Semiconductor Devices & Technology – EE Master Specialisation
What is the specialisation Semiconductor Devices and Technology? The specialisation Semiconductor Devices and Technology is about semiconductors and the devices you can make from them. So a lot of transistors, diodes, chips. Can you explain what a semiconductor device is? Okay. A semiconductor device is an electronic component. So cables come out. And in there, there is a semiconductor.
So, it's a class of material that doesn't conduct well. But, you can modulate how well it conducts. It's very easy to make the semiconductor insulate or make it conduct. And as such, you can make it switch. For instance: in a diode, you have only two wires coming out.
and current can only run in one direction and it does not go in the other direction. So in that case, the semiconductor is telling the current, you can only go that way. So that is one example.
In a transistor, you can make the current flow or you can break it open and stop the current. So the transistor will switch from 'on' to 'off'. And other materials can't do that. It's just a semiconductor that does it.
So, everything we want to do with switching or directing current, we do it with semiconductors. It's both a bit about: how do they work, and how can we make them. And how do these things... How are they intertwined?
To me, I'm discovering: Holy sh*t, this is how it works. It's amazing. That's the joy in semiconductors. It's a confused material. It doesn't know what it wants, but once you give it some direction, it does everything from phones, computers, biomedical, space. For instance, sometimes you want to do a transformation.
Like, when you charge your laptop or your phone, you want five volts. But 220 volts come out of the wall plug. You convert the AC signal to DC signal. And you convert high voltage to low. This you can do very well with semiconductors and is very difficult to do otherwise. So other examples are microchips... The microprocessors, I should say. The chips that do all the number crunching in a computer.
That do all the calculations, like the Pentium or the Itanium and all these very smart chips that Intel produces. Also, the memory chips that are in USB sticks. When students follow this programme, they will find out all the 'ins and outs' about how a transistor is put together, how a solar cell works and they can make one themselves.
They can use our laboratory to make such devices themselves and find out if they work or not. – That also sounds like quite a unique selling point, that the students already can go into the Nanolab. Yes, I think there are three or four places in Europe where you can do that. Where you can do a master studies and actually make devices in a clean room facility. Because these clean room facilities are usually quite secluded.
Only specialists may go inside, and the students are usually kept outside of the lab because the equipment is expensive. Let's act like we wanna make a device. We go a methodical sequence. So we start from an idea, we look at the theory behind it. We then make a design.
And then you start drawing your process flow and see where everything, with all the detailed steps, also everything that's needed to, for that one fabrication step, you need to do the sub-steps and you see, oh, this we can actually not do or that we can... We try to estimate beforehand how the design will perform, for instance, by simulations. So we can do electrical simulation, optical simulation. Then we make it. And then you go into the clear room and try. And then we test if indeed it performs the way we thought.
It's usually not everything in one thesis or... It's usually that you get part of the steps or that there's a really specified type of thing you have to make. Such as the design part is a bit shorter or you really simulate something very well. There are four compulsory courses.
Two are about design and ethics principles. Those are compulsory for all EE students. One of the compulsory courses is the Fabrication of Micro and Nanodevices, where you learn a lot about the technologies that are available for fabrication of chips, for semiconductor things, but also for lab-on-a-chip or bio-applications, sensors, those kind of things. You learn a lot of the tools and the course ends up with... You get a picture of something. Sometimes it has a scale bar, sometimes not.
And you can have to come up with how you could fabricate it. It doesn't have to be the way it was fabricated, but a way you can come up with combining all the technologies and tools that you learn in the course of how you can actually fabricate the thing, which is quite a fun challenge. Sometimes you feel like, "Oh, why did I get this picture?" Sometimes you're like: "Oh, this one makes a lot of sense."
One of the difficulties if you make such a thing is that you don't see what you're doing. So one thing we have to teach our students is first of all to envisage what you're doing. To make a mental picture. And secondly: which techniques are there to find out what you've done? And the other one is Advanced Semiconductor Device Physics, taught by Ray Hueting. And he explains in detail how devices work, how advanced very modern devices work.
Like a bipolar transistor that is used in cell phones for communication. Or power transistors that are used in modern power circuits, such as the adaptor that you use to charge the phone. Those require very special transistors. The adaptor, it connects directly to 220 volts. And most electronics will immediately blow up if you do that. So they have special transistors in there that can hold 220 volts and even more.
Those are the four compulsory courses. But we find it important in the Electrical Engineering programme... ...that you both have experience with research topics and system design topics. So we have given a list of courses that are either very research-type or very design-type. And you have to choose one from that list.
So those are like half-compulsory because there is still a choice. That's five plus five [EC] again. One is Measurement Systems for Mechatronics.
It's a more a system approach course, where we talk about a complete measurement system. How to put that together, how does it work. And especially we focus on the instruments in a measurement system that do the measurements themselves.
It can be a photodiode or gyroscope or an accelerometer. We discuss how these parts work, how these components work, and how you best employ them in a system. You have Micro Electro Mechanical Systems, which goes into the sensors and the mechanical parts.
And you actually design in the group your own mechanical... Microelectromechanical chip, which is fabricated and you measure it. And the great virtue of that course is that you actually design one of these MEMS devices in using the real design environments that also in professional life you will use. Then it is produced in the Nanolab, here in Twente. And you get it back from the laboratory and then you can test if your idea really works. One course that is 10 EC, that is very attractive to take, is called System-on-Chip Design.
And there you learn more the system architecture of a chip, or a combination of chips. And you do that with practical exercises, with real life experiments, and also from theory. There's also, I think Jurriaan will talk enough about it... But with the semiconductors, you can also go more in the solar direction.
Actually in that course they don't go deep but more into application. But I think there will be some more new courses I expect Jurriaan to do. Because it's his new... he has kind of changed his field. He's trying to move from 'reliability of computer chips' to 'solar'. We teach how to harvest solar energy, how to convert solar energy into more useful forms of energy, for instance electricity or heat.
That we want to use in daily life. And of course the solar cell, as we use it in photovoltaics, is an important component. So I spend several hours explaining in full detail how a solar cell works. And how we try to produce it in such a way that it's maximum efficient. Until about 12 years ago, solar cells were always made in the same way.
But then suddenly people came up with new ideas how to make a solar cell differently. And then suddenly every three years or so, people make solar cells in a completely different manner. The level of installed capacity for solar cells is now I think one terawatt or 1.5. But we need to grow further up to 20, 30, 40 terawatt.
If you look around in the city where you live and you see solar panels at the roof, you have to mentalise that actually, we need about 20 or 40 times as many solar cells. Everywhere. In the fields, on the roofs, on the industry roofs, everywhere. And then we satisfied the need for electricity. Some of the obligatory courses are Material Science.
Where you get a lot of the basics which you then apply in the technology course. Nanoelectronics, which is also where they do a lot of neuromorphic things.. Or really on the nanoscale and quantum physics things on chips as well. So that's a fun course I did as well. – Yeah, it sounds really difficult. Okay, the Nanoelectronics one, it's...
Quantum Physics might sound a bit scary, but it's not the fundamentals of it. It's mostly what you can do on a chip and the fabrication. Than you, without having all the math in the course... You can think of a lot of devices which can give a fundamental understanding as well. And it's built up really nicely in the course. So it's not...
It's also for Physics students, but as an Electrical Engineering student, it's not like it's crazy difficult. Then they can follow my course on integrated circuit technology, which is my favourite. In that course, I teach the students how a flash memory is made, How a CMOS digital circuit is made, how a CCD camera is made, how SRAM memories are made and so on. All sorts of chips. I explain how they work and also how the production process goes. It's a very good course. And Jurriaan has this amazing ability to connect to every student and get them excited about something they've never heard of.
He is just amazing. And you can see that students don't want to leave his class after it's done. They want to stay and ask more questions. And then we have a very practical course. It's called Semiconductor Project. And in that project, you will get some samples that we have produced in our own lab or samples that we got from the industry.
And then we have a specific question about those samples. We would like to know, for instance, how quickly they will be destroyed if you put too much voltage on something like that. Those are the topics that our programme director Cora Salm enjoys most. She tells us everybody is allowed to make things in the Nanolab as long as I'm allowed to destroy them again.
You might think that it's just fun to destroy samples, but it's in practice actually really important to understand what are the limitations of an electronic product. And to find out the limitations, you have to go over the edge. – Do I see that as sort of practical case studies that you guys like create this case and like, oh, yeah, just go try it out. See what happens. Yeah, we had one case where we bought RFID tags.
That's a special type of very small chip. And if you keep your cell phone close to it, it will give a number. It will say: I am RFID tag 156 etc...
We did a test with these RFID tags. What happens if you heat them up? Just to find out at which temperature does it go wrong. And we expected that at some point the RFID tags would not answer anymore. But it was different. We found that if you heat up the RFID tag too much, it starts to give wrong numbers.
And that was really problematic, because the whole idea of the ID tag is that it will give the right ID. So you don't want it to give the wrong answer. And then you are at 30 EC.
So the other 30 are basically completely free. Students can choose from the entire offering of the University of Twente which courses they want to take. And because we work together with the TU Delft and TU Eindhoven in the Netherlands, you could even take courses there.
– And what are some of the electives that you like the most in your studies? Oh, the electives that I like the most. That's a good one. I really liked Inorganic Material Science, which is from another study completely.
So perhaps you should go for another one. I really liked the MEMS course. Also including the Advanced MEMS part where we got to design a microelectromechanical system, which was fabricated. We could measure it, but we also did some simulations on it in the advanced part. And it was really nice to get something, some hands on. Hey, we designed this. It's actually there. We can measure it.
And this was one of the courses where you got a lot of practical insights and real life experience already in that first year. Another course I really liked was Advanced Analog IC Electronics. It's from the ICD specialisation, where I gained a lot of simulation experience for circuits.
And it's more for the... And it's about designing the chips that in our specialisation you learn to fabricate. So I think that really complements well.
And I think you learn also a lot about the trade-offs and why you would like to achieve certain things in your fabrication process and which ones are more important, which ones are less. – Are there courses that you really like that we haven't talked about yet? Well, I know that my students really like the course on superconductivity. And it's interesting because electrical engineers usually don't use superconductors.
But of course, conduction for electronics is the most important thing there is. If no current flows, then you cannot make anything. So, of course, they have a generic interest also in superconductivity. Students can choose a bunch of electives that are related to optical devices. Like lasers, LEDs, photodiodes, and also solar cells.
They could also choose to learn more about circuit design. So they can take electives that have to do with the design of chips rather than the manufacturing. But they can also take a mix. They can just follow their curiousity. But the list of courses that we give where they have to choose one is actually a very good list of topics that the students can take.
Basically, all those courses are recommended by us. Otherwise, we wouldn't put them on the list. I just started doing some courses for the master that I liked. And at some point after the first quarter, I was like: "Oh, officially I have to discuss with programme mentors by now my actual programme."
But I was still doubting a bit, because I liked everything that was physics of chemistry related. And I started just with the programme mentor from the group where I did my bachelor's thesis. And he was very kind. He was like, "Oh, but you can always choose later on still. It will be fine." So I just continued doing courses and didn't talk to other programme mentors anymore.
And I actually went through the list of all courses for the EE master. And then just started highlighting everything that I thought was really fun or seemed a bit of fun. And I just compared that to compulsory courses per specialisation. We ask our students to come up with a proposal when they start the master programme. And in some cases, they have difficulty making up their minds. So then we have a personal meeting between the programme mentor and the student to decide what would be good.
And the good thing about this way of working is the students can later on also revise their programme or their plans. If they discover that some topic is much more interesting than they thought beforehand, they can also diverge a little bit from their original plan and come up with a new plan for studies. So the first year of our master programme is all courses.
And it can be practical courses and theoretical courses, depending on the mix that the student choose. The second year is first an internship of 20 ECs. And then the second part is a 40 EC final project. And the internship is normally done in an industry environment.
So they go to a company, can be inside the Netherlands or abroad. And they spend, I think, 15 or 16 weeks doing research in the industry in an R&D department, typically. Yeah, so I did my internship at Tyndall. It's a research institute in Ireland. It was really nice there...
When I looked for an internship I wanted something that felt very close to the specialisation I was actually doing. because there's quite some adjacent specialisations which are also fun. But I felt like I really want to do something with the knowledge I've earned in the specialisation. So, I ended up going to my programme mentor, discussing what would be some fun places for me to go to.
And he had some connections there. And I worked on the efficiency of micro-LEDs. With mostly measuring. They also have a clean room there, but I didn't go in.
I did mostly characterisation and simulation. They had a really cool simulation program where you can also really put in the measurements and then have some modeling of your results compared to the model you've also made yourself. So I was there for four months. Got to meet the vice prime minister.
That was quite cool at some point, because it was a part of a way bigger project that they landed. Yeah, actually, a really good time there. I'm really happy I went there as well. It was very interesting. So really nice people. And it was a really cool opportunity to be able to go there.
And it happens very often that people who do an internship and they chose well, they get hired by that company afterwards. So that is good. The experience is often very good also because we try to hand pick a little bit which student goes to which internship. So we try to find out: OK, is this a good person to send to that location to do that kind of work? So there again, we have one-to-one discussions between the student and mentors or supervisors who have contacts in industry for internships.
In my thesis, I tried to make Tunnel-FET in the Nanolab clean room, here on the campus. – What is that? It's a tunnel field-effect transistor. Most transistors that are used are MOSFETs, Metal Oxide Semiconductor Field-effect Transistors.
They have been miniaturised a lot. But you cannot keep miniaturising forever. At least not with the technology we have today. And there are some fundamental bottlenecks in the way that charge is carried in these MOSFETs. And therefore, there is a search for some other types of transistors that could take over the place that don't have these bottlenecks for some applications.
And Tunnel Field-effect Transistors are one of them, but their fabrication is quite difficult still. There are some steps there that are difficult or it's not completely there yet. And at the same time, there's a lot of planar technologies just on the top of your wafer. Just one layer of transistors. And if you have want more miniaturisation, you could also go in 3D and make more space in such a way. And in my thesis, I try to make a self-aligned 3D tunnel transistor.
Self-aligned is also perhaps relevant that we of course all know ASML nowadays with the lithography steps. But the machines that can print the smallest patterns, they're so expensive that only a few companies can buy them. And if you can make sure things align themselves, you need some less lithography steps.
And you can also make more interesting shapes, that can be relevant and useful also for your device. So some master thesis projects are focusing on understanding components better by using simulations and understanding theory. Other assignments are much more experimental and that can go two ways. Typically, either you go into the laboratory and make something new or you go into the measurement room and you try to find out how does it work and why does it work this way. In that process, you get supervision.
So the students typically talk to their direct supervisor at least once per week to discuss the progress and the plans. But also, for instance, when you go into the laboratory, there will be an engineer who can supervise you and train you on the equipment and show you how things are done. Not only to be effective and proficient, but also to work safely.
That is important to me as well. Of course, the University of Twente is really a technical university. So, we train a lot of people for industry jobs in the end. I think there's quite a variety of what people go and do. And I'm not really sure what I'm going to do yet. So if you get into designing devices, then all my friends sitting behind in this background, they do all the designing of devices.
That's the name of their group. Another field would be to actually engineer materials that will go into devices. So how do you grow them? How do you control them? How do you precisely increase their yields? How do you make more efficient, reliable devices? And then you can also go into companies where you're characterising these devices. When you make something, how do you know if it actually works? How do you know if there are any defects in it? And then there are companies that bridge between: I have a product, now I need to find customers, I need to find markets. Our students have a lot of job offers, within the Netherlands or outside.
So there are great job opportunities when you do this specialisation. – That's so nice. That's very nice. It's a no-brainer, actually.
If you are worried about that, than you can definitely take this specialisation and have sufficient options to go for. For instance, the company ASML in our country and also NXP Semiconductors. They are hiring all the time. And every time I talk to them, they ask me to raise more students, please train more students. And could you please tell them this and could you please tell them that? And can you please send us a few more interns because we still have work to do.
So there is a strong market pool for people with this knowledge. Actually, many of my team and also myself, we have worked in industry before we came to the university to teach. I worked at Philips Semiconductors before. And that team that I worked with is now NXP. So I know a lot of people at NXP that can actually host interns. And that makes it very easy to pick up the phone and say: "I have this student with this profile. He's curious about that.
Do you have something nice?" And that's much easier than waiting for advertisements on the internet for interns. After my master's, I worked at a company called Applied Materials. They make most of the major equipment that's used in semiconductors fabrication. Everything from growing materials to etching them to processing them. Everything that ASML doesn't do. So ASML makes the single most important product. Their lithography machines, it's the most critical. It is the most important. And it's the most expensive.
Companies like Applied Materials, Lam Research, Tokyo Electron, they make the rest of... and KLA. They make the rest of the equipment that's used for fabrication of everything from sensors to detectors, to logic chips, memory chips, modems. Other than a potato chip, they make everything else. So I worked at Applied Materials for about 12 years. – And what made you decide to go away from the work field and start a PhD in this topic? I wanted to transition from being in the industry to getting into more fundamental research in academia. Just go back to making something that's much earlier in the phase where I have a bit more time, a bit more patience, more like an exploring mindset.
I think the freedom of electives and being able to choose the courses that you like and also finding your way while doing the courses. Which are the things you actually really like? And I think also during your thesis when you're trying to make stuff work and it doesn't want to work and you read a lot of papers about people that tried other things and it did work. But they don't say why or didn't work. And then you at some point you actually do get something to work also in the clean room.
That's really rewarding, I think. Once you get there, and it does work. The pure joy that I get is in that search, it's in the hunt. When we find the answer, when we figure it out, I get into depression. It just becomes boring.
It's like we won. Oh, the game is over. I want to play again. I want another problem.
So I enjoy that darkness on the edge of town, like Bruce would say. I think the teachers are also a really big, positive part of the programme, but also this specialisation. It's not a very big specialisation. So even for the entire EE programme, it's quite easy to go by a teacher and ask something. For example, with the MEMS simulations, with person I was doing it with I ended up sitting in the office of the teacher for an hour.
And it was really nice and really useful. I enjoy working with young people to start with. I enjoy to work with smart people also. And I really, I can be super satisfied when I see that a student has learned something from me. When they discover a new world or new ideas, when they manage to make use of techniques or mathematical tricks or whatever.
When I see that they have learned to master something and they are capable of doing something they never did before, this is very satisfying to me.
2024-12-15 02:18
