How long do you wait after changing the temperature?

A small corner of the web dedicated to the pursuit of non-ambient SMX. This rather loosely includes: superstructures, charge density, variable temperature (both ultra high and ultra low), high pressure, gas and vacuum and of course excited state (laser, LEDs and bright lights).
User avatar
Brownml2
Rotating Anode
Posts: 101
Joined: 25 Jul 2014, 22:43
Location: Vancouver, BC, Canada

How long do you wait after changing the temperature?

Post by Brownml2 » 05 May 2015, 17:16

So, if you are doing variable temperature xrd, how long do you wait between each run? I was told 10 minutes, to be safe, and consider that a bit excessive (it is a tiny, tiny crystal, How long can it take to cool?) but today I got into an argument with a fellow grad student who is using 20 minute pauses between each run, with a 20 degree change in between. Is she right, do you have to wait 20 minutes for the crystals temperature to stabilize after changing the temperature? Or is she just throwing away xrd time? Is the a formula for this (X minutes per Y degrees?)

nils
Rotating Anode
Posts: 120
Joined: 20 Aug 2010, 09:30
Location: Zürich

Re: How long do you wait after changing the temperature?

Post by nils » 06 May 2015, 09:01

Under the assumption there is no phase transition I would rather ramp down with a reduced cooling rate. Then you don't need to wait at all.
There's another aspect: if the chiller is not well-serviced it might take longer to cool down than calculated by the controller, i.e. your software already starts measuring while you're still some degrees off (depends on the software and it's no problem if you start manually).
Also some chillers not produced by Oxford Cryosystems ( :D ) don't let you ramp and tend to freak or oscillate right after changing the temperature, so it might be a good idea to throw in 15 minutes to let the chiller stabilize.

User avatar
Helge
Rotating Anode With Optics
Posts: 207
Joined: 26 Jan 2012, 16:04
Location: Dublin

Re: How long do you wait after changing the temperature?

Post by Helge » 06 May 2015, 14:14

My crystal chiller cannot ramp, and hence the temperature takes a little while to stabilise. 10 minutes usually is plenty, but I have to give it at least five minutes, and with 10 I am on the safe side.

User avatar
Brownml2
Rotating Anode
Posts: 101
Joined: 25 Jul 2014, 22:43
Location: Vancouver, BC, Canada

Re: How long do you wait after changing the temperature?

Post by Brownml2 » 06 May 2015, 22:45

It is an Oxford one and can ramp, but it is looking like 30 minutes wasn't enough for what they are doing, which kind of blows my mind. It is a TINY crystal, hard to believe it takes 30 minutes to change a few degrees and be stable! Guess I was totally wrong.

nils
Rotating Anode
Posts: 120
Joined: 20 Aug 2010, 09:30
Location: Zürich

Re: How long do you wait after changing the temperature?

Post by nils » 07 May 2015, 07:10

Phase transition?
What needs to be stabilized, the crystal, the mount or the chiller? How do they measure the temperature that accurately if it's so important? If it's the crystal, in terms of heat capacity I follow your argument.

glazer
Posts: 3
Joined: 06 May 2015, 23:08

Re: How long do you wait after changing the temperature?

Post by glazer » 07 May 2015, 08:09

Why not try this? Ramp down to the desired temperature and then immediately collect some data. Repeat this several times over a period of minutes and see if you can observe any changes. :D

User avatar
wlewis
Rotating Anode With Optics
Posts: 199
Joined: 23 Apr 2009, 19:48
Location: University of Nottingham

Re: How long do you wait after changing the temperature?

Post by wlewis » 07 May 2015, 11:20

It is an interesting question. I am not aware of any systematic studies regarding equilibration time for temperature changes. The 20 minutes for 20 degrees sounds very much like a "rule of thumb".

The first thing that comes to mind:
When collecting an ordinary data collection, if it is collected at low temperature (as most usually are), should you wait for the crystal to equilibrate? In my lab we never do. Waiting for 150 odd minutes for the crystal to equilibrate down to 120K is unreasonable, especially when for a good crystal the collection can only take 10-40 minutes. A good way to test this is to determine the unit cell on the first few frames and then the last few frames and look for significant discrepancies (similar to what glazer suggested).

As mentioned, if there is a phase transition then you may need some time for the reorganization of the crystal to occur, but otherwise it seems like a waste of time to me.

One thing to note, the accuracy of the temperature at the crystal position for any open flow cryo system will only be 2 degrees, although the displayed temperature on the controller may appear to be fluctuating +-0.1 degrees. If they examining very small steps they may not have as accurate a picture as the think (or they should apply statistics to their overall results).

Another note: a rule of thumb for high pressure crystallography (another type of external stimuli) is that the equilibration time should be twice the collection time. Obviously we don't want to apply that to standard low temperature experiments, as every instrument would then spend 2/3rds of its time idle!

nils
Rotating Anode
Posts: 120
Joined: 20 Aug 2010, 09:30
Location: Zürich

Re: How long do you wait after changing the temperature?

Post by nils » 07 May 2015, 11:43

Hm, I think the shown 0.1 K fluctuation is more or less real if the chiller is well-serviced and the sample well-positioned. The sensor is quick and fluctuations wouldn't depend on calibration. Also +/- 2 degrees would really mess you up if you want to cycle around phase transitions or crystallize on the goniometer, as many tribes do.
Often the nozzle is too far away, which leads to high fluctuation at the sample (and usually icing). The absolute temperature value at the sample position might differ from what's shown on the controller, if that's what you mean. The 700 is calibrated for 100K +/-0.1K between 7 and 12 mm. However, down/up to 7/21 mm the temperature deviates only by 1 degree. You can check using a calibrated diode mounted on the gonio head or some substance with a visible phase change at known temperature.
From personal experience, in > 1000 crystal structures from shock-frosted crystals of various sizes at 100K, I've never seen any visible effect after frame #3 of the matrix run. Barring phase transitions, where anything can happen.

User avatar
Marc
God Like
Posts: 191
Joined: 16 Apr 2008, 12:19
Location: Continental Europe

Re: How long do you wait after changing the temperature?

Post by Marc » 07 May 2015, 15:09

I would time T equilibriation of a crystal in ms rather than seconds, certainly not tens of minutes. Equilibriation of the cryo is a different matter.

Return to “Non-Ambient”