There are a number of open source layout and design tools for ICs however here I will just focus on Magic VLSI. Below are the steps to take a design from Verilog through synthesis and layout to the physical mask that is used for fabrication, via the Qflow digital synthesis flow.
The Verilog digital design for this example is a UART interface from www.asic-world.com and is synthesized/routed with the standard SCMOS rules, however a custom .tech file can be specified containing process details and design rules for DRC.
Once Qflow executes successfully, the design can be viewed in Magic after loading the appropriate cells and a GDS file is generated. This is opened in OwlVision GDSII Viewer or similar which can be used to generate the individual mask image files for each layer (active, poly, metal, etc.)
Using the same tools mentioned above, I also designed a simple PMOS chip to test my process. It is scalable to any reasonable size and contains 2 differential amplifier circuits (seen on right and middle) and a number of diodes/resistors and other test features on the left.
The design requires 4 masks for fabrication: active/diffusion, gate oxide, contact, and metal.
Fiducials should be added for subsequent layer alignment. Note: Grateful Dead bears are necessary for the circuit to function correctly.
Instillation and setup of a Silicon Drift Detector EDS system. A SDD is a modern detector for Energy-dispersive X-ray spectroscopy that mounts onto an Electron Microscope and allows the user to analyze the composition of a sample. When electrons from the SEM hit the sample, it emits secondary electrons (along with many other particles and energies) which are used by the SEM to produce the image you see on screen and X-rays which are detected by the SDD to produce EDS information. The X-rays have a specific energy which is characteristic of the material and equal to the energy difference between excited and ground state electron orbitals in the atom.
The SDD (30mm2 in my case) quantizes the energies of impinging X-rays and amplifies them to be read by a computer. As compared to traditional Si(Li) detectors, SDDs are faster, have higher resolution, and do not require Liquid Nitrogen cooling.
The detector is mounted to the chamber at a 35° angle which makes it most convenient for use at a WD of 25mm. When using the SEM at lower working distances, the SDD probe is retracted out of the chamber as to avoid collision with the sample stage. The detector is chilled to -30°C with Peltier modules.
The detector has a built in preamplifier. The output is fed to a pulse processor containing an FPGA which connects to a laptop via ethernet.With the SDD, EDS spectra can be acquired in less than a few minutes at high beam currents.
(Click on image to enlarge)
Thanks to Pulsetor LLC, Rick Mott, Jeff Thompson, David Bono, and John Guerard for their extreme generosity and help with the detector.
An Inficon Transpector2 HPR RGA was purchased from eBay and refurbished. I will use it for leak checking and ion selection experiments. Thanks to Aota Vac and Inficon for software assistance!
Put simply, an RGA is a small mass spectrometer that allows the operator to see the composition of the residual gases left in the vacuum chamber after pumping. Basic theory here.
This RGA is equipped with an electron multiplier so it operates at a maximum pressure of 1e-4Torr and is good down to about 1e-15Torr which is not a limit I will be reaching anytime soon. The spectra shown above indicates a large peak at an AMU of 18 which is H2O. This means that the chamber needs to be baked out to remove water embedded in chamber walls and surfaces until the highest peaks are N2 and O2.
A spectrum analyzer and near field probe was used to determine the main operating frequency of the RGA to approximately 3.02mhz. The RGA is also equipped with an Ar calibration source. Serial logs indicate that it was manufactured in 2007 and has been run thousands of hours past recommended service/filament replacement and the measured total pressure in the chamber was over an order of magnitude too high when compared to a hot-cathode ion gauge, but after calibration everything seems to be in spec.
Thermal evaporation deposition of Al and Ge from Tungsten and Tantalum boats, respectively. A blind hole is drilled and tapped in the chamber bottom plate for a center tap feedthrough (common ground) for the boats. Deposition starts at around 7e-7Torr and ends around 5e-6Torr due to outgassing. Aluminum alloys with Tungsten at the high temperature and causes boat failure, a thicker gauge boat will be used in the future or one made of TiB2-BN or BN. Update: W 0.015″ boat thickness seems OK.
Approx. deposition rate throughout the run was 2.2A/s, with total accumulation of 500A. Much faster than my sputtering setup but yields a worse film.
In situ plasma cleaning is via the red ICP coil seen in the 8th picture. A Quartz Crystal Microbalance (QCM) is used to measure the thickness of the deposited films and current is supplied by a rewound microwave oven transformer. The UV-VIS spectrometer is used to monitor the emission spectra of O2 plasma. O2 is flowed into the chamber via a Mass Flow Controller (MFC) until the pressure is 75-100mTorr and the substrate is plasma cleaned for 5 minutes with 100W RF prior to depositions.
As current pass through the boats, they heat up to 1000 – 1800C and subsequently heat up much of the surrounding chamber and mounting parts. This starts serious outgassing in the chamber and without prior cleaning and bake out quickly raises the pressure to non-workable pressures and the deposition rate slows.
I added a second turbo pump to raise the pumping speed/gas throughput (previously 110L/s and now an additional 50L/s) and to tolerate higher outgassing.
It was also noted that the evaporation of Al with lots of H2O vapor in the chamber (no baking) leads to a reduction of chamber pressure (presumably the formation of Al2O3 with H2O) and the production of H2 as seen on an RGA.
- IEEE Spectrum: The High School Student Who’s Building His Own Integrated Circuits
- Sam Zeloof (1:15) Interview The Greatest Show (and Tell) on Earth EE Journal
- The Bright, Bright Future of Electronic Engineering EE Journal Editors’ Choices for World Maker Faire 2017
- (ARRL) World Maker Faire Visitors Urged to Build, Make, Create, Communicate
- Maker Faire New York 2017 Highlights (Evil Mad Scientist)
RF or DC oxygen plasmas are composed of a number of highly energetic and reactive species that will readily clean most organics off wafers and strip photoresist.
While operating with prolonged high power plasmas residual gas deposition and chamber sputtering collects on most surfaces.
- Clean/prep wafer – Piranha, RCA 1 / 2
- Water rinse
- Remove native and RCA oxide – 1-2% HF dip
- Field oxide growth – 2 hours @ 1200 c w/ water vapor, 5000A blue film
- If wafer in storage, dehydration bake – 10 min @ 220c
- Check wafer hydrophobic if necessary
- Optional spin HMDS
- Spin 3.5mL AZ 4210 resist 30 sec @ 3500 rpm ~3.5um film
- Soft bake resist 3 min @ 105c hotplate
- Expose active area – 18 sec DLP projector
- Develop 1:3 400k KOH:H20 puddle 2 min
- Water rinse (no solvent)
- Inspect wafer, if defect strip resist and retry
- Hard bake 5 min @ 115c hotplate
- Etch active area – 1-2% HF 10 min or until surface hydrophobic
- Water rinse
- Resist strip – Plasma ashing 100 watts RF 5 min @ 125mTorr O2
- Acetone rinse
- IPA rinse
- Water rinse
Tall particles can easily short out the thin gate oxide in these devices, as shown under my SEM. This poses an issue for making such devices in a garage; the gate oxides must be grown thicker to mitigate shorted devices which leads to a higher threshold voltage for the FET.
Recent progress into 1um feature sizes and beyond with a LCoS projector and a metallurgical microscope. 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.
Laser illumination of the projector is avoided due to interference patterns. A SMD 395nm LED is used in some tests and a standard high pressure mercury arc vapor lamp in others.
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.
Basic mask set for a MOSFET. Fiducials should be added to the corners for subsequent layer alignment. Active area is etched into field oxide, active area is doped, gate oxide is etched and regrown as thin a possible without pinholes, contact area is etched into oxide, then aluminum or copper is sputtered or evaporated and patterned.
A small, modular, and versatile chamber was constructed for plasma and other research experiments. The main vacuum manifold can be configured with multiple feedthroughs in either standalone vacuum mode or connected to a large chamber. The front is a 6″ CF viewport that can be swapped for a gas feedthrough assembly.
The chamber is being used right now to study simulate vacuum conductance and pressure gradients across larger chamber systems such as an ion implanter with a turbo pump on one end and an ion source and MFC gas flow far away from the pump. The gas flow creates a higher pressure in the ion source chamber and, in theory, allows for low energy beam transport and acceleration into a much lower pressure substrate/target chamber (in the case of an ion implanter).