@Jorge - The concentration distribution in EDX-SEM will not give you any hint on the structure - with nominal identical composition, you cannot see the different ordering of the cations in the structure! For this, you would need atomic resolution and elemental imaging capabilities ...

Perhaps Rietveld refinement could help. Try to look at the atomic positions, there should be differences in the preferred orientations of the planes.

There are enough differences in the XRD powder patterns of these two crystals that high quality data and Rietveld refinement should be able to separate them. Consider for example the {200} and (004) peaks, which overlap at 33 degrees 2theta in kesterite but are split in stannite, or likewise the {220} and {204} peaks around 47 degrees. Good luck!

You are right Philip. That is true for well crystalline bulk material. My sample is in nanoparticle form, size is below 100 nm and peak broadening in XRD is observed. If I performed Rietveld refinement, it gives a mixture of the two phases with almost identical 'a' and 'c'.

Agree with Philip. If u use high quality XRD data, u can distinguist both. with the difference in intensity and peaks position. For Kesterite that has Zn rich (more electron and bigger in ionic radii), the lattice parameter will slightly expand so the peaks shift to lower angle and also higher in intensity as reconstruct below.

But you must pay attention to sample preparation and instrument calibration.

in principle the difference can be seen in XRD... just remember that theory and practice can be very different especially if you're working with a thin film, in grazing incidence and you have very small domains (so I agree with Philip when he talks about "high quality data", something not always readily available!). Then you can be in troubles. Plus you might have extra phases hidden under the main phase peaks. There has already been a discussion here about a similar issue here and if I remember well, Dirk Luetzenkirchen-Hecht showed that more information can be gained using XAS (Appl. Phys. Lett. 99, 262105 (2011))

So never be sure you have jsut one of them:

https://www.researchgate.net/post/Is_there_anyway_other_way_than_XRD_and_Raman_to_facile_characterize_Cu2ZnSnS4

I have no experience with that minerals, but I believe that you might try with DSC OR TG

I don't have much more experience with this mineral phases. I would like to point out some suggestions for this. If you can obtain, 2D XRD mapping, it may have some differences for those phases. You also can use atomic position mapping too. You can use "electron nuclear double resonance" method for this.

http://www.tandfonline.com/doi/abs/10.1080/00107517208205667#.UgOccay_xkc

http://link.springer.com/article/10.1007%2FBF01747988

Refer this links

Thanks all. I think a careful study of XRD pattern helped me. We did slow scan and saw peak splitting in 220/204 and 112/316 hkl plane positions. Observed 2 peaks for stannite rather than 1 peak in kesterite phase. Thanks all again.

I see that in these phases you have iron and tin, both Mössbauer isotopes. Try to see if Mössbauer spectra of these phases are different, may this help. In fact, Mössbauer spectroscopy suffers less than XRD from the nanosize of particles.

I do not know about these matarials, but you can look to this paper :

Hyperfine Interactions, 1989, Volume 46, pp 695-699

Mössbauer spectra of “stannite group” minerals

A. Baldini, G. P. Bernardini, L. Cianchi, F. Gulisano, M. Mancini, G. Spina

-----------------------------

I also see that the first author of this paper is on Reasearchgate : (https://www.researchgate.net/profile/Alessandro_Baldini/)

best

MTS

iron is just on the "proper" stannite. Here we talk about a compound with the same structure, but with different cations. So just tin can be investigated using Mössbauer spectroscopy.

You are right Matteo! thank you!

Ftir? Maybe hrtem w/saed.

Not HRTEM/SAED, as the visible d-spacing is indistinguishable.

Mössbauer spectroscopy

X-ray Photoelectron Spectroscopy

As Prof Howie stated please revise the XRD cards for both phases in the mineralogical database to make notice, even the intensities as well as small differences in angular values mean differences in the 3D packing of atomic dispersors. Please revise this info to gain understanding regarding this topic. Hope this helps.

I believe neutron diffraction has been used to distinguish these two phases by Susan Schorr, see Solar Energy Materials and Solar Cells 95 (2011) 1482

@ Sasanka Deka ... what is your particle size and how much slow have you gone with scanning speed to see the splitting in XRD?

Do you have possibility to calculate and may be (experimentally use micro-calorimetric device) and compare formation enthalpy (ΔH) of the kieserite phase with that of the stannite phase?

@R. Ahmad, average particle size is ~10 nm. we used various scan rate, for instance 0.25 degree/m, 1 deg/m and maximum 4 deg/m.

Thanks Sasanka

As in the case of CZTS (S instead of Se), EXAFS/XANES may help you (see Matteo Leonis comment above), especially because you may look into the material from different point of views, i.e. you are able to characterize the different environments around the elements SEPARATELY, which may give useful information. In fact, we did XRD as well as EXAFS/XANES on particles that should be CZTS nanoparticles, and it turned out that the XRD are compatible to both kesterite and stannite, with some contaminations of ZnO, while XANES gave evidence that we have more or less ZnO which was contaminated by kesterite and stannite ...

So some additional characterization seems always be necessary in the case of nanoparticles. Neutron diffraction, as mentioned above, could also be useful!

Good luck, Dirk

@Dirk Luetzenkirchen-Hecht?

ZnO is white in Color and CZTS black. How your particles look like? Do you see ZnO in other characterization techniques like Raman, EDX or UV-Vis?

Can you refere some literature about it. Thanks in advance!

The particles are black-brown in colour, but a few % of dark material may be enough to make the mixture appear dark (because we have abot 10-20 nm particles which appear colourless)

And no: We did not performed any Raman or UV-Vis. we did XRF but the light elements are difficult to catch with our system (in air)

Dirk

@Drik L.H. thanks a lot for your prompt response.

Why this high amount of ZnO is not detected by XRD? For ZnS (cubic-blend) I can understand that the peaks are overlapping with CZTS (Kesterite or Stannite).

Oups - I did really wrote ZnO - sorry for this mistake, it must read ZnS - you are absolutely right Rameez - thanks for your comment we (especially I) can remove this bug ... Sorry to all for this mistake!!!

Dirk

@Drik :)

Thanks for clarification.... now with your above comment I can enjoy weekend with peace. CZTS is already complicated with respect to Sulphides by-products and oxides will just be icing on the cake...

After all above discussions and my recent experiments on CZTS and CZTSe, I can come to the following conclusions. ZnS, Cu2SnS3, SnS, SnS2, Cu2-xS and Sn2S3, ZnSe, Cu2SnSe3, SnSe, SnSe2, Cu2-xSe and Sn2Se3 could be present as impurity in CZTS or CZTSe systems (note S or Se). With XRD we cannot distinguish them. But without going to complex characterization technique, we can prove the presence or absence of these impurities with the help of RT Raman spectroscopy. Where the Raman peaks of CZTS/CZTSe are totally different from the rest of those impurities.

Now coming to my original question, we can distinguish stannite and kesterite phase using slow scan powder XRD (our present nanoparticles, 5-50 nm, different batches). Actually we saw splitting of peaks in 220/204 and 112/316. Thus we confirm stannite phase of our CZTSe. UV-Vis and EDAX will not help here. NPs of CZTS and CZTSe both give broad absorption, so difficult to calculate band gap.

Thanks all

@Sasanka D .... RT Raman Spectra is known to broaden and peak-shift, in case of nanoparticles vs. bulk material. The raman signal of ZnS is located around 350-355cm-1 .... CZTS also has a shoulder at 347-348cm-1 which is very close to ZnS.

In case of nanoparticulate systems, I am afraid that even Raman can also be not that conclusive. Probably, only qualitatively it can be said that ZnS is not a major product, which may also mean its just less than 50%.

We recently discussed it in detail here...

http://link.springer.com/article/10.1007%2Fs11051-013-1886-9

This paper might be helpful to you............

http://jap.aip.org/resource/1/japiau/v111/i8/p083707_s1?bypassSSO=1

Hi All, I want to know, what could be the structural reason for this mentioned peak split when we change the stannite-kesterite structure? Thanks in advance for your comments.

The two structures have a slightly different symmetry (different cation ordering pattern) and therefore you can expect a different pattern. Have a look at:

http://rruff.info/uploads/CM16_131.pdf

Thanks for the answer. So, I understand that the disorder in the kesterite structure make an effect that diminishes the distortion in the lattice, so this split is more clear in the stannite structure.

What is the powder XRD experiment parameters (scan rate, time, substrate, etc...) that allow to observe the spilit of peaks in the stannite strucutre?

Thanks

Nessrin, that will depend very much on your equipment and how good your source is. Ask someone who is familiar with your setup what sort of parameters you need to get the required resolution. Then try it, and if you can't see the peak splitting, try a smaller step size or longer count time.

If you have a sufficient mass of material (at least 10 mg), you can perform 65Cu solid state NMR spectroscopy. In the case of the kesterite phase, you will easily see the 2 Cu sites of the structure (only one site for stannite). If you decide to try NMR, note that MAS (Magic Angle Spinning) conditions lead to spectra which are more difficult to interpret than static conditions. You can find very useful information about NMR of CZTS in our publication doi: 10.1039/c3cp51320c "Multinuclear (67Zn, 119Sn and 65Cu) NMR spectroscopy--an ideal technique to probe the cationic ordering in Cu2ZnSnS4 photovoltaic materials."

I hope this will help

The XRD could be give enough, kesrterite and stannite have different chemical compossitions. If you have both compounds, you can try with a EDX-SEM, to evaluate the Zinc and Iron on your sample.

@Jorge - The concentration distribution in EDX-SEM will not give you any hint on the structure - with nominal identical composition, you cannot see the different ordering of the cations in the structure! For this, you would need atomic resolution and elemental imaging capabilities ...

@Jorge you can also have Stannite structure without "Fe" as well.

I wish to take NMR for my Zn, Sn, Cd and Al contend samples. can any one give me the instrument availability details in India

I wish to take NMR for my Zn, Sn, Cd, Al and F contend samples. can any one give me the instrument availability details in India

I suggest to measure and compare Sn119 Mossbauer spectra of both stannite and Kesterite phases

In what ways we find the our synthesized material has kesterite and stannite phases?

Another possibility could be EDX/EELS mapping with atomic resolution using aberration corrected STEM. This has been tried I believe before but the results may have been inconclusive. It may be that in many cases the Cu and Zn atoms are not arranged in either the stannite or kesterite ordered structures but instead are mixed randomly.

Try Mossbauer spectra of Sn119. No isotop enrichment of your smaples is necessary.

At least spectra components would differ in line width values