Problems with CTF for tilted specimens.

jose at samba.cnb.uam.es jose at samba.cnb.uam.es
Tue Jul 5 20:02:13 EST 1994


Hi everyone!!

I'm working on 3D structure determination by electron microscopy of 2D 
crystals,
and right now I am in a dead-end. I have many problems with the CTF (Contrast
Transfer Function) for tilted specimens. I've read some papers, 
and in all of them, the CTF correction for tilted specimenes is carried out 
by multiplying the image by the sinusoidally varying contrast (CTF) as it 
varies across the image due to the gradient of defocus (or by convoluting in 
Fourier space ..., you know).

Well, the problem I have is the following: Why?? Why do we have to multiply
the image by the contrast (CTF)?? 
In untilted specimens the contrast was given by a well-known formula in the
Fourier space. Now the contrast is given by a formula in real space (a formula
that is a generalization of contrast (CTF) for untilted specimens), or at 
least it is applied directly in the real space... Why??
How easy can we go from Fourier space to real space??

In tilted specimens, the difference in height between the two ends of the 
crystal makes a very large change in the amount of defocus. Thus, each part of
the image would have a different CTF. But this CTF is in the Fourier space,
isn't it? However, I don4t see this in the papers. I don't understand. My 
brain is in a mess.

Please, if you have any clue ... would you be so kind to email it to me??
It would be gratefully appreciated.


Thanks in advance.


E-mail: jose at samba.cnb.uam.es


PS: Excuse mistakes I've been able to make. English isn't my natural language.




More information about the Xtal-log mailing list