Taylor D. Grow and Dr. Robert C. Davis, Physics and Astronomy
Cadmium Selenide Nanocrystals are semiconducting particles that exhibit very interesting and as yet not understood quantum properties. These particles also fluoresce with very high efficiency. In order to study how the quantum state of these particles affects their flourescent properties Dr. Davis has designed an experiment where a single nanoparticle needs to be isolated between two gold electrodes. In order to do this I used a Digital Instruments Atomic Force Microscope to develop a procedure to move these particles (Fig. 1).
I started with Cadmium Selenide particles which were capped with Zinc Sulfide in order to increase their flourescent efficiency. Batch 10 was found to have the smallest size distribution and the best fluorescence with the particles being about 7 nm.
The next step was to make a sample that had a good dispersion of particles on a substrate smooth enough to allow the particles to be imaged and manipulated. After being synthesized the particles are in solution with either Hexanes or Chloroform. I tried several different concentrations of the particles as well as several techniques of placing them on the substrate which at first was glass. I found that a concentration of .663 g/L gave a very even dispersion. The simplest way of dispersing the particles was by placing a single drop on the substrate and then touching the left edge of the drop with a chemical wipe. In this manner the solution soaked into the chemical wipe leaving a good spread of particles behind on the right side of the drop while removing unwanted contaminates.
I then proceeded to find a substrate. Glass proved to be to rough so I then started using mica as the substrate which is considerably smoother than glass and allowed for easy imaging and eventual smooth manipulation.
In order to move the particles I first tried using contact mode. In contact mode the tip actually touches the substrate (see Fig. 2). I discovered that though I could get excellent images of the particles the contact mode tips were to dull in order to actually move the particles. Whenever the tip pushed hard enough to overcome the particle/surface attraction the particle simply stuck to the tip. Therefore I was forced to use tapping mode. In tapping mode cantilever is continually oscillating. By measuring differences in the amplitude of the oscillating cantilever which are caused by the tip/surface interaction the height can be measured. Fortunately the tapping mode tips are quite a bit sharper than the contact mode tips which allowed them to overcome the particle/surface attraction without having the particle stick to the tip.
The last step involved developing the actual mechanics of moving the particles. To do this I used a lithography programming environment provided with the AFM which allows the user to control individual movements of the cantilever. I wrote a program that would turn off the imaging mode and center the tip. Then the program lowered the tip towards the surface and dragged the tip along the surface. If the tip is lowered far enough so that the tip/particle interaction is greater than the particle/substrate interaction as the tip moves along the surface it pushes the particle. As long as the tip is not to dull and the surface is not contaminated by dust or oils, when the cantilever raises up to image again the particle will stick to the substrate at its new position. I wrote four programs–one for each direction up, down, left and right.
In order for the program to work there were several steps that need to be completed before initiating it. First an area on the sample with a good dispersion needs to be found and selected. Then a specific particle needed to be chosen and centered in the middle of the imaging area. Finally a capture plane procedure, which is a built in command, needs to be run so that when the program is initiated the tip moves parallel to the sample surface. The capture plane is necessary or otherwise the tip would most likely be damaged by running into the substrate surface that is almost always angled slightly.
The procedure and the four programs developed will be used to manipulate a single particle into an apparatus that will be used later to study the effects of the quantum state on the particle’s fluorescence.