>I have heard that digital filtering brings about the sampling of data points at
>negative times which can be corrected for by a 1st order phase correction.
>
>Xwinnmr does this automatically but how can I do it myself (outside of
>Xwinnmr). How can I calculate the value for the 1st order phase correction?
>
>I have also a related question: Why is the first point in the FID divided by 2
>by default before the FFT (parameter FCOR as stated in the xwinnmr manual). Is
>this also done in the F1 dimension in 2D spectra ?
-------------
The first order phase shift (P1) is caused by the sampling not
started at the time origin. Because we have to give a short delay before
we can sample the FID, the individual magnetization components process away
from each other by an amount that varies with frequency during this time
period. As long as the delay is not very long, you can treat P1 as a
linear function. If the phase difference becomes too big during the
waiting time, the spectrum will be difficult to deal with. {One can
circumambulate the problem by displaying either the magnitude of the power
spectrum, neither of which has a phase dependence. The price to pay is the
loss of the phase information in the spectra.}
The first point in the FID need to be divided by 2 is to get rid of
the baseline. Since the digital FT requires the number of discrete data
points to be the power of 2, if you decompose a FID into the sum of
symmetrical and antisymmetrical parts, you will find that the data point at
the origin will introduce a finite baseline after FFT if this point is not
divided by 2 [A single point (a spark) at the time origin will produce a
baseline in the freq domain via FFT]. The value of the base line is A/N
where A is the original signal amplitude and N is the number of discrete
points.
Hope it helps.
Yang Xia
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~ Yang Xia, Ph. D. ~
~ 274 Hannah Hall, Physics Dept, Oakland Univ, Rochester, MI 48309, USA ~
~ Tel: (810) 370-3420(office), (810) 370-2094(lab), (810) 370-2403(NMR) ~
~ Fax: (810) 370-3408 Email: xia@oakland.edu, xia@vela.acs.oakland.edu ~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~