Disappearing highlights and light brushmarks

My teacher long ago told me to put highlights on thick so they'll last. So final highlights I tend to put on thick.
But not all light marks are highlights. What I think I've found in several paintings is that if make any lighter marks too thin, they actually seem to disappear or fade back towards the darker pigments in the mix. It's as though the oil or the other pigments overtake them as the paint oxidises. Or something. It can be within days or weeks. And I don't use medium any more - this is just all straight from the tube.
Some light marks are not the final, flourishing highlights that I want to be impasto. These are just light hitting a shirt or something. Recently I put blue, almost white highlights on a shirt - now they are just light blue. I remember painting detailed foam between waves and noticing weeks later most of it became almost invisible. Maybe it's lighting, I don't know. I've never heard anyone talk about this.
Has anyone else had this experience? If so, what's going on? I use titanium white.


  • Abstraction

    Oil paint from the tube is about 50% oil and 50% pigment. As the oil polymerises it shrinks, pulling the pigment particles closer together. Small area highlights will be impacted by this effect to a greater degree than the large dark areas.

  • Interesting. So you probably need to check again in a while and touch up those areas.
  • Abstraction

    Yep, even better lay out some titanium white on absorbent paper or board a day or two in advance o treating the highlights. The putty like consistency will have a tendency to lower shrinkage.

    Bulking up the titanium white with an impasto gel or even calcium carbonate (chalk) could do the trick.

  • What would Mark Carder do?
  • @Abstraction – I think it may be useful to think of your entire painting sandwich: substrate, seal, primer, more (alkyd?) primer, tone (mid-value low saturation), paint, sanding down problem areas, more paint, etc.  Please forgive me if I don’t remember the sequence correctly. I suspect the problem, if this is a new problem for you, is located in the grounds and the darker area you are trying to make lighter.

    Since your painting is less than a year old, I think it is not in the oil volume shrinking phase.

    Oil paint polymerizes by first adding and then discarding oxygen molecules. So when it is adding oxygen, it gains in weight and gets dimensionally bigger. When it loses oxygen, it weighs less and gets smaller. The different types of oil (linseed, walnut, etc.) all have well known and documented polymerization changes over time. These changes have been visually graphed as curves. Under normal temperature, humidity, and pressure situations, such as in an indoor studio, these follow a stable pattern.

    Most of the oxygen uptake is done in the first few weeks; the release of the oxygen takes centuries. Here is a good explanation of the polymerization curves and a graph of the different oils. Note that after 60 days, the linseed oil volume is still about 13 - 14% more than when the oil was applied. If I recall correctly, it takes years or decades to approach 110% of the application volume.  

    Weighing In on the Drying of Oils | Just Paint

    @Dencal made a useful suggestion about letting some of the oil leech out of the paint before applying. Monet did this to an extreme. I think adding calcium carbonate will not help much because it is a weak white and does not cover well.

    I don’t tone my surface. I like to paint on a white surface so my lighter areas don’t fade to a darker value. If I have painted a darker area, then change my mind and paint over it with lighter paint, I got to put on a thicker layer or paint a few more layers to cover the darker area.

    If you add to your paint an alkyd with a drier in it, such as Liquin, this really will change the polymerization curve. I find that it makes paint touch dry very quickly – overnight. Since the paint surface is dry to touch, when I put on another layer, this helps keep the new paint from sinking in and becoming duller. In your case, I suspect this would help. You may wish to try applying just white over the area, let it dry or set up, and then, if the white is light enough for you, apply the white-other color mixture on top.

    Please let us know what works out, as we can all learn from each other.

    @kingstonfineart: I don’t know what Mark Carder would do. What do you think he would do? 

  • edited June 21
    Who knows what Mark Carder would do? We have no way of asking him.  I wonder what @KingstonFineArt would do.  Other than offering  a vacuous, platitudinous rhetorical question, I mean.
  • First of all I might watch Mark's videos. 
  • I think Mark wouldn't have his processes so rigidly defined. I think he would paint in a more unified manner. 
  • edited June 21
    Oh, do you, now? Well, I've watched all of Mark's  videos and there's nothing in them that addresses @Abstraction' s particular problem. And what is this about not having the "process so rigidly defined", and  "painting in a more unified manner"? Are you suggesting that @Abstraction's processes are rigid and that he paints in an ununified manner, whatever that means? You need to explain this.  If you have an answer to a problem please give it to us up front instead of acting like some bigshot, know-it-all guru who likes to impress people but who, in reality, knows no more than the rest of us. What, in straightforward terms, is the answer to Abstraction's problem? It's no use resorting to glib rhetorical questions when you don't have a sensible answer. That impresses no one.
  • CBGCBG -
    edited June 21

    You ponder "Maybe it's the lighting"

    Well, perhaps it is.  I suggest you do a little controlling of variables and some testing.  Is it a level of lighting or location of lighting having changed issue?  Is it a physical change in gloss level or depth/shape of impasto "ridges" of the paint that is changing?  or really is it an actual change in pigment color ... and if so what is causing it, e.g. does it change the same amount when painted next to a dark color rather than directly on it?  Does the darkness or lightness of the color underneath change how much it changes?

    Good luck! 
  • @Desertsky It's not confined to this painting. I noticed it in both my seascapes also. When I think about it, i've noticed it mostly when I've painted white or white mix (eg light grey) over BLUE. And different blues, although French ultramarine is my main blue. But it's occurred with prussian and manganese also. So I've painted waves and then painting in the lacing foam between waves until it looks right... then it dissipates. I'll experiment with @dencal Denis's suggestion although do want sufficient oil to create that bond if it's thinly applied. Reading between the lines of your helpful summary I think that because I block in average areas of value and come back later for more detailed highlights it may mean as the oil begins to oxidise in the first few days it may be inherent.
    @KingstonFineArt I've watched many of Mark's videos and haven't seen a mention of this - although may have missed it. The approach to painting I was taught is perhaps quite unified in the sense that I was taught to bring the entire painting along together. Cover the entire canvas quickly with major tonal statements then paint the next biggest difference. Like Mark's style, it also includes protecting your draks. But it assumes you use the layers of paint to adjust, which oils are mostly very good at. I vary my approach all the time. Experiment. Learn.
    @CBG - I don't have a studio or dedicated space. I've got horrid yellow downlights in the room, a table in the way behind me when I try to stand back, daylight spilling both sides and some 5000k lighting and have to pack up because it's a dining area.
  • @Abstraction - My studio is half of the laundry room/pantry. I fantasize that, in the centuries to come, doctoral theses will be written about my paintings and household eras, based on the kinds of dryer fuzz embedded in my paintings: blue bath towel era, brown kitchen towel era, etc. 

    If you leech out the oil from a blob of paint, as long as you subsequently don't add any solvent to it, you should be fine. 
  • I don’t use Geneva paint, but from hundreds of reviews and comments it seems to be fine paint. I don’t know what pigments compose the Geneva burnt umber, but if it is the traditional PBr6, PBr7, or PBr8, it has a lot of manganese in it: up to 25%, depending on the compound. If this paint works for you, I think any manganese compound paint will as well.

    I don’t like using PBr6, PBr7, or PBr8, because they sink in quickly and continue to cause dulling of any color over it. It could be I lack the skill to use these well. In any case, the other manganese compound I use, PB33 manganese blue, does not cause this sinking in.

    The Color of Art Pigment Database: Pigment Brown - PBr (artiscreation.com)

    Paint chemistry is a bottomless hole of exploration. 

  • Abstraction: Could you post some before and after photos so we can get more of an idea what's happening?
  • edited June 23
    Richard_P said:
    Abstraction: Could you post some before and after photos so we can get more of an idea what's happening?
    Yes I realised I need to do that when I wrote this but I only have time to paint on weekends if at all. It's never been consistent enough to suspect it's going to happen and take a photo. It's like... hang on! That doesn't look like it did when I painted it.
    Suez said:
    It’s disconcerting how major technical problems of this sort are popping up here by painters who are using materials and paints other than the Geneva paints Mark Carder recommends for his course. I haven’t seen these problems pop up by anyone dedicated to using his materials and methods.
     People like me who arrived to this forum via Mark’s YouTube videos have no idea, (without searching the forum for individuals stating such problems previous posts), that people having these problems are not following his teachings or using his paints - or his alternate DMP medium for those who can’t purchase his paints in their countries.
    Alert: I don't use Mark's method. I trained in Australian tonal painting which has a long history, is very similar and gets similar results. I have a growing grasp of the technical aspects of painting and there doesn't seem to be anything technically wrong with the method I was taught. Mark's teaching is excellent and I have learnt a lot from him on a range of issues.
    * I use artist grade quality paints.
    * I also paint darks first and highlights later. This problem has nothing to do with that. I consider you should be able to paint white or light opaque paint over an existing medium to light value without it changing.
    Suez said:
    @Abstraction, manganese is very reactive. I’m surprised it’s put into oil paints. It may fall under the new zinc taboo eventually. Just my hunch. I don’t trust it.
    I'm an evidence kind of person. So based on your comment I checked a number of forums such as MITRA (conservators), Painting Best Practices and don't see any problems apparent. In commenting on manganese in earth colours...
    "There is no correlation between darkening of driers and manganese compounds in pigments causing darkening in oil paint." (George O'Hanlon)
    Do you have evidence that the manganese compounds used in PB33 are very reactive? Pure manganese may be very reactive, I don't know - but obviously this isn't pure manganese. Sodium and chlorine are highly reactive, but not in the form we sprinkle them together on our food. :) 
    Manganese blue was made by the BASF company from the 1930s until 1986 (or according to another source the 1990s). The formula is proprietary and is distinguished from barium manganese oxide (also known as 'barium manganate'), which is also listed as PB33. It is rumored that a few artists paint manufacturers purchased the remaining available inventory of the pigment and have been making it for some time. However, this seems unlikely since BASF stopped making the pigment in 1986 or 1990s. The reason BASF stopped producing it was because although the pigment is not considered to be toxic, the process of manufacturing was toxic to workers, and the pigment does not have particularly high tinting strength." George O'Hanlon, Painting Best Practices 23 May 2018.
    But artists everywhere prize it and hunt for the remaining tubes and someone on this site kindly sent me some. It's a lovely colour.
  • Just as a quick test could you try applying some linseed/walnut oil over the whites (once they are dry) and see if you notice a change? Then removing..

    In acrylics a matte surface on lights can make them appear duller than if they were glossy so I'm just seeing if this is the cause here. :)
  • @Suez – the pigment PB33 is not the same compound as you have found on the BASF website.  As far as I know, it is not being made currently on a commercial basis.

    @Abstraction – back to disappearing whites. If I paint with a slow drying paint, and if I paint white (PW06) over a darker value, whether its blue, gray, or brown, the white will fade (become a darker value) a little. If I paint over a dry surface, this doesn’t happen as much. If I use Liquin in my white paint, this doesn’t happen as much. If I paint over a dry surface, AND use Liquin, this happens the least. If I really want a very light white as a highlight, I paint on a white surface and/or use several coats of white paint. This has always happened, and I just adjusted without thinking about it too much. I paint thin, and I think if I put on a thicker layer, it wouldn’t happen.

    For the decades I used Weber’s permalba white (titanium and zinc oxide white, in an alkyd base with a drier added), my white paint did not fade, get absorbed into the underlying paint, or turn color or value. Its freaking great paint. Except for the zinc oxide :smile: I suspect the alkyd stabilizes the zinc oxide, but this is pure conjecture on my part..  

    I am glad you wrote about this, as it made me re-examine my processes. I view this as just a characteristic of oil paint and other materials, and not as a problem or a failure of the materials or process.

  • Painting white highlights with acrylic is even worse as the water evaporates and the whites become more transparent and duller.
  • Folks

    Try using a working fixative on pastel highlights. All the whites turn transparent.

  • edited June 26
    I have noticed some whites are prone to darkening, maybe that’s what’s throwing you off? Julie Beck calls it voodoo darkening and Nicolas Uribe ran into it when using MH quick dry titanium, so it seems to be a titanium + driers problem, maybe. Since companies never disclose driers it’s impossible to tell. I also had a white go slightly transparent on me as it dried, I think it was Gamblin 1980 titanium, it was the same effect as when one loads up with marble dust.
  • edited June 26
    @Abstraction, just a couple of thoughts I had on your problem.

    You said you use Titanium white which is opaque. Is there a possibility that you may have inadvertently used Zinc white which is only semi-opaque? The reason I ask is because last year, whilst doing  one of my snow paintings,  I noticed that I was running out of white on my pallet and I mistakenly laid out some Zinc white from an old tube I should have tossed out ages ago.   I couldn't work out why my white was handling strangely and not giving me good coverage. When I realized what the problem was I had to scrape that section down and redo it with Titanium. 

    Also, I've noticed that even with Titanium, what may have looked like good coverage and opacity when applied, dries to a less opaque state, especially if I've thinned it down with oil to get a smooth finish, like in a pale sky where I don't want impasto brushwork. This happens especially if I'm panting a light sky over dark stained canvas. I'm about to start a new painting with large areas of pale, almost white, sky which is almost flat and I know that I'm going to have to be careful not to thin the paint down, which will reduce covering power, and yet spread it thinly enough to avoid impasto ridges in the paint which I don't want in that section. 

    It's unlikely that you made the same mistake as me and used Zinc white , but it may be that you are thinning Titanium, or mixes made with it, too much.

    Painting is a complicated business!  :)
  • Actually I researched my Art Spectrum Titanium White on the weekend. I noticed they have two Titanium Whites and one is pure titanium, but I've realised the one I have contains zinc. Yes, I'm aware of zinc risks. So I wondered if that's contributing. I've emailed them to ask for the percentage of zinc - haven't heard back so far. Might try a phone call when I'm back in Melbourne.
  • @Abstraction, as well as the archival problems with Zinc white, it is semi-transparent, so I suspect that is at least part of the problem.
  • dencaldencal -
    edited June 27
    Art Spectrum https://artspectrum.com.au/products/artists-oil/titanium-white-series-1/

    Titanium White No 1 
    PW6, Titanium Dioxide. Ground in sunflower oil

    Titanium White No 2
    PW6, Titanium Dioxide. PW4, Zinc Oxide ground in sunflower oil.

    Can’t find any info on quantity of zinc.
     But any zinc is too much zinc.

    Back in 2019, we reported on how many of the oil paintings at the Georgia O'Keeffe Museum in Santa Fe, New Mexico, had been developing tiny, pin-sized blisters, almost like acne, for decades. Conservationists and scholars initially assumed the blemishes were grains of sand trapped in the paint. But then the protrusions grew, spread, and started flaking off, leading to mounting concern. Some paintings have more pronounced protrusions than others, but even when the conservators restored the most damaged canvases, the pimpling (or "art acne") returned.
    paint films consist of slowly evolving heterogeneities at micro- and nanoscales.
    Chemists concluded that the blisters are actually metal carboxylate soaps, the result of a chemical reaction between metal ions in the lead and zinc pigments and fatty acids in the binding medium used in the paint. The soaps start to clump together to form the blisters and migrate through the paint film. "They can form exudates on the surface, which obscure the painting itself, creating an insoluble film or an effect of transparency, so you can look through those layers, which was not the intention of the artist," Marc Walton of Northwestern University told Ars in 2019.


  • @dencal Yes, I'm quite familiar with the background to zinc problems. This effect of zinc soaps even with very old paintings is hit and miss and I've not seen that anyone has fully solved why. I'm also not a fan of sunflower oil either. I inherited this large tube and have so many paints I don't now use I've been reluctant to discard it. May need to.
    @Suez I don't have a problem with getting quality paints. Art Spectrum paints are artist quality and Langridge are professional quality oil paints both made in Melbourne. And most other high quality brands are available also.
  • I've read about the problem many times. The paint is "thinning" and getting more translucent over time, and this seems to be natural. I think that one of mine was affected, noticeable about a year later, but this is not normal to happen quickly. It was painted on a reused panel directly with no overpainting to get plain white background. It looks duller now with less contrast if I compare with a (good quality) photo I took when I was done. But typically it must take decades if not longer.

    The only reliable solution it seems is using more, thicker, paint. Then what is not clear to me is what is affecting the degree of that effect, what is accelerating it. And most interesting, what classic painters, who did thin glazes, were doing to prevent that.
  • edited June 28
     @Suez, I couldn't get Geneva paints in OZ when I first started painting. I tried a few other brands but landed on Langridge professional oil colours. I've been using them for years.  Their Titanium white (PW6) does not contain zinc. It is just titanium dioxide in safflower oil. They use safflower oil because it doesn't yellow over time like linseed. See here:  http://langridgecolours.com/titanium-white-oil-colour/ It is high quality paint as  @Abstraction said.
  • outremer said:
    I've read about the problem many times. The paint is "thinning" and getting more translucent over time, and this seems to be natural. I think that one of mine was affected, noticeable about a year later, but this is not normal to happen quickly. It was painted on a reused panel directly with no overpainting to get plain white background. It looks duller now with less contrast if I compare with a (good quality) photo I took when I was done. But typically it must take decades if not longer.

    The only reliable solution it seems is using more, thicker, paint. Then what is not clear to me is what is affecting the degree of that effect, what is accelerating it. And most interesting, what classic painters, who did thin glazes, were doing to prevent that.
    I think this is closest to what I suspect and aligns with what I was taught originally. When I put final highlights I go thick because of this factor. At the same time I want to understand whether I may be experiencing more than one factor at work.
    The lacing of foam on a seascape took, I suspect, more than a month to reduce - possibly longer. This appears to be that factor. The more immediate - within 1-3 days - appears to be something else. All my painting for 30 years from same tube of white I think. I'm away so haven't had chance to test.
  • Suez said:
    @dencal, Sorry I missed your info before posting.

    I have read that similar pin hole blisters have show up on a few of Rembrandt’s paintings. 
    Rembrandt used lead white. Do you know the cause of the blisters?
  • Suez said:

    Mark Carder say’s titanium in linseed oil is preferable to safflower oil for several reasons. Of course we can do whatever we want.

    "Cautionary notes come with the use of any safflower-based paint. As good of a job as it does warding off yellowing, it does produce a more fragile film, and in the process of drying it shrinks in mass and loses density." https://justpaint.org/williamsburgs-new-safflower-colors/

    It's preferable because it forms a stronger bond than safflower oil. On the other hand Golden Paints have done tests that clearly show less yellowing with safflower, which is more relevant for whites and blues. If we wanted the best of both worlds we could use a titanium in linseed for the body of the painting and switch to smaller tube in safflower for highlights and final layers on lighter colours likely to be affected by yellowing. If I replace my current tube I'd be tempted to go larger tube in linseed and smaller tube in safflower. Not sure I will go that far. Too many paint tubes already. :/
  • edited June 28
    The Williamsburg page states that it is safe to use [safflower oil] "in alla prima techniques where the painting is developed in a single sitting as one continuous wet-in-wet layer, without any of the structural complications that come with the more traditional approach of layering." (That is, with glazing, for example) The DMP method is alla prima and because Geneva paint (and others with which a slow dry medium is used) take a long time to dry, you don't work wet over dry even if the painting takes several sittings to complete. 

    The Williamsburg data indicates that the bond formed with safflower, poppy and walnut oils is not as strong as that formed with linseed oil. I am not in a position to question their data. But how strong is strong enough? I use walnut oil and my surfaces when dry are very strong. You would need a Swiss knife or chisel to prise paint off the canvas.

    Another point to be made concerning white made with safflower oil is that it would be rarely used alone but rather mixed with other linseed based oil colours.

    Given these considerations I think those of us who paint alla prima using the DMP method are safe with a Titanium white made with safflower oil and can rest easy in the knowledge that our paintings will be durable and that our clouds will not turn yellow as the painting ages. 
  • I think this is closest to what I suspect and aligns with what I was taught originally. When I put final highlights I go thick because of this factor. At the same time I want to understand whether I may be experiencing more than one factor at work.
    The lacing of foam on a seascape took, I suspect, more than a month to reduce - possibly longer. This appears to be that factor. The more immediate - within 1-3 days - appears to be something else. All my painting for 30 years from same tube of white I think. I'm away so haven't had chance to test.
    It's not only highlights. In his "Deer haven" Courbet moved one deer and while it was allright for him at the time, we can now see that. And it is not clear highlights, more midtones.

    That's the deer that is a bit to the right of the center. You see a ghost of it, dark outline to the left.

    But well, it's been 200 years...

  • edited June 28
    That is a mystery to me, too, @Suez. I wouldn't use anything with zinc in it.  :)
  • edited June 28
    Thanks, @Suez. I think I'll stick with my Langridge Titanium white. There is no problem with it when used alla prima (See above) and I have paintings that are over 6 years old and there is no sign of cracking or any other problem with them. The surface is tough and durable. That's what I see with my own eyes. 

    I was interested in what was said in the discussion above between yourself and @Desertsky about manganese. I use Manganese violet. I have not noticed any sinking in when it dries. You said "My concern is not about manganese darkening oils. I was thinking about how both manganese and cobalt are themselves driers and how interactions between manganese and titanium are the stuff of batteries, etc."

    I'm no chemist and so I'm wondering what it is about manganese that bothers you. I mix it with Titanium white. My palette hasn't exploded. I had a look here:  The Color of Art Pigment Database: Pigment Violet - PV (artiscreation.com) and it seems safe as long as you don't eat it or snort it. 
  • edited June 28
    Yes, Burnt Umber does have a tendency to dry matte which changes the colour but once you oil out or varnish the colour returns. Still, the colour shift can be disconcerting so I try not to use it. You can make something close to burnt umber with other colours that don't dry flat like Burnt Umber.
  • @Suez

    You wrote: “deferring to experts rather than heeding things we are seeing with our own eyes. At least in my book.” And “just instinct on observing it’s weirdnesses.”

    I have a different perspective. I do not “defer” to experts. I use them as a valuable resource for my own benefit.

    When someone posts here, those posts will be read by an unknown number of viewers, both now and in the future. My goals for my own posts: I do not want to lead anyone astray, even accidentally, so always admit the limitations of what I know, differentiate between my observations and what the experts state, and give links to my sources.

    Oil painting surface pin-holes and acne: The causes of metal soap formation and the problems they cause are known. 

    Shining an infrared light on how “metal soaps” threaten priceless oil paintings | Ars Technica

    Paintings suffer from breakouts, too – but what is ‘art acne’? | Apollo Magazine (apollo-magazine.com)

    Here is a very useful source of information about oil painting materials.

    Resources (udel.edu)

    There are different causes of a light value oil paint becoming more transparent or darkening slightly over time.

    Why do you state: “I also wonder why Landridge sells it’s titanium with zinc to the UK and keeps it’s good stuff for Oz.”  What is your source of information?

  • edited June 28
    It's probably a good idea to question everything. But where the methods used are sound and the resulting data are clear, unequivocal and public,  I go with the scientific experts. Science is our only source of real knowledge. Science is our way of asking the universe questions and getting reliable answers back. Personal experience is better than nothing but it can be misleading as it is uncontrolled and often does not take into account all possible relevant variables. When it comes to the safety, behaviour and archivality of art materials, I can't see any reliable alternative to going with the science. Of course, science is never finished and there will always be more research to do, but it is the only sure way to figure out how and why materials behave as they do in given circumstances.
  • Do whatever works for you, @Suez:)
  • Suez said:
    I  also wonder why Landridge sells it’s titanium with zinc to the UK and keeps it’s good stuff for Oz.
    Long ago it used to have zinc and they removed it. I know because I emailed and the he rang me and talked for 30 minutes. Fascinating to hear him talk about his paints. There were advantages in mixing some zinc with titanium until the zinc potential problems became more known. They had already removed it but forgotten to update their website. So... I think you'll find that in UK, unless they have very old stock, which I doubt, they simply haven't updated their website. Call the store and ask them to check the tubes.
  • @Suez - I checked the Jackson UK website yesterday about the Landridge white oil paints, and could find no indication of zinc oxide being added to the titanium white oil paint. That was the reason I asked you for your source. The information at Jackson's matched exactly that on the Landridgecolours.com website. 
  • It does mention zinc in this particular link. So looks like they didn't update them all. It says PW6 but mentions zinc by name.

  • @Abstraction - you are right! I missed this yesterday.
  • @Suez That's really interesting, thank you - and I'm doing a Caravaggio inspired painting also. I love his thoughtful approach. I won't use egg tempera, though or grind lead! I have lead white and almost never use it.
Sign In or Register to comment.