What's your favourite pigment for painting?

Richard_P

Another discussion question!

What's your favourite pigment for painting?

I love Iron Oxide Red because it is so useful for making skin tones and a range of browns (when mixed with blacks, etc..). It is also super lightfast, extremely opaque and very cheap.

Do you have a pigment you love too? Perhaps you like Ultramarine Blue for the rich colour, or Burnt Umber for how useful it is?

Maybe you like the moodiness of Dioxazine Violet, or the brightness of Phthalo Green?

adridri

I line up with @MoleMan about earthy pigments. I love y. ochre and venetian red. Have to try oxyde red, but I will finish my v.r. first.

Richard_P

Well, oxide Reds cover a spectrum from English Red to Venetian Red through to Indian Red.

tassieguy

Funny you should mention Dioxazine violet, @Richard_P.  I've been making a lot of use of it lately. It's a wonderful pigment for darkening and cooling warm bright colors like yellow and red oxide and for darkening and warming blues and greens. I used to use Manganese violet, but it's weak and so you have to use a lot of it which gets expensive whereas Dioxazine is so potent you only need a tiny bit. Dioxazine also makes a wonderfully rich dark black when mixed with phthalo green and a bit of Quinacridone crimson.  So Dioxazine violet is my favorite pigment at the moment. 

GTO

For three or four years now I’ve used just marks pallet. The only time I’ve added anything different is when I needed a bit of Thalo Green to match the color of a pot.  The thalo green or blue are difficult to work with because they can get out of control an are hard to clean off a brush.  I read somewhere that red paintings and blue paintings are very popular.  In Asia I believe the color red is considered a “happy” color.  It is very popular during the 🧧 lunar new year.
I guess I don’t have a favorite color but I do like that yellow ochre that @adridri posted.in @Desertsky ‘s thread.  But I think paired up with the toned down green of the trees and that blue sky makes it work so well. It’s like the power and interest of the color comes through when combined with another color.

Richard_P

Rob, I am wary about PV23. The lightfastness tests I've seen are not great so I don't use it myself.

tassieguy

That's interesting, @Richard_P. I'm a little surprised because Langridge paints are very high quality. I will look into this. If you have a reference on its lightfastness, it would be helpful.  If it is a problem, I won't continue using it. 

PS From what I've read so far it seems to be a problem mainly for watercolor and the lightfastness varies greatly between brands. I can't find anything specific about it in oil colors, but I doubt Langridge would use it if it had poor lightfastness in oils.

Desertsky

@Richard_P -

You start the most interesting discussion. I thank you (and BTW, could you please start another image a day thread with paintings from people I had never heard of if it wasn't for you? You are helping educate everyone who reads these as well as all those future generations. Thanking you in advance. etc.  )

I have favorite colors that I like to just see - like manganese blue PB33 or yellow ochre. But I don't have any favorite color that I use in a painting because I like looking at the color. The colors I use depend on the image. For landscapes: black, white, some sort of brown, some sort of blue, some sort of green, etc. I usually use only a few colors plus black and white for everything. 

Desertsky

Richard_P

Rob, Handprint.com has a section on it:

Manufacturer and my own 2004 lightfastness tests indicate dioxazine violet actually has better lightfastness than reported by the ASTM, easily reaching "very good" (II) lightfastness in the best brands reported here. The lightfastness reported as 6,7 or 7,7 is equal to or better than the lightfastness I observed in naphthol red, PR170, perylene scarlet, PR149, quinacridone pyrrolidone, PR N/A, perinone orange, PO43, and even some brands of quinacridone rose, PV19, all pigments that are considered acceptable for artistic use. Overall, art materials manufacturers are clearly using pigments from different pigment suppliers, and the pessimistic ASTM ratings are either unrepresentative or flawed.
Examined overall, however, PV23 is a pigment that (1) is not transparently labeled "blue" or "red" by manufacturers; (2) produces highly variable lightfastness test results; (3) may produce unreliable lightfestness test results within a single pigment, given the wide range in lightfastness test results across different grades; and (4) therefore presents a difficult sourcing problem for paint manufacturers, who must themselves do rigorous testing in order to be confident in the quality of the pigment and pigment manufacturer they are dealing with.

This was reinforced for me with the release of GOLDEN's lightfast testing results in LAB format (I asked them for the data). It meant I could look at the colour change visually and see how much of an issue it was. It comes pretty much at the bottom of the lightfastness for their pigments:

 

Richard_P

Desertsky said:

@Richard_P -

You start the most interesting discussion. I thank you (and BTW, could you please start another image a day thread with paintings from people I had never heard of if it wasn't for you? You are helping educate everyone who reads these as well as all those future generations. Thanking you in advance. etc.  )

I have favorite colors that I like to just see - like manganese blue PB33 or yellow ochre. But I don't have any favorite color that I use in a painting because I like looking at the color. The colors I use depend on the image. For landscapes: black, white, some sort of brown, some sort of blue, some sort of green, etc. I usually use only a few colors plus black and white for everything. 

Thank you!

tassieguy

Thanks for the info, Richard. The Handrpint.com info was the first to come up when I did a search.  I will contact Langridge and ask about the lightfastness of the PV23 they use in their oil paint. It depends on the source of the pigment and, as far as I can see, no testing has been done on PV23 in oil paint. It is such a useful pigment that I will be sorry not to have it if it does turn out to be fugitive.

Richard_P

Ultramarine Blue mixed with Quinacridone Violet (the violet form of PV19) should be close as Handprint suggests.

I would take what all manufacturers say about the lightfastness of their paints and pigments (with the exception of GOLDEN) with a large pinch of salt. As far as I understand it Langridge is a small company and won't be able to have done any lightfast testing on their pigments themselves.

Desertsky

Pedantic alert (or maybe not)!

Handprint is a wonderful resource for watercolor lightfastness. This does not have much bearing on oil paint.

The Golden tests were for acrylic lightfastness. This does not have much bearing on oil paint. 

I use the artiscreation.com site as the starting point for LF and other weirdness for oil paint pigments. It also is easy at this site to compare the LF for the same pigment in different media.

I gave away the tubes of PV23 I had a few years ago to a student to play with; I cautioned him about the LF risk. I have PV14 which I like but have yet to use in a painting. It comes in both the blueish (Blue Ridge paints in the US) and the reddish (Michael Harding, and probably others) colors. 

and @Richard_P, you are welcome for the thanks - but I'd really like it if you started up your image-a-day for 2023. 

Richard_P

I disagree with the fact that watercolour lightfastness does not have much bearing on acrylic lightfastness, and so on for oils.

When I spoke to Sarah Sands about releasing their data she said:

While I see no reason we could not eventually provide this information for all our lines, we will likely start with GOLDEN Acrylics. That said, there are only a few pigments that differ in lightfastenss between oils and acrylics, and the pigments we use are generally the same across the lines. So results should be fairly comparable,

Sadly they haven't released the LAB data for Oils as Sarah has retired, but I think her comment stands.

I'm afraid I don't have the motivation or time to start posting an image a day again for 2023. There are plenty of users on this site who could take turns to post something if they were so inclined  

Abstraction

Richard_P said:

I would take what all manufacturers say about the lightfastness of their paints and pigments (with the exception of GOLDEN) with a large pinch of salt. As far as I understand it Langridge is a small company and won't be able to have done any lightfast testing on their pigments themselves.

Langridge partner with Golden and work closely with them - they are very much across the research. I sent an email last year to Langridge querying ingredients on one of their colours and the owner rang me and spoke for nearly 30 minutes about different things. In fact my query alerted him that he had changed the composition of one of his paints based on research but hadn't updated the site. Someone with genuine passion for his work, keeps tabs on the research and thoughtful about his approach. Australia's two main oil paint manufacturers (Art Spectrum being the other) are based in my city of Melbourne.

Abstraction

More recently I've fallen in love with both transparent gold oxide and transparent red oxide (Daniel Smith) so I will say PR101: synthetic red iron oxide. The transparency is brilliant for portraits. I tend to use it to glaze to warm an area rather than in the paints I use for basic flesh tints; good for warming darks; puts wonderful copper tones in hair that can either be kept subtle, like a glint, or deepened. Glazing layers to build a polished wooden chair. Easy to pull in other directions with other transparent or opaque colours.

I've talked about my antipathy to lead white many times. Yesterday I pulled out my cremnitz white PW1: lead white looking for a warmer transparency for faces. It was like a revelation. Suddenly I was able to unify the skin tones I had painted - from slightly patchwork to a unified whole, with a slight glow. I will definitely use it for glazing effects in future in other contexts. I would not use it as my basic mixing white because it is an expensive and insipid mixer that requires high volume and I'm used to titanium. I know lead makes a good paint structure but I'm too messy to be trusted with something so toxic for every day painting so I'll reserve it for when it is useful.

Richard_P

Abstraction said:

Richard_P said:

I would take what all manufacturers say about the lightfastness of their paints and pigments (with the exception of GOLDEN) with a large pinch of salt. As far as I understand it Langridge is a small company and won't be able to have done any lightfast testing on their pigments themselves.

Langridge partner with Golden and work closely with them - they are very much across the research. I sent an email last year to Langridge querying ingredients on one of their colours and the owner rang me and spoke for nearly 30 minutes about different things. In fact my query alerted him that he had changed the composition of one of his paints based on research but hadn't updated the site. Someone with genuine passion for his work, keeps tabs on the research and thoughtful about his approach. Australia's two main oil paint manufacturers (Art Spectrum being the other) are based in my city of Melbourne.

Ok, I was wrong on this one!

Desertsky

@Richard_P - Well, we'll agree to disagree about this. The LF ratings are different for the same pigment - depending on the medium. LF increases from WC (poorest) to oil (middle range) to acrylic (best). 

In the artiscreation tables, if you examine the eighth column from the left, the different LF ratings for the different mediums are shown. Most of the time, the LF for acrylics and oils are the same. Sometimes not. I have no idea how rigorous these ratings are, or how much they vary based on different sources of the pigment, and so on. Or how much this means on personal applications where we all do slightly different things with media and solvents. 

It seems to me that the best way to increase the nonfading of a color is to frame the painting under UV glass.  Maybe UV varnish. 

Could you tell us a little why you think the WC-oil-acrylic LF ratings are mostly the same? 

tassieguy

Abstraction said:

Richard_P said:

I would take what all manufacturers say about the lightfastness of their paints and pigments (with the exception of GOLDEN) with a large pinch of salt. As far as I understand it Langridge is a small company and won't be able to have done any lightfast testing on their pigments themselves.

Langridge partner with Golden and work closely with them - they are very much across the research. I sent an email last year to Langridge querying ingredients on one of their colours and the owner rang me and spoke for nearly 30 minutes about different things. In fact my query alerted him that he had changed the composition of one of his paints based on research but hadn't updated the site. Someone with genuine passion for his work, keeps tabs on the research and thoughtful about his approach. Australia's two main oil paint manufacturers (Art Spectrum being the other) are based in my city of Melbourne.

I'm glad to read this, @Abstraction. I use Langridge products exclusively. I have never encountered better paint. I believe they are very committed to producing a high-quality product. That's what their reputation is based on. I find it hard to believe they would use a PV23 that was not reasonably lightfast in their oil colors. However, I will call them to see what they say about it.

Richard_P

Forget about the LF ratings in this case. What I believe Sarah means is that a lightfast pigment remains lightfast in acrylics and oils, and a poorly lightfast one the same too. Even though the lightfastness ratings might not be exactly the same, the trend is the same. The work Sarah is referring to is their own lightfast testing on their acrylics and oils (Williamsburg) range, and the raw output translated into the LAB colour space so we can see visually the degree of change.

Some of the LF ratings done by the ATSM are decades old and some pigments (like PY3, PV23) vary considerably depending on how it is made.

Abstraction

Interesting to hear what you learn. In this site they say PV23 varies according to manufacturing - I don't know much about the site. Just search the page for pigment number. Also check out the visual for the blue wool scale.

https://www.kimcrick.com/pages/fugitive-pigments-list-lightfast-test-problems-art-supplies

"--- PV23 Dioxazine Violet --- There are some brands using cheap fillers and poor manufacturing processes for PV23 that give it a bad name. Nearly all reputable watercolor brands carry excellent LFI-LFII versions of PV23 (DSmith, M.Graham, Schmincke, RSzmal etc. do not fade)... BUT cheap paints from Superior/Etchr and student grade sets have labeled some purples as PV23 that faded LFIII-IV. Buy this from trusted sources only, or be prepared to test it yourself."

Because this site says PV23 is 8 on that scale for lightfastness -

https://www.ferro.com/products/product-category/pigments-and-dispersions/organic-color-pigments/dioxazine-organic-pigments/dioxazine-violet-pv23

Richard_P

I've seen that site before, very comprehensive. A few things I'm surprised about but all the others I know about and agree with.

It's a complicated area, on that site he found that some pigments fade very quickly when used in a watered down wash in watercolour. Yet Golden when testing acrylics like PY3 and PY73 found much better lightfastness when diluted with acrylic gel compared to white. So is the acrylic medium protecting the pigment, or is the white amplifying the light hitting the pigment and causing the fading?

So much we still don't know..

outremer

As a pure pigment, ultramarine is an absolute beauty. Every time I squeeze it out, I keep staring at it. But itself, not in mixes, it is useless for me. So when I was thinking on which login name to choose for this site, I 've decided on its native name.

I also love Indian yellow. My favourite effect with it is kind of glaze over wet white, to get glowing warm yellow.

Regarding PV23, Bruce McEvoy from Handprint is absolutely right about doing your own tests if you want to be responsible about your work. My goal is not "100 year in museum conditions" but "at least 10 years in my south facing living room". I did test it. And it's lost. I know WN rates it as excellent. Mine was from Daler and Rowney. But keep in mind that MOST manufacturers rate it as "okay' but not "excellent". Not a good sign. Add batch variations here. And then... Here comes the best use of ultramarine! Mixed with any violet leaning red you can forget PV23! (Phtalo won't get you there).

tassieguy

@outremer, I love Ultramarine, too.

Regarding PV23, Daler-Rowney is cheaper, student grade oil paint and not a professional grade. Therefore, it is not too surprising that their PV23 did not pass your test. They will have used a cheaper grade pigment and, from what I've read, a lot depends on the quality of the PV23 pigment used. 

toujours

I have 2 main old favourite pigments and a couple of new ones.
Prussian Blue is an old favourite, as is raw umber.   I could hardly exist without either of these.
New favourites are both Art Spectrum colours: Unbleached Titanium and Olive Green.

MichaelD

I have a turquoise blue that I love, first used it as a background for raspberries on a spoon painting, then for the car colour on the biggest piece I have done a 1964 Falcon Futura convertible.

Holbein-(lots of the lettering on the tube is Japanese as thats where its from

PB28 PG7 PW6



I also love the W&N Artist oil colour Violet Dioxazine—PV23

Richard_P

Interesting mix.. Normally you see a Turquoise colour made with Phthalo Blue and Green, where as the Holbein is using Cobalt Blue with Phthalo Green and White. I bet that's more expensive, but also more opaque..

MichaelD

@Richard_P, You would think so, but I recall it being transparent enough for me to need to mix a bit of green or another turquoise to make it more opaque.

It was pricy, and I had to send off for it. Luckily an American friend visiting a few years ago didnt just bring a tube that I asked her to but gifted me with 3. 

tassieguy

That Holbein Turquoise looks very like Langridge's Cobalt teal for which they give the number PB28 (77346). I understand that the PB28 refers to Cobalt blue but does anyone know what (77346) means? 

Richard_P

I'd guess they are internal product references codes, like these:

tassieguy

You might be right @Richard_P because when I search I can't find any reference to the number as a pigment. But then I wonder what they add to Cobalt blue to get Cobalt teal.

tassieguy

I found this with the number in it but I still don't know what it all means:

Cobalt Blue Pigment blue 28 CI 77346 NIR Reflective Complex Inorganic Color Pigment

Specification

Pigment Blue 28 (C.I. 77346 CAS #1345-16-0)  |  Pigment Blue 36 ( C.I. 77343 CAS #68187-11-1)

Spinel Structure | Spinel Structure

Reddish shade cobalt blue | Greenish shade cobalt blue

Ease of dispersion

Excellent dimensional stability

NIR reflectance (cool pigment)

Light fastness 8 (1-8)

Weather fastness 5 (1-5)

Acid fastness 5 (1-5)

Alkali fastness 5

Solvent fastness 5

Richard_P

Oh, it's a colour index number. That's just an alternative code to the Pxx system.

They add PG7 (Phthalo Green) to the Cobalt Blue to get the Teal colours (as well as a bit of white).

tassieguy

Ah, I see. Thanks, @Richard_P. 

Abstraction

Cobalt teal (PB28 oxides of cobalt & aluminium - Gamblin) is a stunning colour. Probably my favourite in terms of just the beauty of the colour. I used it in one section of my painting of tropical water and it lit with a glow. I don't have a lot of use for it otherwise. I'm not a CMYK colour mixer although I do have the 'teal' in this colour, the magenta, yellow and black.

Desertsky

outremer -  Could you share with us the results of the oil paint colors with which you did the 10 years in a south window test? 

NotACat

I can’t live without holbeins yellow ochre, it’s bright and clean and neither overly warm nor too baby poop green. It’s maddening how such a basic pigment can be so damn different across brands.
Also w&n ultramarine green shade, got so used to it that I don’t know how to paint with regular/red shade anymore haha. It saves me from having to deal with having viridian or pure phthalo on my palette.

osiosbon

My fav could be Transparent oxide red. It's not just how beautiful the color is but how useful it is, almost used all the time. 

outremer

@Desertsky, sorry, missed your question... Of course I did not test for ten years. But in that room, light fastness problems are real. Book jackets fade in a couple of years. Yellows disappear first. I did a simple relative test of an assortment of many paints. Study and artist grade. Watercolor too. Relative means I cannot say which BWS did good ones achieve. Even cheap hansas stood well. It was I think 1.5 years exposure on the  same south side but outside. Only PV23 changed a bit, others were okay. Watercolor was a short test with predictable result, according to light fastness ratings.

Richard_P

I was thinking more about this question in the last few days. I think there are two ways of thinking about and using pigments (or colours).

The first way is where artists like a certain colour, especially a mix of that colour with 1 or 2 others, and use it in their paintings. This style of what I would call limited colour mixing is more common with watercolourists. They tend to use more colours in their palette but less mixing for each colour they want to make.

The second way is where you try to match a certain colour you see in life or on a photo. To do this you use a bunch of pigments or paints to get to that tone. The individual colours you use are less relevant compared to the end result.

I'm in the second camp.

tassieguy

Me too, @Richard_P.  I aim to hit the colours I see. The pigments I have to use to get there don't matter. However, there are some pigments whose colour I really like for some reason. Dioxazine violet is one. Yellow oxide is another. Cobalt teal does it for me, too.