JEOL Probe Users Listserver
Moderator: Ellery Frahm, [log in to unmask],
Electron Microprobe Lab, University of Minnesota
Post a message: send your message to [log in to unmask]
Unsubscribe: send "SIGNOFF PROBEUSERS" to [log in to unmask]
On-line help and FAQ: http://probelab.geo.umn.edu/listserver.html
*
.................................
OK, Gang,
Alfred alluded to this but I want to emphasize a point.
I assume that all of you know that the diameter of an x-ray's analytical volume is mainly a function of the KeV being used. I just want to remind you/emphasize this.
Assume that you are analyzing a particular x-ray in a particular material at some set of operating conditions. The diameter of the analytical volume is 2um plus the beam for this set of conditions. Using a wretched 1000 um! beam means the analytical volume is 2.1um across. Now put in Hlava's magical electron emitter (HMEE - under developement)that produces a one-electron-wide beam. Your resolution has improved, for this signal, to 2.0um! Wow, an improvement of less than 5%. Of course, if you restrict your analyses to boron and beryllium you can lower KeV and get really great x-ray resolution for those signals.
Oh, but you can generate beauuutiful SE's and BSE's to going with the element maps. A problem here is that this may not be a good idea if you give the info to unsophisticated clients. They will disparage your skill because you cannot produce x-ray maps with the same "quality" as the SE's. I've had intelligent (but unsophisticated) Ph.D's do this. I usually use 15KeV for most things and I back off my Wehnelt so I have a crappy beam. My SE's and BSE's are OK up to about 1000x and they match my x-ray maps. Besides this, I get 6 to 9 months out of a W filament. (The first time I needed to change a 9 month filament under this regime, I had to drag out the instruction manual!)
Your choice depends mainly on what you analyze most of the time. Of course, you may have one cutomer that NEEDS this capability and is willing to pay for it. Then go with it. It gives you great bragging rights! But be sure you are honest with the sophisticated group on this listserver.
Paul Hlava
Sandia national Labs.
ps - If you believe in my magical electron emitter (HMEE) you are too naive to be on this list server!
-----Original Message-----
From: JEOL-Focused Probe Users List on behalf of Alfred Kracher
Sent: Thu 9/27/2007 12:16 PM
To: [log in to unmask]
Subject: [PROBE-USERS] LaB6 source
JEOL Probe Users Listserver
Moderator: Ellery Frahm, [log in to unmask], Electron Microprobe Lab, University of Minnesota
Post a message: send your message to [log in to unmask]
Unsubscribe: send "SIGNOFF PROBEUSERS" to [log in to unmask]
On-line help and FAQ: http://probelab.geo.umn.edu/listserver.html
*
Hello Ellery,
We have the LaB6 option. Yes, it needs an ion pump at the top of the column. The required electronics are already buiilt into the 8200; I don't know about the 8900. Our 8200 came with the LaB6 option and diffusion pumps, so I assume that is the default configuration. We replaced the DPs by a turbo in the field; it's what JEOL does when you order turbo.
We have tested both LaB6 and W, and "amazing" isn't the word I would use. The main improvement comes at low voltages, which would be a bonus for us, since we are often analyzing boron, but see below...
I would say at 15 or 20kV the difference is only noticeable if you really push the resolution way beyond what a microprobe needs. You get a few 100nm with W as it is, which lets you see a heck of a lot more than you can analyze. At 10kV and >30nA there is a distinct advantage to LaB6.
At 9/27/2007 12:42 PM, you wrote:
I am also curious if anyone is currently running or has recently run
a LaB6-outfitted machine. Are the concerns about beam stability
valid, or is the issue overblown?
I have a plot (if I can find it) that illustrates the point. We saw current excursions in the percent-range at a time scale shorter than between the cup measurement and the analysis, which would kill any effort at serious analysis. I think these can probably be got rid of (at least to some extent) with conditioning and careful pampering of the filament. You can switch back to W, but then the LaB6 goes into a desiccator, and according to people who have lots of experience with this (we also have a LaB6-based TEM) it becomes useless pretty quickly if is not under vacuum.
Alignment and saturation are also more finicky with a LaB6. A while ago I circulated a question to other users and got two replies from people with LaB6 options on microprobes. Neither one is using it. I will dig their e-mails out for you.
In summary, I would say that the improvement isn't worth the pain unless you are (a) very often running samples that need high spatial resolution AND low accelerating voltage, and (b) you are prepared to put up with LaB6 operation permanently.
Let me know off-list if you want to exchange more thoughts,
Alfred
Alfred Kracher
Ames Laboratory (USDOE)
Iowa State University
227 Wilhelm Hall
Ames, IA 50011-3020
Tel.: 515 294 7097
-----Original Message-----
From: JEOL-Focused Probe Users List on behalf of Alfred Kracher
Sent: Thu 9/27/2007 12:16 PM
To: [log in to unmask]
Subject: [PROBE-USERS] LaB6 source
JEOL Probe Users Listserver
Moderator: Ellery Frahm, [log in to unmask], Electron Microprobe Lab, University of Minnesota
Post a message: send your message to [log in to unmask]
Unsubscribe: send "SIGNOFF PROBEUSERS" to [log in to unmask]
On-line help and FAQ: http://probelab.geo.umn.edu/listserver.html
*
Hello Ellery,
We have the LaB6 option. Yes, it needs an ion pump at the top of the column. The required electronics are already buiilt into the 8200; I don't know about the 8900. Our 8200 came with the LaB6 option and diffusion pumps, so I assume that is the default configuration. We replaced the DPs by a turbo in the field; it's what JEOL does when you order turbo.
We have tested both LaB6 and W, and "amazing" isn't the word I would use. The main improvement comes at low voltages, which would be a bonus for us, since we are often analyzing boron, but see below...
I would say at 15 or 20kV the difference is only noticeable if you really push the resolution way beyond what a microprobe needs. You get a few 100nm with W as it is, which lets you see a heck of a lot more than you can analyze. At 10kV and >30nA there is a distinct advantage to LaB6.
At 9/27/2007 12:42 PM, you wrote:
I am also curious if anyone is currently running or has recently run
a LaB6-outfitted machine. Are the concerns about beam stability
valid, or is the issue overblown?
I have a plot (if I can find it) that illustrates the point. We saw current excursions in the percent-range at a time scale shorter than between the cup measurement and the analysis, which would kill any effort at serious analysis. I think these can probably be got rid of (at least to some extent) with conditioning and careful pampering of the filament. You can switch back to W, but then the LaB6 goes into a desiccator, and according to people who have lots of experience with this (we also have a LaB6-based TEM) it becomes useless pretty quickly if is not under vacuum.
Alignment and saturation are also more finicky with a LaB6. A while ago I circulated a question to other users and got two replies from people with LaB6 options on microprobes. Neither one is using it. I will dig their e-mails out for you.
In summary, I would say that the improvement isn't worth the pain unless you are (a) very often running samples that need high spatial resolution AND low accelerating voltage, and (b) you are prepared to put up with LaB6 operation permanently.
Let me know off-list if you want to exchange more thoughts,
Alfred
Alfred Kracher
Ames Laboratory (USDOE)
Iowa State University
227 Wilhelm Hall
Ames, IA 50011-3020
Tel.: 515 294 7097
|
