> A protein would only scatter but not diffract
or Diffract but not scatter? isn't diffraction a kind of scattering?
But yes, the atoms in the unit cell may seem random in that
distance range (in fact this is assumed in wilson scattering)
but in a perfect crystal they will be the same in each unit cell.
But the scattering rom each unit cell will be systematically out
of phase, so the resultant vector will go round and round the
argand diagram without adding up to anything measureable
(Fig 4.15 in Drenth 1st edition), unless the Laue conditions
are met, and this leads to diffraction spots.
Now if the crystal is not perfect, each unit cell won't be the
same in hi-res details, cancelation will be imperfect, and this may
be related to the "diffuse scatter" which can result from certain
types of disorder.
Still I would expect to see peaks in a wilson plot around bond-length
resolution, similar to the peaks due to secondary structure at lower
resolution.
ed
Tim Gruene wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Dear Jacob,
>
> A protein would only scatter but not diffract, the latter - in my
> understanding - being the result of constructive interference from a
> regular array of unit cells .
>
> A powder pattern is the superposition of many small crystals amongst
> which you don't observe interference.
>
> Tim
>
> On 05/09/12 16:16, Jacob Keller wrote:
>> Dear Crystallographers,
>>
>> the "saxs on crystals" thread reminded me of a question I have had
>> for a while, and never having collected data better than ~1.6 Ang
>> or so, cannot answer myself from experience: I would think that
>> there might be powder-like diffraction rings at distances
>> corresponding to the various covalent bond lengths in proteins
>> (1.2-1.5 Ang), but have never heard of such. My thinking is that
>> the protein itself is essentially a powder sample within the unit
>> cell consisting of many small, randomly-oriented molecules (amino
>> acids) with their covalent bonds. Do the rings in fact exist, and
>> if not, why not? Maybe the electron density is not as "atomic," or
>> discrete, as the nuclei are? I wonder whether generally data
>> collected to beyond ~1 Ang have an intensity "bump" at those
>> covalent bond lengths, as I believe is seen in nucleic
>> acid-containing structures at the base-stacking distance (at the
>> right orientation)?
>>
>> Jacob
>>
>
> - --
> - --
> Dr Tim Gruene
> Institut fuer anorganische Chemie
> Tammannstr. 4
> D-37077 Goettingen
>
> GPG Key ID = A46BEE1A
>
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.12 (GNU/Linux)
> Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
>
> iD8DBQFPqpVlUxlJ7aRr7hoRAhZqAJ9M49u+cYsX+AhzRlYKOCTX8YCsbQCgzFBi
> ZV+cZCyQgGLusrnufngyZ5A=
> =cuQy
> -----END PGP SIGNATURE-----
>
|