Preparing your crystal for data collection.
Mounting your crystal on the goiniometer
Collection single crystal diffraction data.
Collecting Multi-wavelength of Single-wavelength Anomalous Dispersion Data.
This primer is for performing experiments at DND-CAT 5IDB station. I will describe the tools available for MAD phasing as well as some beamline details.
The Amptek detector is an energy-sensitive device. This means that it puts out pulses proportional to the energy of the photons it is detecting, with moderate energy resolution (i.e., for a 10 keV photon in, the detector electronics puts out a pulse of perhaps 4 volts, + or - 0.1 volts). In a MAD experiment, there is an element in the sample which will fluoresce, or put out characteristic energy photons, when the incident x-ray beam from the synchrotron is above the absorption edge energy of the element, but not when the incident x-ray beam energy is below the absorption edge.
To use the Amptek, make sure it is on (the power switch is in the back) and that the detector is pointing at the sample. Also make sure that the MAR shutter is open. The signal from the Amptek is counted in two ways on your counters on lentil1. An integrated count of all the photons hitting the Amptek is given in channel 2. This should be no more than around 3000 counts per second. You may need to put some absorbers (Al foil, for example) in front of the MAR to get the rate down, or move the Amptek further away from your sample.
The counts from a small energy window from the Amptek will also appear in channel 4. It is this window which you want to adjust following the procedure in this note, so that the counts from the fluorescence appear in counter 4. This will allow you to get better statistics when you perform an EXAFS scan of your sample.
The counts in channel 4 come from a Single Channel Analyzer in the center of the NIM bin next to the 5IDB desk. The only adjustment which you may need to make to this unit will be perhaps to change the window size (upper knob, nominal setting 0.25V). We now have control of the position of this window with a computer interface, to a "motor" called sca_level. To properly set this window position, you should use the scanning capability as well as the Multichannel Analyzer (MCA) display on the Compaq computer on top of the rack next to the 5IDB desk.
First set your incident x-ray energy below the absorption edge of your sample. Make sure the total count rate in the Amptek is no more than 3000 cps (check counter 2). Take a spectrum with the MCA (Acquire Menu, Clear and then Start). You should save this spectrum (File Menu, Save As). Now go above the edge, and take another spectrum (again check your total count rate). You can compare this to the first below-edge spectrum with the Compare command (File Menu, Compare). You should be able to see your fluorescence line, which will appear only in the spectrum from above the edge.
Now you want to get the sca_level window to be at the position of your fluorescence line. On lentil1, in the main protoscan window, choose scan actuator. Now choose the motor sca_level. For your first scan choose a starting point of 0 and an ending point of 2000, increment of 20, count time of 2 s, filename sca_level, counter 4 (!). Hit scan. You should now see a plot which should closely resemble (except for noise) the spectrum on the MCA. You may wish to do a more careful, longer scan (i.e., 500 to 800, steps of 5, count 5 s, counter 4 (!)) to get better detail in the region of the fluorescence peak.
Examine the plot to see where the maximum intensity from your fluorescence line appears. This is where you want to set the sca_level detector. If you prefer to get a better position estimate, plot up the scan with xyplot (for example, from a terminal window (not the protoscan window) issue the command xyplot sca_level_003.dat 1 6, (fill in your filename for sca_level_003.dat) this will bring up a xmgrace window in which you can see both the intensity and the position of the cursor). Once you have chosen your target value for sca_level, choose the move actuator menu item (or this screen may already be present). sca_level is at the bottom of the motor control screen - just choose this actuator, enter your target position and choose the move button.
Now it is time to perform an edge scan. Using the main protoscan entry system, put in your scan parameters. (Recommended: background start at –50 eV, edge start at -10 eV, K start 3.0 K, K end 3.1 K, background steps 5 eV, edge steps 1 eV, k steps 0.01 K, count time start 2.0, count time end 2.0, k weighting 1.0, dark count time 0 (! This must be zero unless you have a PF2S2 shutter set up in the hutch!), wait time 1.0s, repeat 1 time). As the scan begins, choose to display mu fluorescence as display number 1. You will see your edge scan displayed.
If your data looks too noisy, change the scan parameters by increasing your counting time, or perhaps choose to repeat the scan several times.
If you would like to examine the data more carefully, or to take the derivative of the spectrum, plot up the data with xyplot (for example, from another terminal window on lentil1, issue the command xyplot pt_edge_004.dat 4 17 (in EXAFS data files, column 4 is the energy, column 17 is the mu fluorescence data). You can view directly the cursor position. If you want to take the derivative of the spectrum, choose data, then transformations, then differences, then forward difference. Your derivative will appear in a different color than the data. You can zoom in or out by choosing buttons on the left side of the display: ZX zooms the x axis (you must draw a box with the mouse to show the zoomed area); ZY zooms the y axis, AX and AX autoscale the x and y axes, respectively.
The current counter setup for lentil1 is as follows:
Ion chamber counts
1.Amptek total counts
2.Ring current (5010 cps/mA)
3.Amptek energy window
The counting time when using the start counting/stop counting buttons is set by the value in the start count time box on the EXAFS menu. If you want to change this time, stop the counters, enter a new time, and then start counting again. The value in counter 3 will indicate the counting time interval.
If you have trouble with lentil1, you may need to log out and log back in again. The username is user5idb. You should make a note of the current password in a secure location. The programs you will need to restart after you log back in are protoscan (start from one terminal window) and servo_piezo_c (start from a different terminal window).
Servo_piezo_c allows you to stay at the same detuning position on the rocking curve of our silicon monochromators (typically the 70% level of the mono_coil scan). With the servo off, position the mono_coil motor at the appropriate position, and then hit the start servo button. You should leave this on for monochromatic data collection, but off for EXAFS scans, and when changing energies (wavelengths).