you could made a bag out of polymer film, such as Kapton film thickness of 7 microns. 

in general light materials, low Z..eg. plastics, carbon etc.

you could made a bag out of polymer film, such as Kapton film thickness of 7 microns. 

You are going to get both effects regardless of the material you use unless as V. Nazmov suggested it is very thin.  Even then I would imagine you will see a hump at low angles due to the polymer.  

You can, to an extent, alleviate the effect by scanning an empty vial or bag and subtract it as a background from your sample data.  This isn't perfect, but it will help.

Do you have access to a horizontal geometry x-ray instrument?  If so you could use a boat where you fill the boat to the brim.  If you are careful you won't spill, and the liquid will be close enough to the lip of the boat to where you can probe at very low angles.  

The contribution of thin kapton foil giving much lower effect than any, can also be subtracted, of course. Both, a bag and a window of kapton because of its high stability against x-rays.

You could also study a high drop of the liquid, accessed directly on the surface providing a high wetting angle

Thanks  mates @ V. Nazmov, Dr. Iva Ninova Kuznetsova, Brian Kitchen...Actually, Katopn fillm vial or tube is the best as it affacts least.We can not use anything empty and straight away subtract  For, if the bottle or vial has compareable or slightly more scattering thaN THE SAMPLE ITSELF THAN EVEN SUBTRACTION WILL NOT SERVE THE PURPOSE. sO, THE ONLY SOLUTION IS USE A MATEROIAL WHICH SCATTERS LEAST, Anyway, ...Valuable discussion and fruitfull convergence......!

Good fundamental query! Great practical tips! I'll contemplate and respond.

There are some simple cups (plastic cylinders of various diameter) both ends of which can be closed with thin foils (6 µm) of Mylar or Kapton.

All great suggestions. Thanks to all contributors!

Summary of discussion thus far:

1. Materials with low XRD signal.

2. Thin and low Z.

3. When possible normalize & deconvolute the effect (hump) of the "vial". Preferably, not "subtract"!

4. Eliminate it altogether if possible by using boat, drop, etc..

Imran! When possible please post your typical liquid diffractogram. I'm interested in the 2Theta XRD range needed, power used, anode material, beam conditioners if any, beam size and resulting signal amplitude as well as SNR (Signal to Noise Ratio). What detector? Glorified Geiger counter?

"Lemon to lemonade" thought! Have you considered a material that was "thin", "low Z", with no signal (hump) in the region of interest (2-theta) for the liquid diffraction pattern? This material may perhaps be an internal standard as well? In my opinion, the question would then be about the observed XRD SNR of such convolution.

Of course it depends on the energy. However if you are looking at liquid scattering than I would not want a sample holder that also gave diffuse scattering due to lack of long range order. Subtracting a polymer or glass signal from a liquid signal? What a nightmare! It would be much easier to subtract the holder signal if it were crystalline. BeO2 is low Z and cubic, so it has few peaks which are sharp. Just remove the 1-2% of the scan which contains those peaks (easily done in most fitting software). The 98%+ remainder of the pattern will not have any contribution from the holder, unlike the case of using a polymer or glass in which case every point of your scan has a contribution from the holder.

The other solution is to use magnetic levitation and measure the liquid containerless.

HI!

I will use quartz capillaries. Of course, it depends on where the expected scattering signal will appear. The magic of such quartz capillaries is that they are very thin and the signal coming from them is at very high angles.

Avoid polymers, since they scatter in the region of molecules. Nevertheless, if you give me more details about your sample I will tell you what material to use.

SALUT!

Tony

I would not use quartz. The amorphous scattering will overlap heavily with the liquid scattering. Quartz (and all glasses) give the most scattering at low angles, not high angles. Even Kapton is better than quartz since it is much thinner that any capillary. You will most likely get more scatter from the quartz than from your liquid since the quartz is some much denser. Definitely not quartz!

The key to success is to use a material with the least signal in the liquid diffraction pattern range. What about toxicity issues with BeO2? Graphene, maybe? Is it available as membrane?

@ Lawrence Margulies....yes , i do agree , Kapton is much better than quartz, especially, in low q -range. I am thinking to use kapton but its really difficult to have kapton tubes. I try few comp+anies here but they are not selling it , on google i can see only chinese traders..................if you can advise anything to buy kapton tubes .

@ Ravi Ananth.....I do apologise that i did not put specifically the x-ray experiment. I will use small angle x ray scattering and most probably the scattering will be in small q -range....guineir , porod regime.... so, as you mention , xrd may be good for quartz but for SAXS,,, i should be more carewfull

For SAXS I would definitely use a low Z crystalline holder. Beryllium oxide is very good since it has very few peaks and none in the SAXS range. Toxicity is not a problem in the oxide state (unless it is ground into a powder and inhaled). It is not like Be metal which you should never even touch. Graphite is also good as well as Boron Nitride, 

You could also use very thin diamond platelet windows - available commercially at a reasonable price, I believe. If they are single crystal, then the Bragg reflections are sharp and located in predictable positions. BeO2 is dangerous - I would not like to bring it into my laboratory. Good luck. 

Dear all,

I am sorry to disagree again. For SAXS measurements (0.05 nm-1 < q < 5 nm-1), quartz capillaries are the best  for liquid samples. Go to any SAXS producers like Rigaku, Anton Paar, Bruker... and you will see what the advice. Guys, even in the WAXS region quartz capillaries can be used (I did it for several years), but it is up to you.

Quartz is inert, clean and affordable material. The others are extremely expensive. Kapton might be a good idea, but only for liquids which do not swell Kapton!!! Do not forget that Kapton is a linear polymer and it is not crosslinked. So, it might happen that you put the solvent and the Kapton... dissolves.

Again, and please believe me, quartz is the best option. Moreover, I have constructed a setup which consist on two mica plates closing a perfluorinated O-ring (to avoid swelling), which also works very well. But mica, quartz, silica... are basically the same!

SALUT and good luck!

Tony

We have always used for SAXS a metal plate with a hole (diameter ~5 mm) inside and capton foil on both sides. The liquid was filles through a hole at the side of the plate, which is closed by a plastic screw. The thickness of the water phase in then approximatly 1mm. This has worked very well for aqueous solutions (with proteins). I guess, it depends on your material and the used energy. Ours was in the high energy regime (10-27keV). E.g. in this paper: http://journals.aps.org/prl/abstract/10.1103/PhysRevLett.106.178102

Dear Irena,

I do not say that Kapton is not a good option. Actually, might be ok for water, but the question did not explained whether was water or any organic solvent. That's the reason why quartz would be the best option (HF is the only problem...). Moreover, you use synchrotron radiation, while I guess that the guy who asked was planning to use a X-ray in house radiation...

SALUT!

Tony

An old recipe for x-ray powder diffraction in a sealed vessel calls for the use of lithium borate glass (Lindemann glass) capillary tubes of negligible X-ray absorption and suitable dimensions  (e.g., 0.2 mm. internal diameter), obtained commercially. I expect different inner diameter tubes are also available, certainly with slightly thicker walls, but still very "transparent" to x-rays.

The alternative is pyrex tubes (boron silicate glass), also readily available commercially in all sizes but with higher absorption and higher melting temperature (approx. 700°C). About 200°C higher than Lindemann glass.

Naturally the procedure would call for running blanks before inserting the (liquid) sample in either case, to evaluate and record absorption and inelastic scattering.