Connect Raspberry Pi to computer with ethernet cable - Raspberryception

Raspberry Pi

Finally I revived not only the blog but also my Raspberry Pi. As my bachelor thesis dealt with this little machine of course I wanted to have my own as well. I bought it already several months ago and even tinkered a little around with it but since I moved to Sweden I let it more or less sleep in its plastic case - until today! Taking only small steps so far but if anyone of you wants to get a Raspberry Pi him-/herself this little introduction might help to get it up and running. :) In any case this will help myself to memorise how to do stuff. ;)

Happening in the following: Flash SD card, initial config and what I did today: plug to PC and access via ss. Sharing the internet connection and further securing your ssh will follow next time.

• Flashing the SD card is really easy by following this tutorial so I leave this one out.
• Initial config can be done at will or following some tutorial from the internet as well so I leave it out as well and concetrate on what I did today: (Only make sure you enabled ssh in the initial configuration.)
• ssh from Windows (I did not manage to do it on Linux so far as I could not find out how to properly connect to two networks simultaneously): Easiest way is to use puTTy. This allows you to ssh onto the Raspberry Pi using Windows. A normal network cable is sufficient, no crossfire cable needed. Before being able to connect via putty you need to give the Pi a static IP. This works the following way:
• Put the Pi SD card into your PC/laptop card reader, navigate to the "boot" folder and open the cmdline.txt with any text editor. To the only line in the file add "ip=<ip-of-your-dreams>" without "". Using Windows you will most probably obtain your IP automatically from a dhcp server so enter the IP as 169.254.#.# where you replace the # with a number between 1 and 254. (Should you not connect to the internet via a dhcp server the IP's first two numbers have to match your PC's IP - can be found from Win-Key + R -> type "cmd" -> type "ipconfig" in the black window -> IPv4 from the adapter you use to connect to the internet.) Save the file and plug the SD card back into your Pi. Boot with ethernet cable connected to both your PC and your Pi.
• Open putty on your Windows machine and under "session" you enter the static IP you just gave to the Pi. For the port use no. 22 if not preconfigured which is the port that is assigned to ssh. If you want to use programmes with graphical interfaces on the Pi you should enable X11 forwarding at the respective place in putty.

That's it! Connect with username "pi" and your password.

(Not even) semi scientific humbug

Incredible cretinism combined with some scientific proof. Or maybe a mind capable of  outstandingly visionary ideas? ;) Welcome to the mind(s) of (a) physicist(s)! Because the following can probably really only happen to physicists:

I could think of nothing more convenient for a story like this ...
((c) Christopher Torres, LOL-comics)

One is in a philosophical mood and posts "There is only one everything." What follows is a deliberate mixing of scientific results in the most queer-headed way possible. All in all it results in the following proof of the contrary which I cannot help but have to break down a little - not sure if out of suspicion towards mathematicians or out of a self-mesmerising-with-possibilities mind ... So the summary of the discussion following the highly philosophical statement was

The universe is infinite and there is an infinite number of infinities.

"Everything" really includes everything so it is the sum of all infinities.

Even an infinite number of infinities can be continuously numbered and

the sum of all integers is -1/12.

Thus, everything is less than nothing.

This might be a little crazy but all of it can be "proven". And here starts the really crazy part. The first statement that the universe is infinite is not yet proven right or wrong. Einstein once brought this down to "I know of two things that are infinite: The universe and human stupidity. Concerning the universe, however, I am not yet sure ..." So it might be infinite as well as finite. So far you can at least imagine that it could be infinite as even beyond matter there could maybe be somthing ... So with this we have to go with the assumption that what is not proven wrong may still be right. ;)

Then about the infinities. Once upon a time four months ago a friend of mine posted a video in his blog about infinities that summarises nice and easy (it is really possible to understand it!) how there are infinitely many infinities, how they are related and that you go crazy if you think about it. (Not kidding, the guy (Gregor Cantor) who made the considerations spent half of his life in mental hospitals because his colleagues were jealous of his genius and mobbed him into insanity.) You can find his post here. (Although the text is in German, the video is in English.)

Last but not least the thing with 1+2+3+4+5+... being -1/12. This is obiously the most crazy thing. But believe it or not this is used in string theory and e.g. the quantum field theory Casimir effect. According to a friend's friend a rigorous treatment of divergent series which the proof of the above sum relies on can be found here. This now is the "suspicion towards mathematicians" part because it is obviously as counterintuitive as it can be that the sum of all integers is -1/12. I tried to understand the principles of the linked proof and it seems that there really is a proof making use of cutoff functions and the like. (There are many more proofs to which you have to add quotation marks (-> "proofs") that are quite easy to understand but usually there are obvious faults in them.) However, for this more rigorous proof my application-oriented mind is definitely not twisted enough so there will have to be at least the discovery of the world formula involved before I will give it a second glance ... Until then I will stay with my physics and kepp on juggling math in ways that makes sense. ;)

Revolutions made in Lund

Completely out of context: Our most recent lecture hall.
Once again it feels like Hogwarts ... Life is good!

As far as I know if you try to patent something and it was mentioned somewhere before the patent will most likely not be granted. I do not know if you can do any harm by unofficially publishing scientific results as well so I prefer to be careful. ;) Today our professor opened the lecture by showing us data which might be world record in electronics. The results are from yesterday so he was quite excited. Of course the results have to be further investigated but they look very promising so far! Professorquote: "World records are always fun!" I agree. ;)

The rest of the lecture was about toxic gases in WW I. No, wait. Not completely. But in one of the methods to fabricate semiconductor monolayers AsH_3 is used. This was indeed used as a toxic gas in WW I. Which is why they do not use this method in America due to administrative approval problems. Instead they prefer a method which is a few million (probably ;) ) dollars more expensive. I expected that in a country where children chocolate eggs are forbidden in favour of machine guns a toxic gas in science would be something to cheerfully approve ... Summarising comment of a class mate: "Crazy people ..."

Still after the toxic gases the professor drew some "funky quantum wells" (his expression ...) on the board. By combinig different materials it is possible to form the weirdest band structures. As he said: "I do not know what this might be good for but it is possible." Well ok then, it is always better to ask "Why not?" than "Why?" in research. Although while still studying it is more the latter that you ask ... Quite desperately sometimes. ;)

Nanowires

Now that my new semester started I thought I'd share a short excerpt about what "we" do here in Lund - what is special for Lund research. It is everywhere here. If you study here you might even forget that there is something else going on (in physics). It is nanowires. In principle they are exactly what the name implies. Wires. Small. But first of course the always most important question first: Why would you investigate them?

They show some very interesting properties that conventional wires do not show. These properties lead to promising application possibilities such as (of course) wires in ever modern devices which are getting smaller and smaller. But here besides being small they already show further promising advantages. By combining materials in a way described further below they might allow transmission of two signals in opposing directions at the same time or introduce barriers or quantum wells to quantise conductance. Furthermore, junctions of nanowires might allow to study Majorana fermions which is an interesting scientific field. (Should I roughly understand what Majorana fermions are I might write an article about them but so far I only understand that they seem to very interesting ...) There are some possibilities for medical sciences as well but I do not know much about this. (E.g. something with detecting cancer I think.) Not to forget applications in power generation as I described in the article "The future of solar cells?".

Figure 1: Examples of nanowire
structures consisting of different
materials. The right image is a
cross-section through a wire.
A structure of this kind allows to
transport two signals at the same
time.

Now how are they fabricated and why do they have further interesting properties than just being small? And what actually is small? Nanowires again already implies the size: The diameter of nanowires usually ranges from a few to a few ten nanometers. Their length is usually at least 20 times as long. Otherwise it would not be a wire ... The small diameter now gives the wires a large surface to volume ratio. And this is the interesting part!

If you grow different elements on top of each other in crystalline structures often the top layer will break after a few nanometers. This is because different elements have different distances between the atoms in the crystals which causes strain in the growing layer. After a few nanometers this strain will be too large to allow a monocrystalline structure and the crystal breaks. You cannot use it for devices anymore. With the large surface to volume ratio the nanowires can balance the strain as there is a lot of space for the atoms to gradually change from one crystal structure to another and the different crystals do not break. This can be used to grow heterostructures in ways shown in figure 1.

The principle of fabricating nanowires is as easy as elegant. (Note: the principle.) On a substrate seed particles are distributed by nanoimprint (basically using a nanostamp) or from a gas. Afterwards the nanowire material is applied as gas as well. It will settle on the seed particles, diffuse through them and push the seed up. So underneath the seed there grow the wires. Afterwards branches can be grown and stuff like that. Which is basically what happens in Lund all day. ;)

Figures 2a) (top) Seed particles on a substrate. 2b) (bottom) Different stages of growing nanowires.

Hacking under the christmas tree

Just in time for the end of 2013 my laptop broke down. So now I finally had an excuse to put the plans that I already had for a few months into practice. The basic plan was to install Windows and Linux in parallel on the same machine and then do some fancy stuff with it. But this time "for real" and not with the help of a Windows installer. Because some people occasionally asked me about installing operating systems (OS) and partitioning hard drives now I wrote down what I did as a summary and a rough direction guide as this is usually what I am missing when I want to try something new.

First thing to do was reinstalling the Windows OS. For that I used the built-in recovery partition of my laptop. Although I knew that this would install a whole bunch of crap programmes that I would remove immediately afterwards I wanted to keep the nice power management tools that I would have lost if I had not used the factory setting recovery. Windows easy peasy done in half an hour. ;)

Next step was preparing the hard drive for the second Linux OS. For that I added three partitions with Partition Wizard: One for the actual OS (ext4 file system), one for data (ext4 file system) and one for exchanging files between Windows and Linux (FAT32 file system). Actually this separation of OS and data is not necessary, it is more a "good style" to do it. You should do it for your Windows system as well although I did not do it for my laptop as I tend to back up everything anyway. Nevertheless I should feel bad for this inconsistency and I do. There would have been just too many partitions ... I will probably use this exchange partition as a small network hard drive as well when I am back in Sweden so that I can exchange files within the home network as well without a USB stick.

Afterwards it was time to choose one of the around 600(?) available Linux distributions ... As you only learn stuff if you do it on your own and do it properly I chose Arch Linux. This comes along with nothing except the basic Linux OS. No graphical user interface (GUI), no drivers, no internet, no nothing ... Yay, so much fun ... ;)

I burnt the Linux image to a CD (yes, my laptop is old enough that it still has a DVD drive) to boot it with ImgBurn. Before I tried to boot it from SD card but my laptop seems to refuse booting from SD card in principle. -.- So CD it was. From the Arch Linux live environment I then installed the OS onto the hard drive following this tutorial. (Actually that is all the magic behind becoming a computer "expert": Read and try.) Most of the tutorial worked fine, only for the wifi part I had to change some of the commands by trial and error - it seems that the author mixed some old commands with some new effects ...

Anyhow, afterwards I installed the GNOME GUI following the Arch Linux Wiki Beginner's Guide. This took me as long as installing the OS itself. -.- This time it was for my stupidity and impatience which is inherent to every physicist/computer scientist nursery child that always prefers trying on his own rather than following boring guides. ;)

Actually I did not want to install a GUI at all to internalise "how to command shell". ;) But then I realised that I will have to read a lot in future and that I will not always have a second machine to use a browser so I installed a GUI ... Looks really nice by the way so it pays off optically as well. And installing it teaches you a lot about how OS work. :)

Last of course (which I did before installing a GUI but you can do it afterwards as well) I needed a boot loader that recognises both the Linux and the Windows OS. I just used the standard GRUB boot loader for that. The installation guide above includes this.

So after some hours of blood, sweat and tears (and coffee) I really managed to install my favoured dual boot system. No big deal once you know how to do it but first you have to try it on your own. Mission accomplished. :)

You should try installing a dual boot system as well! ;)

The future of solar cells?

As mentioned in one of the last posts we had a project at the university last week which we will finish tomorrow with a presentation. In this project we measured the reflectance of nanowire arrays depending on their diameter. What? And why?

One of the most promising energy sources ever is the sun. Solar cells are used to take advantage of this mega-source. But common solar cells are not that efficient. (20 - 30 percent I think.) So some intelligent physicists had the idea to investigate different approaches than using silicon for the cells. Especially here in Lund nanowires are of big interest. And I have to say the guys here are very good (means respected world-wide ;) ) at  growing nanowires. Now what are nanowires and why are they special? The name implies already what is behind the technique: Nanowires are wires with a very small diameter from a few nanometers up to maybe 100 nm. (And yes, this is true, we saw them under a scanning electron microscope!) The special thing about them is that they can be formed out of semimetals that have very nice electrical properties but cannot be used to form layers or bulk samples. This is because if they are present in solid form the distance between their atoms are very different for different metals. That causes strain in the material and this again causes cracks. Not good, solar cell broken. In nanowires this does not happen as the strain from the different atom spacing can be compensated due to the large surface-to-volume ratio of the wires.

Figure 1: Schema of the arrangement of the wires on the sample.
Indium phosphide was used during this project. On the right there
is a schema of the sample that we measured. Each array on the
sample contains nanowires of different diameter.

But back to nanowires in solar cells. As mentioned these semimetal compounds have nice electrical properties which promise higher efficiencies than common solar cells. If they are placed for example in arrays like shown in figure 1 they can - put simply - absorb sunlight which can be used to generate energy. Just like in common solar cells, only more efficient. What we did in the project was to measure the reflectance of such nanowire arrays. If energy is generated by absorbing sunlight it is of course very desirable that the cells/arrays absorb as much of the light as possible. And thus show a reflectance  as low as possible. That was what our measurements were about. We varied the diameter of the nanowires and measured reflectance spectra. Some are shown below (figure 2). Every colour represents a different nanowire diameter and you can see that the reflectances vary depending on the wire diameter.

The reason why the reflectance depends on the diameter is that the absorption and thus the reflection of light depends on how strongly the electromagnetic light fields couple into the nanowires. This again depends on the diameter of the wires so the reflectance depends on the diameter as well. Now only somebody needs to make this applicable for industrial fabrication and here we go - energy crisis solved. ;)

This is very recent research by the way! An extensive paper on this topic which we used to prepare for the project was released only in May this year!

FIgure 2: Reflectance spectra of nanowire arrays. Wavelength on the x-axis,
reflectance on the y-axis. Diameter of the nanowires was varied from 30 nm to 80 nm.
The length of the wires was 1.1 µm.

Trapped single states

Figure 1: Sample structure: Different
semiconductor materials stacked.

Today I have been working on a lab report all day ... Only after some time I realised how awesome the physics is that is behind the experiment! (At least I think it is awesome. :) ) We trapped single electrons in a quantum well! I mean really single electrons, not the assumed single electrons or only few single electrons. So as I am working on this anyway I thought I should share parts of it in a very high-level (that means not too much crazy science, only a little ;) ) way. I hope it is understandable. :)

The setup consists of a stacked semiconductor structure as shown in figure 1. The gallium arsenide and and the aluminium gallium arsenide have different band gaps which the electrons in the device see as a well in energy. This is like if you jump into a well in the ground - you see the walls around you once you are in the well. (See figure 2 for the wells.)

Figure 2: Two bands resulting from
the different band gaps in the structure
from figure 1 and possible ways
to excite an electron from the bottom
to the top. Laser!

The next prerequisite is to know that physicists assume that there are differen bands in semiconductors that define the energies that electrons can have. (In which they can move if you want.) Now there are not only bands for electrons but also for holes which are the conception of missing electrons. Just like if you take a chocolate marshmallow out of its box leaving a blank space. These holes can move as well just like electrons.

So if there is an electron in one of the bands (the valence band to be precise) it can be moved to another band (the conduction band) by giving energy to it. (For example by shooting at it with a laser - pew, pew!) From before now it is known that this will leave a hole behind.

Figure 3: Spectrum recorded
during the experiment.

After some time the excited electron (the one that was shot to the other band) and the hole recombine because it is more suitable for them. (Particles always want to reach the lowest possible energy state.) If they recombine they emit the energy that the electron got from the laser before. This is visible as light so it can be detected. By a detector ...

Because we are not sure (unless we calculate it before, as I just had to for the lab report ...) which energy this light will have we let the detector detect a lot of possible energies. For some of the detected energies there will be nothing special but for some other energies there will be peaks which means that at these energies there is something going on. For example recombination of electrons and holes. This measurement of different energies is called a spectrum. Figure 3 shows an original spectrum from our experiment. :)

But how do we know that we trapped only one single electron in our well? That we know from the calculations. The movement of particles can be described by the Schrödinger equation. The setup with the quantum well gives boundary conditions to this equation so we get discrete solutions for it. As the semiconductor setup used is a little more complicated than a "normal" quantum well (whatever a >normal >quantum well may be ... <.< ) the solutions for the Schrödinger equation could not be obtained analytically. Instead we used a graphical method. The clue behind this is to plot the graph of the equation. Only at points where the graphs intersect a bound (trapped) state can exist. If you look at figure 4 you see yourself that there is only one intersection point. So one single trapped electron.! BÄM, science! ;)

Figure 4: Graphical solution of the Schrödinger equation. There is only
one intersection point (blue and red) so there is only one(!) single(!) electron
trapped in the quantum well!

So this was a very much broken down overview of what happened. I hope you could understand something. I am still fascinated that something like this is possible and that we did it. O.O

Last thing: Why do we do stuff like this? Well if the behaviour of the electrons in the well is better understood one can build some fancy devices out of this! Maybe new transistors which result in faster computers? Maybe quantum computers? Maybe better solar cells? (Actually I participated in another project where we measured nanowire solar cell arrays! I hope that I will tell something about this as well.)

If I am that fascinated this seems to be the right stuff to study! :)