Paper: Transparent yttrium hydride films prepared by reactive sputtering

Last week the first scientific paper where I am the first author became available on the internet. That is of course a grand step for me on the way to obtaining my PhD degree.

It was a bit more than one year ago that I had just started working with a new material in my sputtering machine. I had an idea that I could make some nanometer-thick layers of a semiconducting material called yttrium hydride by putting layer by layer of yttrium atoms on a surface, while the whole process was going on under the presence of hydrogen gas. Yttrium is a somewhat rare metal (not so rare as to be extremely expensive), and by reacting with hydrogen it transforms into yttrium hydride. This yttrium hydride can appear metallic, black or yellow-transparent, depending on the amount of hydrogen absorbed. The transparent state was what interested me as a potential material for solar cells.

Thin layers of yttrium hydride had already been made by others, and was first reported in 1996 by a group in the Netherlands. However, as it is difficult to make yttrium take up hydrogen directly, the films were made by putting a thin layer of palladium on top of the yttrium film, and then exposing them to hydrogen gas. The palladium is helping to take up the hydrogen, but there are some problems. Firstly, it is an extremely expensive material, which has a similar price to gold. Secondly, in addition to help to take up hydrogen, it also helps the hydrogen to escape, so if you take the yttrium-palladium sample out of the hydrogen gas, it does not remain in the same state. These were my reasons to try this new method reactive sputtering, which had not been utilized before for this exact material.

Since this was a new method for this material, my expectations were not so high. First, I started just laying layers of yttrium atoms on some glass, and got some metallic-looking films that had the expected properties for yttrium metal films. Then I put some hydrogen on, and I was content to observe that I obtained films that were darker, similar to the black state of yttrium hydride. Then, increasing the hydrogen pressure a little bit more, I got a film with the transparent state of yttrium hydride! I was very happy to have been able to make this kind of material by a completely new method, and surprised by the facility with which I had done it.

Figure 1 - I was very happy to see how easy it was to form transparent, black and metallic (left to right on the photo), just by adjusting the hydrogen pressure in my process.

The last year, I have spent working on these samples. Unfortunately I have not been able to make any solar cells of this material, but I have discovered a lot of other interesting things. Some of them, like the finding that these films have a cubic crystal structure as opposed to hexagonal in other findings for the transparent yttrium hydride material, can be read about in the present paper. And for the people that are not metal hydride geeks, I can already tell you that some much more interesting reports will come out soon, subject of one or more papers I hope to wright in the near future.

Published in Journal of Alloys and Compounds, DOI: 10.1016/j.jallcom.2010.12.032

Get the paper (open access PDF).

Scrap metal solar cell

T. Mongstad, S. Zh. Karazhanov, D. M. O. Heggø

A quest for new materials that can be suitable for solar cells has been going on over the last 50 years. Some good candidates have been found and even successfully commercialized, but the rarity of elements that are essential for these technologies is eventually expected to be their Achilles heel. In order to make a serious contribution to the transition from fossil fuels to renewables, a solar cell technology has to be based on abundant elements. IFE is now investigating a new class of materials, which could result in a highly efficient solar cells made of scrap metal!

Limitations for thin film photovoltaics

Currently the most successful alternatives to crystalline silicon (Si) are copper indium gallium selenide (CIGS) and cadmium telluride (CdTe). The last 10-15 years has brought these so-called thin film solar cells from the research labs to the factories. You can now buy both CIGS and CdTe solar panels with efficiencies that are approaching that of crystalline silicon (Si) solar cells, at a lower cost than for Si cells. The efficiencies are still increasing, which is promising. However, there is a show-stopper. An important constituent in the CIGS cells is indium, which is a rare and expensive element. For CdTe cells, which at present is the cheapest solar technology measured in cost per watt, the tellurium (Te) gives a similar problem. The limited abundance of these elements puts a roof on the production, and these technologies may run into difficulties in as little as 10 years [1].

Expensive 3^rd generation cells

Crystalline and multicrystalline silicon solar cells are still dominating the market. The efficiency-to-cost ratio is continuously getting better, but we are bound to reach a limit. The theoretical limit for silicon solar cell efficiency is around 30%, and it is highly unlikely that it goes beyond 20% for reasonably priced silicon solar cells. In spite of this limitation, silicon cells will still be around for a long time. But at some point we will move into more complex technologies, and one solution is tandem solar cells. Tandem solar cells selectively absorb light in different materials to get the most out of every photon. This technology has actually proved to give more that 40% efficiency in laboratory cells, but unfortunately this has only been achieved with the use of extremely expensive materials.

Metal + hydrogen = metal hydride

Metal hydrides can be the solution. By adding hydrogen to different metals we can get materials that are semiconducting, which means that they can be capable of generating electricity in a solar cell device [2]. Semiconducting materials based on abundant metals in combination with hydrogen might be the solution to the problem with rare and expensive elements. And by choosing different metals and alloys, we also can easily make materials that absorb different parts of the sunlight.

Using metal hydrides we might be able to make a highly efficient tandem solar cell out of “scrap” metal. IFE has ongoing experimental work on making thin films of semiconducting metal hydride films for photovoltaics, and we have so far showed results with magnesium [3] and yttrium [4] hydrides.

Figure 1: The author with a transparent semiconducting yttrium hydride sample (left) and a metallic yttrium sample (right). The only difference between these samples is the content of hydrogen!

Further reading:

[1]       B.A. Andersson, “Materials Availability for Large-scale Thin-film Photovoltaics,” Progress in Photovoltaics: Research and Applications, vol. 8, 2000, pp. 61-76.

[2]       S.Z. Karazhanov et al., “Hydrides as materials for semiconductor electronics,” Philosophical Magazine, vol. 88, 2008, pp. 2461-2476.

[3]       C. Platzer-Björkman et al., “Reactive sputtering of magnesium hydride thin films for photovoltaic applications,” Materials Research Society Fall Meeting, Boston: 2009.

[4]       T. Mongstad et al., “Transparent yttrium hydride thin films prepared by reactive sputtering”, Journal of Alloys and Compounds, (in preparation) 

Papers and journals

I'm an experimental scientist. I spend much time in the laboratory, but still I think I spend most of my time reading scientific papers. Scientific papers are texts written by scientists, for scientists. People not working in science would generally not understand anything of such a text, and it may even prove difficult for a scientist to understand a paper about a subject that she is not working with especifically. I mean, I'm a physisist, and I would of course not understand so much a scientific paper about lung cancer, but also papers about different areas of physics can be super-greek to me.

Being a scientist is revolving around these papers, and a scientist is generally evaluated on the basis on the papers she has written. Applying for a future job, she needs to refer to good papers in good journals with a lot of citations. In some countries you will get bonuses and raises as a scientist based on the papers you have written. Good papers and citations will also make it easier for you to establish new research projects and apply for official funding.

As a PhD candidate, the main objective is actually to write these papers. I need to write about 4-6 of them for my PhD degree to be approved. A paper can be anything from three to twenty pages. It sounds easy, but it's a real pain. For writing a paper of three pages I have to spend half a year in the laboratory pulling out my hair and another half in the office reading the thousands of pages other scientists have written before me about similar subjects.

Scientific journals are really not journals any more. I have not seen a journal since I started working on my PhD about a year ago. Well, many people have heard about Science and Nature, that are scientific journals that you might actually find on the shelf in the library. However, most researchers just use the on-line versions, which are databases of papers. If you know the author, the journal and the year a paper was published, you can find it.

But these journals are actually not so easy to get to. People outside universities that try to find a scientific paper, might find it, but would normally have to pay 50 dollars or so for downloading the document. If you are in a university you can download it for free, but the universities pay extremely large amounts of money to have this access. Who gets the money? Certainly not the actual scientists. I will not get any money publishing a paper, and I would actually have to pay for publishing it if I send a paper with color photographs.

And there is also the choice about the journal which you want to publish in. The journals are rated, they have what is called impact factor, which says something about how many people read the papers in this journal. Here is a list of impact factors for journals that are relevant for me:

• Nature: 34.5
• Science: 29.7
• Physical Review Letters: 7.8
• Solar Energy Materials and Solar Cells: 3.9
• Applied Physics Letters: 3.6
• Physical Review B: 3.2
• Europhysics Letters: 2.9
• Journal of Alloys and Compounds: 2.1
• Journal of Applied Physics: 2.1

Well, highest is of course best and everybody wants to publish in Nature. But that's nearly impossible for a normal scientist as me. When you send a paper to a journal, the editor will take a look at it, and if it's not a total crap she will send it to a set of reviewers, which are experienced scientists within the same field. They will evaluate if it's a good enough paper for publishing, and then probably send you back some comments about things you should change or clarify. 

Scientific papers are normally written by 3-10 co-authors. That means in practice that one PhD student writes the paper, and another five people just want to put their names on the paper for their own convenience. Well, it's not so bad, having experienced scientists reading through your paper and keeping themselves informed can be to great help to a PhD student.

So, there is a lot of things to keep track of. I just submitted my first paper, I sent it to Journal of Alloys and Compounds, which as you see is on the bottom of the list of journals. I have to start somewhere!

Life in a cleanroom

A cleanroom is a room within the laboratory with extra low concentration of dust and contaminating particles. To be in the cleanroom I have to dress up in a funny yellow suit to protect the environment from my dusty body. I can not bring normal paper inside, and I can not use my cellphone without a special bag to put it in, for example. Some times I work the whole day in here, but going in and out, dressing up in the funny yellow suit 15 times a day just to sit in front of my piece of lab equipment for hours without even beeing able to bring a book or a printed paper to read. But it gives me some time to think about stuff, it gives me a certain distance to the rest of the world. There is a window here, where I can see the rest of the laboratory from, I can see my colleagues running around in their white lab coats while I stand there in my funny yellow suit waiting for my laboratory processes to finish.

Financial crisis in photovolatics

Growth in 2008, growth in 2009, growth in 2010. I am not a financial analyst, but it seems to me that we are on the right way. Recharge reports that a 40% growth is expected in 2010, which is more or less the mean value for growth in the number of installed solar cells per year over the last ten years. If this growth continues for the next 20 years, the annual production in 2030 of solar energy will be of more than 40.000 TWh. That is approximately 1/3 of the total energy consumption per year in the world today!

Spinning electrons

I have now been in Grenoble in France for a couple of weeks, where I am doing a course on what experiments can be done with synchrotron x-rays and neutrons. Here in Grenoble, one of the most powerful synchrotrons in the world is situated, which I guess is the reason for the course taking place here. We are a group of 75 students on the course, most of us are PhD candidates, but there are also some senior scientists. The course is called HERCULES, and it has been on each year for 20 years now. That means that quite a bit of the research community working with synchrotron and neutron radiation has been through this course, and they have a pretty good line-up of lectureres.

Anyway, what's a synchrotron? Well, it's a big ring where electrons run around at a speed close to the speed of light. They are kept going around in the circle by magnetic fields that are guiding them around. When these electrons are pulled around by the magnetic field, they start emitting very high energetic radiation in the same direction as they are propagating. This radiation is mainly consisting of x-rays, which can be pretty useful for many things. The main thing about x-rays, as you probably know already, is that they go through things that normal light does not go through. They are also smaller (have shorter wavelength) than light, so they can see smaller things. And last but not least, if they propagate through a crystal, the atoms in the crystal can spread the x-rays into a special pattern that gives a lot of information about the crystallic structure. That's what's important to me, and many other researchers. A lot of things are crystalline, and certainly most semiconducturs, which is what we are making solar cells of. But you can discover crystalline structures even in chocolate, as some of the participants in the course have been able to see in their practicals.

Neutrons techniques is the other subject of the course. The neutrons can do similar things as the x-rays. Although neutrons constitute about half of the matter on earth, they are not so easy to get out from the atoms. You actually need a huge thing to kick the neutrons out, as for example a nuclear reactor. The fission of uranium gives neutrons flying out in every direction, which in energy reactors could be considered a problem. However, the neutrons has the property that they penetrate through things that not even x-rays would consider possible, so they can be quite
useful for structural analysis.

Here's a photo of the synchrotron ring and the reactor here in Grenoble, from a photo I took on my mountain hike this Saturday:

Metal hydride switchable windows and mirrors

The metal hydrides I am trying to make solar cells of, can also be used to make switchable windows. When hydrogen is added to a thin film of some of the metal hydrides, an abrupt transition from metallic or black to transparent can be observed. It has not yet reached any high grade of commercialization, but switchable windows are believed to become tomorrow's curtains for offices and lazy people wanting to shut the curtains with an application on their iPad.