Heat Seeker: UV Fluorescence as a Gemological Tool
By Richard W. Hughes & John L. Emmett
One of the greatest challenges gem dealers and gemologists face today is being able to accurately determine if a stone has been heat treated. While a 100% reliable answer to that question is a job for a major gem lab, there is a simple and inexpensive tool that can often give an important indication.
So what is this miracle tool? We speak of the lowly ultraviolet light.
It wasn’t long ago that ultraviolet (UV) fluorescence was considered the poor stepchild of the gem lab, a pint-sized pea shooter when compared with the high-caliber cannons available in modern labs. But with the rising importance of treatment detection, the humble UV lamp is making a comeback.
Many heat-treated rubies and sapphires will display chalky short-wave (SW) fluorescence. This reaction is practically never found in untreated corundums and was first noted by Robert Crowningshield (1966, 1970). It is actually the colorless portions of the stone that fluoresce (a reaction similar to Verneuil synthetic sapphires). Since colorless areas follow the original crystal’s growth structure, the fluorescence will follow the same pattern as the gem’s color zoning. In addition, other trace elements in corundum may produce fluorescent reactions, from the well-known red glow of ruby to other reactions that are still not completely understood. Many will be illustrated below, both known and unknown.
Show and tell
So how does one go about checking for this reaction? The first step is to obtain a combination LW/SW lamp. You will also need a pair of protective glasses (SW light can burn your eyes with prolonged exposure). A viewing cabinet is also a plus. Finally, you will need a small lens to magnify the stone. Figures 6 –8 show one setup for both viewing and photographing fluorescence.
When observing fluorescence, the idea is to hold the stone with tweezers and bring it as close as possible to the lamp and view it under magnification. Examine the stone from all angles; many times the key chalky areas are confined to tiny portions of the stone.
One of the authors (Hughes, 1997) has suggested a lens be incorporated as an integral part of the viewing cabinet, but sadly instrument manufacturers have yet to produce such a unit.
CaveatsThis test does require a bit of knowledge. If a ruby or sapphire shows a chalky fluorescence in SW, it is probably heat treated. If it is inert, that does not mean it’s unheated. Also be careful that the stone is clean. Soap and other chemicals can also produce chalky fluorescence. And while this test is a tool that can be extremely useful, it is not a substitute for a complete gemological examination in a fully-equipped laboratory. Finally, keep the exposure times of corundum to SW fluorescence to a minimum. SW irradiation does create a yellow color center that can alter the color of the gem; even five minutes exposure can do this (see Figure 9). While this color fades with prolonged exposure to daylight, it can turn a blue stone more greenish (not good if it’s your stone and you’re trying to sell it).
One further caveat concerns a type of chalky green SW fluorescence sometimes seen in natural, untreated blue sapphires (particularly those from Madagascar). The fluorescence tends to be weak, and extremely superficial, being limited to thin layers at the surface. In addition, the fluorescent patches tend to have sharper boundaries than the reaction in heated stones (Figures 10–12).
Breaking down fluorescence
In its most basic sense, fluorescence is the emission of visible energy of a longer wavelength when bombarded by energy of a shorter wavelength. The stimulating energy may be x-rays (x-ray fluorescence), ultraviolet light (UV fluorescence) or even visible light. Ruby provides an excellent example of the latter.
When a ruby is put into daylight, certain electrons are excited to higher orbitals, producing absorption of the corresponding wavelengths. But instead of falling straight back to the ground state, the electrons fall in steps. In most cases, the release of energy from each of those steps is in the form of phonons to the crystal lattice (vibrational heat), and thus invisible to the human eye. But in the case of ruby, some emissions fall into the red (at 692.8 and 694.2 nm). This is what makes ruby so special; not only does it possess a red body color, but that red body color is supercharged by red fluorescence. This is what led the ancients to believe ruby had a fire burning inside.
UV fluorescence can be an extremely sensitive indicator not only of trace impurities, but also the conditions under which the gem formed. Indeed, it is not unusual for fluorescence to be easily seen from strongly-fluorescing ions at concentrations in the range of 0.01 parts-per-million (ppm). For lay people, that’s an itsy-bitsy amount, completely beyond the detection limits of all but the most sophisticated and expensive analytical equipment.
Speaking of sapphire
While the red fluorescence of ruby is detailed in many gemological texts (c.f. Hughes, 1997), the cause of the chalky fluorescence has not been covered. Let’s take a look at it.
Sapphire generally shows no fluorescence to visible light. But that changes if we expose it to short-wave UV. This is most clearly seen in synthetic colorless sapphire, which displays a bluish white (‘chalky’) emission in the range of 410–420 nm.
This blue fluorescence in synthetic sapphire has been observed at least since 1948. While it has been generally ignored in the gemological literature, it has been the subject of numerous scientific papers (c.f. Evans, 1994).
Evans surmised after reviewing the data that the 410–420 nm fluorescent peak was due to Ti4+ charge-transfer transition. That was later confirmed by Wong, et al. (1995a and 1995b). Isolated Ti4+ ions, or Ti–Al vacancy pairs produce this fluorescence.
The Ti4+ charge-transfer transition in corundum is so strong and the efficiency so high that the fluorescence is easily observed by eye at even just 1 ppm Ti4+. Most of the synthetic sapphire in the market contains at least one ppm of Ti4+ from the Al2O3 starting material, if not more, and thus fluoresces. The fluorescence peaks at about 415 nm at very low Ti4+ concentrations, but as the concentration increases, the fluorescent band broadens and the peak shifts to as high as 460 or 480 nm, making the fluorescence appear more greenish-blue or whitish-blue.
Why this chalky fluorescence occurs relates to the growth temperature and Ti4+ concentrations relative to other impurities. In synthetic corundums, the high growth temperatures and high Ti4+ concentrations produce the chalky fluorescence. In certain heat-treated sapphires with low Fe levels (such as those from Sri Lanka), high-temperature heat treatment creates similar conditions to the synthetic. Thus the chalky fluorescence.
But what about natural, untreated sapphires? Why don’t they fluoresce blue or bluish white? The reason relates to growth temperatures and time. Natural sapphires grow at much lower temperatures, so Ti4+ is much less likely to pair up with Al vacancies.
These lower temperatures also allow easier pairing of Ti4+ with other ions (usually Fe2+ or Mg2+) that prevent fluorescence. Another damper is the presence of Fe3+, which also kills fluorescence. And finally, as the crystal sits in the ground for millions of years, diffusion slowly takes place, allowing the Ti4+ to slowly pair up with other ions, thus killing the fluorescence.
Why then, do some heat-treated blue sapphires fluoresce chalky blue to green or white, and what causes the difference in appearance?
When blue (or geuda) sapphires are found in nature, they usually contain exsolved rutile. Titanium is concentrated in these rutile micro-crystals. When the stone is heat treated, the rutile dissolves into the corundum by diffusion, but because diffusion is slow, the local concentration of Ti4+ can be quite high. In the high concentration regions the Ti4+ concentration will exceed the local charge compensators (Fe2+ or Mg2+) and thus free Ti4+ ions will form. In addition, the dissolution of rutile will locally force the creation of some aluminum vacancies and some of the Ti4+–Al vacancy clusters will form. These types will fluoresce and thus some heat-treated sapphire will fluoresce somewhat like synthetic Ti-bearing sapphire. Because the original distribution of rutile (and iron in solution) occurred in zones, the distribution of the fluorescence will reflect that zoning. The fluorescence will be most intense where Fe is lowest and Ti4+ is highest, i.e. in areas of minimal color. The high iron-content basaltic sapphire (such as that from Australia, Thailand, etc.) will not fluoresce after heat treatment, as the iron concentration is much higher than the Ti4+ concentration everywhere.
All of the above is summarized in this table:
The appearance of chalky fluorescence in a corundum depends strongly on both the Ti4+ and Fe3+ concentrations. Considering Ti4+ first, it is important to note that the charge-transfer absorption in the UV per ion is extremely high. If we look at the fluorescence of a piece of synthetic sapphire with several ppm of Ti4+, it seems to glow blue throughout the volume. This is because the total Ti4+ charge-transfer absorption is low enough that the UV photons can penetrate into the bulk of the sample. When the Ti4+ concentration is higher, the fluorescence seems to be coming from a thick layer near the surface because that is as far as the UV photons can penetrate. At high Ti4+ concentrations, only a thin surface layer is penetrated by the UV and the fluorescence appears as a chalky surface layer (see Figures 15 & 16). The charge transfer absorption of Fe3+ is also very high. Thus iron will contribute to limiting the penetration of UV into the sample also. Thus the very different appearance of the fluorescence of some synthetic sapphire and some heat-treated natural sapphire is not a different phenomenon, just a difference in impurity concentration.
One of the authors (JLE) has used a Schott BG-12 filter to heighten the superficial chalky fluorescence often seen in heat-treated ruby. This filter eliminates the red fluorescence and transmits the Ti4+ blue fluorescence (Figure 17).
The traditional gemological versus the laser junkie
While laser junkies focus strongly on the subject of fluorescence of ions in crystals, it is an orphaned topic in gemology. Let’s look at the different approaches.
With gemology, fluorescence is typically only visually observed with either LW or SW UV radiation, with the results recorded in terms of just brightness, color, and the presence or absence of phosphorescence.
In the study of ions in crystals, the parameters measured are more extensive. Typically the spectral distribution of the fluorescence is measured, as well as the spectral distribution of the light that can excite that fluorescence (excitation spectrum). In addition, the temporal decay parameters of the fluorescence are measured using a short pulse light source. Sometimes the decay curve is a single exponential indicating a single site or a single ion. Other times the decay curve is a combination of two or more exponentials indicating multiple sites or multiple ions. All of these parameters are often measured as a function of temperature.
With UV fluorescence, we have something all too rare in gemology today: an inexpensive test that is as sensitive as even bomb-science level analytical equipment.
Now what does this mean for a gem dealer? With a small UV lamp, one can quickly check potential purchases. Any stones that show a chalky SW fluorescence are most likely heat treated. Total equipment outlay? The lamp alone costs less than $300. Heh, heh, heh, we can already see you smiling.
And for the laboratory gemologist? This technique has been under-appreciated in the gemological community. Fluorescence might provide an avenue to determine if some sapphires have received heat treatment at temperatures lower than those normally used for geuda. But this will require adopting some of the techniques and instrumentation of the laser junkie. While expenditures for such instrumentation are hard to justify without a guarantee of just what may be learned, the increasing need to stay abreast of gemstone treatments requires an expansion of our array of techniques. Sophisticated fluorescence instrumentation is far less expensive than SIMS, LA-ICP-MS or LIBS analysis. Perhaps not quite so sexy (nor nearly as expensive), but when it comes to utility, this still looks like a pretty decent dance partner.
RWH wishes to thank John I. Koivula for his encouragement and suggestions during the writing of this article.
About the authors
Richard Hughes is the author of the classic Ruby & Sapphire and
over 100 articles on various aspects of gemology. His writings can be found on
his personal web site, www.ruby-sapphire.com.
This article came about following RWH's plunge back into serious gemology in January 2005, when he joined the AGTA GTC. While checking the SW fluorescence of a heated sapphire, he decided to call JLE to inquire about the cause of the chalky fluorescence in heated and synthetic sapphires. "Interesting that you should ask," Emmett replied. "I've been doing much thinking about that same subject of late." And so it was that RWH and JLE began sharing thoughts on this subject.
Penned in bits and pieces in the first half of 2005, an edited version appeared in The Guide (Sept.–Oct. 2005, Vol. 24, Issue 5, Part 1, pp. 1, 4–7. Pieces also appeared as part of the AGTA GTC's regular Laboratory Updates.
This page is <http://www.ruby-sapphire.com/beryllium_sapphire.htm>
First posted October 2005; last updated 7 March, 2013