Resolution Enhancement

Several options are provided for resolution enhancement:  

lg - Loretnzian-Gaussian
tf - Traficante filter
jt - Traficante filter
ms - mutliply sine
tm - trapezoidal multiplication
sgf - Savitzky-Golay filter
p2d - positive second derivative

LG -- Lorentzian/Gaussian resolution enhancement

This command multiplies the FID by a function which combines a Lorentzian using a negative line broadening with a Gaussian. The composite function has the shape shown below. Two parameters must be specified before LG can be executed: LB (line broadening, the same parameter used by the EM command) and GF (Gaussian factor). LB must be negative for the Lorentzian/Gaussian function. If LB is set to a positive number, NUTS will use the negative of that value. GF is a number between 0 and 1 and defines where the maximum of the function will be, as a fraction of the acquisition time. (This is the same as the Bruker GB parameter). The default value of GF is 0.3, and can be set with the GF command.  By default, the command is applied to all slices.  It can be applied only to the current slice with "lg this".

Reference: A.G.Ferrige and J.C.Lindon, J. Magn. Reson., 37, 337 (1978).

GF -- Gaussian Factor for LG command

This is one of the 2 required parameters for Lorentzian/Gaussian resolution enhancement. It must be a number between 0 and 1. (If its value is set outside these limits and the LG command is executed, NUTS will reset the GF value.) The GF value sets position of the maximum of the Gaussian function, expressed as a fraction of the acquisition time.

TF -- TRAF resolution enhancement

Performed on a FID. The function has the shape shown below. The user must input a value for LB, which is an estimate of the natural linewidth. The TRAF function provides resolution enhancement with minimal loss of signal-to-noise.

Reference: D.D.Traficante and D.Ziessow, J. Magn. Reson., 66, 182-186, (1986).

JT -- S-TRAF resolution enhancement

Performed on a FID.  The function has the shape below.  The user must input a value for LB, an estimate of the natural linewidth.  Shown below is LB = .3

MS -- Multiply Sine

Multiply the reals and imaginaries by a window function which is the first half of a sine wave. Executing the command twice gives a sine squared window function.  The command can have one argument, which is the desired phase shift for the sine function, a value between 0 and 90.  If the command has no argument, the current phase value will be used.  The phase value can be set with the S# command.

These 2 figures show the shape of sine and sine squared functions.

The sine function may have a starting phase different from zero. 

The next 2 figures show the shape of sine and sine squared functions with phase (S#) of 45.

The next 2 figures show the shape of cosine and cosine squared functions, which are sine functions with phase (S#) of 90.

S# -- Phase shift for sine multiplication

Used in conjunction with the Multiply Sine (MS) apodization function. The phase shift is entered in degrees. This can also be set within a macro, for example, for 2D processing.

The only valid values for S# are between 0 and 90, inclusive. Entering a value less than 0 or greater than 90 will cause NUTS to set the value to 0 or 90, respectively.

TM -- Trapezoidal Multiplication

Multiplies FID by a trapezoidal shaped function defined by parameters T1 and T2. The first T1 number of points are scaled linearly from zero to one. The last T2 number of points are scaled linearly from one to zero to avoid truncation of the FID. Other points are unchanged. 

It is first necessary to set the values of T1 and T2, expressed as percentage of the data points.  For example, the following series of commands results in application of a TM function with T1 and T2 set to 10% and 60% of the data points, respectively.

tm 1 10
tm 2 60
tm

SGF -- Savitzky-Golay filter

This can be applied in 2 different ways, for different purposes.  [Details forthcoming.]

The Savitzky-Golay filter method essentially performs a local polynomial regression to determine the smoothed value for each data point.

Used differently, it is a resolution enhancement technique based on taking the second derivative of the spectrum.  In our hands, the p2d command gives better results;  see below.

P2D - Positive 2nd Derivative

This command takes the second derivative of the spectrum, inverts the real half of the data, then takes the absolute value of all data points.  An example is shown below.  The red spectrum is a multiplet resulting from FT with no window function.  The blue spectrum is the result of p2d. 

Last updated: 6/24/08