proposes a compound, synthesizes it in the lab, tests its properties, finds it falls short, and starts again. Multiply this cycle by dozens or hundreds of experiments, and you understand why bringing a genuinely novel material to market can consume a decade or more. For industries chasing affordable medicines, efficient energy storage, or alternatives to scarce raw materials, this pace is a luxury they cannot afford.
Fraunhofer ISC, a German research institute with deep expertise in chemical synthesis, has long understood this constraint. Classical laboratory work is thorough but slow. What if unsuitable candidates could be eliminated before a single beaker was filled? What if a computer could map the landscape of possible materials and flag the most promising ones for human researchers to pursue? Digital simulation offers that possibility—but the mathematics involved is ferocious. Describing how quantum systems actually behave, how electrons dance and interact across a molecule, demands computational power that even today's supercomputers strain to deliver.
Enter quantum computing. A quantum machine, in theory, could simulate these quantum effects directly, without the brutal approximations classical computers require. The speed advantage could be transformative. Imagine screening for rare-earth-lean magnets—materials that deliver high performance without dependence on geopolitically fragile supply chains. Quantum simulation could accelerate that search dramatically, opening up regions of the materials space that classical methods would never reach in reasonable time.
Algorithmiq, a Milan-based quantum computing firm, has spent years building algorithms designed specifically for chemistry and molecular science. The company does not wait for perfect quantum hardware. Instead, it deploys a hybrid strategy: quantum processors handle the genuinely quantum parts of a problem—the hard quantum effects—while classical computers manage optimization and data analysis. This pragmatic approach acknowledges that today's quantum machines are noisy and error-prone, yet still capable of useful work when paired with classical systems. In April, Algorithmiq won a $2 million Wellcome Leap prize for demonstrating this hybrid workflow in drug discovery, proving the concept could yield real scientific advantage in the near term.
Now the company is extending that success into materials science. Algorithmiq and Fraunhofer ISC have formed a strategic partnership to apply quantum-classical simulation to chemical materials development. The collaboration pairs Algorithmiq's algorithmic sophistication with Fraunhofer ISC's hands-on knowledge of synthesis and materials characterization. Miriam Unterlass, director of Fraunhofer ISC, frames the ambition clearly: simulations can reveal "white spots" in the materials landscape—compounds no one was explicitly seeking but whose properties could prove invaluable. More broadly, the partnership aims to compress the discovery cycle by replacing some of the trial-and-error with prediction.
Unterlass envisions a future in which the digital twin becomes as central to materials development as the laboratory itself. A researcher would not need to trace every experimental step in person. Instead, simulations would predict how a material would synthesize, what properties it would exhibit, and how it could be recycled at end of life. The human creativity that drives innovation would remain essential, but the toolkit would become radically more digital.
Sabrina Maniscalco, CEO and co-founder of Algorithmiq, pushes back against a common misconception: that quantum computing's value lies mainly in the hardware itself. "Hardware alone is not enough," she argues. Without major advances in algorithms and software, quantum computers risk remaining scientifically impressive while delivering no industrial benefit. Algorithmiq is building the algorithmic layer that transforms quantum machines from curiosities into tools—tools for chemistry, for drug discovery, for materials science, for problems that matter.
The partnership represents a shift in how the quantum computing industry thinks about itself. The conversation has dwelt too long on processor speed and qubit count. What matters is whether quantum machines can solve real problems faster than classical alternatives. For materials discovery—a field where time and cost are measured in years and millions of euros—that question is no longer theoretical. The work has begun.
Notable Quotes
Simulations can help identify 'white spots' in the materials space—materials we may not have been explicitly looking for, but whose properties could be highly promising.— Prof. Dr. Miriam Unterlass, director of Fraunhofer ISC
Hardware alone is not enough. Without major advances in algorithms and software, quantum computers risk remaining scientifically impressive without delivering meaningful industrial value.— Prof. Dr. Sabrina Maniscalco, CEO and co-founder of Algorithmiq
The Hearth Conversation Another angle on the story
Why does materials discovery take so long in the first place?
Because you're working mostly blind. You propose a compound based on theory, synthesize it, test it, and if it doesn't work, you start over. There's no shortcut through that cycle—not yet. You need to actually make the material and measure it.
And that's where simulation comes in?
Exactly. If you could predict which candidates are worth synthesizing before you touch a beaker, you'd eliminate most of the dead ends. Classical computers can simulate some of this, but quantum systems are so complex that the math becomes intractable. You hit a wall.
So quantum computers solve that wall?
In theory, yes. A quantum machine can model quantum effects directly. But here's the catch—today's quantum computers are unreliable. They make errors. So Algorithmiq doesn't wait for perfect hardware. They use a hybrid approach: quantum processors do the hard quantum work, classical computers handle the rest.
Has this actually worked?
They won a $2 million prize for demonstrating it in drug discovery. Now they're applying the same logic to materials science, partnering with Fraunhofer ISC, which knows how to actually synthesize things.
What changes if this works at scale?
The entire discovery process becomes faster. You're not replacing the lab—you're replacing the guesswork. Researchers still need creativity and intuition, but they're no longer burning years on experiments that simulation could have ruled out in hours.
And the materials that matter most—the ones we actually need?
That's the real prize. Rare-earth-lean magnets, efficient batteries, alternatives to scarce materials. These aren't academic problems. They're economic and geopolitical. Speed matters.