Why do sapphires change color in different light?

[LightBright]

Wby do sapphires change color in different kinds of light?

I have, for example, a bezeled sapphire (sold to me as Natural Ceylon - not a definitive description as it has no cert) that is a very transparent unsaturated grayish blue in dim indoor light. In regular indoor light the stone appears more saturated and starts to take on a more deep electric blue where indirect light hits it. In direct outdoor light the stone is transparent and bright electric blue with no purple tint. In shaded outdoor light it becomes saturated loses transparency and is very deep saturated blue with a purple undertone. In the car it is deep intense saturated blue.

I’ve seen a thread here on a nearly transparent antique cut sapphire that changes color in various kinds of light; my sapphire is similar.

Is this something that natural sapphires only do, or can lab created sapphires do that as well? Do heated sapphires change color the way unheated sapphires do? Is there some sort of internal mechanism at play (like microscopic inclusions that reflect various wave lengths of light)? Thanks for any info you might have.

 

[Starstruck8]

Your detailed description of colour shift is wonderful. My bluish, purple and magenta sapphires behave in a similar sort of way, shifting towards blue in skylight and towards red in indoor light. Note, I don’t have a highly saturated true blue sapphire, so these may be different.

Actually, most stones are at least a bit shifty, usually but not always towards blue in daylight. “Colour change” stones like alexandrite just take it to an extreme. Colour shift is not restricted to gemstones – we have all bought clothes in a shop and found that they looked different when we got them home.

The colour shift of a stone depends on its absorption spectrum. It is not directly related to heating or inclusions, or to whether the stone is natural or synthetic, except to the extent that these things affect the spectrum.

Colour shift is a by-product of the way our visual system tries to compensate for different illumination. Very roughly, our visual system makes something like the white balance adjustment familiar to photographers. But that is not enough to fully compensate all colours.

It took me a lot of reading and thinking to get to the bottom of this. I wish I could link to a succinct account that started from the basics, was easy to follow, and was factually correct. Unfortunately, I don’t know of one. Be warned, there are a lot of misleading and half-baked pseudo-explanations out there.

 

[LightBright]

Your detailed description of colour shift is wonderful. My bluish, purple and magenta sapphires behave in a similar sort of way, shifting towards blue in skylight and towards red in indoor light. Note, I don’t have a highly saturated true blue sapphire, so these may be different.

Actually, most stones are at least a bit shifty, usually but not always towards blue in daylight. “Colour change” stones like alexandrite just take it to an extreme. Colour shift is not restricted to gemstones – we have all bought clothes in a shop and found that they looked different when we got them home.

The colour shift of a stone depends on its absorption spectrum. It is not directly related to heating or inclusions, or to whether the stone is natural or synthetic, except to the extent that these things affect the spectrum.

Colour shift is a by-product of the way our visual system tries to compensate for different illumination. Very roughly, our visual system makes something like the white balance adjustment familiar to photographers. But that is not enough to fully compensate all colours.

It took me a lot of reading and thinking to get to the bottom of this. I wish I could link to a succinct account that started from the basics, was easy to follow, and was factually correct. Unfortunately, I don’t know of one. Be warned, there are a lot of misleading and half-baked pseudo-explanations out there.

Thank you for the explanation! The shifting colors make wearing the sapphire both interesting and enjoyable.

 

[Garry H (Cut Nut)]

At a trade fair the rubies will be displayed under halogen warm lighting and the sapphires under cold white fluorescent or LED. If you take any sapphire to the ruby side you will see a color change.
But there are color change sapphires with more pronounced changes like Alexandrite as a result of some rare earth trace elements.

(I always take the rubies to sapphire light and the sapphire to ruby lights - some dealers frown,but you get to see what is really happening in a range of lighting)

 

[LilAlex]

Some is fluorescence.

Some is that the light in different environments just has very different qualities. I think the following was from a manufacturer of LED bulbs. Only two of these have a lot of blue incident light. I'm sure the difference is even more extreme in the shade or on a bright, overcast day (my favorite for colored stones). Candlelight (not shown) is even more red-balanced than incandescent.

LED is really a mixed bag -- even before you go down the rabbit-hole of color-rendering index, etc.

 

[VividRed]

Some is fluorescence.

Some is that the light in different environments just has very different qualities. I think the following was from a manufacturer of LED bulbs. Only two of these have a lot of blue incident light. I'm sure the difference is even more extreme in the shade or on a bright, overcast day (my favorite for colored stones). Candlelight (not shown) is even more red-balanced than incandescent.

LED is really a mixed bag -- even before you go down the rabbit-hole of color-rendering index, etc.

Exactly this, it’s all about the spectrum of the light source and the atomic structure of the stone. Trace elements in a stone do not “give color”, they change the way lightwaves are absorbed. That’s why one should never buy colored stones in a jewelry shop without walking out with it.

Cool white LED for blue sapphires,
Warm white LED (or incandescent) for rubies and emeralds…to our eyes they look similar
But their composition is so different, as shown in your graphs

 

[Starstruck8]

Some is that the light in different environments just has very different qualities.

That’s part of the puzzle, but only part.

Incandescent light (for example) is ‘redder’ than daylight. But that does not in itself explain (for example) why white paper looks approximately white under either, but many sapphires look distinctly ‘redder’ under incandescent light. With white paper, our visual system adjusts properly for the different illumination, but for the sapphires, it doesn’t. It is this difference that needs explanation. It is an interaction between the illuminant spectrum, the stone’s spectrum and our visual system.

From one point of view, there is no mystery – the effect can be calculated from standard colour theory. I have read GIA articles that do just that. But an easily grasped account in plain words is harder to come by.

 

[LilAlex]

With white paper, our visual system adjusts properly for the different illumination, but for the sapphires, it doesn’t.

Can't think of any biological or physiological reason why my brain would color-adjust for everything in that scene -- except for the sapphire. I don't think that is it. Queue the PTSD from the black-and-blue dress meme...

 

[Garry H (Cut Nut)]

Even the background hue tone and saturation can change what you perceive.
And yes the blue black dress is an amazing example.
And if you close one eye you will see less dots here:

 

[Starstruck8]

Can't think of any biological or physiological reason why my brain would color-adjust for everything in that scene -- except for the sapphire. I don't think that is it. Queue the PTSD from the black-and-blue dress meme...

Your visual system colour-adjusts everything in the scene. (The official name for this is colour constancy.) But its algorithm works better for some things than for others. It works very well for white paper, not quite so well for most ordinary objects, and less well again for most sapphires. For alexandrite, its results are just weird - hence the colour change. Why the differences? Basically, different object spectra.

Here is a G&G article on the causes of colour in corundum:

https://www.gia.edu/gems-gemology/spring-2020-corundum-chromophores

For each chromophore they discuss, they work out an absorption spectrum. They use this to calculate the resulting perceived colour (in certain standard conditions, and assuming standard colorimetric theory) under daylight and incandescent light.

Check out Figure 11, for Fe-Ti, the most important chromophore in blue sapphires.
Note that the incandescent colours are darker and less saturated. See also Figure 4, for Cr, the basic red chromophore, The incandescent colours are a bit lighter. So for a stone with a mix of these chromophores (which would be plausible for blue/violet/purple sapphires), the incandescent colour would be redder and less saturated. This is in fact what we see.

In a sense, this explains the colour shift. But the result falls out of an opaque calculation. It would be good to have an intuitive story that identified the particular features of the absorption spectra that lead to the colour shift.

 

[Garry H (Cut Nut)]

From Grant Pearson FGAA in an article - will try to find online

 

[LightBright]

At a trade fair the rubies will be displayed under halogen warm lighting and the sapphires under cold white fluorescent or LED. If you take any sapphire to the ruby side you will see a color change.
But there are color change sapphires with more pronounced changes like Alexandrite as a result of some rare earth trace elements.

(I always take the rubies to sapphire light and the sapphire to ruby lights - some dealers frown,but you get to see what is really happening in a range of lighting)

This is so cool. My sapphire (not a great one) really does change incredibly, it is a pale almost gray and flat blue in the early morning inside. Then I can take it into a car in the afternoon and it turns intensely and deep blue, then in the direct light it is electric bright translucent light blue. Same stone, it projects different colors.

Now I’m wondering what your sapphires do in the light optimized for rubies, and vice versa...

 

[LightBright]

Some is fluorescence.

Some is that the light in different environments just has very different qualities. I think the following was from a manufacturer of LED bulbs. Only two of these have a lot of blue incident light. I'm sure the difference is even more extreme in the shade or on a bright, overcast day (my favorite for colored stones). Candlelight (not shown) is even more red-balanced than incandescent.

LED is really a mixed bag -- even before you go down the rabbit-hole of color-rendering index, etc.

This is very interesting and complicated. Somewhat related - when I wear my blue blocking clear glasses, my sapphire looks less blue, same with my sunglasses (for obvious reasons). So on my outdoor hikes I try not to wear my glasses for some of the time so I can admire the intensity of the blue of the sapphire (It is not always this blue).

 

[LightBright]

Exactly this, it’s all about the spectrum of the light source and the atomic structure of the stone. Trace elements in a stone do not “give color”, they change the way lightwaves are absorbed. That’s why one should never buy colored stones in a jewelry shop without walking out with it.

Cool white LED for blue sapphires,
Warm white LED (or incandescent) for rubies and emeralds…to our eyes they look similar
But their composition is so different, as shown in your graphs

I’m wondering what the Cool White LED versus the Warm White LED does - I’m not understanding the science. Is it that the Cool White limits the spectrum to blues, and Warm White removes blue? (and thus the reflections emanating from the stone?)

 

[LilAlex]

Check out Figure 11, for Fe-Ti, the most important chromophore in blue sapphires.
Note that the incandescent colours are darker and less saturated. See also Figure 4, for Cr, the basic red chromophore, The incandescent colours are a bit lighter. So for a stone with a mix of these chromophores (which would be plausible for blue/violet/purple sapphires), the incandescent colour would be redder and less saturated. This is in fact what we see.

This is clear, of course, and aligns with what the rest of us are saying. But it does not speak to my brain's ability to color-adjust for everything in the scene except a sapphire or an alexandrite. That's the only place where I disagree with you. I do not think that is the explanation, even partially. My "brain" does not know whether I am looking at a sapphire or a chunk of glass, and the color does not change in my mind when I get close enough to the jewelry case to learn that it is sapphire, say.

 

[LightBright]

Your visual system colour-adjusts everything in the scene. (The official name for this is colour constancy.) But its algorithm works better for some things than for others. It works very well for white paper, not quite so well for most ordinary objects, and less well again for most sapphires. For alexandrite, its results are just weird - hence the colour change. Why the differences? Basically, different object spectra.

Here is a G&G article on the causes of colour in corundum:

https://www.gia.edu/gems-gemology/spring-2020-corundum-chromophores

For each chromophore they discuss, they work out an absorption spectrum. They use this to calculate the resulting perceived colour (in certain standard conditions, and assuming standard colorimetric theory) under daylight and incandescent light.

Check out Figure 11, for Fe-Ti, the most important chromophore in blue sapphires.
Note that the incandescent colours are darker and less saturated. See also Figure 4, for Cr, the basic red chromophore, The incandescent colours are a bit lighter. So for a stone with a mix of these chromophores (which would be plausible for blue/violet/purple sapphires), the incandescent colour would be redder and less saturated. This is in fact what we see.

In a sense, this explains the colour shift. But the result falls out of an opaque calculation. It would be good to have an intuitive story that identified the particular features of the absorption spectra that lead to the colour shift.

I’m in awe. Thank you for this explanation, and I intend to study this paper now. This is fascinating.

 

[LightBright]

From Grant Pearson FGAA in an article - will try to find online

So this proves that even synthetic sapphire or other stone could change dramatically in various types of light.

 

[LilAlex]

Then I can take it into a car in the afternoon and it turns intensely and deep blue, then in the direct light it is electric bright translucent light blue. Same stone, it projects different colors.

Exactly! I love seeing my spouse's blue when we hike on overcast days. (Back when she wore it, she would humor me by wearing it hiking.) And when the sun hits it when her hands are on the steering wheel.

I think jewelers should have a separate room with a fully-illuminated ceiling that mimics a bright cloudy day. Every stone I review is a wild-card until I can see it outside on a cloudy day. Several sharp jewelers/gemologists think I am insane when I mention this. Some things have looked way better outside at home than they ever did under the blazing halogens of the shop. Those are the true keepers, imo.

 

[Starstruck8]

This is clear, of course, and aligns with what the rest of us are saying. But it does not speak to my brain's ability to color-adjust for everything in the scene except a sapphire or an alexandrite. That's the only place where I disagree with you. I do not think that is the explanation, even partially. My "brain" does not know whether I am looking at a sapphire or a chunk of glass, and the color does not change in my mind when I get close enough to the jewelry case to learn that it is sapphire, say.

The way I have been describing it, the differences noted in the G&G article are the result of imperfect colour compensation. No, of course your brain doesn’t decide, ‘This is an alexandrite, I’ll really mess with its colour’. Rather, it applies the same algorithm to everything. The algorithm just happens not to work very well for alexandrite. (Because it has a weird spectrum.)

But maybe that way of describing it was unhelpful. Here is what I think actually happens. [Warning: totally unqualified amateur at work.] Our brains infer colour (leaving a lot of complications aside) from the relative outputs of the three types of cone cells. These outputs depend on the spectrum of the light hitting the retina, which in turn depends on both the reflectance spectrum of the object and the intensity spectrum of the source. So if we want to see objects as about the same colour under different light sources, our brain has somehow to compensate for differing source spectra. How we do this is not entirely known. But the assumption used in colorimetric calculations is that we scale the outputs so as to make white objects look white, and use this scaling for all other objects. (This is just like setting camera white balance.)

This ensures that objects with flat reflectance spectra always look neutral, as they should. But for objects with other spectra, there are no guarantees. For any such object, there will always, in theory, be some pair of source spectra that will give at least a small colour shift. For most ordinary objects, and for most ordinary light sources, the colour shifts are small, so we don’t usually notice them. Or if we do, we get used to them. But for many sapphires, and for alexandrite, the shifts are larger and we do notice them.

Maybe you think this story is wrong? (Quite likely – the above disclaimer is not false modesty.) Or maybe you think it’s right, but not helpfully summarized by ‘imperfect colour compensation’?

 

[Starstruck8]

This is so cool. My sapphire (not a great one) really does change incredibly, it is a pale almost gray and flat blue in the early morning inside. Then I can take it into a car in the afternoon and it turns intensely and deep blue, then in the direct light it is electric bright translucent light blue. Same stone, it projects different colors.

Now I’m wondering what your sapphires do in the light optimized for rubies, and vice versa...

I love your enthusiasm and your inquiring mind. In your descriptions, I can see my (also not so great) sapphires.

From Grant Pearson FGAA in an article - will try to find online

I wasn't able to google the article, but the picture itself is worth a thousand words!

I think jewelers should have a separate room with a fully-illuminated ceiling that mimics a bright cloudy day. Every stone I review is a wild-card until I can see it outside on a cloudy day. Several sharp jewelers/gemologists think I am insane when I mention this. Some things have looked way better outside at home than they ever did under the blazing halogens of the shop. Those are the true keepers, imo.

That's a great idea - in Fantasyland, alas. My favourite jeweller always invites me to take stones outside. Of course, the weather can't be guaranteed, but any sort of daylight makes a big difference.

 

[chrono]

Even the background hue tone and saturation can change what you perceive.
And yes the blue black dress is an amazing example.
And if you close one eye you will see less dots here:

This is super cool. I see the filled dots change to empty dots instantenously as my eyes move around the "chart" and vice versa. Does anyone else see this too?

 

[ItsMainelyYou]

This is super cool. I see the filled dots change to empty dots instantenously as my eyes move around the "chart" and vice versa. Does anyone else see this too?

Yes, with a shifting color gradation of the smaller gray dot always in the periphery when they eye is resting or in movement. Fun!

 

[Garry H (Cut Nut)]

This is super cool. I see the filled dots change to empty dots instantenously as my eyes move around the "chart" and vice versa. Does anyone else see this too?

The point is that with one eye closed you see a different effect. Human vision is not like a camera. But the color of the CZ chips in different lighting will appear different to our eyes. Plus of course we each have different acuity for different parts of the color spectrum / rainbow. Aka color blindness etc.

 

[Garry H (Cut Nut)]

So this proves that even synthetic sapphire or other stone could change dramatically in various types of light.

Never done this before - but I think I can attach an article. This was published in the Australian Gemmologist 2016 / 2017.
There are spectra and all sorts of high tech stuff.
Grant and me are doing some fluorescence diamond studies at the moment.

if anyone wants a copy find my email address (not hard) and I will shoot it thru

 

[kenny]

This is super cool. I see the filled dots change to empty dots instantenously as my eyes move around the "chart" and vice versa. Does anyone else see this too?

No.
But I'm afraid you have a rare and fatal eye condition called Ocular Giveupgemitis.

It can only be cured by shipping all your jewels to me.

 

[LilAlex]

The point is that with one eye closed you see a different effect.

Same with two eyes closed. So weird.

 

[Garry H (Cut Nut)]

Same with two eyes closed. So weird.

Cheeky!

 

[LilAlex]

Any vendors use these lights from Waveform? Gotta assemble it yourself or have a pro install it. But CRI 99 is amazing. Second image is the spectrum -- which looks pretty great!

Anyone want to crowd-fund me for a Sekonic Spectramaster ( ) so I can really see what natural-light spectrum makes sapphires pop (bright overcast day)?

With some of these LED vendors, you can even custom-select the emission wavelength profile by varying the phosphors used. You could have different ones for the sapphire case and the ruby case and...

 

[Garry H (Cut Nut)]

Any vendors use these lights from Waveform? Gotta assemble it yourself or have a pro install it. But CRI 99 is amazing. Second image is the spectrum -- which looks pretty great!

Anyone want to crowd-fund me for a Sekonic Spectramaster ( ) so I can really see what natural-light spectrum makes sapphires pop (bright overcast day)?

With some of these LED vendors, you can even custom-select the emission wavelength profile by varying the phosphors used. You could have different ones for the sapphire case and the ruby case and...

It is rare to see any LED that has an output below 420nm?

 

[LilAlex]

It is rare to see any LED that has an output below 420nm?

Yes! It still misses the UV but creates a violet peak (centered on 420 nm) and catches a piece of the near-UV.

From their literature (I have zero connection to this company but was just looking for the highest-CRI LEDs out there):