How to collect a chromatic aberration correction profile:

The Optima AUC is capable of collecting multi-wavelength data. However, all instruments we reviewed exhibit some level of chromatic aberrations. These aberrations seem to be instrument dependent, and are different for each instrument. Therefore, they need to be measured and properly calibrated and associated with the instrument to make sure each wavelength is properly shifted to the position that was obtained for the radial calibration, and the wavelength at which the radial calibration was performed.

To collect a chromatic aberration correction profile, do the following steps:

  1. Perform a radial calibration at 3,000 rpm and note the wavelength at which the calibration was performed (say, 350 nm). Turn off the radial calibration checkbox on the instrument, the instrument will save this information internally for future runs and this does not need to be repeated unless changes to the optics are made.
  2. Fill two cells with water, in both channels, about half-way full.
  3. Place one cell in hole 4, and one cell in hole 2
  4. Create a run profile to scan both cells in UV-intensity mode from 190-800 nm with 10 nm increments (62 wavelengths) at 10 micron radial resolution, 20C, covering the radial region where the menisci are located. Spin the rotor at 14,000 rpm. Wait for 1 hour while the rotor is spinning at 14,000 rpm before starting to scan to allow the rotor to equilibrate for rotor stretch and temperature (adiabatic stretch-cooling). Repeat each scan at each wavelength 5 times.
  5. Import the data using the Optima Data Viewer (Utilities:View Raw Optima Data)
  6. Export OpenAUC data
  7. Click on the "Zoom" button in the plot window that shows the cells. For each cell and channel, draw a zoom box around the downward facing peak of the meniscus (in intensity mode)
  8. In your $HOME/ultrascan/imports folder will be a directory with the run name. In this directory, you will find 8 new files:
    1. 2A.wavelen.radpos.dat
    2. 2A.wavelen.speeds-meniscus.dat
    3. 2B.wavelen.radpos.dat
    4. 2B.wavelen.speeds-meniscus.dat
    5. 4A.wavelen.radpos.dat
    6. 4A.wavelen.speeds-meniscus.dat
    7. 4B.wavelen.radpos.dat
    8. 4B.wavelen.speeds-meniscus.dat
  9. Import the four *.wavelen.radpos.dat files into Excel or other plotting program and average the 4 file by adding the 4 radial positions from each wavelength and dividing by four. For example, if your column A is the wavelength, and columns B-E are the radial positions for each wavelength, you could use a formula like Fn = (Bn+Cn+Dn+En)/4 where n is the row number.
  10. Subtract the AVERAGED value at 350 nm (where you did the radial calibration) from all other averaged values. Typically, some of the differences will be positive, some will be negative values. The value at 350 nm should be zero.
  11. Using a plotting program, fit these average values to some well parameterizing function (which one you should choose depends on the shape of your curve).
  12. Export the wavelengths and corrected values of this parameterizing function from 190-800 nm in 1 nm increments to a datafile (using this ASCII format - see example here).
  13. In UltraScan, and as user with admin privileges (userlevel=3), load this file into the instrument record. (Edit:Preferences:Instrument Preferences:Change:Edit Current Entry:Load Chromatic Abberration Array) and set the Radial Calibration Wavelength to 350 nm.

You now have a chromatic aberration correction file in your LIMS database instrument record that will be used each time you import data from this instrument to your disk with the Optima Data Viewer. The program should inform you that it is performing these corrections while exporting to OpenAUC data format. Each radial position will be adjusted by the increment specified in this file to make sure the proper radial position is recorded regardless of wavelength used.

Please note that this procedure does not care about the absolute values recorded for the meniscus positions, it only needs the relative differences between measurements made at different wavelengths to generate this file. Therefore, the absolute rotor stretch value here is not relevant (just in case you wondered).