GreenSolvent Perovskite Ink: DMF/NMP-Free Precursor for Ambient-Air Module Coating

Concept

A qualified perovskite precursor ink whose solvent package is γ-valerolactone (GVL) / 2-methyl tetrahydrofuran (2-MeTHF) / dimethylsulfoxide (DMSO), sold as a replacement for the DMF- and NMP-based inks perovskite coating lines use today. The product is the ink plus a process-of-record — coating window, drying recipe, lead-source hydration spec, and a pre-dissolved passivation additive — qualified for ambient-air operation on large-area lines. The wedge is regulatory: manufacturers cannot scale a process built on reprotoxic DMF/NMP under tightening EU REACH scrutiny, and this ink removes that constraint without giving up module performance. Because the same ink is engineered to crystallize defect-minimized films in ambient air, it also cuts the dry-room/glovebox capital that conventional perovskite coating demands.

One sharpening this iteration: the evidence says a solvent swap alone is not sufficient. Both new ambient-air demonstrations reached their results through complementary chemistry — PEAI passivation plus solvent-additive strategies in the 320 cm² sub-module [1], and a hydrated non-halogenated lead source plus a resurfacing treatment in the all-scalable cells [2]. "Drop-in" therefore has to mean a co-qualified system (solvent blend + lead-source spec + passivation package), not a bare solvent substitute.

Why now

The anchor result stands: 7200 cm² modules at 17.2% total-area steady-state efficiency certified by NREL, produced in ambient air, passing all IEC 61215 standards certified by TÜV Rheinland [0]. Two new independent demonstrations now corroborate the ambient-air, scalable-coating premise:

• Ambient-air sub-module with LCA backing. A 320 cm² sub-module (201 cm² aperture, 93% GFF) was fabricated in ambient air by hybrid slot-die/blade meniscus coating with air and green antisolvent quenching, reaching 16.13% efficiency and >3000 h ISOS-D-1 dark-storage stability, with the sustainability of methods and materials demonstrated by life-cycle assessment [1]. The LCA matters commercially: it gives module makers a documented eco-profile for the green claim, not just a solvent substitution.
• All-scalable ambient-air cells with moisture tolerance. Using only slot-die and blade coating in ambient air, MAPbI3 devices reached 19.91% (rigid) and 17.4% (flexible), with oxygen/moisture tolerance (ISOS-D-1) and stable maximum-power-point operation at ~20% RH without encapsulation [2]. Robustness came from a hydrated non-halogenated lead source — evidence that humidity sensitivity is an ink-chemistry problem that can be engineered out, which partially answers the open question on site-to-site variability (with the caveat that MAPbI3 is not the FA-rich composition most module lines run).

Honest bounds on the claim:

• Scale-up losses persist. The 320 cm² sub-module still gives up ~7% efficiency relative to its small-area cell [1]; the ink's defect-minimization claim should be read as narrowing the lab-to-line gap, not eliminating it.
• Competing green routes exist. The sub-module work used a green antisolvent quench step [1] rather than a one-step solvent-engineered ink. The GVL/2-MeTHF/DMSO system's differentiation is a single qualified solvent package (2-MeTHF can serve the antisolvent function within the blend), but the process-of-record should be qualified to work with an air/green-antisolvent quench where line conditions require it, rather than assuming one-step coating always suffices.
• No conflicting evidence emerged this iteration, but the commercially decisive questions remain open: shelf-life of the green blend vs DMF/NMP, evaporation-rate and boiling-point profile vs line throughput, cost-per-liter delta, VOC-abatement compatibility with existing ovens, and whether REACH scrutiny eventually extends to GVL or 2-MeTHF. These are the qualification workstreams for the next 12 months, not reasons to wait.