In a lab in Toronto, a small team is asking a chemistry question that, until recently, would have taken a supercomputer cluster a very long weekend to answer: which novel semiconductor materials, when dropped into contaminated water and hit with sunlight, will rip apart pathogens and organic pollutants without any added power, filters, or chlorine? Xatoms, founded in 2024, is betting that quantum chemistry simulations paired with machine learning can shortcut the search, and that the resulting photocatalysts can clean drinking water in places where the grid is a rumor and a pump is a luxury [BetaKit].
The company's pitch, in plain language, is solar-activated water purification. CEO and co-founder Diana Virgovicova, alongside Kerem Topalismailoglu and Shirley Zhong, is using computational chemistry to discover materials that absorb visible light and catalyze the breakdown of contaminants [Xatoms]. The wedge is the discovery engine itself: rather than synthesize and test thousands of candidate compounds in a wet lab, Xatoms screens them in silico and only makes the ones the model says should work. According to the University of Toronto Entrepreneurship office, the company has secured roughly $3 million to scale that approach [University of Toronto Entrepreneurship].
That $3 million pre-seed is unusually well-decorated for a company barely a year old. Backers include Alexis Ohanian's 776, BoxOne Ventures, BDC, Genesis Ventures, and Quantacet, with angel checks from Joe Gagliese, Evan Kubes, Alex Challans, and Jennifer Francis [Private Capital Journal][Tech Funding News]. Xatoms is also a participant in the 776 Climate Fellowship and the Compute for Climate Fellowship, the latter giving early-stage climate companies access to the kind of computing budget that quantum chemistry workloads tend to eat for breakfast [The Recursive].
Why the bet is interesting
Roughly two billion people drink water from sources that are not safely managed, by the WHO's long-standing count, and the dominant disinfection technologies (chlorination, UV lamps, reverse osmosis) all assume infrastructure: chemicals, electricity, replacement parts, trained operators. A photocatalyst that works on sunlight alone, sprinkled into a tank or coated onto a surface, sidesteps most of that. The catch has always been finding a material that is cheap, stable, non-toxic, and active under visible light rather than UV. That is exactly the search problem that quantum chemistry, run at scale, is suited to.
Here is the back-of-envelope. A single person needs about 20 liters of safe water per day for drinking and basic hygiene. Disinfecting that with UV consumes roughly 40 watt-hours, or about 14.4 kilojoules. A square meter of midday sunlight delivers around 1,000 watts. Even at a generous 2 percent end-to-end photocatalytic efficiency, one square meter of treated surface could in principle sanitize 20 watt-hours of water in about an hour, meaning a coated panel the size of a car hood could cover a household's daily needs with nothing plugged in. The unit economics depend entirely on the cost per square meter of the catalyst and how many cycles it survives before degrading. That is the engineering question Xatoms has to answer, and it is a more honest question than most early climate pitches get to.
The team and the runway
Virgovicova's origin story is unusual for a deep-tech founder. She entered the Stockholm Junior Water Prize circuit as a teenager working on water purification in Slovakia [Markets Insider, 2019] and later moved through the University of Toronto's engineering and entrepreneurship pipeline before founding Xatoms [University of Toronto Engineering Podcast]. Topalismailoglu, the CTO, is, per his own LinkedIn, currently hiring [LinkedIn]. Matt Damon, through his water-focused philanthropy, has been publicly associated with the company alongside Ohanian [BetaKit].
Pre-seed raised | 3 | $M
776 Climate Fellowship | 1 | program
Compute for Climate Fellowship | 1 | program
What bears will say, and what bulls answer
The credible bear case is that photocatalytic water treatment has been an academic field for thirty years and the graveyard is full of titanium dioxide variants that worked beautifully on a microscope slide and never made it into a village. Materials degrade. Real water has silt, organic load, and biofilm that fouls surfaces. Field deployment is a different sport than discovery. Xatoms's most plausible answer is that the bottleneck has historically been the search space, not the physics: if quantum simulation genuinely expands the candidate pool from a few hundred well-studied compounds to tens of thousands of plausible ones, the odds of finding a material that survives contact with reality go up materially. The fellowship with Compute for Climate suggests Xatoms is taking the compute side of that argument seriously [The Recursive].
The other open question is the business model. Photocatalysts can be sold as a consumable, licensed to filter manufacturers, or deployed through NGO and government channels in low-income markets. Each path has very different gross margin and very different sales cycles. The company has not publicly committed to one, which at pre-seed is fine, but it is the question that the next round will turn on.
What to watch
The next twelve months will likely bring a lead-material announcement (the first compound Xatoms believes is ready for field testing), a pilot deployment somewhere the company can show before-and-after pathogen counts on real water, and probably a seed round in the $6M to $12M range to fund synthesis and field work. Hiring is already underway on the technical side [LinkedIn]. If the company can publish a peer-reviewed result showing a novel, AI-discovered photocatalyst outperforming a known baseline on real contaminated water, that is the moment the thesis stops being a deck and starts being a company.
The incumbent Xatoms has to beat is not another startup. It is LifeStraw, the Vestergaard-owned hollow-fiber filter brand that has spent two decades becoming the default answer for off-grid drinking water at NGO procurement desks. LifeStraw is cheap, proven, and shippable in a backpack. A sunlight-activated powder will need to be cheaper per liter over its lifetime, or genuinely easier to deploy at village scale, before the procurement officers switch.