Wentao Zhang
I grow thin films one atomic layer at a time and build the models that explain how they behave. PhD candidate in Chemical Engineering at Yale.
≈ 1,400 ALD cyclesat 0.05 nm per cycle
The color of a silicon wafer under a transparent film, computed from thin-film interference for thicknesses from 0 to 300 nm. Cleanrooms keep a chart like this on the wall for reading oxide thickness by eye. TiO2 and Al2O3 are films I grow by atomic layer deposition; SiO2 is the thermal oxide the wall charts were made for. The plot shows how much light the wafer reflects at each wavelength, and the color is what the dips leave behind. Drag along the strip, or use the arrow keys, to pick a thickness; the address bar keeps it, so a copied link opens on the same color.
About
I am a PhD candidate in Chemical Engineering at Yale University, in Shu Hu’s Photocatalysis and Reaction Engineering Lab, which I joined in 2021. My work sits between semiconductor processing and electrochemistry. I develop protective and passivating coatings for III–V semiconductors such as InP and GaInP, do the nanofabrication that turns them into devices, and write the transport models that connect a recipe to a result.
Before Yale I studied chemical engineering at the Cooper Union and at Columbia University, where my research followed ion transport in membrane-coated electrocatalysts.
What I work on
- Atomic layer deposition
- Process development for TiO2, Al2O3 and GaN films and multilayer stacks, tuned for passivation and interface chemistry. I also run and maintain the group’s plasma ALD system.
- Nanofabrication
- Electron-beam and photolithography, reactive ion etching and PECVD for sub-100 nm III–V structures, checked with SEM/EDX, XPS, AFM and ellipsometry.
- Modeling
- COMSOL and ANSYS Fluent models of transport, reaction and plasma, used to plan experiments and to explain their results. One example is the coupled transport and pore-diffusion model in our paper on electrified polyethylene pyrolysis.
- Solar fuels
- Photoelectrochemistry and photocatalysis on coated semiconductors, from water splitting to turning dissolved carbon into fuels.
Publications
- Enhanced Photocatalytic Charge Separation in Coating-Protected Porous Silicon Quantum Dot Films With Stability Approaching 100 Hours Advanced Functional Materials, e02774 (2026)
- Selective Electrified Polyethylene Upcycling by Pore-Modulated Pyrolysis Nature Chemical Engineering 2, 424–435 (2025). doi:10.1038/s44286-025-00248-0
- Solar-Driven Selective Conversion of Millimolar Dissolved Carbon to Fuels with Molecular Flux Generation Nature Communications 16, 1558 (2025). doi:10.1038/s41467-025-56106-3
- Models and Measurements Quantify Photon Recycling, Charge-Carrier Diffusion, and Photon Scattering Contributions to Photoluminescence in InP Nanowire Arrays Journal of Physical Chemistry C 129, 8270–8283 (2025). doi:10.1021/acs.jpcc.5c01618
- Photoemission Study of GaN Passivation Layers and Band Alignment at GaInP(100) Heterointerfaces ACS Applied Materials & Interfaces 17(4), 7087–7097 (2025). doi:10.1021/acsami.4c17453
† equal contribution * corresponding author
Education
- Yale UniversityPh.D. candidate, Chemical Engineering, 2021 to present
- Columbia UniversityM.S., Chemical Engineering
- The Cooper UnionB.Eng., Chemical Engineering
Outside the lab
Burgundy first, then Napa, with detours through Bordeaux. I organize blind tastings with friends, everyone bringing a bottle, and keep my tasting notes, 175 so far, on CellarTracker. Sake when I can find good sake, and beer, which I check in on Untappd.
- In the glass
- 214 bottles since 2020: 97 from Burgundy, half of them from the Côte de Nuits, 50 from Napa and 18 from Bordeaux.
- In the cellar
- 85 bottles, counting those still on the way, and two in three from California, mostly Napa cabernet: Realm Cellars, and six vintages of Fait-Main’s Beckstoffer Las Piedras. Twenty from Burgundy, from Bourgogne rouge up to de Vogüé’s Musigny.
- Tastings
- Ten Clos Vougeots from ten producers, 1989 to 2020, in one evening; thirteen Burgundies from 1978 to 2021 in another. Realm and Fait-Main side by side, double-decanted six hours ahead. Opus One, Promontory and Stag’s Leap Wine Cellars at the wineries.
- Highest marks
- 97 for the 2017 Promontory and the 2019 Fait-Main Las Piedras; 96 for the 1996 and 2008 Latour; 95, and wine of the night, for the 1993 de Vogüé Musigny.
Winter weekends go to skiing. I track every day on the mountain with Slopes, where you can add me as a friend.
A few days from that log. My biggest day is still Copper Mountain in late November 2019: 18 runs and 27,306 vertical feet. At Breckenridge in 2022 I got above 12,800 feet, and at Jackson Hole on the second day of 2023 the Aerial Tram took me 4,107 feet up in one ride. Big Sky has the other two: my top speed, 53.1 mph, in 2025, and the Big Couloir off Lone Peak in February 2026.
滚落线Fall Line Slope angle, 20° to 50°+ A 3D ski map I built after FATMAP went dark. It shades every slope by its angle and the way it faces, computed in your browser from open elevation data, draws runs and lifts for ski areas around the world, replays GPX tracks and keeps working offline on the mountain. In Chinese and English. Open the map 两清All Square Who owes whom A bill splitter I built for group trips. Log the house, the lift tickets and the gas with who paid and who shares each, and it works out the fewest transfers that settle everyone up. Share the ledger as a link and the whole group keeps one copy up to date. In Chinese and English. Split a tripRock climbing, mostly at the Gunks in New York.
Mostly film. 35 mm on a Leica M3 and a Nikon 35Ti, panoramas on a Fuji TX-1; 6×6 on a Rolleiflex 2.8F; 4×5 on a large-format camera that also takes an Instax Wide back. Provia and Portra, and Kodak 5294 in 35 mm. I develop the film myself, scan it on an Epson V800, and print in the darkroom under an enlarger. Astrophotography when the sky cooperates.
显影Developed
