Assume you have two FIB system with almost the same specifications, and the only difference is that one of them can provide a probe current of up to 60 nA and the other one up to 100 nA.
But the first one has other advantages over the second system.
So, I want to know how important the FIB probe current is (for 3D EBSD and EDS analysis). Is there a significant difference between 60nA and 100nA, when it comes to milling speed and resolution?
60 nA is enough for wide spectra of materials. If you are planning fibbing of diamond samples (a lot of samples, large area of milling) I would say 100 nA is better. The best way - try both machines and choose. Ion current is not the single important parameter of FIB that should be under consideration.
Thanks Sergey. Yes I read that 100 nA can be somehow a bridge between nano-milling and micro-milling. Well, we might have some large area applications, but at the same time we need high resolution milling.
If somebody asks you what are the 3 key factors for choosing a good FIB system (for high resolution and fast FIB, also for relatively large areas), what would you answer?
High resolution = low ion currents, large areas needs time => high resolution at large area = low ion current + time =>NOT FAST. Appropriate FIB system depends on your task. For example if I need lamella preparation the key factors are fully automatic stage, stable manipulator, low ion accelerating voltage and automatic preparation software. For other tasks they can be different.
By analyzing technical descriptions of different FIBs it is extremely hard to choose machine. The best way - take our samples to the factories and try to do fibbing (or you can send it by post).
...and even then you still do not know whether you are affecting the microstructure in an inaaceptable way or not. The idea with 3D EDS is cartainly nice but if I compare the EDS resolution with the size of the FIB prepared sample I am not sure how informative (representative, statistically relevant) your results are. May major complain however is the influence of the ion beam on the microstructure. It is clear, we have a amorphisation layer which can be reduced, but for EBSD we get only the signal from the first nanometers so that it is quite questionable what I measure...
Agree Gert. minimization of the amorphous layer by ultra high vacumming at the FIB source can be an option... but still not very convincing.
In my opinion you should not use the maximum current for 3D tomography applications since the beam gets extremely broad and you lose a lot of resolution. Working at a few nA, maybe not more than 10-20 nA, is better. For less amorphisation, it is necessary that you can also work at ion acceleration voltages of down to 5kV.
Furthermore, a good drift correction is needed for such long time measurements.
So for your applications, these are the key factors, in my opinion: long-time stable beam with high resolution at medium currents and with energy down to 5keV and automatic drift correction.
Another idea: If you need to mill large areas, you could think about a plasma FIB like the Tescan Fera. It uses Xe ions and reached beam currents as high as 2µA. It mills 50 times faster than a Ga FIB. I used it once and it is a quite impressive instrument!
During our last EBSD meeting in Duesseldorf Dr. Krause from Halle presented a laser ablation technique for fast preparation and comperatively huge areas. It was planned to combine it with a FIB, i.e. removes much faster and more gentle material. As B. Hoffmann already recommended, it is a good idea to think about some alternatives which work faster and less destructive than FIB. As last step under very gentle condition it is certainly OK and unbeatable for local prep. But not an allrounder... We shouldn't believe every advertisement.
I found a talk from a colleague from Tescan who presented 3D-EBSD with a Xe-FIB on large copper wires. Take a look, link attached!
http://www.emez.ethz.ch/education/pastevents/pastev2014/fibw/Jiri_Dluhos
Thanks all for your comments. I have seen the specifications of TESCAN, but for some specific reasons, they might not fit to our requirements for a new FIB/SEM.
I have seen ZEISS promoted a FIB system combined with Rapid Laser Ablation.
The other option is Broad Ar Ion Beam milling (BIB), which is kind of fast for large area milling and is cheaper than a laser ablation attachment.
Any idea on this alternative? which one you prefer? BIB or laser ablation?
Dear Ehsan,
I am pretty sure that the Zeiss Auriga with the laser ablation is not a good choice for your measurements. The laser is mainly used for massive sputtering to open buried layers or structures and to open complete chips, for example. Also, the laser introduces massive damages due to heat transfer. Typically, you get a very rough surface which needs to be polished again by FIB to enable EBSD. I have two presentations from Zeiss about the Auriga. If you give me your email adress, I can send them to you. The Auriga is an impressive microscope, but, however, maybe not the ideal instrument for 3D-EBSD.
If your main application is going to be FIB tomography for large areas, you don't only want to consider the maximum beam current, but the beam current density. This can vary significantly depending on the quality of the FIB optics. So, just because one manufacturer claims to be able to provide higher current, if that beam has huge beam tails it will be of limited practical utility for 3D reconstruction applications. You can estimate this by simply putting the beam into spot mode for a controlled time on a sample (gold film on Si will provide good contrast) and measure the beam spread.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Probability
- Probability Theory
- Reliability