User:Brian P. Josey/Notebook/2010/06/07
|Project name|| Main project page|
Previous entry Next entry
Continuing from Friday
I am going through and running my calculations from Friday. So far, it appears that magnetizing a needle and placing it into a solution of ferritin is not as promising of an idea as I had hoped. The main issue is that the field is much weaker than when using the large yoke, and the effective range from the needle is very short. This might be advantageous, in that I could directly control a very small area, or not.
I choose a sewing needle over a syringe to capitalize on the solid volume, and thus greater magnetic moment, of the solid needle. I also skipped over a wire to take advantage over the sharp tip of the needle. Creating my model, I simplified it, running it twice, once with a 52 MGOe neodymium magnet, and another time with a 32 MGOe magnet. The 52 MGOe serves as the upper limit of what I can expect, and the 32 serves as a better approximation of a magnetized needle.
The above table is the values I recorded for the forces a magnetized needle would exert on a single ferritin protein, I used my usual notation for the sets of data, namely the top three go vertically from the tip of the needle in increasing distance, and the horizontal is what I expect if I held the cover glass of a flow cell against the tip of the needle. The issue here though is that not all of the points are really relevant beyond demonstration. The key points of data are those nearest to the tip, which represent ferritin floating freely by the needle. As the numbers show, it is clear that the needle comes nowhere near the force of the yoke, and small size of the needle dramatically reduces the force. For the 32 MGOe needle I ran two simulations. The first, like the 52 MGOe model, was as close of a representation of a #12 sharp needle as I could create. Needles of this type thicken from the tip to about a third of the way down. This slowly leaks out the magnetic field. The second one, featured a much sharper tip that comes form machining, did not allow as much field escape. Here is a side by side comparison:
Note to self the distances for indexing in the FEMM analyzer code are: