Dear Gert if put two pictures one before your sample exposed to the electron radiation and one after that ordeal is over for that poor specimen!! in your laboratory!  I presumed that if the inelastic scattering  of  electrons does occur  substantial enough then their effects on the micro size silicate particles should be observable.  This is some type of local heating or fusion associated with the impinging high energy particles,  which may generates sub nano-metric - surface roughness that affects transparency of those tiny quartz particles. One observes periodic surface undulations  on those semiconductors surfaces   exposed to the bombardment with ionized particles to remove or to detach target material.

 Depending upon on the target material,  temperature, ion beam energy, angle of incidence, and various other parameters, one generically observes the rough surfaces, cellular patterns, and in particular under oblique incidence for the ion beam, the formation of regular ripple pattern etc.   BEST REGARDS 

Professor Dr. R. Mark Bradley of Colorago State University is an expert on this subject, he is the member of RG. 

@ H.M.B. Darwish: which ion beam? There has no ion beam used. The sample is SiO2 (either quartz or agate) and investigated in fast scan making the image and slow scan (the damaged area). 

@ Tarik: Dear Tarik, the sample was before EBSD (slow) scanning everywhere flat and the microstructure looked everywhere either like quartz or like agate. During EBSD measurement the agata was damaged. It definitely depends on the interaction of the electrons with different SiO2 modifications. Since both are linked SiO4 tetrahedra it is quite difficult to understand that agate will be damaged, but quartz won't, but this is the experimental result. The horizontal stripening is not unusual (although during common measurements not that nicely visible). The e-beam is not that stable as expected. I assume because of local charging or decharging. Nevertheless, you don't get any pattern (although the grain size is sufficient) becuase of the immediate damage. The same happens, if you are imaging the same area over a longer time, e.g. for focus optimization. As reported, I've seen a similar effect last week for calcite as well, although there I see at least in the beginning indexable patterns. But if I start a scan it becomes soon so bad that patterns are no more detectable. There also the charging effects are worse than at the "real" calcite.   

Hi  Gert

10 kV is very high voltage with slowly  scan that can raise the temperature of the sample  which make it easy to be like that. From my opinion you have to less the voltage and start with very low voltage with rising the voltage gradually and look to the result experimentally at every voltage you use. Because you know every setup has special situation. I did etching for diamond by TEM but we put special filter to prevent  sample from ioninization with the time. Just we think with you and  you are the best one can know the circumastnaces around your experiment and your device.

Best wishes

For the EBSD pattern formation you need a certain amount of electrons. If you reduce the kinetic energy of the electrons you need to pump more electrons in the sample what increase the danger of charging and heating as well. The interesting fact is not that it appears (this has been observed already on ceolites that the electron beam destroys the sample), but that it appears at materials which are assumed to be equivalent (or at least very similar) is quite surprising (at least for me). I did measurements down to 7kV and it appears also there (because I need to keep the beam perhaps 10 or 20 times longer than at 20kV). Anyway, thanks for the comment.

Dear Gert, it seems to me that only the regions between large grains are affected, which you call them agate. There is only one plausible explanation, namely;  those regions some how got more hot then the rest. The reason for their over heating may be due to their  irregular  surface orientations in micro-scale with respect to the beam  , which effectively increases the surface area exposed to the radiation., and plus the multiple scattering. The rough surface involves residual stresses (compressive because of the  quenching action when you turn off the beam) especially the trough regions has much higher stress concentrations compare to the crest regions, which accelerates   the mass transfer from trough to the crest region that makes the surface groovy and even may cause the formations micro cracks according to our computer simulations. BEST REGARDS

I am just speculating!

Hi Tarik, I've read a bit about agata. It can be quite porous which enables the faked colours often sold on markets (cobalt blue, pink etc). I guess that the local heating affects the content of these pores and finally their reaction destroys the structure. however, this doesn't explain the similar reaction of calcite. 

The obvious, visible difference between agate and quartz is the grain size. In so far I agree that this might affect the interaction, but seriously I've never heard that grain boundaries have an influence on the interaction with electrons. In other words, it is never discussed that the glass or a nano-crystalline material reacts differently under electron beam. 

Coming back to your stress argument: SiO2 glass (quartz glass) has a very small expansion coefficient what is the reason that it is used preferably in chemical labs. Therefore I did not expect there any problem. But maybe I am wrong.

Best greetings

Gert

Dear Gert, not the grain boundary itself but the second phase particles such as agates segregated all along the intra granular region are affected by the electron beam and its local heating, which eventually results rough  surface appearance. Porosity near the surface definitely  enhances cracking and its penetration deep into the bulk.  l am talking of thermal expansion of porous agate not sio2 . That should be  definitely different. 

Definitely as you  have suggested properly, the relatively energy high  electron beam is interacting or damaging the surface region of agates in such an  extend that   it can be observable.

Best Regards to All

Dear tarik, I am also not expert in agate but if I look in Wikipedia or some mineralogical books, agate is described as micro- to nanocrystalline modification of quartz. From this I conclude that from the bonding and even structure agate IS quartz. Or am I wrong?

Best greetings

Gert

Hi  Gert

Should be the beam of electrons does not carry  enough energy  to ionize the sample. So the radiation  of electron should has sufficient energy only for excitation and the movement of an electron to a higher energy state. but if the electron has a higher frequency and shorter wave length than nonionizing radiation and long time of exposure it can burn the sample. These can attributed the answer for Prof Oğurtani   that your sample is rough and not flat enough and therefore the top of the roughness  area of the sample  will be near the beam  of which make some of the sample  burned and damaged and the irradiation with high voltage can caused color changes also. otherwise there is another interpretation for this feature It has been found experimentally  that the exposure to higher the energy of radiation decrease the melting point  of the sample and it can move  toward the lower  temperature area  and  changed the color. If we have a look to your micrograph it looks like the material moved toward the edge.

So from my opinion before the experiment there are some requirement have to do first

1) You have to polish the sample very well.

2) Also you can do long scan (I.e. the exposure time must be very slowly may it takes hours why not) but you have to decrease the kV.

May this can help

Good luck

Dear Gert I don't trust what those guys are saying about agate. What they mean by saying  'micro-nano crystalline modification of quartz' ? May be not XRPD but  the laser  Raman Spectroscopy would be best experimental method  to reveal whether the agate  has a long range order or not , and to discern the characteristic diffraction lines from the very similar pattern of quartz. Only we have some information about  Moganite, which is monoclinic SiO2. I think people spelling the word  agate very sloppy way to call the crashed quartz micro-crystalline particles in practice. Even though they use same word for the crashers that our grandmothers were using in their  Kitchen:  Regards

Regarding the sample roughness: believe me, for EBSD the sample are damn smooth after polishing 30min with colloidal silica. you would see roughness by the backscattered signal I showed. If there would be any roughness you wouldn't see any color anymore (please note, this is an image collected at a sample tilted by 70 degrees and acquired in SEM sollected with three BSE detectors and mixed). If you see there any color you can be sure that the beam sees at this position crystal orientations! At least during "comparatively" fast scanning (dwell time 30ns). For imaging this is typically not very fast but in comparison to EBSD it is.

Well, Tarik, you are certainly right. This was one reason why I was investigating the agate; in order to find out what's going on. Unfortunately I had to give up because i do not see any diffraction anymore during slow scanning. I remembered this since I observed the same at calcite. Nice contrast images for clearly bigger grains, but fast degredation of the structure whereas a few 100µm away the crystal become bigger and here you can map without any problem. Not mysterious but interesting... We know the solutions somehow but cannot test it. The sample was provided by one current  expert for agate.

Everyone here is just talking about accelerating voltage, but what about the beam current? Local heating and destruction depends on the delivered power which is a combination of dose and electron energy. If you lower your beam current, the power in a defined amount of time becomes less. At a certain current the material may not be damaged anymore, because it is capable of transporting the heat away. Of course, for EBSD you then need do increase the dwell time, but the decreased power could help.

I tried a few options (and hope that I mentioned this somehow). We went down to 10kV and also checked 30kV. The damages appeared higher with higher kV. The problem with lower kV were then charging-related effects (drift, focus lost, changing astigmatism...).

I meant the beam current, for example try to go down from your typical 5 nA to 500 pA.

Edit: This will also decrease the charging problems.

Yes, we also change the aperture (for ZEISS Gemini the only option to change the current). The acquisition time became then dramatically longer which cause defocusing during measurement. Since this was finally a less acceptable option since we need the maps I stopped the measurements. My question was also more related to the fact whether somebody found similar effect and why it happens at agate as well as in calcite? Nevertheless, thanks Björn.

Björn is partially right but he is also slightly off.  Dear Gentlemen here the power intensity,  exposure time and the angle of incidence are three  important parameters for the local heating to create observable amount of damage on the surface. The first one  is  directly related to the amount of energy delivered  by the electron beam per unit area per unit time, which may be calculated by  P = I V /A , I: current,   V : voltage,   A: cross sectional area of the beam if it would be perfectly  coherent  ray or better measure  would be the area of the specimen that is exposed to the beam. 

Not sure this can help but I have been working with shale and limestone at the FE-SEM. I often collect BSE maps after polishing my samples in a ion-beam cross polisher -you can't get a better polish than that. I always have problems with calcite charging similar to those you describe. I work below 5kV  and 1.2nA current but when I have a lot of calcite I still need to apply light coating to prevent charging. The larger the grains of calcite the worse the problem. Light coating (I use chromium 5 sec) is the only solution I found but I'd love a better one.

Dear Deborah, thanks for the tip. It is funny...we did a lot of work on eggshells but never had remarkable problems with calcite. Also the typical hydrothermal calcite works well but at the same sample along the contact zone to the host rock (sorry, I am no geologist or mineralogist) where we  have very fine grained calcite which will be "destroyed" during scanning. It seems that I observed the opposite at my samples. Do you know whether there is any "different" calcite? As discussed above, agate is assumed to have nano pores which are perhaps filled with water which could be the reason for the observed effect, but in case of calcite? Many thanks again for your comment. 

 I work with sedimentary rocks and I'm most familiar with calcite from fossils and from diagenesis and cementation of fossils. Intra-porosity is very common in calcite from fossils. I have cases where even though I do not see pores at the SEM, I see formation water come out of calcite grains -which indicates nanopores are present. It looks like the grains are sweating. Because the pores are not visible (my spatial resolution is ~16 nm) I cannot tell if these pores are always present in calcite or only in certain types.

I don't think you have experience with eggshells, do you? In case you havn't couldn't you give it a try how different calcite behaves?

Thanks for your comment!

I never worked with eggshells. Right now I'm not going to be able to look at them for a number of reasons. If I get a chance soon I will let you know -I'm curious to see:)

Take an ostrich egg. Easier to prepare :-)

Hey Gert,

your FSD image looks more like charging than phyiscal surface modification from my humble experience. Did you try this EBSD scan with low vacuum like >5 Pa Nitrogen or so? If you don't have a device for that maybe I can do a scan for you? Just send me the sample.. you know where to find me :D

cheers,

Tom.