Update 8/26/17: Developing Patterning Process for Homemade Microelectronics
The general idea is to use a modified presentation projector and reduction optics to transfer an image to the photoresist on a wafer without the use of expensive masks. Below are descriptions of the 4 iterations of my photolithography setups:
Automated DLP submicron stepper for 2″ (50mm) wafers with LabView control, computer alignment, and wafer vacuum chuck
The second and third images above are composed of 4 precisely aligned exposures which enables submicron resolution over large areas. This “true” stepper operation uses a closed-loop feedback system and computer visual alignment.
Manual LCoS submicron stepper with red laser alignment illumination. Given a numerical aperture of 0.98 on the microscope objective and with an exposure wavelength of 365nm the simple calculated resolution is 0.227um however the actual resolution is probably around 0.5um due to diffraction limitations inherent in this projection system. The depth of focus @ NA = 0.98 is calculated to be approximately 1.8um but is likely worse.
Manual DLP projection aligner >10um features. Color wheel is removed because it did not transmit much light below 400nm. Emulation (relaxation oscillator) circuits were made to reproduce the signals that the projector expected from the color wheel back EMF motor drive/sensor and photodiode.
Exposure times calculated by integration of total UV dosage measured at different wavelengths with the radiometer. To calculate exposure time for AZ4210 resist, for example, the datasheet is consulted to see a recommended dose of around 135 mJ/cm^2 for a 3.5um film thickness. If exposed with a 5x objective on my system, the exposure time @ 410nm is (135 mJ/cm^2)/(4.05mW/cm^2) = approx. 33 seconds. This is a bit longer than I would like but given that it is a positive resist that is to be expected.
Proof-of-concept DLP setup
Installation of Varian Navigator TV 141 Turbomolecular Pump on a 6″ CF to ISO 100 elbow – custom Lesker part. Pump controller is integrated and is powered by 240v single phase. Backing line is KF16 to KF25 conical adapter and the KF25 stainless hose runs to a heated vapor trap and then to an Edwards oil rotary vane pump. Turbo pump speed is ~120L/s!!
Drilling and tapping 1/8 NPT in a 2.75″ CF blank flange to mount a Hastings Instruments DV-23 thermocouple gauge. This, along with a Baratron pressure (capacitive) transducer, will be my main low to mid vacuum gauge.
Installation of a Perkin Elmer hot cathode ionization gauge on my PVD chamber. I applied current to the filament briefly at atmospheric pressure and the filament did not burn out so it may have a yttria-coated iridium filament rather than tungsten or any other refractory element. This will be my main mid to high vacuum gauge, used in conjunction with a thermcouple gauge and a Baratron pressure traducer for backing pressure measurement. The gauge mounted up to a 2 3/4″ conflat. I have a HP 59822B ionization gauge controller which provides filament current and reads collector current and displays it in torr.
Thanks to Charles Alexanian for this additional information:
“…your assumption about your gauge is correct in that it is an iridium filament. That being said the yittria coating can be poisoned by a great many things requiring a re calibration of your system depending on which type or controller you are using. The tungsten filament versions stayed popular in chamber research because you could simply boil off anything that might condense on the filament… I have changed over my Bayard Alpert type gauges. (That is the technical term for the gauge you are showing) for inverted magnetron types, particularly the MKS903 type units because they give a analog voltage output and require no external controller. Additionally they can be easily disassembled and cleaned. The also have a higher starting pressure where I no longer need thermocouple gauges.”
The quartz thickness monitor is a Maxtek TM-100 that I picked up on ebay for under $100. The board is very corroded so after a lot of contact cleaning I was able to get it working. Displays rate in angstroms per sec and integrates to find total thickness. Surprisingly simple circuity, it’s all based around a single counter chip and of course 7 segment display drivers and such.
A brief introduction to semiconductor fabrication processes and terminology. It is not intended to be an in depth view of any single process, but rather an overview so that provides enough information for someone to get started with making diodes and transistors at home.
Tour of my home chip fab setup in early 2017. I’ve been accumulating this equipment since October of 2016.
Step by step FET fabrication
High vacuum basics
Buck/Boost converter design for a solar glider I was working on last summer, at 16v it reached 98% peak efficiency and 95% average in boost mode. Also tried to make a linear power supply board using a laser engraver to cut away the copper, it worked pretty well to replace the normal ferric chloride etch step. Continue reading Switch Mode Power Supply Design + laser cut/etched PCB