2D experiments can be loosely grouped into three major categories:
homo-nuclear, hetero-nuclear with conventional detection and hetero-nuclear
with inverse detection. Ordinarily the correlation is between spins with
scalar coupling (J) or exchange of magnetization (NOE or chemical exchange).
Two basic methods are used to generate a complete 2D data matrix, leading to
either an absolute value (magnitude mode) spectrum or a phase sensitive
representation (Pure Absorption Phase).
Absolute value experiments are more and more being replaced by phase sensitive versions for a number of reasons which are beyond the scope of this section. In most cases, absolute value spectra are relegated to survey type, routine spectra.
Phase sensitive 2D spectra can be generated in two ways: Time Proportional Phase Incrementation (TPPI) where the phase of (usually) the first observe transmitter pulse is incremented by 90° for each set of FID’s, and the method proposed by Haberkorn, Ruben and States where two sets of FID’s with a 90° phase shift are collected for each increment of the evolution period. The choice between the two seems to be a matter of personal preference.
Parameter files are set up to provide a starting point for some of the most often used
experiments. In many cases the basic parameter files can be used for other experiments
with only minor modifications. By convention, the first dimension is referred to as F2,
the second dimension F1.
qtocsy.400 Quick TOCSY, producing a COSY-like spectrum. The mixing time is short;
only direct couplings show. 1 scan per increment.
Spectra can be run in less than 10 minutes!
cosy45.400 Proton Homonuclear COSY, magnitude mode. 45° detection pulse.
Suitable for quick survey spectra of small molecules. Qtocsy should work
better and take less time.
cosydftp.400 Phase sensitive COSY with double quantum filter and TPPI. Can be used to
set up NOESY experiments as well; change the pulse program to, e.g.,
noesytp and check the delays.
roesytp.400 Phase sensitive ROESY experiment for exchange measurements with a CW
spin lock pulse.
invcar.400 Inverse (proton-) detected proton-carbon chemical shift correlation.
xhcar.400 Proton-carbon chemical shift correlation.
Since it is not possible to provide a set of parameters suitable for all cases,
some further customization might well be needed. Usually, a 1D spectrum
(in the case of hetero-nuclear experiments, spectra for both dimensions might
be desirable) is first obtained to establish the values needed for the
following parameters:
SFO1 (BF1 AND O1)
SW (spectral width, for both dimensions)
RG (receiver gain)
SFO2 (BF2, O2 if applicable)
OFFSET (for axis calibration, for both dimensions)
Frequently, the spectral width can be reduced to the minimum width required,
particularly in the second dimension. Using the minimum width results in
better digital resolution for a given number of increments. To obtain the
same resolution with a larger width requires more increments, thus
increasing the experimental time requirement. Use OW2D to set the F2/F1 sweep widths.
Once the 1D spectra have been acquired with optimized parameters, they can be
saved for later plotting together with the contour plot.
After selecting a new data set name (edc) and reading in a 2D parameter set,
use ‘eda’ or [Param] > [Data/Acqu/Proc] > [eda] to access the 2D
acquisition parameter menu. Parameters for both dimensions can now be set up.
Minimally, the following acquisition parameters need to be checked and altered as needed:
F2 F1
NS
TD TD
SW SW
SFO1 SFO1
SFO2 IN0
RG ND0
Depending on the software version, changing SW(F1) may require that the value
for IN0 must be adjusted. The comments in the pulse program state how to set
IN0 based on the values of ND0 and DW(F1). All delays and pulses need to be
checked as for any experiment (use ‘ased’).
Before the acquisition is started, the processing parameters should be
set up as well for the F1 parameters to be properly recognized during
subsequent processing. Use ‘edp’ (on the same menu as eda) to
access the processing parameter menu. The following parameters need
to be checked:
F2 F1
SI SI
SF SF
OFFSET OFFSET
WDW WDW
Note that WDW is used to select the window function applied to the FID.
Only the parameters for the selected functions need to be correct.
If, e.g., the WDW = SINE function is selected, only the SSB value is
used, the values for LB, GB etc. are ignored during processing.
After all parameters are set, start the acquisition with ‘zg’. Once
the acquisition has been started, the progress of the experiment can be
followed after a minimum of 16 sets of FID’s have been acquired. The
processing parameter TDeff for F1 determines how many sets of FID’s
are to be used for the transform. Set TDeff to a value equal to or less
than the number of sets acquired. Click on [Param] > [data/ACQU/Proc] > [edp]
and select TDeff or type ‘1 tdeff’. Type ‘xfb’ or click on [Proc] >
[xfb] to start processing.
NOTE: Do not forget to restore TDeff before processing after the acquisition
is completed! To use all of the sets, set TDeff to be 0.
Caution: On occasion, processing a 2D while still acquiring has crashed
the system! (requiring exiting uxnmr to run killcops)!
It is best to process data only to verify that the experiment is working.
Transfer the Data Set to a workstation for further processing, and let the
acquisition continue.
The 2D display mode provides many features ranging from simple operations
required before plotting to more sophisticated manipulations. With the mouse
positioned over {thresh}, hold the middle button down while moving the mouse
to select the lowest level to be displayed. The right button can be used to
adjust the number of levels to be used for the display.
At the very least, the 2D spectrum will probably need final adjustment
of the axis calibration. Click the left button on {Calib} to enter this
routine. After positioning the cross hair, click the middle mouse button
and enter the values for both dimensions.
The display can be expanded by first clicking the left mouse button and
moving the mouse to one corner of the area to be expanded. Click the middle
button, position the mouse at the opposite corner desired and click the
middle button again. Finally, click the left button again, move the mouse
to {zoom} and once more click the left button.
If the plot parameters have not yet been adjusted (if necessary),
use ‘edg’ ([Param] > [Plot/Print]> [edg]) which now brings up the
2D plot menu. The projections for F1 and F2 should either be turned off
or changed to reflect the desired selection, e.g., internal vs. external
(including the 1D file name) projections.
If internal projections will be used, click the left mouse button on {1D}
to enter this mode. Click the middle button on {F1Proj} and {F2Proj} to
initiate the calculations. Click on [Return] to return to the regular 2D
window. After possibly re-adjusting the level display, click on {DefPlot}
to store the level selection and to specify the number of levels to be
plotted. Click on [Output] > [Plot] > [plot] to initiate plotting.
Phase sensitive 2D spectra may need a phase correction in either one or both
dimensions. The need for a phase correction can be determined by examining
positive and negative levels separately. For experiments such as a ROESY the
diagonal normally is 180° out of phase compared to the cross peaks
(except for cross peaks due to chemical exchange or residual J correlation).
The corrections can be done using the Phase Correction Subroutine (details
are as described in section 19.2 of the Bruker uxnmr manual). It is also
possible to use the manual procedure described below.
2D Phase Correction Subroutine
The 2D Phase Correction Subroutine allows mouse selection of rows and columns
to be copied to three windows. A window is activated by clicking the left mouse
button inside of it. The phase corrected rows and columns are shown in real time.
Xf1p, xf2p or xfbp can be used by clicking on [Phase]. [Store] saves the phase
parameters for later use by the phase commands after returning from the phase
correction subroutine.
Manual Zero Order Only Phase Correction
Zero order phases errors for F2 and F1 can be separately corrected with
the following procedure:
From the 2D window, click on {1D}. Select row (for F2) or column (for F1)
display by clicking on {Row} or {Col}. Move the mouse to the desired position
and click the middle button. This selection can be repeated for different sections.
The vertical scaling can be adjusted as needed. After selection of the row or
column, click on {1D-Mode}. In the regular 1D mode, phase correct the row or
column and click on [Store2D]. Return to the 2D window by clicking on {2D}.
To apply the phase correction, select [Proc] > [Phase] > [xf2p] for F2
(xf1p for F1) phase correction. Typing ‘xf2p’ or ‘xf1p’
is also possible.
If you find that the phase in F2 consistently differs from a 1D
spectrum by 90° or 180°, the particular pulse program can be
edited by adding 1 (or 3 for -90° ) or 2 to the receiver phase in the pulse program.
Manually Correcting Both Zero And First Order Phase
Correction of both zero and first order phases can be done for both
F2 and F1. If a row or column contains peaks near both the beginning
and the end, the first order phase can be adjusted at the same time
as the zero order phase (see above). If not, the following procedure,
shown for F2, can be used. For F1, substitute “column” for “row, “xf1p” for “xf2p”.
The method below is used to extract a row from one end of the 2D
spectrum, add it to one from the other end and apply zero and first
order phase correction to the result. A subsequent ‘xf2p’ applies
the corrections to the 2D spectrum.
In the main 2D display, click on {1D-Mode}. In the 1D window, select
a row on one side of the spectrum and click {store} to save the row to
disk. Select a different PROCNO, e.g., 200 (as in F2, zero order). Select
a row from the other end of the spectrum and click {store}. Select PROCNO of, e.g., 201. Click on [Return]. Click on {to1D} to enter the normal 1D mode window. Use
‘edc2’ to select the PROCNO of the rows stored. Set PROCNO for
second data set to 200, set PROCNO for third data set to 201. The
processing parameter DC (multiplication factor of data set to be added)
should be set to 1. Click on [Param] > [Data/Acqu/Proc] > [edp], set DC = 1
(or type ‘dc’). To add the two rows together, click on [Proc] >
[Algebra] > [add] or type ‘add’. Click on {Phase} and phase correct
in the usual manner. Click on [Store2D] to save the new phase parameters.
Click on {2D} to return to the 2D window. (It is possible the you may need
to specify the data set again.)
‘xf2p’ now uses the saved values to be added to the zero and first
order corrections.
2D plots can be done in several ways. All the options are described in
the full uxnmr manual. A general procedure could be:
After processing the data, adjust the threshold levels. Move the mouse
cursor to {thresh}. Press the left button to select the threshold level
while dragging the mouse. Use the right button to select the number of
levels while dragging the mouse. By adjusting the number of levels to be
the same as the number of pens to be used for the plot, you can get an
indication of how the contours will appear on the plot.
After adjusting the levels, click on {DefPlot} to store these levels and
answer the questions regarding the number of pens.
For large spectra, plotting speed can be improved by selecting the REDRES
parameter in edg. If set to “yes” the 512 by 512 display will be used
rather than a calculation from the entire matrix. Much expanded regions
should be plotted with REDRES = ”no”.
If projections are to be plotted along the side of the contour plot,
they must first be calculated (if internal). The type projection or the
file name(s) of 1D spectra must be specified in edg.
Plotting is started the same way as a 1D plot. Plots and flplot can
also be used to combine plots. Currently, no plot preview feature
(as the 1D inter-active plot set-up) is available. It is possible to
switch to the contour display to see how the levels will be plotted.
Although the disk unit has a very large capacity compared to earlier
instruments, it will sooner or later be filled, particularly if a
number of 2D spectra are stored. To free up disk space, it is possible
to delete processed data while retaining the raw data as well as the
complete set of acquisition and processing parameters. A subsequent xfb
results in a spectrum which is identical to the one before deletion in
every respect. To delete processed data only, use [Data] > [Delete] >
[del2d], mark the ones to be deleted and click on “Execute”.
Be sure that the center bottom panel of this menu is in the “del
data only!” mode. Click on the spectrum to be deleted.
Deleting 2D spectra is also recommended before archiving data sets
to reduce the time it takes to copy a data set and to use the archive
media more efficiently. The same applies to transfers to other computers.
Click Here to return to the NMR Facility Home Page.
Click Here to return to the College of Chemistry Main Menu
Questions, comments to: Rudi Nunlist
rnunlist@bloch.cchem.berkeley.edu
Last Update: 11/23/94.