User:Brian P. Josey/Notebook/2010/05/25
|Project name||Main project page|
Previous entry Next entry
Various Magnet Set Ups
I've put together a couple of simulations to test which one would be the most useful set up to have mounted on the microscope stage so that I can observe the ferritin while it is under the influence of the magnet. After running these simulations, I was able to calculate the average force a single ferritin would feel in four different cases:
Here are the forces that I calculated for the various models:
In addition to the standard yoke, which comes in 0.1" and 0.4" gaps here, I created four other models that best represent the different ideas that I came up with, or represent a simple idea.
Note: I measured the thickness of the microscope's stage at 4/10 inch, and all of the models were designed with this in mind.
I know that magnetic force on a dipole is dependent on the gradient of the field, but it has not registered completely with me yet. So I created a simple model of a single two inch long, 0.5 inch diameter, N40 neodymium magnet. This represents the simplest magnets that I have. From here, I simulated the field from the surface at the center of the flat surface moving outward (in the same direction as the magnetization). Then I for the lateral force in a flow cell.
When I calculated the force, I found that there force nearest to the magnet is about 1/81 the strength the 4/10 inch gap yoke, and the other points follow a similar trend. The only difference is that on the edge of the cylinder, the force for the points nearest the magnet is stronger, but this is simply the result of the gradient. Here the magnetization is slightly weaker, but the magnet is larger, and the field is also concentrated into a sharp corner. Looking at the numbers, there is no real advantage to the cylinder magnet beyond this one place where the force is stronger. Even then there is the practical problem of getting the edge close enough to the sample.
The second model that I created was of two cylinders, each identical to the one in the single cylinder case, held along the same axis of magnetization with a 4/10 inch gap between them. In Gary's paper, he noted that if the magnets had different poles (ie. north and south) facing each other, they could potentially create a line or ferritin, while having the same poles (here I simulated south and south) it could create pockets of ferritin on each side of a cell placed half-way between them. In both cases, I found similar results to a single cylinder magnet, in that there was no real advantage to the cylinders, except very near the edge. In that case, a single magnet was favorable.
For the next case, I simulated two cone magnets attached to the cylinder magnets in the above cases, and held them so that the tip of the cones were separated by 4/10 inch. From my data, there is a very slight, ~0.01 fN near the tip, advantage for using this model. However, this advantage is so slight that choosing between the two cones and the yoke appears to be simply a matter of preference. The advantage of the yoke is that the magnets are held in place, while the cones do not have the issue of magnets running into the stage that the yoke does.
Here, I positioned two of the saddle magnets that I designed on Friday facing each other with their magnetization in the same direction. As I found on Friday, the force is stronger near the cylinder magnets on the side, but the force near the cone is close too that I calculated for the yoke. The advantage of the saddles however is that they are smaller and more compact than either the yoke or the two cones.
Single Particle Tracking
I want to show that I am moving individual ferritin proteins soon, and to that effect, I am starting to look into ways of doing that. I found one review paper, "Tracking bio-molecules in live cells using quantum dots" by Yun-Pei Cheng at UCLA, link to paper, that I found useful to read through. Her paper gave a brief overview of several different methods to track cells and macromolecules, culminating in why she believes that quantum dots are a good choice for tracking structures in a cell.
Essentially the argument comes down to three things: photobleaching, size of particle and absorption and emission spectrum. For the absorption and emission, florophores have a short range of absorption and a broader range of emission, while in quantum dots, the trait is reversed. The advantage in this is that quantum dots are more readily able to absorb different frequencies of light, and emit a very narrow band. This allows for specific quantum dots to be attached to specific structures and be differentiated. Quantum dots obviously win in the photo-bleaching category. It was noted that the one scientist was able to track diffusion of glycine receptors in living neurons for twenty minutes using quantum dots, while the fluorophore CY3 last five seconds before photobleaching. She also noted that a typical life time is around ten seconds.
I asked Andy about flourophores and photobleaching. He said that if I photobleached one I cannot recover it by allowing it to sit, which is what I assumed. All of the fluorescent dyes have benzene rings, where the electrons resonating in pi and sigma bonds are the cause for the fluorescence. He wasn't sure about the exact mechanism, but light knocks off one of the electrons, and this chemically makes it impossible for the dye to fluoresce again.
The other issue is with size. When they are originally synthesized, quantum dots are 2-10 nm in diameter and hydrophobic. They are then coated in an amphiphilic substance, before they can be used in the aqueous environment of the cell. This pushes the diameter up to 12nm for CdSe/ZnS quantum dots, and 15-20 nm for streptavidin quantum dots. They can reach upto 25 nm in diameter depending on what is attached to it. On the level of the cell, or of whole sub cellular structures, this is not that important, but it is significant when dealing with proteins. In the case of proteins, this is actually pretty large and could create steric strain that affects the proteins functionality. Also, in my case, the diameter of ferritin is about 12 nm, so attaching a quantum dot to a single ferritin could dramatically affect my results. If I wanted to track individual proteins I would have to use an organic flourescent dye and not a quantum dot. However, this does not rule out the "salad dressing" idea.