University of Ottawa NMR Facility Web Site

Please feel free to make suggestions for future posts by emailing Glenn Facey.

Friday, June 4, 2010

E.COSY and the Relative Signs of Coupling Constants

Spin-spin coupling constants can have values greater than or less than zero. The absolute sign of the coupling constants cannot be discerned from the simple examination of a 1H NMR spectrum. The E.COSY1 (Exclusive COrrelation SpectroscopY) technique is one method which can be used to determine the relative signs of coupling constants. E.COSY is a phase sensitive COSY variant which produces off-diagonal signals showing only the active coupling (i.e. the coupling directly responsible for the cross-peak) as 2x2 antiphase square tetrads displaced in both the F2 and F1 domains by an amount equal to the passive coupling constants (i.e. the couplings not directly responsible for the cross peak). The slope of a line drawn through the cross-peaks is used to determine the relative signs of the passive coupling constants. The sign of the slope depends on whether the signs of the passive couplings are the same or whether they differ. The figure below shows the gradient E.COSY spectrum2 (using the Bruker pulse program,"ecosygpph" ) for the ABX three-spin system in phenylalanine.The cross-peaks highlighted in yellow in the top panel of the figure are expanded in the bottom panel. Black contours are positive and red contours are negative. The panel on the bottom left shows the cross-peaks for AX and BX. In the case of the AX cross-peak, the line drawn through the cross peak has a positive slope indicating that the passive couplings JAB and JBX are of opposite sign. The line drawn through the BX cross peak also has a positive slope indicating that the passive couplings JAB and JAX are of opposite sign. From this we can deduce that JAX and JBX are of the same sign. The panel on the bottom right shows the cross-peak for AB. In this case, the line drawn through the cross peak has a negative slope confirming that the passive couplings JAX and JBX are of the same sign. In conclusion, the geminal and vicinal coupling constants are of opposite sign.

1. C. Griesinger, O.W. Sorensen & R.R. Ernst, J. Magn. Reson. 75, ; 474 - 492 (1987).

2. The ecosygpph pulse program produces spectra similar to those described in reference 1 as "complimentary" E.COSY spectra. The slope of the lines through the cross peaks in "complementary" E.COSY spectra are of opposite sign to those obtained from the E.COSY spectra described in reference 1.


John Trant said...

Interesting, but what information does the sign of the coupling constant provide about the molecule?

Glenn Facey said...

Hi John,

Thank you for your comment and question. The sign of the coupling constant can indeed provide information about a molecule. It is known ,for example that three-bond vicinal proton coupling constants are positive whereas two-bond geminal proton coupling constants are negative. If one was trying to identify an unknown sample and found that a particular coupling constant was negative then one might have reason to believe that the protons were geminal to one another rather than vicinal.


Bhaswati said...

What is active and passive coupling???I searched but didn't find a good definition.

Glenn Facey said...


In the COSY spectrum, the active coupling is the coupling giving rise to the COSY cross peak being considered and the passive coupling is not. For a example in an ABX spin system, if one is considering the cross peaks between A and B then the active coupling is JAB and the passive couplings are JAX and JBX.


Anonymous said...

Hi! Can we use gDQFCOSY (from Varian pulse sequence library) for the same purpose?

Glenn Facey said...

I do not know the answer to your question.

Anonymous said...

can you specify the experimental parameters for ecosygpph bruker pulse program .. i am not able to reproduce the spectra using PHE in D2O..


Glenn Facey said...


Pl1 = 0
pl10 = 9.63
p1 = 11.88u
p2 = 23.76u
p16 = 1000u
p17 = 2500u
nd0 = 1
ns = 4
ds = 16
td0 = 2k
td1 = 256
Fnmode = TPPI
gpz1 = 40 %
gpnam1 = SINE.100


Anonymous said...

Thanks Glenn, it's very helpful. Do you know if there's a way to assign spins A and B stereospecifically for a similar molecule?

Glenn Facey said...


I don't think this E-COSY experiment will allow you to make such an assignment. I would suggest a NOESY measurement if there are enough protons in the system.


AJAY said...

Hello sir,E.COSY is applied for strong coupling also..or not

Glenn Facey said...

Dear AJAY,
Thank you for your question. Although I have not looked into the matter in any depth, I suspect the ECOSY method can be used to measure the relative signs of coupling constants for strongly coupled spins - at least to a point. If the coupling is much larger than the chemical shift difference then I suspect the analysis would be non-trivial.


AMAN SIDHU said...

Hello sir, I am working with a molecule having five coupled nuclei (four 19F and one 1H). Do you think I can use E.COSY to find out the relative signs of only 19F nuclei while 1H is decoupled during acquisition?


Glenn Facey said...

I think you should be able to use this method to determine relative signs of 19F-19F coupling constants however I think you will have to use 1H decoupling for the entire sequence (except the recycyle delay).

YK Nalli said...

Dear Glenn Facey!
We recorded E-COSY spectra of a compound on Jeol-600 MHZ using default pulse sequence (ecosy_phase.jxp).

My purpose is to calculate J-values to fix the stereochemistry of compound, but, in the spectra all the signals (Off-diagonal peaks) are simmetrical with same J value for all protons. Could you please suggest me if any parameters below to be corrected.


x_domainn : Proton

x_offset : 5[ppm]

x_sweep :15[ppm]

x_points : 1024

x_prescans : 4

mod_return : 1

y_points : 256

x_acq_time : 0.11375[s]

x_resolution : 8.79159[Hz]

y_acq_time : 28.43628[ms]

y_resolution :35.16635[Hz]

Glenn Facey said...

Y.K. Nalli,
I am not familiar with the JEOL software or parameters so not in a position to offer advice. I suggest that you contact one of the JEOL application chemists. They are very knowledgeable.


Rajendra Thakur said...

Hi Glenn,

I completed PhD in solid state NMR with Dr PK Madhu, TIFR, Mumbai, India in 2009.
I was working with YK Nalli to obtain ECosy spectra of phenylalanine. After optimising parameters with jeol engineers we got a reasonable spectra but all the slopes are of one sign. We used L-phenylalanine as sample.
Do you think using D-phenylalanine can help? Should be try L-phenylalanine.HCl as a sample.
I can run the same experiment on a Bruker spectrometer.
Please let me know if there is another trick I am missing.

Glenn Facey said...

I do not know why the slopes are all the same. I don't think it has anything to do with which optical isomer of Phenylalanine used.