America has decided to build its own solar supply chain, and it is not being quiet about it. Tesla filed paperwork this month for a $10.1 billion cell and module campus outside Houston. A presidential proclamation signed on August 6 drops a hard price floor under every imported panel starting December 4.<\/p>\n
Europe ran this play first. The result is a building outside Catania, Sicily, roughly 2.15 million square feet of it (200,000 square meters), tooled to turn out 14,000 solar modules and 800,000 cells a day.<\/p>\n
The equipment is all installed. The line is running. It has never run full.<\/p>\n
That last sentence is the part worth carrying into any conversation about American solar factories. 3SUN, the Enel company that runs the plant<\/a>, has the technology, about a billion euros of investment behind it and an Italian tax code that currently makes its modules the only realistic option for certain state incentives. None of that has filled the hall.<\/p>\n The site sits in the Etna Valley, the semiconductor cluster southwest of Catania, and it does both halves of the job under one roof: cells and modules. That vertical integration is the whole point, and it is rare in Europe.<\/p>\n The numbers come from Marina Foti, Head of Advanced Technology Development, R&D and Technology Transfer at 3SUN, who laid out the ramp at a technical conference covered by TaiyangNews<\/a> in March. Roughly 200,000 square meters of floor. About 800,000 cells and 14,000 modules of daily capacity. A design target of 3 GW a year.<\/p>\n Total investment runs to about \u20ac1 billion, of which roughly \u20ac200 million came through European and Italian public programs. Enel’s own corporate page puts the 3 GW target at around 15,000 modules a day and lists 600 employees, on the way to a stated goal of 1,000 direct jobs.<\/p>\n Foti’s own account was the honest part. All the equipment has been commissioned, and the plant has not moved to full-capacity operation, because of where the market currently is. Enel was still advertising line-operator posts in Catania as recently as this spring.<\/p>\n So the 3 GW figure is what the building can do on paper. It is not what came off the line last month, and anyone quoting it as output is quoting a blueprint.<\/p>\n Most of the world’s solar cells are TOPCon. 3SUN builds heterojunction, which sandwiches a crystalline silicon wafer between thin layers of amorphous silicon, and the manufacturing differences are not cosmetic.<\/p>\n HJT runs below 200\u00b0C. TOPCon processes can go past 800\u00b0C. HJT takes around six process steps where TOPCon takes roughly fourteen. Fewer steps and lower heat mean tighter process control and more automation, which is exactly what you want when your labor is European and your competitor’s is not.<\/p>\n The cells are symmetric, which lets you go thinner on the wafer and use less silicon. 3SUN’s cell efficiency sits above 24%, and its B60 module runs 610 to 640 watts with bifaciality up to 90%. A bigger B66 with 132 cells and up to 730 watts has been slated for this year.<\/p>\n The catch is silver. HJT has always eaten more of it than TOPCon, along with indium in the transparent conductive layer, and silver is not getting cheaper. The industry response is copper: at a TaiyangNews equipment conference on August 25<\/a>, DKEM’s Cong Chen put silver-coated copper paste for HJT at 3.4 mg per watt. 3SUN is working the same problem from silver-copper pastes toward straight copper metallization and indium-free coatings.<\/p>\n Good technology, then. Which has never been the thing that kills a European solar factory.<\/p>\n Meyer Burger was the European HJT champion, the company whose SmartWire designs got licensed around the industry. It opened a 1.5 GW module plant in Goodyear, Arizona in 2024, the only HJT manufacturer in the United States.<\/p>\n Then D. E. Shaw Renewable Investments terminated a supply agreement covering up to 5 GW in November 2024, and the whole thing came apart fast. The German subsidiaries filed for insolvency on May 31, 2025. Arizona production stopped the same day and the US workforce was let go. Chapter 11 followed on June 25, 2025, and the assets were sold off inside 90 days.<\/p>\n American appetite for the technology did not die with it. TOYO and SEG Solar have both pushed HJT cell and module plants forward in the US this year. But the cautionary tale is sitting right there: the technical case for HJT was never the problem, and it did not save anybody.<\/p>\n Here is where Catania stops looking like a factory story and starts looking like a policy one.<\/p>\n A 2023 Italian decree set up a register at ENEA sorting EU-made modules into three categories. Category A covers EU-made modules at 21.5% module efficiency or better. Category B needs EU cells and modules at 23.5% cell efficiency. Category C requires bifacial silicon heterojunction or tandem cells made in the EU, at 24% cell efficiency or better.<\/p>\n An amendment in December 2025 knocked category A out of Italy’s 2026 super-depreciation benefit, leaving only B and C. Category B currently has no modules registered at all. Category C is nearly empty of anything actually in production, because Meyer Burger is still listed and no longer manufacturing.<\/p>\n Which leaves, in practice, one supplier. Eleven European manufacturers filed a complaint with EU institutions objecting to what they called a possible de facto monopoly, according to Italian energy outlet QualEnergia<\/a>, which has tracked the rule since it appeared.<\/p>\n No EU body has ruled on that complaint, and no authority has found 3SUN or Enel in violation of anything. The rules were written by the Italian legislature, not by the company that benefits from them. But it is worth sitting with the arithmetic: a factory with a legally protected home market still has not filled its building.<\/p>\n The reason anyone outside Italy pays attention to Catania is perovskite. 3SUN stacks a perovskite top cell on an HJT bottom cell in a two-terminal design, developed jointly with France’s CEA at the INES campus.<\/p>\n The lab results are real. CEA and 3SUN certified 30.8% efficiency<\/a> in January 2025 on a 9 cm\u00b2 cell, up from 28.4% a year earlier. Most tandem records get set on 1 cm\u00b2, so the larger area matters for anyone who wants to eventually build these by the million.<\/p>\n The timeline is the part that has moved. Enel Green Power was once talking about offering tandem modules from 2026. The current target, per Foti’s March presentation, is initial industrial deployment before the end of the decade, with the technology moving from lab cells onto a pilot line running full wafer formats.<\/p>\n The wider industry roadmap agrees with the slower version. Presenting the latest ITRPV findings at that same August conference, Markus Fischer projected tandem reaching about a 15% share of the market by 2036, at mass-production module efficiencies around 31%. HJT and TOPCon are still doing the work in the meantime.<\/p>\n So Catania is not shipping tandem panels. It is running heterojunction on the main line and perovskite on a pilot next door, and telling you honestly which is which.<\/p>\n Tesla’s filing with the Texas Comptroller, signed July 22 and surfacing publicly on August 6, asks for a ten-year property tax limitation on a 3,050-acre site near Richmond. The capital number is $10.116 billion, split into about $1.5 billion of real property and $8.6 billion of equipment, with 9,712 permanent jobs projected and commercial production targeted for the first quarter of 2029.<\/p>\n The application does not state a nameplate capacity. Joe Hennessy, market research analyst at PV Tech Research, called it as big as anything the sector has seen<\/a> for a single site, and estimated it could clear 10 GW. Tesla says it is still weighing a competing out-of-state location.<\/p>\n The gap this is meant to close is real. US module assembly capacity was running around 60 GW in early 2026 while domestic cell capacity sat under 15 GW, which is why the December 4 proclamation sets separate floors at $0.22 a watt for cells and $0.38 for modules, plus a 15% tariff on the derivative products.<\/p>\n Catania is what the far end of that road looks like when the building goes up before the demand does. Sicily has the vertical integration Texas is proposing, the technology Arizona lost, roughly \u20ac1 billion spent, and a national tax code that currently steers buyers toward its modules and almost nobody else’s.<\/p>\n It still has not filled the hall. The building was never the hard part.<\/p>\n","protected":false},"excerpt":{"rendered":" America has decided to build its own solar supply chain, and it is not being quiet about it. Tesla filed … <\/p>\nThe plant is tooled for 14,000 modules a day<\/h2>\n
Heterojunction is a genuinely different way to build a cell<\/h2>\n
The last company that bet everything on heterojunction is gone<\/h2>\n
Italy wrote the factory into its own tax code<\/h2>\n
The tandem cell is still a pilot line in the same hall<\/h2>\n
Texas is proposing to spend ten times as much<\/h2>\n