Windows That Generate: Transparent Solar and the Building as Power Plant

A window has always had one job in the energy balance of a building: to let heat and light in, and to let conditioned air stay in. It has never, in the ordinary sense, generated anything. Photovoltaic glazing — glass that keeps functioning as glass, transmitting daylight and a view, while also converting some of the sunlight passing through it into electricity — proposes something genuinely different: a facade that is no longer only a boundary the building's energy systems have to work around, but a small part of the system generating that energy in the first place.

This isn't a single technology so much as a competitive field of them, at very different stages of maturity, chasing the same difficult trade-off: transparency and electricity generation pull against each other, because a photovoltaic cell's job is to absorb light, and a window's job is to transmit it. This article looks at what's actually been built — including a 70-story residential tower going up in Miami with 500 photovoltaic-integrated windows on its west facade — and at what the underlying materials research says is realistically achievable, versus what is still, honestly, a promise for the glazing of the future.

"Transparent photovoltaics" and "photovoltaic glazing" cover a family of related but distinct technologies, and it's worth being precise about which one is being discussed, since they differ enormously in how transparent they actually are and how much power they actually produce.

Semi-transparent thin-film and crystalline glazing is the most commercially mature category. Companies like UK-based Polysolar manufacture glass using established photovoltaic materials — cadmium telluride (CdTe) thin film or monocrystalline silicon — deposited or laminated in patterns that leave gaps for light to pass through. Polysolar's own published specifications describe a CdTe product line reaching roughly 12% efficiency at up to 80% light transmission, and a semi-transparent crystalline-silicon line reaching around 20% efficiency but only about 10% light transmission — a direct illustration of the trade-off: the more of the glass a cell covers to boost power output, the less see-through the glass becomes.

Genuinely transparent photovoltaics are a newer and much less mature research category, aiming for the visible transmission of ordinary window glass (70%+) while still harvesting usable energy — typically by capturing wavelengths the human eye can't see (ultraviolet and near-infrared) rather than absorbing visible light at all. The most developed version of this idea is the transparent luminescent solar concentrator (TLSC), pioneered by Richard Lunt's research group at Michigan State University: a transparent coating embedded with molecules that absorb invisible UV and infrared light, re-emit it at a different wavelength, and guide it sideways to solar cells hidden in the window's edge. Lunt's group reported a fully transparent version of this concept in 2013.

Perovskite and organic photovoltaics are a third, still-experimental category increasingly discussed alongside transparent glazing — semiconductor materials that can, in principle, be tuned and layered to trade off transparency against efficiency more favorably than silicon. As of the most recent research reviewed for this article, tandem semi-transparent perovskite cells have reached roughly 12.7% efficiency at 77% transparency in laboratory conditions — the best combination of the two properties reported to date, though still a laboratory result rather than a shipping product.

At the physical level, every photovoltaic technology relies on the same basic mechanism: sunlight striking a semiconductor knocks electrons loose, and if the material is structured correctly (with a junction between differently-doped regions), those freed electrons flow as an electric current rather than simply falling back into place. What differs between a standard opaque solar panel and a transparent one is which part of the sunlight spectrum the material is designed to absorb.

Standard silicon solar cells absorb across most of the visible spectrum, which is exactly why they look black or dark blue and block light — they're designed to capture as much of it as possible. A semi-transparent panel, like Polysolar's products, achieves transparency by literally leaving gaps in the light-absorbing material — either using a thinner, partially light-permeable film, or by patterning discrete solar cells across the glass with clear space between them, similar in principle to a picket fence.

A true transparent luminescent solar concentrator works differently: it lets essentially all visible light pass straight through, and only intercepts the ultraviolet and near-infrared light the human eye can't see anyway. The captured invisible light is absorbed by luminescent molecules embedded in the material, re-emitted at a specific wavelength, and channeled sideways through the glass itself (by total internal reflection, the same optical principle used in fiber optics) to strips of ordinary photovoltaic cells hidden at the edges of the pane. Because it isn't trying to absorb visible light at all, this approach can reach very high transparency — Lunt's group has reported figures around 86% — but at a steep cost in efficiency, since UV and infrared light carry only a fraction of the sun's total usable energy.

At 1428 Brickell in Miami, the 500 photovoltaic-integrated windows on the tower's west-facing facade sit closer to the semi-transparent category than to a fully clear TLSC — the coverage figures reported (roughly 20,000 square feet of "solar backbone" glazing) are consistent with a partially light-transmitting product rather than a near-invisible one, though the project's own materials don't name the specific glazing manufacturer or publish transparency and efficiency figures, which is a genuine gap in the public information about the building.

The appeal is straightforward: building envelopes are enormous. A tower like 1428 Brickell has far more vertical glazed surface area than roof area, and most of that facade currently does nothing for the building's energy balance beyond, at best, being well-insulated and shaded. If even a modest fraction of that surface could generate power without meaningfully compromising the daylight, views, and thermal performance a curtain wall is already expected to provide, the energy math for a building changes — not because any single window generates much electricity, but because there's so much glass to work with.

It also reframes a question this publication returns to often: whether a building can move from consuming resources to participating in a cycle that gives something back. A photovoltaic facade that only offsets the building's own draw is a Regenerate-capability story on its own terms — reducing net consumption. But 1428 Brickell's developers and several of the industry sources reviewed for this article go further, describing power flowing "back to the grid" or a building's excess capacity as something a city's electrical infrastructure could draw on — which starts to look like the Connect capability: a building as a small, distributed node in a city's energy network rather than an isolated consumer. It's worth being honest that this second claim is more aspirational than demonstrated at the facade-glazing scale specifically, even though grid-tied net metering itself — a building feeding surplus rooftop solar back to the grid — is a well-established, everyday practice with conventional panels.

Regenerate. This is the direct, uncomplicated connection. Photovoltaic glazing, at any of the maturity levels discussed here, is architecture generating renewable energy as part of its own fabric rather than through a separate rooftop installation — a facade participating in an energy cycle instead of only consuming from one. This is exactly the kind of resource-cycle thinking the Regenerate capability tracks, and it's the strongest, least speculative claim this article can make.

Connect. The idea of a building sharing surplus energy with its surrounding city — rather than only meeting its own needs — is where this technology starts to reach toward Connect: a building functioning as a node in urban infrastructure and networks, not a sealed object. It's a real and legitimate extension of the Regenerate story, but this article treats it carefully rather than as an accomplished fact: the sources reviewed describe potential grid contribution in aspirational terms (a building "capable of" producing a given output, energy that "can be... fed back to the national grid") rather than documenting an operating building at meaningful net-export scale specifically from transparent or semi-transparent facade glazing. The distinction matters — a building offsetting its own demand and a building meaningfully powering its neighbors are different claims, and conflating them is exactly the kind of unearned leap this publication tries to avoid.

Where the connection is weak: none of the technologies discussed here sense, think, respond, adapt, or grow in any meaningful sense as installed — a photovoltaic window doesn't monitor conditions, make decisions, change its own configuration, or grow. (Some "smart glass" marketing language blurs photovoltaic glazing with electrochromic dynamic tinting, which is a genuinely different, Respond-capability technology; this article is specifically about the energy-generation function, not dynamic shading, and the two shouldn't be conflated even though some manufacturers market them together.)

Technology maturity differs sharply by category, and treating "transparent solar" as one thing would flatten real distinctions worth preserving.

Semi-transparent thin-film/crystalline glazing (Polysolar and similar products): EMERGING. These have been commercially available for over a decade (Polysolar's own materials cite a 2010 product launch), carry 25-year warranties, and have been installed in real projects — but remain a niche product, roughly double the cost of standard glass by industry estimates, with payback periods commonly cited in the 5–20 year range. This is real, deployable technology, not speculation, but it has not become a mainstream building-envelope default.

1428 Brickell specifically: EMERGING, with an unusually large information gap. The tower is real, under construction, and financed (reported completion in 2026) — a genuine building project, not a rendering exercise. But none of the sources reviewed name the photovoltaic glazing's manufacturer, publish its transparency percentage, or specify which underlying technology is being used. More concerning for a publication that tries to fact-check its own sources: the building's own promotional materials contain internally inconsistent energy figures, still unresolved as of this writing. The developer's site describes the system as "capable of producing up to 170,000 kilowatts, or 170 megawatts, of clean energy per year" — confusing power (a rate) with energy (an amount over time) — and cites this as offsetting 915 average U.S. homes. New Atlas, working from the same underlying figures, calculates roughly 170 megawatt-hours per year — equivalent to only about 6 average U.S. homes, a discrepancy of more than two orders of magnitude. This article is not able to state a verified annual energy output for this building, and flags that as a real limitation of the available public information rather than picking whichever number sounds most impressive.

True transparent photovoltaics (TLSC, high-transparency organic/perovskite): EXPERIMENTAL to SPECULATIVE. The efficiency numbers are genuinely low in absolute terms — TLSC technology achieves roughly 86% transparency but under 1% efficiency, meaning a very large area of glass is needed to generate a modest amount of power. The best laboratory combination of transparency and efficiency reported in the sources reviewed here — a tandem semi-transparent perovskite cell at 12.7% efficiency and 77% transparency — is a meaningful research result, but perovskite photovoltaics as a broader category face well-documented, ongoing challenges around long-term stability and degradation under real-world UV and moisture exposure that would need their own dedicated research before any claim about building-scale durability could be made.

The "buildings sharing surplus power with the city" vision specifically: mostly SPECULATIVE at facade scale. Grid-tied net metering is established technology for conventional rooftop solar. But a facade of transparent or semi-transparent glazing generating enough surplus, beyond the building's own substantial energy demand, to meaningfully export to a city grid is not demonstrated by any source reviewed here — it appears in marketing and industry-overview language as a capability or aspiration, not as a documented, metered outcome from an operating building.