January 24 Through February 4—TUCSON, ARIZONA: Annual show
Welcome to the GemologyOnline.com Forum
A non-profit Forum for the exchange of gemological ideas
It is currently Thu Mar 28, 2024 4:57 pm

All times are UTC - 4 hours




Post new topic Reply to topic  [ 20 posts ]  Go to page 1, 2  Next
Author Message
 Post subject: more ruby comparison
PostPosted: Tue Apr 28, 2009 4:54 pm 
Offline
Gold Member
User avatar

Joined: Mon Apr 16, 2007 12:05 am
Posts: 1499
Well it has been a while, so some new ruby work. I guess I'm going by my made-up rule that you can never see too many spectra of ruby.

Here you can compare spectra from (a) a natural ruby from North Carolina, (b) a flame-fusion ruby, and (c) a laser ruby rod. These are transmission spectra collected using a UV-VIS light source. Click on the picture to get a bigger version.

First up, the North Carolina ruby. I've not posted its spectrum before. Everything but a small pinch of green light absorbed from 250-600 nm. Fluorescence peak at 694 nm, not resolved into two separate peaks. You can see some broad fluorescence structure at slightly longer wavelength.

Image

Next up, the synthetic ruby Frank sent me. This may have been posted before. You can just barely tell that absorption occurs in three bands, one centered in the SWUV, the other in the violet, and the last in the green. Fluorescence peak at 694 nm, not resolved into two separate peaks. Lots of other broad fluorescence features at slightly shorter and slightly longer wavelengths... reminiscent of the "organ pipe" fluorescence in spinel.

Image


Now for a spectrum collected down the 2 1/2" length of ruby rod. Here you can clearly see the three absorption bands, the SWUV, the violet, and the green. The light travels a long enough distance through the rod, all the light going to the metastable state that fluoresces has been used up, and so the metastable state gobbles up its own excitation wavelength... thus producing the narrow absorption at 694 nm! (Remember kids! Whatever emits light can also absorb light!)

Also we see a lot of other narrow structures that aren't present in the other two spectra... at 320 nm, 470 nm, 660 nm, and an interesting absorption dip at 750 nm. We figure this structure could be due to some metal impurity introduced during the manufacturing process, but really we have no idea.

Image


Now for one more spectrum of the ruby laser rod, this collected from a 1/4" slice (thanks John!) of the rod. Through this short length, we manage not to use up all the SWUV, violet, and green light, so we can identify the central wavelengths of the absorption bands: 250 nm (more or less), 405 nm, and 555 nm. Since we didn't use up all the violet and green light, the metastable state is able to fluoresce at 694 nm instead of absorbing. Also we can see again the narrow structures that we believe are due to some other metal impurity.


Image

So let's review... all three rubies show strong absorption in the SWUV, violet and green regions of the spectrum. It is well known that the violet (which extends into the LWUV area) and the green absorptions lead to the 694 nm red fluorescence. Shine either a green or violet LED on a ruby and it will glow red. Shine a LWUV lamp on it and again it will glow red. However, apparently the deep absorption in the SWUV does not lead to the 694 nm red fluorescence... inert when exposed to SWUV lamp.

So... if you wanted to use a monochromatic source of light to excite only the fluorescence, what wavelength would you choose? Well, either something around 405 nm or around 555 nm. As it happens, DPSS lasers producing light at 532 nm are pretty cheap, and that wavelength is close to the absorption minimum.

So... next up...
Fluorescence spectra of these three rubies exposed to 532 nm laser light.


Top
 Profile  
Reply with quote  
 Post subject:
PostPosted: Tue Apr 28, 2009 9:39 pm 
Offline
Platinum Member
User avatar

Joined: Wed Apr 26, 2006 12:44 am
Posts: 2056
Location: San Francisco
Nice work brian.. were those other samples ones I did as well?? Maybe Gearloose's batt lap made it into the spectra?? :) Boil in H2N03 for a while etc..

Now that 532nm pumping is really wild.. First you use a 890nm laser IR to pump the 1062nm NdYag then a KTP crystal to frequency double the 1064nm to 532 and now your going to pump a ruby to emit 694nm...

Well, if it gets you there from here.. I hope your students appreciate the convoluted fun path your taking them down... :)

_________________
Custom Gemstone Cutting and various other activities!


Top
 Profile  
Reply with quote  
 Post subject:
PostPosted: Wed Apr 29, 2009 3:14 am 
Offline
Site Admin
User avatar

Joined: Fri May 12, 2006 11:24 am
Posts: 7523
Location: Rome, Italy
THANK you DR. Brian, nice work indeed :D
Can't wait to test some samples by myself....
ciao
alberto

_________________
GemmoRaman-532 - GemmoFTIR - GemmoSphere - EXA


Top
 Profile  
Reply with quote  
 Post subject:
PostPosted: Wed Apr 29, 2009 5:38 am 
Offline
Site Admin

Joined: Thu Jan 05, 2006 4:27 pm
Posts: 1750
Thanks Brian,

very interesting...I remember discussing in the chat that the ruby rod was too long to allow fluorescence all along it's length. Nice to see it pictured in a graph.

Frank


Top
 Profile  
Reply with quote  
 Post subject:
PostPosted: Wed Apr 29, 2009 8:13 am 
Offline
Gold Member
User avatar

Joined: Fri Oct 19, 2007 5:04 pm
Posts: 1642
Location: Walnut Creek, CA
Just a curiosity:
in diamond anvil cells used to study materials at high pressure people usually put in the sample chamber a tiny crystal of ruby.
When you apply pressure to the sample you can have an estimate of the pressure in the sample chamber by exciting the ruby fluorescence with a blue laser (510 nm, I guess) and looking at the shift of the R1 (and R2...) fluorescence peak (originally @ 694 nm).
It's quite convenient...


Top
 Profile  
Reply with quote  
 Post subject:
PostPosted: Wed Apr 29, 2009 10:43 am 
Offline
Gold Member
User avatar

Joined: Mon Apr 16, 2007 12:05 am
Posts: 1499
Maialetto: Yes, using ruby fluorescence as a measure of pressure is quite interesting. In fact, my research interest is straying into materials whose fluorescence can be used to measure temperature!

Frank: Yep, we have some different lengths, so that we can show the fluorescence peak decreasing as length increases... and then turning into absorption as the length continues to increase.

Alberto: I can't wait either!!!

John: yep you polished the window in the North Carolina ruby.

Also I don't think the BATT causes the strange fine structure to show up in the laser ruby spectrum, or else there would be some evidence in the natural ruby spectrum. That fine structure and the wide dip at 750 nm bug the heck out of me. At first I thought I was dividing out the incident light incorrectly, because it has some structure, but no... I have repeatedly looked at the incident light's spectrum only to see over and over that structure is in the wrong place. Now I'm beginning to think maybe it is due to fingerprint oil. Whatever it is, it really bugs me.

And yep, I agree, it is indeed a convoluted path to isolate the 694 nm fluorescence light. Much easier to produce stimulated emission... just flood the ruby with white light!

Here is the fluorescence spectra of the three rubies, using 532 nm laser excitation. I think the proper name for this sort of spectrum is a luminosity spectrum, but don't quote me on that.

Image

These spectra were collected with my VIS spectrometer with about four times better resolution than the UVVIS spectrometer used to collect the transmission spectra. So VIS spectrometer easily resolves the R1 and R2 fluorescent lines. In fact, the overlap between the two peaks is due primarily to their natural widths... the measured full-width-half-maximum of the peaks is about 0.8 nm (slightly worse for the synthetic), while the spectrometer resolution is about 0.4 nm.

The VIS spectrometer has been expertly calibrated by my students this semester, and so the peak positions shown in the spectra above line up perfectly with the known values. In all three spectra, the R1 line peaks at 694.27 +/- 0.04 nm and the R2 line peaks at 692.91 +/- 0.04 nm.


Top
 Profile  
Reply with quote  
 Post subject: Re: more ruby comparison
PostPosted: Sun Jan 16, 2011 10:43 pm 
Offline
Active Member

Joined: Thu Apr 16, 2009 4:36 am
Posts: 61
Hi Brian, I am curious as to why you collected this spectra in tranmission and then explain about is as a function of amsorbtion. Becasue transmission spectra is not linear, it is difficult to resolve most the the important ruby features as they seem to be saturating your instrument. Can you post absorption spectra for these three samples also? Absorption spectra might more clearly resolve those "unknown" features you are seeing in the laser rod. Just my .02 and being from NC, Id be interestedin seeing the absorption spectra for the NC ruby.

Cheers


Top
 Profile  
Reply with quote  
 Post subject: Re: more ruby comparison
PostPosted: Mon Jan 17, 2011 3:28 am 
Offline
Gold Member
User avatar

Joined: Mon Apr 16, 2007 12:05 am
Posts: 1499
hi geniegems

It is really a matter of preference whether one works with a transmission spectrum or an absorption spectrum. If you want absorption, just flip the graph upside down and replace the "zero" at the bottom of the transmission spectrum with a "one" at the top of the absorption spectrum. When fluorescence is presence, though, it seems more proper to present a graph showing what light is coming out, rather than what light is not coming out.

What do you mean when you say the transmission spectrum isn't linear?

The light signal through the rubies does not saturate the spectrometer; if it did, then I'd just turn down the light source or integrate over a shorter time period. Maybe you are speculating that the change from fluorescence peak to absorption dip is due to saturation of the spectrometer? That is not the cause. I've discussed why the transition from peak to dip occurs in this chat.

I now think the "unknown" features in the ruby rod's spectrum is probably a systematic error due to a bit of the source light skirting around the edge of the rod. With such a long rod, our attempt to isolate the spectrometer from leakage light was quite awkward, and I now think it was somewhat unsuccessful. This systematic error is not present in the spectra of the other two rubies.


Top
 Profile  
Reply with quote  
 Post subject: Re: more ruby comparison
PostPosted: Mon Jan 17, 2011 11:07 am 
Offline
Active Member

Joined: Thu Apr 16, 2009 4:36 am
Posts: 61
Hi Brian, when I saw that transmission is not linear I mean that it is a logarithmic scale where absorbtion is a linear scale. In transmittance mode, the closer you approach 0 then you loose all resolution of your absorption features as it just becomes a big blur. Case in point, with two of your samples there is no detail below about 600nm as they are practically at 0% transmittance. Therefore, there is no information there. If you ran those samples in absorption mode and they were sufficiently thin not to exceed the dynamic range of the instrument, then you could resolve the peak height of the 560(U)band, the chromium lines at ~460nm (B) band and the strong feature at ~400nm (Y)...by the way, you'll notice is you add those band notations to the R line and reverse it, it is clear what those bands are named for :wink: . Also, in absorbance mode you might be able to resolve iron features. Just some food for thought.
Cheers
Genie


Top
 Profile  
Reply with quote  
 Post subject: Re: more ruby comparison
PostPosted: Mon Jan 17, 2011 11:42 pm 
Offline
Gold Member

Joined: Mon May 14, 2007 11:00 am
Posts: 1133
Location: Monterey, CA
Brian wrote:
It is really a matter of preference whether one works with a transmission spectrum or an absorption spectrum. If you want absorption, just flip the graph upside down and replace the "zero" at the bottom of the transmission spectrum with a "one" at the top of the absorption spectrum.

Brian,
Three hours ago, I sent my printer my latest (60 page) manuscript titled, "Pragmatic Spectroscopy for Gemologists". I expect to have it ready for Tucson, also in CD form.

I vote for transmission spectra as one can relate them to what one actually sees with a hand spectroscope.

You should get a kick out of my ruby spectra showing dichroism. You and the other nay sayers might be surprised at the results one can get with the GL Spectrometer. Also, I have a section on Spectra of Colorless Gemstones.


Top
 Profile  
Reply with quote  
 Post subject: Re: more ruby comparison
PostPosted: Tue Jan 18, 2011 3:37 am 
Offline
Gold Member
User avatar

Joined: Mon Apr 16, 2007 12:05 am
Posts: 1499
hi Genie

Now I see what you are defining as linear.

The transmission is the ratio I/I0, where I and I0 are intensity of transmitted and incident light, respectively. So one would say that transmission is linear with respect to intensity. In contrast, absorption is defined as -log(I/I0), so that it is not linear with intensity. As can be seen in this well-known absorption spectrum, the vertical scale is plotted logarithmically.

However, absorption is linear with concentration of of absorbers. So, as you say, an absorption spectrum may be able to tease out contributions from low concentrations of iron ions. That is, if the noise variation can be reduced enough. I don't know that it is worth the trouble, since we already know the iron concentration must be very low, because of its tendency to poison the chromium fluorescence. For example I have chromium garnets without iron that fluoresce beautifully, and I have a chrome pyrope with a barely detectable iron line (in transmission, of course) that won't fluoresce at all.

geniegems wrote:
If you ran those samples in absorption mode and they were sufficiently thin not to exceed the dynamic range of the instrument, then you could resolve the peak height of the 560(U)band, the chromium lines at ~460nm (B) band and the strong feature at ~400nm (Y)...by the way, you'll notice is you add those band notations to the R line and reverse it, it is clear what those bands are named for :wink:

The basic problem here is that putting enough light through the first two rubies in order to get well defined absorption curves will push the fluorescence way beyond the spectrometer's dynamic range. It can be done as long as one is willing to throw away the range above 600 nm, but I guess the purpose here was more to highlight the fluorescence rather than the absorption that gives rise to it. However, maybe this time around, I'll have my students make a separate measurement to obtain nice absorption curves that locate the peak absorption wavelengths. I'll make sure they use the NC ruby.

And thanks for the absorption band names, very cool.


Hi Bill

Bill Hanneman wrote:
I vote for transmission spectra as one can relate them to what one actually sees with a hand spectroscope.

Clearly I agree.

Bill Hanneman wrote:
You should get a kick out of my ruby spectra showing dichroism. You and the other nay sayers might be surprised at the results one can get with the GL Spectrometer. Also, I have a section on Spectra of Colorless Gemstones.

I bet I will enjoy those ruby spectra. And a section on colorless gems, eh? Do some of these colorless species have diagnostic spectral features that allows one to either identify an unknown as that species or eliminate that species from consideration? I guess so. :D

As for the GL Spec, my quibble isn't its use by an expert who understands its limitations. And it still isn't appropriate for photoluminescence, which kills my interest in it.


Top
 Profile  
Reply with quote  
 Post subject: Re: more ruby comparison
PostPosted: Tue Jan 18, 2011 3:57 am 
Offline
Active Member

Joined: Thu Apr 16, 2009 4:36 am
Posts: 61
I guess this is what they mean by "different strokes for different folks"....although im not quite sure who "they" is. Anyway, If transmission works for your application then it is valuable. For me, I use absorbtion spectra to see what color causing impurities are present and to calculate their concentrations. So, tranmission spectra isnt particularly useful to me.

Also, I am curious as to what type of intrument you are using and what the path length through the NC ruby is. Also, you might be able to minimize the effects of fluorecence by moving the sample farther away from the detector. (this i relize is probably dependant on what type of machine you are using and your resources for sample preparation, but just a thought)


Top
 Profile  
Reply with quote  
 Post subject: Re: more ruby comparison
PostPosted: Tue Jan 18, 2011 5:31 am 
Offline
Gold Member
User avatar

Joined: Mon Apr 16, 2007 12:05 am
Posts: 1499
Ah, so I was correct, measuring absorber concentration.

So Genie, you brought up some good points. We'll accommodate your request with the NC ruby... absorption spectrum, path length, experimental details. Check back mid-May.


Top
 Profile  
Reply with quote  
 Post subject: Re: more ruby comparison
PostPosted: Tue Jan 18, 2011 12:35 pm 
Offline
Gold Member

Joined: Mon May 14, 2007 11:00 am
Posts: 1133
Location: Monterey, CA
Brian wrote:
Ah, so I was correct, measuring absorber concentration.

The latest fad is showing plots of Absorption Coefficients (per cm.) with values on the ordinate ranging
from 0 to over 25.

Can someone explain to me what value that information has to practical gemology and how one determines path length through a rough or faceted stone?

I am referring to an article which compared spectra of heated and unheated stones. The caption reads "The spectra have been offset vertically." and the text says " Abundant cracks were obvious in almost all of the heat-treated stones but in only a few of the unheated samples..." GIGO. :lol:


Top
 Profile  
Reply with quote  
 Post subject: Re: more ruby comparison
PostPosted: Thu Jan 20, 2011 3:08 am 
Offline
Established Member

Joined: Sun Sep 26, 2010 11:01 pm
Posts: 46
Hi Brian,

I understand that you did alot of UV-VIS-NIR. Recently I am doing some testing with synthetic ruby vs real ruby and emerald vs synthetic emerald. I read from books that synthetic emerald and synthetic ruby(flame fusion) transmit short UV than natural ruby or emerald. However my UV-VIS-NIR measurement shows that all the stones NO NOT TRANSMIT short UV very well. Could you shed some light if it is wrong results? I am also wondering how can we do a proper UV-VIS-NIR measurement since the gemstones are of different cut/size, do you have a special jig for your measurements? Thank alot!


Regards,
Zen


Top
 Profile  
Reply with quote  
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 20 posts ]  Go to page 1, 2  Next

All times are UTC - 4 hours


Who is online

Users browsing this forum: No registered users and 13 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group
Gemology Style ported to phpBB3 by Christian Bullock