Why ATO matters in 2026
Antimony tin oxide is a transparent conductive oxide that absorbs near-infrared radiation while passing visible light and remaining transparent to radio frequencies. That last property is why automotive OEMs are migrating away from indium tin oxide and metallic silver-stack solar heat control: ITO and silver block 5G, V2X, LiDAR, and radar. ATO does not.
The catch is that ATO performance in glazing is dominated by particle size and shape. A 50 nm spherical ATO particle scatters very little visible light. A 200 nm faceted fragment scatters a lot. The synthesis route decides which one you get.
Two synthesis routes
Top-down ATO is made by ball-milling bulk antimony-doped tin oxide ceramic down to nano size. The process is mature, the equipment is cheap, and the yield is high. The problem is geometry: milling produces broad particle-size distributions with irregular shapes, sharp facets, and surface defects from mechanical fracture. Final particle size in commercial milled ATO sits in the 80-200 nm range with a long tail.
Bottom-up ATO is synthesised from atoms. Antimony and tin precursors react in controlled chemistry that grows the oxide crystal directly at the target size. The process is harder to develop and engineer to scale, but the output is structurally different: narrow particle-size distribution, near-spherical morphology, no mechanically induced surface defects. Final particle size sits in the 20-50 nm range with a tight distribution.
Kriya synthesises ATO bottom-up under proprietary chemistry. The bottom-up ATO deep-dive documents the route and its IP boundary. The 1,000,000 kg/year plant design at Chemelot Industrial Park is engineered for this chemistry, not for milling.
Synthesis comparison
| Synthesis attribute | Bottom-up (Kriya) | Top-down (milled) |
|---|---|---|
| Process | solution-phase atomic growth | mechanical milling of bulk ceramic |
| Particle size (D50) | 20-50 nm | 80-200 nm |
| Size distribution (D90/D10) | narrow, ~2x | broad, ~5-10x with long tail |
| Morphology | near-spherical | irregular, faceted |
| Surface defects | few; clean crystal surface | many; fracture-induced |
| Crystallinity | high, controlled by synthesis | variable; milling can amorphise |
| Antimony doping uniformity | homogeneous | depends on starting material |
Performance: where the gap shows up
The synthesis difference becomes a measurable performance gap in two places: haze and NIR-blocking efficiency at a given visible transmission. The data below is from the Kriya bottom-up ATO deep dive, validated against a representative top-down ATO product commonly used in automotive PVB masterbatch.
| Performance parameter | Bottom-up (Kriya) | Top-down (milled) |
|---|---|---|
| Haze after PVB lamination | <0.3% | >1.7% |
| Automotive haze acceptability | meets OEM spec | fails OEM spec |
| Effectiveness at equal SHGC | 5x at equal solar heat gain | baseline |
| SHGC tuning range | 0.2-0.7 | narrower; scattering-limited |
| VLT at 515 nm (250 W/m² blocked config) | 64.4% | lower at equal NIR block |
| NIR transmission at 1400 nm | 1.8% | higher; requires more loading |
| 5G / RF / LiDAR transparency | 100% | 100% (intrinsic to ATO chemistry) |
| Cost vs ITO solar heat control | ~60% | ~60% if haze acceptable |
The haze gap is the difference between automotive-acceptable and not. OEM windscreen haze specifications typically cap haze at 1.0% or 0.5% depending on safety class. Milled ATO at 1.7% post-lamination cannot meet that spec — which is why milled-ATO solar control has historically lived in architectural and aftermarket window film, not in OEM glazing.
The 5x effectiveness number deserves explanation. At equal solar heat gain coefficient, bottom-up ATO requires roughly one-fifth of the loading needed for milled ATO. That is partly because narrow particle-size distribution puts more particles in the optimal absorption window, and partly because clean crystal surfaces have higher free-carrier mobility, which strengthens NIR absorption per gram.
Dispersion stability
ATO has to live in a PVB masterbatch, a window film coating, or a sol-gel direct coating for months between manufacture and lamination. Stability is the difference between a usable batch and scrap.
Bottom-up nanoparticles disperse more cleanly because the surface chemistry is controlled during synthesis. Surfactant attachment is uniform. Settling rates are low. Re-agglomeration is slow because the particles do not have sharp edges to interlock. Kriya ships dispersions with a six-month minimum guaranteed shelf life under recommended storage conditions, every batch filtered to 0.8 micrometres.
Milled ATO presents larger particles with irregular surfaces. Higher settling rates. More aggressive re-agglomeration. Shorter useful shelf life. The industry work-around has been heavier surfactant loadings — which then complicate downstream chemistry and add cost.
Calculation Model 887: what the numbers translate to
Kriya’s Calculation Model 887 converts ATO performance into cabin temperature and air-conditioning load. With NIR blocked at 250 W/m² the model predicts an interior temperature reduction of 9 °C, an AC power reduction of 35%, and a total solar transmittance (Tts) of 41%. The formula:
Car Interior T (°C) = 1.025 × Ambient T (°C) + 0.036 × Solar Radiation (W/m²) + 8.67
Validation r² is 0.999 against Grundstein et al. field measurements. For premium EV segment vehicles (average 85.6 kWh battery, 442 km WLTP), the AC load reduction translates to up to 16 km of WLTP range extension — or about 8.5 kg of battery savings (~€325 per vehicle) if range is held constant. For compact EVs (average 39.3 kWh, 227 km WLTP), the extension is up to 7 km.
These outcomes are only achievable at automotive-grade haze. Milled ATO cannot deliver them because the haze gate closes the door before the thermal benefit can be collected.
Application-suitability matrix
| Application | Bottom-up | Top-down (milled) |
|---|---|---|
| OEM automotive windscreen / sunroof PVB | qualified at automotive haze spec | fails haze spec |
| EV thermal management (range extension) | up to 16 km premium EV (Model 887) | marginal; loading-limited |
| Aftermarket / retrofit window film | premium grade | commodity grade |
| Architectural glazing | excellent; large area | feasible at architectural haze tolerance |
| Direct sol-gel coating on lightweight polymer | excellent | not viable; particle size too large |
| 5G / V2X / radar transparency | 100% | 100% (chemistry-intrinsic) |
| ITO replacement on cost grounds | ~60% of ITO cost | ~60% if haze acceptable |
Three delivery formats from one platform
The same bottom-up ATO dispersion ships in three formats depending on integration point.
- PVB masterbatch — for mass-market automotive laminated glazing. Integrated by the Tier-1 PVB partner. Volume play, full SHGC tuning range.
- Window film coating — for retrofit and aftermarket. Higher per-square-metre price, faster path to market.
- Sol-gel direct coating — for lightweight polymer glazing on electric vehicles where every kilogram of mass matters. The same ATO chemistry, deposited onto polymer instead of dispersed in PVB.
Manufacturing readiness
The 1,000,000 kg/year masterbatch plant design is engineered, ATEX-compliant, and execution-ready. The location option is a PVB-facility-adjacent site at Chemelot Industrial Park, Geleen, Netherlands. Batch flexibility runs from 40 kg R&D samples to production scale. Quality control covers up to 10 critical-to-quality parameters per batch under ISO 9001:2015 with a Certificate of Analysis on every shipment.
When milled ATO still has a place
Milled ATO remains the right call for cost-driven architectural applications where haze above 1% is tolerated and the loading is high enough to overcome lower per-gram efficiency. It does not have a place in OEM automotive glazing in 2026. The OEMs that are deploying solar heat control at scale have already migrated to bottom-up chemistry, and the supply chain has organised itself around that decision.