Friday, December 5, 2014

1D Selective 1H - 19F HOESY

2D Heteronuclear Overhauser Effect SpectroscopY (HOESY) is an effective way to determine whether or not a pair of heteronuclear spins are close to one another in space.  It is particularly effective for 1H and 19F where both nuclides are 100% naturally abundant.  2D 19F detected 19F - 1H HOESY data are typically obtained which provide all NOE correlations.  2D experiments, however,  can be quite time consuming, especially when only a few NOE correlations are sought after.  In such cases, 1D 1H detected 1H - 19F HOESY experiments1 are very desirable and can save a great deal of time.  When only one 19F resonance is present, they can be obtained by using hard 19F pulses.  This was recently illustrated well by Dr. Michael Lumsden of Dalhousie University.  When more than one 19F resonance is present, one can use a selective 19F pulse and repeat the experiment selecting each type of fluorine.  An example of this is shown in the figure below.  Selective 1D 1H detected 1H - 19F HOESY spectra were collected for 2,3-difluoropyridine using a selective 19F pulse.  The simple 19F spectra are shown on the left with the selected 19F resonance color coded.  The upper two spectra on the right are the HOESY spectra while the spectrum on the bottom right is a simple 1H spectrum.  One can see that when the fluorine in the 3-position is selected, there is a strong NOE to the nearest proton, C.  Alternatively, when the fluorine in the 2-position is selected, there are no strong NOE's as there are no adjacent protons.



1.  L.E. Combettes, P. Clausen-Thue, M.A. King, B. Odell, A.L. Thompson, V. Gouverneur and T.D.W. Claridge. Chem. Eur. J. 18, 13133 (2012).   

24 comments:

Unknown said...

What standard Bruker library sequence could be used for such an experiment?

Unknown said...

How could one obtain such a pulse sequence for a Bruker Avance II spectrometer?

Glenn Facey said...

Hi Bogdan,

Sorry for the delayed response. As far as as I know there is not a program in the Bruker library for this measurement. I have used the program below.

Glenn

;hoesyfhgp1d
;avance-version (07/04/04)
;1D heteronuclear 19F-1H NOESY
;
;
;L.E. Combettes, P. Clausen-Thue, M.A. King, B.Odell,
;A.L. Thompson, V.Gouverneur, and T.D.W. Claridge
;Chem. Eur. J., 2012, DOI: 10.1002/chem.201201577
;
;$CLASS=HighRes
;$DIM=1D
;$TYPE=
;$SUBTYPE=
;$COMMENT=

;1H observe 19F-1H HOESY
;selective version
;no 19F CPD during 1H observe
;with grad purge and 180 pulse BPP
;with ZQ cancellation after mixing time

;Sequence written for shared 1H/19F transmitter

#include
#include
#include


"p2=p1*2"
"p4=p3*2"
"d11=30m"
"d12=20u"


"TAU1=d8*0.5-p16-d16-53u"
"TAU2=d8*0.5-p16-d16-d12-p3*2-53u"


1 ze
d11 QNP_X
2 30m
d12 SWITO_F
d1 pl2:f2
(p3 ph1):f2
50u UNBLKGRAD
p16:gp1
d16
p11:sp1:f2 ph1:r
p16:gp1
d16
50u BLKGRAD
(p3 ph2):f2
d12 SWITO_H
(p1 ph1):f1
50u UNBLKGRAD
TAU1
p16:gp2
d16
3u
(p2 ph1):f1
d12 SWITO_F
(p4 ph1):f2
3u
p16:gp2*-1
d16
50u BLKGRAD
TAU2
(p3 ph5):f2
(p3 ph6):f2
d12 SWITO_H
(p1 ph4):f1
go=2 ph31
30m mc #0 to 2 F0(zd)
exit


ph1=0
ph2=0 2
ph4=0 0 0 0 1 1 1 1 2 2 2 2 3 3 3 3
;ph4=0 0 0 0
ph5=0 0 0 0
ph6=0 0 2 2
ph31=0 2 0 2 1 3 1 3 2 0 2 0 3 1 3 1
;ph31=0 2 0 2


;pl1 : f1 channel - power level for pulse (default)
;pl2 : f2 channel - power level for pulse (default)
;pl12: f2 channel - power level for CPD/BB decoupling
;p1 : f1 channel - 90 degree high power pulse
;p2 : f1 channel - 180 degree high power pulse
;p3 : f2 channel - 90 degree high power pulse
;p4 : f2 channel - 180 degree high power pulse
;p11: f2 channel - selective 180 degree pulse
;p16: homospoil/gradient pulse
;d1 : relaxation delay; 1-5 * T1
;d8 : mixing time
;d12: delay for power switching [20 usec]
;d16: delay for homospoil/gradient recovery
;NS: 16 * n
;DS: 16



;for z-only gradients:

;gpz1: 40%
;gpz2: 15%

;use gradient files:
;
;gpnam2: SINE.100


;$Id: hoesyph,v 1.6 2007/04/11 13:34:30 ber Exp $

Anonymous said...

Is it possible to run this experiment without gradients? Do you have the pulse program?

Glenn Facey said...

Anonymous,

Yes, it should be possible to collect data without the use of gradients. I do not have a pulse program for it.

Glenn

Anonymous said...

Hello, Glenn! It's Andrew from Ukraine!

What shape did you use to excite 19F? Can you give approximate values for p11 and sp1? I have tried to do this experiment with Gauss1.1000, but it was not successful..

Glenn Facey said...

Andrew,
I used the Reburp.1000 shape. Typical values of SP1 and P11 were 22.7 dB and 5000 microseconds, respectively.
Glenn

Anonymous said...

Thank you very much, Glenn!!!

Andrew.

Old NMR guy said...

Hi Glenn. I can't tell you enough how much your blog has meant to me over the years. You've helped me tremendously and you are always a great source for information. I tried the selective HOESY pulse sequence, but it failed no matter what type shape pulse I used, even after a lot of careful calibration of power levels. I asked someone at Bruker and he suggested the change as can be seen in the enclosed sequence. It just involves setting the power level for the 19F channel to 0 (PL0 = 0W)before the selective pulse. It then turns it back up to the normal high power before the next high power 90 deg 19F pulse. This made all the difference in the world and it works beautifully using the Reburp.1000 shape. The pulse sequence is below. Thanks again for everything.

Jim

;A.L. Thompson, V.Gouverneur, and T.D.W. Claridge
;Chem. Eur. J., 2012, DOI: 10.1002/chem.201201577
;
;$CLASS=HighRes
;$DIM=1D
;$TYPE=
;$SUBTYPE=
;$COMMENT=

;1H observe 19F-1H HOESY
;selective version
;no 19F CPD during 1H observe
;with grad purge and 180 pulse BPP
;with ZQ cancellation after mixing time

;Sequence written for shared 1H/19F transmitter

#include
#include
#include

"p2=p1*2"
"p4=p3*2"
"d11=30m"
"d12=20u"


"TAU1=d8*0.5-p16-d16-53u"
"TAU2=d8*0.5-p16-d16-d12-p3*2-53u"


1 ze
d11 QNP_X
2 30m
d12 SWITO_F
d1 pl2:f2
(p3 ph1):f2
50u UNBLKGRAD
p16:gp1
d16 pl0:f2
3u
p11:sp1:f2 ph10:r
3u
p16:gp1
d16 pl2:f2
50u BLKGRAD
(p3 ph2):f2
d12 SWITO_H
(p1 ph1):f1
50u UNBLKGRAD
TAU1
p16:gp2
d16
3u
(p2 ph1):f1
d12 SWITO_F
(p4 ph1):f2
3u
p16:gp2*-1
d16
50u BLKGRAD
TAU2
(p3 ph5):f2
(p3 ph6):f2
d12 SWITO_H
(p1 ph4):f1
go=2 ph31
30m mc #0 to 2 F0(zd)
exit


ph1=0
ph2=0 2
ph4=0 0 0 0 1 1 1 1 2 2 2 2 3 3 3 3
;ph4=0 0 0 0
ph5=0 0 0 0
ph6=0 0 2 2
ph10=0
ph31=0 2 0 2 1 3 1 3 2 0 2 0 3 1 3 1
;ph31=0 2 0 2

;plw0=0 watts
;pl1 : f1 channel - power level for pulse (default)
;pl2 : f2 channel - power level for pulse (default)
;pl12: f2 channel - power level for CPD/BB decoupling
;p1 : f1 channel - 90 degree high power pulse
;p2 : f1 channel - 180 degree high power pulse
;p3 : f2 channel - 90 degree high power pulse
;p4 : f2 channel - 180 degree high power pulse
;p11: f2 channel - selective 180 degree pulse
;p16: homospoil/gradient pulse
;d1 : relaxation delay; 1-5 * T1
;d8 : mixing time
;d12: delay for power switching [20 usec]
;d16: delay for homospoil/gradient recovery
;NS: 16 * n
;DS: 16



;for z-only gradients:

;gpz1: 40%
;gpz2: 15%

;use gradient files:
;
;gpnam2: SINE.100


;$Id: hoesyph,v 1.6 2007/04/11 13:34:30 ber Exp $

Rookie chemist said...

This is a great blog Glen! Thank you very much.

I had a question about the 19F-1H HOESY experiment. How close together should the Hydrogen and fluorine be to show up in spectra?

Glenn Facey said...

Rookie Chemist,
Thank you for your kind words. A precise answer to your question is difficult. The NOE is proportional to the square of the product of the gyromagnetic ratios between the two spins and falls off as the reciprocal 6th power of the distance between spins. For protons, one often hears that there will be an insignificant NOE at a distance greater than 5 Angstroms. This would suggest that 1H-19F NOE's would be insignificant at distances greater than about 4.7 Angstroms.

Glenn

Anonymous said...

When I try and use the above modified pulse program from Jim (old NMR guy) I get syntax error messages. Any help, please ?

Glenn Facey said...

Anonymous,
You may have to make modifications to the pulse programs above depending on your hardware and software. I have not tried the pulse sequence given above by Jim nor do I know the specifics of your hardware and software configuration nor the specific syntax errors, therefore I am unable to comment.
Glenn

Old NMR guy said...

Anonymous,
I noticed that a couple of things were missing after I pasted in the sequence. Just replace the three lines with "include" with those below and it should be fine.

#include
#include
#include

I have used it on 400, 500 and 600 AVIII spectrometers, both with shared amplifiers for H/F and with individual amplifiers. It's convenient, since the QNP command is ignored if there is no QNP module present. The 5000 us Reburp.1000 does work OK, but when I want to be more selective, I successfully use a 80000 us Gaus1_180r.1000 if those pesky fluorines are too close together. Good luck.

Jim

Old NMR guy said...

Anonymous, I don't know why certain characters won't show up when the comment is posted, but they didn't make it for my previous comment. The first #include needs to be followed by Avance.incl in between a "less than" sign and a "greater than" sign. The next #include line needs to be followed by Grad.incl in between a "less than" sign and a "greater than" sign. The third #include is followed by Delay.incl in between a "less than" sign and a "greater than" sign. I'm very sorry for the confusion.

Jim

Anonymous said...

I appreciate the responses - and had another specific question - i am really struggling to get my selective pulse to be "selective" even though my Fluorines are 20ppm apart. I create (using Gauss1_180r.1000) and save my pulse using stdisp, (it is for about 100Hz window at -118ppm in the 19FNMR). My question is how do I calculate the power to use ?

I know for my instrument, 19F 90deg pulse is 19W for 17.26 uSec.

If my shaped pulse is 21340 uSec long, the power should be...?

How do I do this, or where can I read about it ?

Glenn Facey said...

Anonymous,

You can set up a selective zg experiment (pulprog = selzg) with your shaped pulse and do a simple calibration as in this link.

http://u-of-o-nmr-facility.blogspot.ca/2014/02/deternining-90-and-180-soft-pulses.html

I think there is a way to get the newer versions of 'shape tool' to give you the power level as well based on the power and duration of your hard pulse. Check out the 'shape tool' manual in topspin.

Glenn

Glenn

Miguel said...

Dear Glenn,

Is there any chance to get this experiment working on a Varian spectrometer? I have not been able to find the pulse sequence.

Regards,
Miguel

Glenn Facey said...

Miguel,
I'm sure it is possible to run this experiment on a Varian instrument with the necessary hardware however, I do not have a pulse sequence for you.
Glenn

Unknown said...

Hello Glenn,

I think your blog is great and helped me and colleagues at my lab. I am trying to adapt the above program to use on a 1H / BB(19F) probe that we have here on a Bruker instrument.

I assumed that this would mean to remove the SWITO and QNP lines, but I also had errors kicked back for the (un)blkgrad's, which means that the #define is just missing somewhere. I guess is there anything else I should keep mind of (Tau computation?) or am I reinventing the wheel when I am doing this?

-David

Glenn Facey said...

Hi David,
I don't think you have to remove any of the SWITO stetements from the sequence to get it to work on a BBOF probe. I used a BBOF probe in the data in the BLOG post. Did you look at any of the suggestions from the others who commented on this post? Do you have the proper #include statements which define gradient parameters?

Glenn

Unknown said...

Hello Glenn,

After a bit of tinkering and a bit more learning I found it was easy to reproduce the non-selective pulse sequence listed in the cited paper. The resulting spectra were coherent with 2d HOESY spectra obtained for some test samples.

I think my issue is getting the selective pulse sequence to work right. It seems clear that there is more to just dropping in the pulse program and hitting play. If you have any advice on this, I'd love to hear it.

Thanks kindly,
-David

Glenn Facey said...

David,
Could it be that you just have to calibrate the selective pulse? This can be done through the Bruker shape tool or the old fashioned way described here:
https://u-of-o-nmr-facility.blogspot.com/2014/02/deternining-90-and-180-soft-pulses.html

Glenn

Unknown said...

Hello Glenn,

Judging by the charts on your linked post, this is probably the issue-- I will try this next tomorrow.

-David