(A
PDF version of this page is available from the Handouts site)
MTR
Resources
The Nuclear Overhauser Enhancement Spectroscopy (NOESY)
experiment is the 2-D equivalent of the transient-NOE experiment
(NOESY1D). It is important to remember that
the NOE's generated in this experiment are NOT equilibrium values
and thus will be less than those measured using a traditional 1D
steady-state difference NOE experiment. Furthermore, the % enhancement
obtained from this experiment and from the NOESY1D are NOT directly
comparable to those obtained from steady-state measurements. NOE's
are plotted as the cross-peaks between two coupled nuclei. NOE's
arise between two spins that possess dipolar coupling (i.e. through-space
coupling) and not from scalar coupling (spin-spin coupling which
produces the multiplets in proton spectra). Since NOE depends on
the inverse of the distance to the sixth power, short-range dipolar
couplings are heavily weighted. This dependence on distance allows
for an estimation of the internuclear distance. Care must be taken,
however, as many processes lead to reduced NOEs including spin-lattice
relaxation, temperature, increased solvent viscosity, increased molecular
weight, and dissolved paramagnetic impurities including oxygen. Furthermore,
it is difficult to completely remove unwanted COSY-like cross-peaks
(due to zero-quantum coherences), which at best makes interpretation
difficult. In fortunate cases, these COSY-like cross-peaks, which
have the same phase as the diagonal while NOESY cross-peaks have
the opposite phase, will not overlap with NOESY cross-peaks making
interpretation less cumbersome. At other times, however, the COSY-like
cross-peaks can coincide with the NOESY peaks, which will cancel
the NOE signal.
Note: Bold text represent
boxes you should click. italic text represent text you should type
and hit RETURN.
Determine Through-Space
Interactions for Stereochemistry (NOESY, NOESY1D):
Running a NOESY experiment:
Make sure your sample is free from particulates.
It is best to deoxygenate your NMR sample
to ensure the maximum NOE. A simple somewhat crude method of degassing
the sample is to slowly bubble nitrogen through the solution using
a thin hard plastic tube (PTFE 1/32 inch diameter). Make sure you have
a slight excess of solvent to conpensate for evaporation. Bubbling
for 5 minutes should suffice. A more thorough method is freeze-pump-thaw.
Acquire a single scan proton spectrum (nt=1): transform,
phase, and reference.
Set the temperature to the desired value and let it
equilibrate for at least 20 minutes.
Determine the 90 degree proton pulse. Click HERE for
the procedure.
Set pw to the 90 degree value and acquire a spectrum.
Type df dscale. Note the time (dscale increment)
your FID first comes to the scale. Typically, this is around 0.500
to 0.800
seconds. This value will be used to set the mix time in the NOESY experiment. NOTE: Optimal NOE occurs approximately when the mix time equals the
T1 for a given spin. Since different protons have different T1's, no
single
mix time will be the best for all the protons. Furthermore, short range
NOE's will reach their maximum at earlier times than do longer range
NOE's. To get information on a variety of NOE's, you will need to run
the NOESY experiment with several different mixing times.
In the same workspace, type NOESY to load the standard
NOESY 2-D experiment:
Set mix to the time you determined in the single
scan proton FID: type mix=proton FID time (e.g. mix=0.700).
If you don't have a time, use mix=0.600. Longer mix times are better
for seeing
long-range coupling, but this is at the expense of overall lower
S/N.
The number of scans must be at least 8 to ensure
proper phase cycling (nt=8), nt=16, 32 or 64 is not uncommon. Increase
in multiples of 8 for dilute samples.
The default value for ni, the number of
increments or points in the F1 dimension, is 200, which is about
22 Hz resolution on the VXR-500 or 0.044 ppm.
With
zero-filling
and linear prediction, apparent resolution is about 6 Hz/point.
This can be increased if you require higher resolution in the F1
dimension.
To increase,
type , for example, ni=256.
Type time. If the time is longer than your reserved
time, decrease ni.
Type pw? and make sure that is set to the 90-degree
value you determined.
Spinning is not recommended for any 2-D experiment;
therefore, turn spinning off and optimize shims.
Type go: this acquires your NOESY spectrum.
Data Manipulation:
NOESY is a phase sensitive experiment, so both F1
and F2 must be phased.
Phase F2: this is simply achieved by phasing the
first increment of the 2-D dataset.
Type wft(1): this transforms the first increment.
Click Phase and adjust so that the spectrum
is inverted (i.e. all peaks point down).
Type cfpmult to automatically calculate
the fpmult (first point multiplier) value. This is important
to minimize 'F2 ridges'.
Phase F1: this is a little more tricky and will
be done after the 2-D transform.
Type setLP1 gaussian wft2da: this applies
linear prediction in the indirect dimension, gaussian multiplication
in both dimensions, and transforms the phase sensitive 2-D dataset.
A 2-D color map is now displayed on your screen.
If you are unfamiliar with manipulating the color map, click
here for instruction.
The phase for the F1 dimension is done by phasing
the right-hand side of the spectrum and then the left.
Click on Trace and place the cross-hair
over the right-most diagonal peak.
Type ds. You will have a 1-D spectrum
with the red cursor on the diagonal peak.
Click Phase and phase the diagonal
peak such that it is inverted. DO NOT ATTEMPT TO PHASE ANY OTHER
PEAKS!
Type dconi and then select a diagonal peak on
the left-hand side of the spectrum.
Type ds. Click on the right-hand side
but DO NOT PHASE. Click on the left-hand diagonal peak and phase
so
that it is inverted. DO NOT ATTEMPT TO PHASE ANY OTHER
PEAKS!
Type dconi. Your spectrum should be well phased.
You may phase the F2 dimension, if needed.
Type pmode='full' and type trace='f2'. Type
dconi.
The F2 axis is now on the bottom. Phase as
described for F1.
Autoprinting your NOESY with Projections
generated from NOESY:
Display the region of interest and click Autoplot:
This plots your NOESY with the projections generated from the
1-D data subsets. The indirectly detected dimension will have
low resolution and thus, the projection for that dimension
(usually F1) will have broad peaks. For better resolution projections,
use the following procedure.
Printing your NOESY with 1-D Spectrum
as Projections:
Open your NOESY Spectrum. Click Main Menu=>Display=>Size=>Center.
Join another experiment and retrieve the 1-D
spectrum:
Type jexp1 or any other experiment
number.
Load the 1-D spectrum; Fourier transform
(wft), and phase (aph).
Type jexp2 or join the experiment number
where your NOESY resides:
Click Display=>Contour and
expand, scale, etc. the region of interest (refer to table
for interacting with the color or contour map).
Type plgcosy and follow the directions
on the screen: This is our in-house macro that prints
your desired 2-D spectrum.
NOESY1D is a selective 1-D experiment that
uses shaped pulses to selectively excite specific resonances in
order to observe isolated dipolar couplings. This is a transient NOE
technique and will have NOE's that are less than those obtained from
steady-state experiments. The NOE signals will have opposite phase to
the inverted peak (i.e. the
inverted
peak,
when
phased properly,
will be pointing down in the spectrum while the NOE signal will be up).
Signals with the same phase as the inverted peak are typically a result
of exchange coupling (see EXSY), but can arise from
relayed NOE's (NOE's alternate sign and lose intensity as they are relayed
down a chain; i.e., positive, negative,positive). With the shaped pulses
and
gradient
selection, there
are essentially
no interference
peaks, which makes identification of small NOE's feasible.
Interpretation of NOESY1D spectra is straightforward and quite a bit
easier to understand than NOESY. When you are
looking
at NOE's from a relatively small number of signals (less than 10) or
for very small long range NOE's, NOESY1D is preferred.
It must be noted, however, that the % enhancement
obtained from this technique is not directly comparable to those obtained
from traditional steady-state NOE experiments (i.e. the classic NOE difference
experiment). Furthermore, the % enhancement is dependent on the
parameter settings (careful calibration and multiple experiments are
necessary to obtain accurate values), which means that
you should either note and report those settings if you intend to
site
% enhancement
using
this
technique or simply report relative values (i.e. small, medium, or large
enhancement).
Setup a 1H NMR experiment in exp1 and type gain='n'.
Make sure your sample is free from particulates.
It is best to deoxygenate your NMR sample
to ensure the maximum NOE. A simple somewhat crude method of degassing
the sample is to slowly bubble nitrogen through the solution using
a thin hard plastic tube (PTFE 1/32 inch diameter). Make sure you have
a slight excess of solvent to conpensate for evaporation. Bubbling
for 5 minutes should suffice. A more thorough method is freeze-pump-thaw.
Acquire a normal 1H spectrum. Phase it well and save
the FID fur future use.
Set the temperature to the desired value and let it
equilibrate for at least 20 minutes.
Determine your 90 degree pulse if desired or type pw90? to
get an approximate value and record this value.
Type tpwr? and record this value.
Type mf(1,2) to move the FID to exp2.
Type jexp2 to join experiment 2.
Type wft to Fourier Transform the spectrum.
Type NOESY1D('ds') Note: Use capital letters for NOESY1D.
Type pfgon='nny'. This ensures that the pulsed
field gradient is on.
Turn off the spinning and increase lockgain and lockpower
so that the lock level is about 70. Be sure not to use too much lockpower.
You have too much lock power if the lock level is not stable (decrease
lockpower until lock level becomes stable).
Move the 2 cursors to bracket the first peak of interest.
Click Select.
Move the cursors to the second peak (if needed). Click
Select. Repeat for all peaks of interest.
Click Proceed.
At this point, you may be required to answer the
following questions:
Enter reference 90 degrees pulse width (usec):
Type the value you determined previously
(from pw90? in exp1)
Enter reference power level:
Type the value of tpwr from exp1 (from
tpwr? in exp1)
Type nt=32 or higher. Be sure to use multiples of 8.
Type gain='y'.
Type go. This will start your acquisition.
Data Manipulation:
Type wft to Fourier transform your data. This will transform
all of your NOESY1D('ds') spectra.
Type vp=80 to move the spectrum up.
Phase the largest peak (this is the irradiated peak)
so that it is negative (i.e. it points down). The small NOE peaks will
be opposite phase to the irradiated peak.
To view all spectra, type dssa.
To view individual spectra, type ds(#), where
# is the number of the desired spectrum. For example, I decided to
do 5 different NOE experiments irradiating at 9, 6, 5, 3, and 1 ppm.
I selected the peaks in the order given, which means that typing, ds(1),
will give me the spectrum with irradiation at 1 ppm. Typing ds(4) would
give me the spectrum with 3 ppm irradiation.
Integration and peak picking are done exactly like 1H
spectra.
Printing your NOESY1D Spectra:
To print all spectra in the array:
Type dssa and look to see
if the spectra overlap. If so, type ds(1) and reduce the
scale of the spectrum. Type dssa. If necessary, readjust
by typing ds(1), adjusting scale, and typing dssa.
When satisfied with spectra, type pl('all')
pscale page.
To print individual spectra:
Type ds(#), where # is the spectrum number you wish
to print.
Phase, peak pick, and integrate as desired.
Type pl pir pll pltext(150,150) page, for example.
Running an EXSY Experiment:
The EXSY (EXchange SpectroscopY)
experiment is used for the investigation of dynamic processes within a
given molecule. One is able to discern which resonances are exchanging
by quick inspection of the 1-D or 2-dimensional spectrum that is obtained.
Extraction of rate constants is possible, but can be a bit tedious. Qualitative
interpretation, however, is rather straightforward as long as one knows
the COSY spectrum. The pulse sequence is identical to NOESY for 2-dimensional
and NOESY1D for
1-dimensional.
For the 2-D EXSY, the cross-peaks due to EXSY will have the
same phase as the diagonal and are generally quite large as compared with
the NOESY peaks. For the 1-D EXSY, the EXSY peaks will have the same phase
as the irradiated peak (i.e. the irradiated peak and the EXSY peak will
both be down).
Depending on how readily exchange occurs in your molecule,
you may need to increase the temperature. If you are unsure, start at room
temperature and then work upwards.
Max
gets NOESY 
