We shall perform this experiment on the either the standard D2O sample or the CHCl3 in acetone sample. It really doesn't matter which. Lock and shim and set the O1 such that the compound in question is approx. 50 to 100 Hz off-center. Set the receiver gain to a reasonable level and acquire a one scan spectrum. Increase the on-screen amplitude (using the arrow keys or the coarse and fine adjustments) so that the FID covers about 1/3 of the screen.
We are now going to switch to a four-scan experiment. Acquire and observe how the amplitude of the FID grows throughout the experiment. Now, go into the phase program for the receiver (AQ) in the pulse program section and change the second scan to 180 degrees opposite (e.g. if it is x make it -x, if it is y make it -y). The phase programs are built on the following convention: x = 0, y = 1, -x = 2, -y = 3. Acquire a new FID. What do you observe? Change the phase diagram such that the signal goes away completely. Write down the phase program which accomplishes this.
Change the phase program back to what it was originally (if you have forgotten, just call in the parameter set again). Do the same experiments but, this time, play with the phase of the 90 degree pulse. After you have played with these parameters, you should start to feel confident in predicting the effect of a change in the phase program.
Now let us try to do something useful with the phase program. Set the pulse program so that it pulses and acquires from the same direction for each of four scans (e.g. Pulse x x x x Acquire x x x x). Acquire a spectrum. How bad is the O1 glitch in the center of the spectrum? Play around with the phase program so that the "real" peaks are maximized (4 x one scan) but the O1 glitch is hopefully gone. What is the optimum phase cycle to accomplish this task? Is there more than one way to do it? Can you do it in two scans? Do eight scans improve it (with repetition of a four-scan cycle or with an additional change in the phase cycle)?
I will look tomorrow to see if there is a way to turn off automatic quadrature detection. If there is, we will do that as well and see how well phase cycling can remove quadrature images.
All right, if you have questions about this experiment, let me know.