Noise Considerations

This experiment is very sensitive to noise because the signal that the technique tries to pinpoint essentially looks like noise itself. That makes it so it is very easy for one to confuse noise with accurate DLS results. This section is dedicated to some of the problems that are easy to run into when running a DLS experiment.

External Light

External Light is one of the most detrimental things to a DLS experiment. Any free photons that hit the photodiode will add to your fluctuations that should just be reserved for photons scattering off the spheres. Turning off the computer monitor, all lights, blocking off LEDs, anything that gives off any light should be turned off, blacked out, or blocked. A Faraday cage over the sample and the photodiode reduces the chance of this happening, but less light will always be better.

60 Hz Noise

60 Hz noise popped up repeatedly through the duration of this experiment. Blocking out the fluorescent lights in the room, turning off the computer monitor, working the amps off of battery power are all entirely necessary. If none of those are done 60 Hz noise will flood the spectra that is calculated. Oddly enough, there seemed to be 60 Hz noise that occurred when the oscilloscope was plugged in on occasion. Whenever this appears it is best to look for something that is on and plugged into a wall outlet.

Circuit Noise

With the equipment that this iteration of the experiment was run, the circuit noise is dominated by the photodetector by a factor of two or three times both of the amplifiers combined. To give a rough estimate. Noise from the circuit with just the amplifiers contributing hovered around 6 to 8 millivolts. While the noise with the amplifiers and the photodiode’s detector blacked out was around 20 to 27 millivolts. The sensitivity of the measurements that will be made with this equipment is almost entirely a function of the photodiode.

Concentration Dependence

DLS measurements rely on being in the single scattering regime. If the sample placed in the holding cell is too concentrated, then photons will start scattering off multiple particles before reaching the photodiode which changes the result since the derivation of the width of the signal’s spectrum specifically relies on one photon with one particle interactions before scattering off. Even if data looks good it should be checked with several concentrations above and below to ensure that the data that has been taken isn’t dependent on the concentration of the spheres in the solution.

Hopefully this information is useful to anyone wishing to carry out the experiment in the future.