— BLOG —

What Is Cadmium-Free Yellow?


Last year, Liquitex issued a new line of cadmium-free paint.  They held a promotional campaign in which they mailed out sample tubes of the new paint along with a normal tube of cadmium paint for artists to test out.  The twist was that the tubes came labeled as “A” and “B”.  It was unclear which tube had the cadmium and which was cadmium-free.   The point here was Liquitex thought they had done such a good job making a nontoxic replacement for cadmium paints that painters would not be able to tell the genuine from the replacement.


There have long been nontoxic replacements for cadmium, but most of them have been unsatisfying in that the replacements tend to be transparent and weak mixers.  Cadmium yellows, oranges, reds, and even greens are consistently opaque and strong in mixtures.  For painters who get used to using cadmium colors while students, it is a hard thing to give up years later.  That noted, cadmium is a toxic, heavy metal.  Only a few years ago (as covered in “The Metals Run Dry” here on this blog), cadmium paint came close to being completely banned in the European Union.  As much as I love cadmium, especially cadmium red medium, a truly satisfying replacement would be no small thing.


My paints from Liquitex were labeled as “Cadmium vs. Cadmium-Free Yellow Medium” A and B.  Before even unscrewing the tubes, I noted the heft of them.  Cadmium, being a heavy metal, is, well, heavy and dense.  The first tell for most cadmium replacement paints is that they are too light when you pick up the tube.  Interestingly, tubes A and B felt to my hand to be the same weight.  The tubes were of course identical volumes.  To test out the paints, I made a series of swatches on primed paper.  These include tints, shades, mixtures with red and blue, as well as glazing the paint over light and dark grounds.  The masstones for A and B were both very close.  As dollops right from the tube, B seemed slightly paler than A, though B’s color deepened slightly upon drying, diminishing the contrast.


Where the paints were most different was in color mixtures.  The tint of A with white seemed to be lighter and more lemony whereas the tint of B looked slightly warmer in temperature.  The shade of A with black produced an olive green slightly more saturated than when B was mixed with black.  In the case of mixtures with blue (I used cobalt blue) or red (I used cadmium red), the A again yielded more saturated mixtures.  The overall difference between the A and B was definitely small, even when viewed under indirect sunlight.  Had I not been somewhat clinically making these swatches, I am not sure I would have noticed the difference.


Swatches of paints A and B


To complete the test, Liquitex invited painters to go online to their website and tell them whether they thought the A or B tube was the real cadmium.  I guessed that the B was the cadmium-free paint, which was correct.


Screenshot from Liquitex website on 11/6/2017


Of course, it also appears the plurality of the responders also guessed correctly, if not a majority.  I make this point not to dismiss Liquitex’s cadmium-free paint, because I think they have managed to make a replacement paint so close that all but the most exacting cadmium lovers will enjoy it.  Especially for those painters who avoid cadmium because of toxicity concerns, this is only a good thing.  There are too many inferior cadmium replacement paints, so there is a real need for a line of paint like this.


I did increasingly become curious as to what pigment (or pigments) were in this tube of cadmium-free yellow medium.  Typically, cadmium yellow hues tend to have some mixture of arylide yellows, which are often too transparent and cool in mixtures.  Golden Paints, a competitor of Liquitex, has a cadmium yellow medium hue that is a mixture of the pigments diarylide yellow HR-70 (PY 83), nickel titanate (PY 53), and arylide yellow 10 G (PY 3).  Winsor & Newton sells an acrylic paint in the Galerina line called cadmium yellow medium hue that is a mixture of arylide yellow GX (PY 73) and diarylide yellow HR-70.  The arylides (and diarylide) are all synthetic organic pigments, and none of them are anywhere near dense or heavy enough to be confused with cadmium, as my “B” tube was.  Normally, this information about pigment composition would be located on the side of the paint tube, but the Liquitex cadmium-free line of paints are not labeled in this way.  Their website is equally vague, so I decided to email the company.


Mark Cann, a chemist for Liquitex, responded opaquely to my cadmium query:


“I am one of the chemists that was involved in the development of these cadmium free colors.  Unfortunately, I cannot give you any guidance on this question.  All I can say is that we have taken a lot of time and effort to ensure that these colours are the very closest they can be to the original cadmium in both colour feel and specific gravity and because of this it’s been decided by the company not to disclose how.”


This, unfortunately, did little to make me less curious.  The rest of this article is what I will call my educated guess as to what I personally think Liquitex cadmium-free yellow medium really is.

I keep a CIE style color wheel of paints in my studio for reference.  The closest yellows to cadmium yellow (PY 35) that are not synthetic organic pigments are nickel titanate (PY 53) and bismuth yellow (PY 184).  Going back to the density of the paint, cadmium yellow medium has a solid bulk density of 4.5 – 4.9 g/ml (Mayer, 120).  Nickel titanate’s solid bulk density is 4.4 – 4.7 g/ml, which puts it roughly at the same density as cadmium (Mayer, 125).  Bismuth yellow’s solid bulk density is much lower, however, at 0.5 – 0.6 g/ml (Nayak, 834).


Yellows in the CIE color wheel


The density here argues in favor of nickel titanate, but unfortunately, the pigment is a bit too pale to be an exact match.  To my eye, bismuth yellow is a much closer hue.  In order to better measure the actual color of the paint, I obtained a small, handheld spectroscope so that I could do my own spectral analysis.  I already had spectral diagrams of what these pigments should look like, so I knew what I was looking for.


For cadmium yellow medium, there is little reflectance of visible light in the violet to blue part of the spectrum, but high reflectance in the yellow to red.  The reflectance curve for cadmium yellow has a signature sharp rise in reflectance around 500 nm (blue to green), and most of the arylide yellows that get used as replacements also have something very similar.  Bismuth yellow also has this sharp curve up after 500 nm.  Interestingly, nickel titanate does not.  Rather, nickel titanate’s reflectance curve continues to go up from 400 nm (violet) right up until about 550 nm (green).  It is a slower, smoother curve.


Reflectance curve for cadmium yellow medium (Mayer, 120)


Reflectance curve for bismuth yellow (Kim)


Reflectance curve for nickel titanate yellow (Mayer, 125)


When looking at cadmium yellow through a spectroscope, this sharp curve looks like a relatively sharp edge at the start of green, and there is no blue visible.  The smoother curve has the edge of the green extend out further, fading in from a smooth gradient of black to green.  The Liquitex cadmium-free yellow medium has just this smooth gradient to green when viewed through a spectroscope.  Given that the cadmium-free paint has a similar bulk density to nickel titanate (and no other yellow pigment similar to cadmium yellow is this dense), a reflectance curve that resembles nickel titanate in the green portion of the visible spectrum, and that at least one other competitor to Liquitex uses nickel titanate in its cadmium yellow replacement, I think it is very possible nickel titanate is in Liquitex’s cadmium-free medium yellow.  Also, I think it is possible that the paint includes some bismuth yellow, as it is closer in hue color and its reflectance curves to genuine cadmium than the nickel titanate.  Possibly, there is a mixture here of the two, and what might be truly new about Liquitex’s cadmium-free paint is that they are creating new replacement mixtures for traditional inorganic pigments without the use of any synthetic organic pigments.  Again, this is my speculation.  While nickel titanate and bismuth yellow have been around for a few decades, they have only recently begun to gain popularity among paint manufacturers.  I suspect what sets Liquitex apart is that these cadmium-free paints make a point of using lesser known metal but nontoxic pigments.  This leads to a paint that is close to the original not just in terms of hue, but also in opacity, mixing characteristics, and density.


Unfortunately, my spectroscope and painting studio/”chemistry lab” are not so sophisticated to do a truly definitive analysis.  I very much encourage anyone who is so inclined to take the task up and tell me what they discover.  At this time, it seems clear that Liquitex is not going to disclose this information, even though being mysterious about the contents of a tube of paint in my opinion runs counter to being able to advertise the paint as nontoxic (it’s nontoxic, but we won’t tell you what’s in it).  Also, I would love to hear if anyone else participated in the Liquitex cadmium-free challenge and what they thought about it.  Apparently, other people got a cadmium-free red instead of yellow, and I have no idea at this time what pigments might be in those tubes.




Kim, Tae Woo, Yuan Ping, Giulia A. Galli, and Kyoung-Shin Choi.  Simultaneous enhancements in photo absorption and charge transport of bismuth vanadate photoanodes for solar water splitting.  Nature Communications, October 2015.


Mayer, Ralph.  The Artist’s Handbook of Materials and Techniques, Fifth Edition.  Viking, 1985.


Nayak, Rangadhar, A. Suryanarayana, and S. Bhanojee Rao.  Synthesis, Characterisation and Testing of Bismuth Vanadate – An Eco-friendly Yellow Pigment.  Journal of Scientific & Industrial Research, Vol. 59, October 2000.