Talk:YInMn Blue

Can the color be displayed?

Can a sample of the color be displayed? Are the RGB or other color coordinates known, or does this exceed ordinary color gamuts? Is this actually a new color, or is it a new pigment? Thanks --Lbeaumont ([[User talk:Lbeaumont|talk]]) 02:37, 29 June 2016 (UTC)

@Lbeaumont: It's both a color and pigment, like ultramarine. I couldn't find any info on CMYK figures, i guess it's still too early. ~ Elias Z. (talkallam) 10:28, 29 June 2016 (UTC)

@Lbeaumont: @Elie plus: Color and other pigment characteristics have been published in the journal, "Dyes and Pigments." From the abstract, "CIE color L* (∼34) and b* (∼-39) values, and increased a* (∼10) values in an acrylic coating." (D65 10° standard observer). --Mccomstock (talk) 18:22, 18 January 2017 (UTC)Mccomstock

I'm on it @Mccomstock: @Lbeaumont: ~ Elias Z. (talkallam) 06:48, 19 January 2017 (UTC)

Thanks for this. I am assuming that the color coordinates given above are using the LAB color model. I used these in Photoshop to specify a color and it is a darker, duller blue than I expected to see. The Corresponding RGB coordinates (according to the Photoshop conversion) are #3E4D8E (Hex) R:62, G:77, B:142 (decimal). See:

Hex Code	Sample
#3E4D8E

for a sample of that color. I am surprised by this and wonder what I have wrong. Thanks!

Thank you for pointing this out! I used an online converter and completely forgot about photoshop, let me check again. ~ Elias Z. (talkallam) 06:17, 20 January 2017 (UTC)

@Mccomstock: I tried photoshop, using the coordinates for the pigment that was judged optimal by the cited paper (YIn_0.8Mn_0.2O₃). The color coordinates are L=34.6 a=9.6 b=-38.9. I used the following tool to convert the values to corresponding HEX and RGB coordinates. Photoshop only accepts integers while the tool above accepts decimals. Can you access the paper? if not i will forward it to you via email or other means so you can verify the coordinates i got. ~ Elias Z. (talkallam) 06:43, 20 January 2017 (UTC)

@Mccomstock: I just realized you're on the paper 07:07, 20 January 2017 (UTC)

Ha ha --Mccomstock (talk) 13:29, 24 January 2017 (UTC)

@Lbeaumont: @Elie plus: The color of the powder or pellets that are shown online looks brighter than the pigment dispersed in a medium. This is pretty common. Also many of the online pictures show unmilled samples which almost always look more chromatic than milled samples.--Mccomstock (talk) 13:29, 24 January 2017 (UTC)

OK, so I used those LAB numbers and the tool you linked to and got Hex #2e5090 R:45.82, G:79.92, b:143.51 (decimal) See:

Hex Code	Sample
#2e5090

for a sample of that color. I notice it is slightly lighter and brighter. It might be helpful to include a table of sample color patches in this article for the various pigment formulas studied in the paper. Thanks for all this. Lbeaumont (talk) 21:58, 24 January 2017 (UTC)

Are you sure you have the color right? The pigment, when seen in real life, is a vivid blue (similar to the photo, but with a vivid quality that isn't quite captured in the photo, in a similar manner to how royal blue has a more vivid look than it does in photos). The RGB color sample shown on the page looks more like what the pigment in acrylic once the acrylic paint has dried would look like (which tends to dull colors, make them look more gray, something that doesn't happen to the same extent with oil paints). — al-Shimoni (talk) 20:31, 28 May 2017 (UTC)

There are colorspace conversion issues happening here: For the acrylic medium: ProPhoto RGB values are (scaled by 255) 62.5543, 62.8504, 117.9158; AdobeRGB are 60.492, 80.9126, 140.3348; sRGB, which most people will be using, is terrible at blues and is 45.8252, 79.9163, 143.521, which will already fall outside the range of the average cheap 6bpc+frc sRGB TFT panel which can only display an approximate 7 bits of half-assed information so it's going to lose a relative ton of accuracy. These are D65 with no colorspace->device space conversion, which is a requirement that the web didn't bother thinking about and most browsers don't support even for images with colorspace specified. It looks like sRGB is borderline / probably can't display the wavelength (473.1nm) / xy coordinates. I'm on my phone right now but I'll check whether it's out of gamut for sRGB by doing a soft-proof on a 16bpc CIELAB in a moment, and post the perceptual and relative colorimetric values, as well as whether it's even in gamut for AdobeRGB or the panel native which goes somewhat higher than that. The increasing divergence in R and G as the representable colors narrow from ProPhoto with its imaginary colors that extend outside the region of blue this color is in make both of the narrower spaces suspect, but I'll know when I do conversion. A Shortfall of Gravitas (other machine) (talk) 05:23, 5 April 2021 (UTC)

I should add I grabbed those numbers from [1] A Shortfall of Gravitas (other machine) (talk) 05:31, 5 April 2021 (UTC)

@Lbeaumont:@Talkallam:@Elias Ziade:@A Shortfall of Gravitas (other machine):@Mccomstock:

Hi everyone, according to the paper that was referenced above, the specific mixture of YInMn blue that they are discussing in the abstract, designated YIn_0.80Mn_0.20O₃, has L* = 34.6, a* = 9.6, b* = -38.9. Giving a chroma of ~40. Higher chromas are possible with lower concentrations of manganese. There is no specification in the paper about which white point or which observer was used to calculate the Lab values, so I must assume that they were D65, 2-degree. They do mention the ASTM method used but this only describes the measurement geometry for reflectance measurements. It doesn't appear to include details about specific method for converting reflectance to CIELAB.

With my assumption that the whitepoint is D65 the CIELAB coordinates convert to an RGB values of [0.194 0.317 0.557]. This color is in-gamut and the hex code #31518E  . This would be the masstone color of the specific chemical designation mentioned. However, lower concentrations of Mn result in a more chromatic and brighter color. YIn_0.95Mn_0.05O₃ has CIE lab coordinates of [45.4 10.9 -51.0]. A much higher chroma of approx ~50. This is also in-gamut for sRGB and translates to an RGB color [0.189 0.417 0.752], hex: #306AC0  . These lower concentrations of Mn are probably the more common in artist applications.

Thanks, please ping in reply. TDcolor (talk) 16:40, 18 August 2021 (UTC)

Incorrect temperature conversion

In the comment about synthesis, the conversion of Celsius to Fahrenheit is incorrect. As written it should read, "heated to 1,093 °C (1,999 °F)." The conversion looks to use some markup formula, which seems like it's not adding 32 °F, as in: degrees F = (9/5) * degrees C + 32. --Mccomstock (talk) 18:34, 18 January 2017 (UTC)Mccomstock

Fixed; the template was falsely used to display it as degrees difference so the +32 didn't come into play. I've opted to have the °F given as 2000 rather than 1999, under the assumption that that was the original measurement and the 1063(.33...) °C is a derived value, rather than the other way around. Zeyes (talk) 01:36, 8 June 2017 (UTC)

VICE article

For reference (it's a blog, so not exactly WP:RS): [2] 27.115.113.102 (talk) 08:26, 14 July 2016 (UTC)

AMD use on hardware

Seriously, does this marketing gimmick belong here? — Preceding unsigned comment added by 2001:14BA:3FC:5C00:D485:8EBA:EF8A:D037 (talk) 07:22, 2 August 2017 (UTC)

I believe you are right. Moved it to an "In Popular Culture" section, to mark that it AMD using YInMn is not a "real" thing, they just like the color... (which by itself has zero heat dissipation effects, but that's THEIR problem) YamaPlos talk 00:07, 11 October 2018 (UTC)

The name

I find Crayola's name suggestions to be too kiddy. If anything, we could name this as Yttrium Blue or Subramanian Blue in tradition of naming colors after artists who made it such as Hooker's green or Payne's grey.Alnair93 (talk) 17:35, 5 September 2017 (UTC)

Should add to article about the price

Because it uses both rare metals yttrium and indium, it's price obviously has to be significantly higher than usual industrial pigments without rare metals. Yttrium oxide price is about 40$/kg, indium is roughly at half price of silver at 250$/kg currently. (Though, cobalt for cobalt blue too isn't at dirt price, but definitely is less than indium) So, this pigment would be acceptable for Lego pieces, where buyer in any case is being ripped off with thousands times inflated price of product, but would not exactly be something to paint a ship or bridge with. 5.179.29.84 (talk) 11:04, 12 May 2018 (UTC)

it appears they were selling samples at $10 for 10 grams, expecting eventually perhaps $1,000 per kilogram. Alas, I've been looking at so many sites trying to find info on this, that I don't know where I got this datum, sigh. Mark it to talk, then... — Preceding unsigned comment added by Yamaplos (talk • contribs) 22:12, 10 October 2018 (UTC)

Samples are cheap, then you gotta pay applies here. Indium has gone up and Yttrium is needed more as dopant for LED phosphors I believe. Kremer pigments has it for $69/10g, and $295/50g (eq. $5900 / kg, they don't offer that quantity). For reference their purest grade lapis lazuli is sold up to kg quantity for $30,643.00, just in case $6k seemed high. A Shortfall of Gravitas (other machine) (talk) 06:10, 5 April 2021 (UTC)

Crayola marketing

Apparently (personal communication from Shepherd Color), Crayola wanted to use YInMn, but was not allowed to. So they made a color "based" or "inspired" by YInMn, but actually containing none of it. There seems to have been an article on the Sheperd Color website mentioning this in some very delicate way. As is, the Crayola mention was misleading, and actually irrelevant to actual YInMn (based on? cmmon!), but I am afraid I haven't yet found a public document that clears this matter up in plain language. Please help YamaPlos talk 22:10, 10 October 2018 (UTC)

I rebuilt a few things. AMD and Crayola are just taking advantage of the YInMn viral thing of last year, and their presence in this article is just "cultural", not factual and related to YInMn, they use none of it (though probably Crayola would want to use it if allowed). YamaPlos talk 00:05, 11 October 2018 (UTC)

Some problematic claims

In the article the pigment is claimed to be a "near-perfect blue" and also to be much more durable than ultramarine. Which standards are applied in these cases? Does the "near-perfect" simply mean it is very saturated? Ultramarine itself is very permanent, so in what way is YInMn Blue a relevant improvement?--MWAK (talk) 19:19, 19 December 2018 (UTC)

Don't know what near perfect means either, but ultramarine decomposes above 400°C, while YInMn is fine even above 1400°C. More pertinent to it's stability in typical use, ultramarine is unstable below pH6, where it blanches and releases hydrogen sulfide. Given that oxides of nitrogen and sulfur from industrial pollution produce acidic water and air pretty much everywhere there are people, this is a real issue. There are newer acid resistant forms where particles are coated with silicates. These may still be subject to color fading that has been observed in old paintings using ultramarine (see https://www.sciencedaily.com/releases/2006/10/061002214727.htm). YInMn is extremely stable under conditions that ultramarine is not and which arise under real world conditions. I'd argue that the claim of greater chemical stability is justified. Born2clone (talk) 20:16, 16 June 2022 (UTC)

Misleading information

A widespread, inaccurate, oft-repeated statement is contained here, too. In the Uses section, this comment is unsubstantiated: "It is additionally safer than cobalt blue." Cobalt aluminate is non-toxic. Here is a link to a page where you can find information about some toxicological studies on cobalt aluminate blue spinel, CoAl2O4 blue: https://echa.europa.eu/registration-dossier/-/registered-dossier/15286/7/3/2. In fact, cobalt aluminate has a better (less toxic) LD50 than table salt, https://echa.europa.eu/registration-dossier/-/registered-dossier/15467/7/3/2. --Mccomstock —Preceding undated comment added 15:24, 7 March 2019 (UTC)

Why limit to one displayed author per cite?

@Huntster: regarding "Limit to one displayed author per cite", why do it? I don't really see why this is any better than listing all authors, given the page has the "data" to do it. BernardoSulzbach (talk) 14:30, 25 October 2019 (UTC)

BernardoSulzbach: purely for ease of reading and avoiding clutter. The underlying data remains, and is just as machine readable, but any citation I do that to will have at minimum a DOI, if not a bibcode, arxiv, and occasionally a Pubmed ID to fall back on in terms of complete record. It simply isn't necessary here to display them all, and it's the common convention in bibliographies in the wider world to do this. — Huntster (t @ c) 16:39, 25 October 2019 (UTC)

I agree that when a DOI is available listing all authors is irrelevant. Doing what you did seems like a good idea. BernardoSulzbach (talk) 18:24, 25 October 2019 (UTC)