Articles by: maz

GMR and the Nobel Prize

A computer hard disk reader that uses a GMR sensor is equivalent to a jet flying at a speed of 30,000 kilometers (19,500 miles) per hour … at a height of just one meter above the ground, and yet being able to see and catalogue every single blade of grass it passes over,” [Ben Murdin, a physics professor at the University of Surrey in southeast England] said.

Impressive, eh?  This year’s Nobel in physics was awarded for something that you actually use on a daily basis…but never think about. Albert Fert and Peter Grünberg get to split $1.5Million for their discovery of Giant Magnetoresistance (GMR).  I know; not only does it sound amazingly cool, but it’s actually useful?  I mean, anti-particles, darkmatter, parallel universes and string theory are sexy and all, but come on.  This tech. actually is called Giant.  Not to mention that the discovery of GMR led to the field of spintronics, and the principles behind your iPod and 500 gigabyte hard drive.

On to the science.

To start this discussion, I have to explain a little about spin and it’s interaction with mag. fields.

Let’s review.

So if you sent a beam of electrons through a magnetic field and had some quantum trash bags to collect the electrons upon exiting this mag. field, you’d see that 1 bag was full of spin up electrons, and 1 bag was full of spin down electrons.

One way you can attempt to rationalize this (and I know this is a little hand-wavy, but it’ll do) is to imagine an infinitesimal charge, dq on the surface of the electron.  If we see that the electron is spherical and spinning, either clockwise or counterclockwise, then this dq rotates about as well.  So would all the other dq’s on the surface of the electron.  You know that the flow of charge generates a magnetic field.  Through Maxwell’s eq. you could show that if the electron is spinning clockwise it’s z-axis, you’d see a magnetic field pointing in the “-z direction” (remember, no magnetic monopoles (we think) so the field lines could be thought of as initially going out the bottom of the electron. and then curling back up to arrive at the top), and if it was spinning counterclockwise, you’d see the field pointing in the +z direction.  These two dipole moments we call spin up and spin down, and they either align in parallel with the external magnetic field, or antiparallel with it.  As you’re probably guessing, the parallel alignment is slightly lower in energy then the antiparallel.  This is known as the Zeeman effect.


So GMR is, in my opinion, one really big and really cool trick.  I say so because it’s one of those things that seems to me as obvious but not trivial.  There are a few different types of GMR applications that operate off of the same basic principle.  Here I’ll talk about the most used one, the spin-valve.

If you take a ferromagnetic layer and polarize it, the unpaired electrons in the layer will align themselves to the external magnetic field.  These electrons are now what we call “spin polarized”.  Do this for a second layer and place them next to each other, but separated by a non ferromagnetic layer.  Slap a potential difference across the two layers, and the electrons will maintain their polarization while moving through the circuit.  But when these electrons hit a material with a mag. field opposite their spin direction, they get flipped.  And here’s the rub:  flipping the spins requires extra energy…in other words the electrical resistance is increased when the magnetic materials are polarized in the opposite directions (anti-parallel alignment).

Who cares?  Well, you do, you just might not see it yet.  Being able to change the like this electrical resistance, which is easily detectable, is what computer guys call “non-volatile”.  Meaning you don’t require power to keep the changes made.  If you let the “low resistance” represent 1 and the “high resistance” represent 0, you can get digital logic.  Imagine doing what we just discussed, hundreds of billions of times on a 3.5″ disc…and you’ve got yourself a hard drive.  This is a relatively simple type of spintronic device, which is why they say that the GMR discovery gave birth to the field of spintronics.

Spintronics is, in general, the development of technology which allows you to make use of the spins of electrons. If you can generate a current of like-spinned electrons, i.e. polarized electrons, and send them through a device that can detect and act based on the spin of these electrons, you have created a spintronic device.  It is sensitive not to voltage or charge or mass, but on the electron’s spin.  That’s a really big deal.  After-all, you’ve got an intrinsic 1 or 0 right there.

So the next time you strap that overpriced, little white noise machine onto your arm and go for a jog, say thanks to a couple of physicists who discovered something that turned out to be useful…in your lifetime.

Note:  This was by no means a rigorous discussion, and if anyone is interested in the topic, you could start a thread in the forums and see what grows.  For the blog though, this is maybe already too indepth.  I should have added some pretty pictures.  Tongue


By October 11, 2007 0 comments science news

Energy Discussions 1: Water Cars

Since I have seen a number a of energy related topics on the boards in the past few weeks and Mitch decided to talk about cold fusion, I decided my next few posts will be related to new/renewable energy alternatives to fossil fuels.

This particular post was inspired by the many, many electrochem. questions by Walman who also reminded me about Stanley Meyers’ work on water cars.

Yes thats right…water powered vehicles.

Now I know you are all rolling your eyes saying things like “electrolysis takes about 5 times more energy in then you get out” or “If he really invented some new form of electrolysis why don’t we all drive water cars?” and of course “Who cares? Maz is nuts anyways!”.

I would ask you naysayers to wait a bit and take a look at some of this guy’s evidence. NOTE: I don’t really buy it, but you never know. It may be possible.

So lets begin with a basic review of electrolysis, which is the separation of certain bonded atoms or molecules by running electric current across them. We are concerned here with the electrolysis of water which goes like this:

2H2O –> 2H2 + O2 Where the H2O was in liquid phase, and the hydrogen and oxygen products are in gas phase.

When you put enough electrical current across water, you add enough energy for the water to split into its ionic components. Hydrogen, being positively charged, moves toward the cathode and oxygen, being negatively charged, moves toward the anode. When hydrogen cations hit the cathode, they get reduced and form H2 gas. Oxygen hits the anode and gets oxidized, forming O2 gas.

Now the quantity of the separation is proportional to the amount of electric charge you send across. This means that the more current you send through, the more hydrogen gas you get (within limits of course). So we can all have electrolytic cells producing hydrogen to burn for our cars and homes, right? Well…not exactly. The amount of energy you get out from burning the products of the electrolysis is not greater then the amount of energy it takes to do the separation. Classical theory predicts the maximum efficiency to be between 80 and 94% See here for details

This key point is where Stanley Meyers claimed to make a breakthrough. Using his own design of an electrolytic cell, he said he gets somewhere around 1700% efficiency.

His design for the cell is different from contemporary cells in that they utilize tiny amounts of current. Half an amp, for his 1700% efficient design. The trick, it seems, is to use high voltages with low current and PULSE the current using large surface area electrodes.

Why does this supposedly work? You’ve got me there. Perhaps there’s some weird interaction driven by the strong force at the electrodes? Maybe you cold fusion enthusiasts ought to look into it with high pressure confinement. Maybe then you’ll see your fusion.

Whatever the case, and whatever your current opinion is, first watch these two videos. The first is just 2 minutes long, the second is a more serious 17 minute clip. THEN formulate your opinion. Of course I would also say you should visit the wikipedia article on Stanley Meyer

Video 1:

Video 2:

Obviously, I think that his ‘water fuel cell’ is pretty much a vat of crock. I am sure it is a conspiracy theorists dreamland, but then again, all the supposed “free energy” inventions are.

Except for mine of course. But that’s a secret.

P.S. If anyone reading this understands Japanese, could you please tell me what they are saying in this video?


By April 25, 2007 0 comments science news

Plasmonics Part II

Sorry for the long wait for this second post in the plasmonics series, but I got extra busy with school and the lab.

Anyways, before I start talking about what plasmons are and what they are good for, I want to thank “plasmonicsfocus” for pointing out a fact I failed to mention in my previous post.  The 10-6 transmission factor does not exactly represent the light passing through the tip.

See, classical diffraction theory predicts that the transmission of electromagnetic radiation through an aperture, of radius d, in a perfectly conducting plane obstacle should be given by,

T = (8/27)(pi)2(kd)4 + O[(kd)6]

So when the radius of the aperture is much smaller then the wavelength of the incoming wave, (d<<lambda) the transmission of energy through the aperture dies like (d*lambda)-4.

The catch here is that this analysis assumes the transmission is measured far from the aperture (far field diffraction). Far-field diffraction theory neglects those terms in the radiation field that die off faster than x-2 in the distance from the aperture.  So there is in fact, more energy that is interacting with the sample before the wave disperses to the far field.

For example, part of the transmitted energy emerges in an evanescent wave that travels along the reflecting interface between the media. Now it is possible to couple to this wave, despite the fact that its intensity diminishes exponentially with distance from the interface, and this allows one to create surface plasmons in thin metal films.

However I get ahead of myself.

Plasmons are plasma oscillations, or you could think of them as charge density waves.

Simply imagine a sphere comprised of many discrete, evenly-spaced positive charges, which we can approximate this as a positive charge distribution.  Now imagine a free electron cloud hovering just on top of the positive charge distribution, but not in contact.  The electrons are interacting with each other, and can be approximated as a negative charge distribution.  You have arrived at what is known as the Jellium model in metals.  These are plasmons in metal, and when the electrons move around, they are pulled back toward the positive charge distribution.  We assume they are not energetic enough to escape the E-field by the positively charged nuclei.  Thus they end up oscillating over/about the positive charge distribution.

In case you didn’t want to read all that, suffice it to say that plasmons in metal are basically vibrational modes of the electron gas density oscillating about the metallic ion cores.

Here is a picture to help you visualize it.

The oscillation frequency of the plasmon (wp), can be derived fairly easily after making the assumption that the E-field confining the electrons is harmonic.  Note how in the bottom right of the following picture, the frequency/waveguide (w/k = w*lambda) asymptotically approaches c.

Now surface plasmons describe the special case in which the charges are confined to the surface of the metal. In this case, the electric field is strongest in the plane of the metallic surface.  This is because the strength of the E-field dies off exponentially as you get further away from the charge distribution.

Plasmons confined to a plane do not radiate light, but when the local planar symmetry is disturbed, they can.  So unless you have a perfectly planar sample, and I really mean perfectly planar, the plasmons can radiate.  Surface plasmon-mediated emission from defects in metal surfaces has been observed, for example here.  These features have been described as “plasmon flashlights”; they are a localization of photon emission in a region generally experiencing non-radiative, collective oscillation of the surface electron density. Hence, It is possible to design structures that take advantage of this emission of surface plasmon radiation.

Lets take a look at a little more rigorous derivation of plasmon-light coupling:

NOTE: The slides in this post were made by Jim Shuck, LBL. 

That is really the main point to take from this discussion.  Surface plasmons can oscillate at relatively low frequencies for relatively small wavelengths.  That’s why if you can turn light into plasmons (couple them), you can do sub-diffraction imaging.

Next time I’ll wrap all this up by talking about plasmonic nano-antennas.  Until then, if you’re interested you might want to look at these papers.

Toward Nanometer-Scale Optical Photolithography: Utilizing the Near-Field of Bowtie Optical Nanoantennas

Diffracted evanescent wave model for enhanced and suppressed optical transmission through subwavelength hole arrays 

By March 19, 2007 3 comments materials chemistry


Have you ever felt more then a little dissatisfied with your SEM images?  Black and white just not doing it for you?  Wish you could see height differences at that small of a scale?  Just plain tired of babying that danged machine?  I know that more then once I’ve dreamed of using a bench top microscope to optically analyze our samples.  Some people have even tried (much to amusement of the rest of us) to jury-rig a set up to make this possible, but so far I’ve yet to here of any success (Mitch).

However, there is work being done on a method for being able to OPTICALLY analyze samples on the NANOMETER scale.  Yeah, it’s not bench top, but it has color, shadows, and did I say color?

And no, I’m not talking about NSOM, the method that is so lousy, inefficient, slow, and delicate that it seriously makes you question your commitment to science.

For those of you who don’t know what NSOM is, I’ll do a quick overview.  NSOM (Near Field Scanning Optical Microscopy), is a method of doing sub-diffraction imaging.  Remember that light is actually a self-propagating electromagnetic wave.  Like all waves, light diffracts when shoved through a tiny space (on the order of it’s wavelength).  Back in your highschool or lower division physics classes, you probably messed around with single slit diffraction and saw neat patterns like this.

Well, to shine very coherent light “spots” on your samples so that you can see them, you need a very very small aperture.  However when you go to use one that has a diameter close to or less then the wavelength of the light you send through, you see the light diffract.

So if you are trying to look at a sample and want to resolve details with <1000nm with visible light, your in trouble.  NSOM “gets around” this though, by sticking the NSOM tip very very close to the sample.  As in around 25-50nm away from it.  Why does this work?  When you get the tip that close to the sample, it is acting like a light source because you are inside the first diffraction maximum.

The setup kind of looks like this:


Jim Schuck, Molecular Foundry, LBNL


Now that all seems fine, but inherently you’re hitting another problem here.  The tip is so much smaller then your wavelength that it greatly prohibits the photons from getting through.  Of course some will, but now it is a matter of statistics.  The math has been cranked and the transfer efficiency of NSOM is only about 10-6!!  That means 1 in 1 billion photons sent at the tip gets through!  LOUSY

So now we get to the brighter solution.  heh.

Or actually, we won’t just yet.  This post is starting to get long, and I am starting to get tired.  I think I’ll do this in a 3 post manner.

1.) NSOM and why is bites.  2.) What the heck are Plasmons?  3.) Plasmon-light coupling and Nano-antennas.

Stay tuned…


By March 7, 2007 0 comments materials chemistry