Author Topic: Offset problem for shaped pulse  (Read 4984 times)

srbh.sxna

  • Member
  • *
  • Posts: 10
    • View Profile
Offset problem for shaped pulse
« on: October 15, 2012, 12:59:04 PM »
Hi..
I am using the Spinach (Version 1.1.1054) and I found that the function "shaped_pulse" does not enforces the offset given by the user. I tried this with varying offsets and pulses ..and surprisingly it assumes the offset to be at 0 every time.

For very simple spin system (60 spins "13C" distributed from 35ppm to 180 ppm )..to observe the effect at any desired offset(say 35 ppm) I had to shift the shift the chemical shift (0-chemical shift=-35ppm))..and then this works perfectly. But this can't be done in practice!!

could it be the bug or i am making any mistake?
thanks..

kuprov

  • Administrator
  • Member
  • *****
  • Posts: 58
    • View Profile
Re: Offset problem for shaped pulse
« Reply #1 on: October 15, 2012, 02:00:32 PM »
Hmmm, that shouldn't happen. Looking at it...

srbh.sxna

  • Member
  • *
  • Posts: 10
    • View Profile
Re: Offset problem for shaped pulse
« Reply #2 on: October 16, 2012, 02:51:54 PM »
thanks.. :)
For your ref.. I have attached two 1D spectra to show the offset error. I have used off resonance sinc pulse (CO 180 of CA) on 62spins(13C).

Pulse sequence= shaped_pulse + step(spin_system,Ly_C,rho,pi/2)-->acquire;

Case1:
inter.zeeman.scalar=num2cell(linspace(0,180,62));
parameters.offset=35 %(in ppm)
result: wrong_offset.eps (attached)

Case2:
inter.zeeman.scalar=num2cell(linspace(-35,145,62)); %[(0-35),(180-35)]
parameters.offset=0 %(in ppm)
result: offset_corrected.eps (attached)

of course..it is not in good CO range..but this offset error can play a big role in 2D exps!
« Last Edit: October 18, 2012, 12:12:52 PM by srbh.sxna »

kuprov

  • Administrator
  • Member
  • *****
  • Posts: 58
    • View Profile
Re: Offset problem for shaped pulse
« Reply #3 on: October 17, 2012, 06:52:08 PM »
Could I have the code that produced those? It would be a lot easier to see what goes wrong there. If you look inside the shaped_pulse function, you can see easily that the offset is accounted for, so these errors really should not appear.

kuprov

  • Administrator
  • Member
  • *****
  • Posts: 58
    • View Profile
Re: Offset problem for shaped pulse
« Reply #4 on: October 17, 2012, 06:54:02 PM »
Ah, looking at your code fragment -- Spinach expects offsets in Hz. Does that solve the problem?

kuprov

  • Administrator
  • Member
  • *****
  • Posts: 58
    • View Profile
Re: Offset problem for shaped pulse
« Reply #5 on: October 17, 2012, 06:59:19 PM »
The other thing is that parameters.offset is printing offset, it has no influence on the spin Hamiltonian. The pulse offset must be specified during the call to shaped_pulse function itself.

srbh.sxna

  • Member
  • *
  • Posts: 10
    • View Profile
Re: Offset problem for shaped pulse
« Reply #6 on: October 18, 2012, 11:38:07 AM »
thanks for help :) ..

Yes..indeed.. the center of the spectrum changes when we change the offset...so setting the offset is ok(shaped_pulse function seems correct)....but there seems to be some issue related to the effect of the "off resonance pulse" at "on resonance".

yes..The offset was put in Hz (I just wrote in ppm here to avoid cluttering). Also the parameters.offset was given as an input argument to the shaped_pulse function.

I have attached the code for your reference..kindly guide :)
« Last Edit: October 18, 2012, 12:49:07 PM by srbh.sxna »

kuprov

  • Administrator
  • Member
  • *****
  • Posts: 58
    • View Profile
Re: Offset problem for shaped pulse
« Reply #7 on: October 18, 2012, 07:42:17 PM »
I've cleaned up the code a bit and I am getting a fairly neat inversion around 55 ppm. Is that what you expect? (see the attachment)

I've emailed you the code and the current version of Spinach (yours is a bit old).
« Last Edit: October 18, 2012, 07:55:16 PM by kuprov »

srbh.sxna

  • Member
  • *
  • Posts: 10
    • View Profile
Re: Offset problem for shaped pulse
« Reply #8 on: October 19, 2012, 11:25:25 AM »
Great...it works :) thanks a lot :)