liquid handling for solution SAXS measurements

[bshilton]

A biological SAXS beamline at the Canadian Light Source is in the planning stages. We would like to include an automatic system for handling samples so that people do not have to go in and out of the hutch for each measurement. Desired features are temperature control, ability to mix two samples (i.e. to record a concentration series, or to titrate one protein with another), a UV absorbance detector close to where the sample passes into the beam, and a capability for making time resolved measurements. I have attached a "cheesy" and very preliminary conceptual picture - not sure what the real thing would look like.

Perhaps such a system is impossible to engineer, but seeing that advanced sample environments such as this are part of the SAXIER project, I am looking for advice on the design as well as leads to companies or groups that might be able to handle this type of project.

Thanks for any help or comments - Brian

[AL]

Hi Brian,

at the X33 EMBL BioSAXS beamline located at DESY (Hamburg, Germany) we have an automated sample changer, in user operation since September '2007. It doesn't have the ability to mix two samples. But the fact that people don't have to go in and out of the hutch for each measurement speeds up the whole process a lot, also the number of user mistakes can be reduced. The sample changer can store up to 2x96 samples (200 ul) and 2x24 buffers (1.5 ml) under Peltier cooling 1 to 60 °C (with separate temperature control for the vacuum sample cell). It performs reliable sample loading and automated cleaning/drying to avoid cross-contamination. The overall experience is quite positive, however we found out that tube length is important: transport time for samples and cleaning time is proportional to tube length; also long tubes result in losses from sample volume. Depending on physical properties of the sample foaming and bubble formation can be a problem as well (which is what I am currently working on).
The sample changer was manufactured by the Fraunhofer Institute (Stuttgart, Germany).

[Bente]

The X33 robot moves the sample through quite tubing to the cell (pushed by air/liquids). This has positive and negative consequences: cleaning is VERY good, and it works very reliably. On the downside is that not all proteins like transport through tubing, and there's a small challenge with positioning the sample in the cell. The SIBYLS beamline have another solution where they move the needle itself from sample vial to the Xray cell. However, if you wish to have your cell in vacuum (which is of course optimal) this poses a serious challenge for the positioning of the needle. Volumes are going down in both cases (rumours have that Adam Round at the new ESRF ID14-3 will have an X33-lookalike with very low volumes, maybe he can comment also).
Finally there's of course the option to go for the full monty, and use microfluidics which WILL mix your samples in very low volumes. That's absolutely not an off-the-shell solution (yet) but it's very elegant and you can add UV-cells etc. Check out Toft et al (2008) Analytical Chemistry, 80, 3648 (ask for more details if needed)

[ARound]

We are working on a new prototype system in a collaboration with the ESRF and EMBL Hamburg and Grenoble outstations. The design is based on the concept of the X33 sample changer (pulling samples through tubes). One important lesson learned is that the effects on the sample caused by the tubes is proportional to the contact surface area. By making the device as small as possible and reducing the length of tube through which the sample must travel we hope to improve speed, reliability and reduce the required sample volume (we are hoping for volumes in the range of 10 micro-liters per measurement).

As for mixing of two samples it is a consideration but will not be rapid mixing. Our concept could select two or more samples sequentially and allow mixing by diffusion during the transport to the measurement cell. It could also be possible to transfer multiple samples into a mixing cuvet prior to loading for long term reaction experiments but this is not our main focus.

I am sorry but for the moment we are still testing possible methods and cannot give too much detail. We hope to have a prototype of the new system installed for testing on ID14-3 early in 2009. Please feel free to contact me if you have any more questions.

[ARound]

The new generation sample changer designed through the trilateral collaboration of ESRF, EMBL Grenoble and Hamburg outstations, has been in succesful operation at the ESRF bioSAXS beamline (ID14eh3) for several months and will soon be operational at Petra III.

The new system is optimised for 3rd generation synchrotron sources and can load Sample volumes down to 5 microliters at a temperature range between 2 and 60 degrees. Samples can be flowed through the beam position to help mitigate radiation damage (increased volume needed for faster flow) and mixing of samples can be achieved piror to loading via pippetting functions.

[Kamma-Lorger]

Is there an estimate how much it costs the new generation sample changer? Or where we can get more information on specifications? I have been searching the web, but cannot find much...

[AL]

Is there an estimate how much it costs the new generation sample changer? Or where we can get more information on specifications?

Please contact Florent Cipriani for the details.