Tuesday, 6 August 2019

Katarina Cabral - Summer STEM Work Placement Student - Weekly Reports

Work experience is great and it provides students with many benefits whilst gaining skills and helping choose the right future career path for pupils. Having work experience on a Curriculum Vitae will make a person stand out to employers as they will notice they have motivation and a real passion for work. It will also help to boost a students confidence in themselves! 

The STEM placement students are sending us weekly blogs all about their experience in the world of work! 

See how Katarina is getting on with her STEM placement at the University of Surrey! 

Week One 

"I finished my first week at my placement. I started by being introduced to the coding and simulation programmes I would use for my project and got to grips with it on the first day by completing tasks where I had to run simulations and plot graphs, experimenting with different work functions and gate voltages to find the optima for various transistors. I then began my project with Radu who sent a sketch of the structure I needed to make using the TFT (thin film transistors) as a starting point, altering the structure and mesh. 

We then played with the doping so that it built the structure Radu wanted.  I had some problems running the simulation but Eva showed me that the issue was that I didn't have mesh points for some of my region points. I plotted and looked at things like electron concentration and potential. We played more with the structure, changing gate voltages and increasing doping, then I learnt how to do cut lines and plot the graphs for those cut lines." 

Week Two 

"I began with altering the structure to shorten the source length and creating a file with a shorter gate overlap. I then showed transfers, outputs and took some cutlines. We then wanted to compare this to using the MOSFET so I created one from my code and overlayed this with the outputs and transfers of my previous structure. We explored using p-type doping at the source but this didn't work. Instead we changed the n-type doping and reduced the char value. The cutlines showing conduction band energy looked good so Radu suggested we start simulating with lots of combinations of bulk and n-typing doping. Each simulation takes forever! However, on Friday we came across errors in the code where I hadn't changed some variables so had to start simulating from the beginning. Luckily I get to hog all the licenses and will be back on track in no time."

Week Three 

"I ran my new simulations this week and we noticed that for high doping sometimes there was current leakage at zero gate voltage. The device wasn't turning on other times because too much positive charge means fewer electrons, so no current. We then decided to alter the structure and add some more bulk doping. The new simulations were even better than before and the plots looked just like textbook examples of how it should saturate. We then tried another change in structure with the doping to see the effects. They turned out well too. 

After this we needed to change the simulated material from amorphous silicon to a higher mobility material and tried running with impact ionisation switched on or off. Including impact ionisation would model what would be expected from devices with higher mobility materials because the electrons can travel faster. The same day we also had the faculty BBQ! On the Friday, I made a table of all the simulations I had done so we could clearly see what we have for comparisons. There was then a bit of trouble where some files kept disappearing (windows related) but it was all fine in the end." 

Week Four 

"For the impact ionization turned on in my simulations, the outputs were showing an increase saturation. This was because the positive charge was making the source inject more electrons. I compared outputs for different drain voltages and then took some cutlines. I had to start thinking about writing my report so Eva did a racap lecture on the physics behind the devices. We then tried some new structures, altering electrode length. I had accidentally run some of them without impact on so had to redo those. Sometimes the device was struggling to ramp properly because the steps were too big so we reduced them. One result seemed suspiciously wrong so we ran it again and that time it behaved. On my last couple of days, I began writing my report and enjoyed the end of my placement, concluding my results and reflecting on my placement. It has been great!" 

- Katarina Cabral 

No comments:

Post a comment