Zirconium Blue and Cerulean Comparison

Zirconium cerulean (PB71) watercolor in center, surrounded by similar blues on Arches 140 lb paper.

I was asked recently about zirconium blue, PB71, which is called zirconium cerulean by Kremer Pigments because of its similarity to cerulean. Unlike normal cerulean it doesn’t contain cobalt, so I presume that it’s less toxic, but it does cost about the same or a little more.

In the chart above zirconium blue is in the center, surrounded by all of the watercolor ceruleans (both PB35 and PB36) that I have access to and also some cobalt teals that I thought would be similar, plus manganese blue. The tartan blue was a limited edition variety of cerulean that I don’t think is made anymore, and the pigment for manganese blue isn’t made anymore either.

Zirconium blue is basically a heavily granulating turquoise or greenish teal. Some of the ceruleans and cobalt teals granulate more than others, but zirconium blue is only matched by genuine manganese blue. It seems to be a bit weaker tinting than the other paints here. The lifting is partly affected by the medium used to make the paint, but what I made here does lift very well, as does the “cobalt blue light” (actually cerulean) seen at the center left and the cerulean grey from Blockx in the bottom left.

The medium that I used to make each of the self-made paints was an experiment. Each time I make watercolor medium it’s a little different to explore how changes in proportions affect the paint. This time I used 5.5 grams of gum arabic powder dissolved into about 12-13ml of water that was heated to 140-150°F (60-65°C). Then about 2-3ml of light colored honey and 1-2ml of glycerin was added, which is less of both of those than I normally use. After shaking it gently but thoroughly 4 drops of synthetic ox gall from QoR were added. As a preservative I dipped the end of a brush handle into a bottle of clove oil, getting just a thin coating of it on the tip, and I put that into the medium. So far the medium seems to be working well. All of the self-made paints were just made justing a palette knife to combine the pigment with medium. Update- the paint was a little dry, so I approximately doubled the amount of glycerin.

Yellow Ochre and Mars Yellow Oil Paints

There’s a large variety of both natural and synthetic yellow ochres, and as one least expensive pigments it’s been a way for me to try different brands while also getting a useful paint. The natural version is normally called yellow ochre, and the synthetic version is Mars yellow, but paint makers can label their paints with whatever name they want. This includes using the name yellow ochre for paint made with the synthetic pigment. I’ll just call them all yellow ochre here.

Obviously most of the common yellow ochres, seen in the top half of the chart, are very similar to each other in color, so I would suggest just choosing only one or two at the most. Synthetic ones tend to have higher tinting strength, which is seen by how well the paint retains its color when mixed with white, and are also almost always more opaque. The exception would be the transparent mars yellows at the bottom.

Other aspects of the paint, such as how loose or stiff it is, may be very different even between paints that are similar in color. For example, the Old Holland paints above are the stiffest, while the M Graham and Winsor & Newton paints were much looser. Blockx was smooth and buttery, but not as stiff as Old Holland. The paints I made were loose and had no stabilizers added.

The more exotic ochres at the bottom aren’t seen as often in paint lines but have a greater variety of opacity and hue. A few of them I made myself from dry pigments. In some cases, such as the orange ochre from Rublev, those can also be bought as already made paint. I just don’t have those paints myself and my efforts to make paint shouldn’t be taken as good examples of what the already made paint would be like.

A couple of especially notable paints in this chart are the lemon ochre for its brightness and clarity and the orange ochre for how similar it already is to certain skin tones when just mixed with white. If I were to just choose two, those would be my favorites.

All of these swatches were mixed with Williamsburg titanium white. I tried to mix equal amounts but could have been a little off. The photo was taken in sunlight and white balanced in Photoshop, but may not be perfectly accurate. Some of these tubes were old and may not represent what these brands currently produce.

A similar comparison of red ochre oil paints can be found here.

Asphaltum, Van Dyke brown, Cassel earth, and Anthracite

In this brief comparison I have a variety of unusual brown paints that each have long histories but are seldom used today.

Van Dyke brown is one whose name is commonplace even if the genuine paint is not. It’s named after Anthony Van Dyck, a famous Flemish portrait artist who made extensive use of it in the 17th century. It’s also known as Cassel earth because of the deposits of it near the town of Cassel, Germany, and carries the pigment designation NBr8 (natural brown 8) because it contains natural organic matter.

The last paint in the chart above is a modern imitation of Van Dyke brown. It’s actually a mixture of pigments – in this case burnt umber and ivory black – that only imitates the color but not the handling of the original pigment. There’s many other modern pigments and mixtures that also borrow their names from some historical pigment that’s either no longer used or is only rarely used, such as indigo, sap green, vermilion, etc. As is often the case, part of the reason to make a modern imitation is because it’ll have better lightfastness than genuine Van Dyke brown.

Although it obviously has much higher tinting strength than the other paints – the third tint is a 1:7 mixture with titanium white and is close to the 1:1 mixtures for all the others – it also loses its limited chroma very quickly when tinted to higher values. The Gilsonite and anthracite may be weaker tinting but they’re also much more chromatic at high values.

I also made an area, not shown here, of evenly spread test strips of each paint that I check each day to see how long it takes for each to dry. All of this was started nine days ago and as of the time I’m posting this only two have dried because most of these are slow drying paints. I’ll keep checking and update this when or if the others dry.

These three I made from dry pigment, using linseed oil from Rublev. Nothing else was added, and I don’t know for certain how well they’ll dry on their own. They probably could have benefitted from a little bit of umber (PBr7) being added, because that contains manganese and acts as a drier for oil paint.

Gilsonite is a bitumen-impregnated rock that’s mined in Utah, USA. Bitumen (NBk6), also called asphaltum, has had many uses since ancient times, such as waterproofing for boats and ships. Today it’s still used for things like road construction, hence the term asphalt. Another example of use is that Ford model T cars were black because they were coated in a laquer made from Gilsonite. There’s many other possible applications for it. This is the most chromatic of all the paints here, but I’m unsure of its longevity as bitumen gained a reputation in the early 19th century for causing oil paintings to crack.

Anthracite, a high purity grade of coal, was used along with lower grades of coal by some artists in the past as a black pigment. Almost three years ago I made some watercolor with it, seen here.

The tube of Williamsburg’s Van Dyke brown was given to me by Williamsburg a couple of years ago and is one of the two varieties of this pigment they offer. Only two other brands – Rublev and Vasari – make paint from genuine Van Dyke brown. It has raw umber in it, but just enough to decrease the drying time to about a week without having much impact on how the paint looks or handles. The test strip took seven days to dry.

French Cassel earth is Williamsburg’s other Van Dyke brown. My tube is their old version from before they reformulated it to also have a small raw umber addition. Both of these Williamsburg paints have a large pigment grain and produce unique results that would be impossible to produce with modern imitations. Although they may look gritty in this photo, it’s not difficult to use them.

For comparison, M Graham’s paint here is a modern paint that has been given the name Van Dyke brown. Even if modern imitations are very similar in color they may be very different in handling, as you can see in how it has more uniformity and its covering power made it comparatively difficult to smudge thinly. In this case I think it’s due in part to the smaller particle size of the pigments used here. This paint was dry the first day that I checked the swatches, only about 24 hours after making them. That’s even though this is the only paint here that’s made with walnut oil, which normally takes a little longer to dry than linseed oil.

M Graham’s paint is itself an excellent paint, but there is a large difference between its handling and that of genuine Van Dyke brown. Whether this difference is an advantage or a disadvantage depends on the situation and the painting style.

Cobalt Blue Comparison

In a forum thread we were discussing cobalt teal and I posted this photo of various cobalt blue pigments in oil paint, which I thought I’d share here too. I can’t guarantee color accuracy in the photo, but I think it’s close.

Regular cobalt blue is normally made from the pigment PB28 and it’s a good, but often expensive, middle blue. It dries fast because of the cobalt content. Compared to ultramarine it’s a little more opaque and has a little less red while being lighter in masstone. Although this particular pigment is called by the name cobalt blue, all of the above paints contain cobalt. The difference between them being which other metals are included, such as aluminum or chromium, and in what amounts.

There’s two pigments that are labeled as cobalt teal by the paint makers that offer them. One is an uncommon and very opaque teal version of the standard cobalt blue pigment, PB28, and the other is a teal version of one of the cobalt greens, PG50. They’re similar enough that if you have one you won’t need the other. I almost never see other artists talk about having teals like these on their palette, so they must not be popular. I think they can be useful for painting green hills far in the distance, and definitely tropical water, but until now I’ve also rarely used them myself.

PB36 is another pigment that comes in a large range of varieties. Although it’s not the original cerulean blue, PB36 can be so close to the original in appearance that it’s often given the name cerulean. At the other end of its range are blue turquoise and green turquoise varieties. Shown here is a green turquoise. I like the turquoise color, but again I rarely use it.

The original cerulean is PB35. It typically dries to a more matte surface. Either this or the PB36 version labeled as cerulean would be good for painting skies, especially closer to the horizon where the sky has more green, as they are both slightly greener than cobalt blue. I actually prefer the PB36 version of cerulean because the color has a little more intensity to it. The name cerulean probably comes from the Latin word for heaven or sky.

Lastly, in the right column I mixed all of these with an equal amount of nickel titanate yellow, which is a somewhat dull lemon yellow, to see how they’d behave.

Terre Verte / Green Earth

Getting a good photo was hard, and even this isn’t perfect. Whenever I’ve made oil paint from this pigment it’s extremely short paint when it’s freshly made. Even without any stabilizers added the consistency of this paint is the closest there is to being like butter. I think I could even make a sculpture with it.

Genuine terre verte is very transparent and weak tinting. It’s good for glazing, for example. Some brands make an imitation terre verte by mixing pigments like phthalo green with burnt sienna, but the imitations are always far too opaque and are much stronger tinters.

Cool terre vertes, like the Nicosia green earth and the French terre verte at the top, are hard to find. Everyone seems to prefer making the warmer varieties. Sometimes I like to add a little terre verte to ultramarine to make it closer to a middle blue. 🙂

 

Ultramarine, Lapis, Indigo, and Indanthrone Blue Charts

A while back I had posted this chart on a forum in response to a question about what these blue paints looked like compared to each other. The color isn’t 100% accurate but it’s close enough to show how they compare. The box on the right shows color averaging from the first 1:1 mix with white for each paint for comparison.

Ultramarine
This is a workhorse paint. It’s a very good, strong, easy to use, and useful color that’s reliable to not fade. Every brand of paint sells this and it’s always one of the least expensive from any of them. It’s hard to find a more perfect paint. Almost every paint set for students includes this. For most artists that I’ve seen, if they only have one blue on their palette, it’s usually this one. This is sometimes referred to as “synthetic ultramarine” while talking about lapis lazuli, which is natural ultramarine.

Lapis Lazuli
Natural ultramarine is the opposite of synthetic ultramarine in almost every way. It’s weak, dull, very expensive, only available from a few brands, and difficult to use in many painting styles because it’s so transparent and so easily overpowered by anything you mix into it. You can see this from how light it becomes when mixed with the same amount of white as any of the other paints on the chart were. It’s good for glazing because of its high transparency, and in watercolor it’s actually useful for its high granulating texture. Other than that I think a lot of artists tend to have high expectations for this famous paint that’s spoken of so highly in much of art history but when they actually try it themselves the paint doesn’t match the hype. There’s different grades of pigment depending on the purity and to make this from the highest purity would be extremely expensive. I think it’d be better to just use synthetic ultramarine.

Indigo
Natural indigo is known to fade and I don’t think anyone makes paint with it anymore. Synthetic indigo is PB66, only used by a few brands, and I’ve read conflicting reports about its lightfastness. Almost all paints named “indigo” today are just mixes of blue and black, sometimes including other pigments such as a little violet, with each brand using their own mix. The synthetic indigo in this comparison is very slightly bluer and less grey than the mix below it (which is made from phthalo blue, lamp black, and ultramarine), but if you wanted the mix to be bluer it would be easy enough to just add a small amount of extra blue to it. Doing this would avoid any question of whether synthetic indigo is actually lightfast and you wouldn’t be confined to just the few brands that make it. If you want to use a so-called “Zorn palette” (usually yellow ochre, vermilion or light cadmium red, ivory black, and white) you could consider indigo instead of pure black to give the palette a little more flexibility. Greens would be slightly easier to mix and the cool greys would be a little more believable as “blue” when placed next to a bright red than simply mixing black with white.

Anthraquinone, Indanthrone, or Indanthrene Blue
The name depends on the brand, but I call it indanthrone. The popularity of this slightly reddish blue seems much lower than many of the others because I rarely see it on anyone’s palette. I think it’s usually somewhere in the middle of a brand’s price range and is clearly not as intensely chromatic as ultramarine. It has much more tinting strength than ultramarine though, so a smaller amount of this paint will go further in mixes, possibly offsetting the price. It also seems a little more opaque than most blue pigments. I used to use this a lot early on when I started painting and was exploring different pigments, but I haven’t used any for a long time now. Maybe that’ll change.

A Perilous Climb (revised)

Oil on canvas, 5″ x 7″

I had wanted to call this done before, but maybe it was a little too rough. In this revision I’ve added German black (natural PBk11, Williamsburg) to the sky and more of the previous Davy’s gray and my own white that I was working with in the first version.

Here’s a comparison of the white I made against a titanium white from Blockx-

You can see that the white from Blockx in the bottom row, which is Titanium white pigment in poppy oil, has a distinctly cooler/bluer color when mixed with German earth black. It’s also a shorter paint consistency, more opaque, and higher tinting.

My own white that I mulled is a mix of these pigments- titanium white (Blue Ridge), zinc white (Gamblin), and marble dust (Daniel Smith).
The oils used are walnut (M Graham) and refined linseed (Gamblin).

I didn’t measure the amounts used, but if I had to guess it’s around 60-70% titanium white and of the rest it’s probably more than half marble dust. You can see there’s a little bit of oil separation, but not much.

I actually like my own white paint for a couple of reasons. First, pure titanium white is normally very opaque and has very high tinting strength, far more than it really needs to be for artists. The addition of zinc white and especially marble dust has weakened the paint enough that it’s a little easier to control it in mixes. Second, titanium white has a reputation of having a “cooling effect” on other colors. I tested that out by comparing titanium white with zinc white in this post and found that the difference between their effects on color temperature to be situational, with some colors more affected by others, and at times nonexistent. Still, in this case my own white was clearly warmer and so its tints with the black were closer to neutral.

Red Ochre Oil Paints

PR101 and PR102 (Pigment Red 101 and 102) are among my favorite pigments for paint and they come in a large variety of colors. PR102 is a natural red iron oxide (red ochre) and PR101 is the synthetic version. Below is my collection of oil paints using either of these as a single pigment paint with nothing else added. I photographed this in direct sunlight during the afternoon. All of the paints were mixed twice with about an equal 50/50 mix with white.

Old Holland Yellow Ochre Burnt
This is my only natural red ochre in oil paint. As I understand it this pigment is made by roasting natural yellow ochre pigment until it turns reddish. It could make a good alternative to burnt sienna. It’s a little lighter and brighter, and has a stronger color.

Winsor & Newton Transparent Red Ochre
I wasn’t really sure what to expect when I got this because the name sounds like it’d be what’s normally called Transparent Red Oxide, which is the variety of PR101 that W&N already uses in their “Burnt Sienna”, but this is really very different. I think it’s close to the same as the Yellow Ochre Burnt above it but this is a slightly cleaner and more pure color. As the name suggests, this paint is semi transparent. The tinting strength isn’t as overpowering as some of the others in this list so it’s a bit easier to work with. This is one of those paints that I wish I had gotten years ago. It might even be my favorite red ochre.

Rembrandt Burnt Sienna
Although real Burnt Sienna is made from either PBr6 or PBr7, there are a few brands of paint that use a Transparent Red Oxide version of PR101 instead, which has a much stronger color, higher transparency, and higher tinting strength than actual burnt sienna. It doesn’t show up well in the photo but this particular one is lighter in masstone than the “burnt sienna” below it from W&N.

Winsor & Newton Burnt Sienna
A nice example of Transparent Red Oxide. It has good transparency. I tend to prefer paints like this over actual burnt sienna for the greater color strength.

Blue Ridge Transparent Red Oxide
Very similar to the above paint, but not as transparent. This one also seems to have a slightly higher tinting strength, but I can’t guarantee that I got the mix with white an exact 50/50 mix on any of these.

Old Holland English Red
Now we’re getting into the slightly cooler temperature colors for red ochres. This is a very high tinting strength paint, and easily overpowers most other paints that I mix it with unless I use extremely small amounts of this. Because of that, I don’t think I’ll ever run out of this one when you consider how little of it is needed in each paint mix. This one it so similar to the venetian red below that I actually mixed up the labels on the image at first. I’m 99% sure I have it right now since I kept the tubes lined up in the same order that I placed the paint on the palette and because this paint has a thicker, more viscous consistency.

Winsor & Newton Venetian Red
There’s a little bit of variety of colors that are referred to by different brands as “venetian red” but this name is generally given to a semi-cool earth red that’s not as cool as something like an Indian Red.

Blockx Mars Violet
As far as I know this is about as cool as red ochres can get. I bought this one from Blockx because when I looked at a paint chart someone else had made of different brands of mars violet the one from Blockx looked like the coolest. On an earth palette of red and yellow ochres this could be a very useful.

A similar comparison of yellow ochre oil paints can be found here.

Williamsburg Black Paint Assortment, Plus Lemon Ochre

It’s been awhile since I’ve done this, but over the past few months I’ve gotten various new paints from several brands in both oil and watercolor and I thought I’d make some swatches with them over the next few days.

This time I have the 1 ochre and 4 black oil paints I got from Williamsburg over the summer. Each swatch was first spread from a small blob of paint and then mixed in a roughly even mix with M Graham titanium white (PW6 + PW4). I’m pretty sure that’s the white I used. It was actually yesterday that I did these. That mix was then spread to the right of the original paint. A small amount of that was mixed again with an equal amount of white, and then again once more, for a range of increasingly lighter swatches that show the tinting strength of the paint.

All of these photos were taken today in the afternoon during overcast clouds. I can’t guarantee perfect color accuracy.

German Earth (PBk11) is natural black iron oxide. As expected of this pigment, it’s an opaque paint and a strong tinter. Williamsburg’s website describes it as “bluish” when mixed with white, but I’m really not seeing it. I’d say it’s fairly close to neutral.

Slate Black (PBk19) is actually very similar to German Earth, but less opaque. It’s apparently made from slate from Pennsylvania. Just a little bit gritty.

Davy’s Gray Deep (PBk19) is also slate from Pennsylvania but much lighter and with a warmer color. It’s very smooth and spreads surprisingly thin. I think it’d be very good at glazing. The photo doesn’t show it the best, but this was lighter than the first two in masstone. Its tinting strength is pretty low. Apparently slate comes in several colors, and as far as I can tell Williamsburg might be the only paint company that currently makes oil paint with slate. Many other brands all produce a “Davy’s Gray” of their own that is always one of various mixtures of pigments. A very unique paint.

Graphite Gray (PBk10) is made from ground graphite. It’s another unique paint that I’m glad I got. I think it looks a little lighter here than it really is because of how the light was reflecting off the paint, but this is the lightest of the four. It’s just slightly iridescent because of the graphite. I really like this one because it’s easily the bluest black paint I have.

This Lemon Ochre (PY43) is another natural iron oxide and is part of the Native Italian Earths set. I only have this one from the set. It’s a very strong earthy yellow. I really like the color a lot. Kind of gritty though.

As you can see in this closeup the vertical streaks are where tiny particles of grit were dragged along under my palette knife as I spread the paint thin. I had heard that Williamsburg’s earth pigments tended to be gritty, and at it seems least a couple of them are. While I can’t say this is a positive thing, it’s not necessarily a big deal. I mostly notice the grit in any paint from any brand, if it has any, while spreading and mixing it on my palette with a palette knife. When I’m actually brushing the paint onto a canvas I don’t really notice as much because of the different tool and surface texture.

Since I recently got some dry pigments from Natural Pigments and one of them was their version of Lemon Ochre I thought I’d make some paint myself and see how it compared. The color between the two was nearly the same, although the Williamsburg paint on the left was slightly lighter. The paint I made had a little bit higher tinting strength but I don’t have enough experience with making my own paint to know what the best proportions of oil to pigment are and I may have used too much pigment. The paint I made had a tiny bit of grit too, but noticeably much less.

Here I made an earthy green with a mix of Lemon Ochre and Graphite Gray. I really like the subtleness of it. A very natural looking green.

Just to show it, here’s what’s left of the white in the corner of my palette. Each time I get another dab of white I thoroughly wipe off my palette knife first so it doesn’t get the white messy.

Overall I really like all of these paints. My collection of black paint still doesn’t quite have all of the black pigments out there, but I’m now a lot closer to it.

Zinc White vs Titanium White in Tints

I posted this on an art forum a little earlier in response to a thread asking if the long held belief that zinc white causing mixes to be “cooler” (i.e. bluer) than when using titanium white is correct.

Titanium white (PW6) is much stronger so it only takes a small amount to have a big effect on the paint you’re tinting with it. It’s more opaque and is often times said by artists (from what I’ve seen especially by those who prefer lead white, such as PW1) to make tints seem “chalky” or dull. In this comparison I’m also trying to see if this is the case.

Zinc white (PW4) is more transparent and much weaker when it comes to lightening other paints in tints, so you’ll need more of it to have the same effect. For this reason it’s sometimes called mixing white, because you can more easily make small adjustments to a paint’s lightness without accidentally making it too light. It’s also said to become hard and brittle if allowed to dry on a canvas by itself and not mixed into anything else.

This was photographed yesterday in late afternoon outdoor sunlight. The image should be mostly self explanatory. I mixed each row to be about the same value (amount of lightness/darkness), and it did take far more of the zinc white to get the same result.

The color samples were averaged in Photoshop over a 31x31px area of each swatch and the results are shown in Lab color space.

As you can see, the “cooling” effect of zinc white was minimal, being only noticeable by a small amount in the middle and dark values of grey. I chose charcoal black because it’s close to being neutral black, so I thought if anything would show a slight blue shift then that would be it.

The blue showed almost no difference between the two in the photo, although in real life I can see a slight difference with the lighter value being a little duller on the titanium side.

The red was easily the biggest difference, with the lighter value being very noticeably duller on the titanium side. Interestingly, the green/red results had almost no difference but blue/yellow did.

It appears people who say titanium white makes chalky tints actually may have a point, as it definitely affected the red.

This photo was taken the next day in about the same lighting as the previous one. I thought there’d be a difference in the greens or yellow, but they’re almost 100% exactly the same. I got most rows to be nearly the same value, except the last couple on blue. This time some samples were averaged from 11×11 squares because small ridges of paint across a few samples that might throw off the accuracy.

The conclusion that I’m coming to includes three things-

1. The idea that zinc white “cools” colors in mixes doesn’t seem entirely true, at least not with this tube.

2. The idea that titanium white produces “chalky” tints also seems to be false, given that most of the tints from both tests are nearly the exact same whether it was made with titanium or zinc. Maybe the people who say it’s chalky are comparing it to a lead white, but I avoid lead paint and can’t do a proper comparison with it.

3. Either of the above conclusions may have specific exceptions, and those alone may be the origin of these ideas. The lightest cadmium vermilion tint was much duller with the titanium than zinc, and the charcoal black did show a small shift toward blue with zinc.

My camera itself and the lighting may also have played a role. Even so, from what I’m seeing here it seems that there is in most cases no more than a minuscule difference between zinc and titanium whites when both mixes have been mixed to the same value, and many cases where there is no visible difference.