Q. Why build the beam source on a silicon wafer? Traditional beam writing technologies do not easily scale to the large number of parallel beams necessary to write wafers at high volume manufacturing throughput (over 100 wafers per hour). By using multiple chips with linear arrays of millions of proton beams, the ProScan can accommodate any width of wafer.
Q. Why use protons? A 3.5KeV Wideroe style linear accelerator for protons is only 60mm long. By comparison, electrons would require an accelerator about 2.6m long. Also, the 16V p-p voltage used by the Wideroe design is well below dielectric breakdown voltage.
Q. Is a special resist needed for proton exposure? Standard PMMA can be used as a positive resist at low dosages. Using protons to expose the resist has many advantages over electrons or photons. Protons are 1800 times more massive than electrons and the resist is more than 100 times more sensitive to these protons than to electrons at equivalent energy. The effective proton wavelength of 0.5nm is much smaller than any expected feature size. The resist is exposed by collateral electrons produced as the proton decelerates after hitting the resist layer. Resolution is high as these low energy electrons do not scatter very far. 3.5KeV protons do not produce the resist-fogging backscattered electrons that plague 50KeV e-beams.
Q. Why are "cold" protons necessary? Hot ions with 10 to 100 electron volts of transverse energy will create blur when exposing resist. The lower the lateral energy, the less blur.
Q. How are "cold" protons produced? It is best to use a method that never heats up the protons. We are investigating various ion conductive membranes that can withstand the pressure differential and conduct protons at room temperature.
Q. Why start with a 20nm pixel? Simulations show that a 50nm channel on a 100nm pitch should make a 20nm beam. Beam size can be adjusted by changing the distance from the print head to the wafer and by changing the number of protons per pixel. This design should scale to smaller pixels in the future, but 20nm is a good place to start.
Q. How did Transmute get started? Founders Kim Hailey and Bob Conn started with the idea of building very high flux proton accelerators based on silicon technology. The possibilities of integrating hundreds of thousands of accelerators became obvious. Both of the founders have decades of experience as senior technologists in the semiconductor industry, so lithography became the top application of this technology.
Q. What is the biggest advantage of ProScan? Just like any integrated circuit, the development cost is substantial, but the incremental cost is very low. Once a prototype is proven, hundreds of systems can be produced quickly. It is possible that ProScan could reduce the cost of lithography by 10x.
Q. What about setup and maintenance? The ProScan system is self aligning. Each print head is mounted with its own nano 5 axis positioners and interferometers. The print heads are designed to be field replaceable.