the material attributes of paints
Paint can be used more effectively once you understand the material attributes of paint and how these attributes affect its behavior. Ignorance in this area makes it harder to learn, understand and control your color materials.
Paints as material substances can be described on more than 20 different physical characteristics. Many (such as gloss or chalk resistance) don't usefully describe watercolor paints. I've chosen to emphasize the following six pigment attributes:
lightfastness (which depends on the pigment particle size and chemical structure)
refractive index (which affects the paint's hiding power) and transparency) and
Contents of the paint vehicle determine other paint traits, such as how quickly the paint dissolves in water, drying time, viscosity and bronzing.
Even painters who have been formally trained in art school often do not understand the physical basics of their materials and or how to evaluate them, even though it is easy to test paints yourself. You don't need experts, or chemists, or technical documents, to find out whether a paint is well made, high quality, or suitable for your use.
pigment & paint attributes
This section explores the material attributes of pigments that affect the color and handling attributes of paints. Once you understand these attributes you will be able to use paints with more skill and diagnose the cause and remedy for specific painting problems.
First, let's clarify the difference between two types of colorants: dyes and pigments. A dye completely dissolves in water as individual dye molecules, which are typically only 5 to 10 times the size of a water molecule. The water molecules firmly attach themselves to each dye molecule, like a life preserver, which lets the dye molecules swim in the water almost indefinitely. Dyes do not have a particle size, and all dyes are transparent.
In contrast, a pigment does not dissolve in water. The smallest pigment particles are 30 to 200 times larger than a water molecule, and they are not broken down by water into individual molecules. The water molecules do not attach to the pigment surface but simply bump against it, so the pigment jostles among the water molecules and gradually, inevitably sinks to the bottom of the container (or the tube of paint, which causes the separation of vehicle and coarse grained pigment found in unused tubes of watercolor paint). All paints consist of these tiny undissolved pigment particles, which are suspended in the paint vehicle and then, when diluted, in water.
Because of their small size, dyes can either dissolve in, bond with or be absorbed by many porous or organic materials, foods and beverages in particular. When they can't, as usually happens with the cellulose fibers of paper or cotton, the dye must bond to the material through an intermediary chemical called a mordant. In either case, these bonds cannot easily be undone or edited after they have formed, which makes dyes unsuitable for most artistic work.
Instead, artists use pigments, which are either (1) clumps (aggregates or crystals) of insoluble colorant or (2) dye molecules that have been laked bonded chemically to very small particles of a colorless, transparent, chemically stable, insoluble salt that in some cases acts as its own mordant. (The term vat dye refers to dyes that are applied to fabrics presoaked in a vat of mordant solution.) The paint binder (gum arabic in watercolors, acrylic resin in acrylics) then binds these complex but chemically stable particles to the support (paper or canvas).
The substrates used in laked synthetic organic pigments include alumina (aluminum hydrate or transparent white), chalk (calcium carbonate or calcium sulfate), china clay (hydrated aluminum silicate or kaolin) or alum (aluminum potassium sulfate, used to lake pigments since the Roman era). However many other insoluble salts of iron, chromium, tin or other metals are used in laking modern synthetic organic dyes to produce pigments with complex color or increased hiding power (opacity).
Confusingly, the same colorant can be referred to as a dye when discussed as an individual molecule or chemical compound, and as a pigment after it has been laked to a transparent substrate or crystallized into larger particles. All tube and pan watercolor paints, and some liquid watercolors, are made with pigments, not dyes.
Color Appearance. The most salient pigment attribute for painters is its color appearance in artists' media.
Pigment color is fundamentally created by the atomic structure of substances, and in pigments typically arises from the absorption of specific wavelengths of light by electrons oscillating across double chemical bonds (called chromophores) or flowing across the atoms in a dye molecule or metallic crystal.
The material light absorbing and reflecting attributes of a pigment are exactly described by a spectral reflectance curve, and for that reason the guide to watercolor pigments provides the reflectance curve of all major pigments, linked from the spectrum icon .
Using the methods of colorimetry, the reflectance curve can be translated into three visual colormaking attributes that describe our color perception under normal conditions of lighting and display. These are hue (the spectrum name of a color, such as red, orange, yellow, green, blue and violet); lightness (the value of the color in comparison to pure white) and chroma (or saturation, the richness or intensity of the color in comparison to a colorless gray of the same lightness).
These colormaking attributes fit the way we naturally think about colors, and are much simpler to interpret than reflectance curves. However different reflectance curves can produce exactly the same color appearance, and this metamerism means we cannot determine the material reflectance curve of the pigment from its visual color alone.
This is why mixing paints can be so unpredictable. Paint mixtures produce color at the wavelengths where both paints in the mixture reflect some light. But we can't tell, just by looking at the color, which wavelengths a paint absorbs or reflects. So we can't tell, just by looking at the color, that two paints with the same color will mix with other paints in the same way. Usually, they don't. Paints with the same color appearance can produce different color mixtures, if their underlying reflectance curves, their material color, are different. This is a fundamental fact of all subtractive color mixing that I call substance uncertainty.
So the material color of paints appears in their mixtures. This is an important difference in outlook between novice and experienced painters. Novices think about the pure color of the paint as it comes out of the tube, and choose paints that "look pretty". Experienced painters think about the mixture colors that a paint produces with all other paints on the palette, and choose paints that are versatile in mixtures.
The visual color of a paint is an inadequate description of its mixing behavior. Painters who learn "color theory" in terms of the visual colors of paints only learn how to mix hues. As you become familiar with paint mixing, you must study the chroma and lightness of the mixtures whether they are intense or dull, light or dark. This is the intuitive way that a painter comes to understand a paint's material color, as defined by its invisible reflectance curve.
Pigment Concentration. The color appearance of a paint is affected by how thickly the pigment it contains is concentrated in one area, and there is a specific terminology to describe these differences.
Full shade is the pigment appearance in a sample that is infinitely thick, meaning that 100% of the light incident on the pigment actually strikes a pigment particle and is either reflected or absorbed by it. This occurs whenever the color or pattern of any material behind or below the pigment is completely hidden.
Full shade samples are often measured on pure pigment powder that has been compressed into a small tablet or disk. Watercolorists routinely see the full shade color in their pan watercolors or in thick piles of dried tube paint, which appear much darker and duller than the color of the paints on paper. Dull yellows looks green or brown, deep yellow is orange, red violet is purple, and phthalo blue or green can appear black.
full shade color of paints around the hue circle
Painters usually visualize a paint color in terms of its masstone or top tone, which is its color appearance when applied on a pure white surface as a layer that mostly but not completely hides the surface below. In watercolors this typically means any high concentration of paint in water, including in some cases paint applied to paper "full strength" from the tube, without any added water.
However, the undertone or tint is the color appearance of the paint applied on a pure white surface as a highly diluted solution (low concentration of paint in water), or as mixed with a large quantity of titanium oxide (PW6) paint.
The color of a paint changes in three ways between the top tone and undertone: (1) the undertone has a lighter value, (2) the undertone has a lower chroma or less saturated color, and (3) the undertone usually has a slightly different hue (indicated as the hue shift in the paint ratings). However, in many dark valued red, purple, blue or green paints, increasing dilution actually increases the chroma up to a point.
Most artists think in terms of the masstone color, as that is normally how paints are applied to paper in paintings. So the masstone standard is used to describe paint color in the guide to watercolor pigments and the artist's color wheel.
A second aspect of pigment concentration is the pigment load, or how densely the pigment is concentrated in the paint. This is determined as the volume of pigment in proportion to the total paint volume.
In watercolors the proportion of pigment to paint ranges from less than 10% up to around 20% for finely divided, strongly tinting pigments such as the phthalocyanines, red quinacridones or dioxazine violet; from 20% to 30% for prussian blue, carbon blacks, the "raw" (uncalcinated) black and red iron oxides, zinc white, yellow quinacridones, benzimidazolones and most other synthetic organic pigments; 30% to 40% for the yellow iron oxides, viridian, ultramarine blue or ultramarine violet, and the finer grained cobalt pigments (blue, turquoise, green); 40% to 50% for the cadmium yellows, the coarser cobalt pigments (cerulean, violet) and "burnt" (calcinated) red and yellow iron oxides; and 50% or more for cadmium orange, the cadmium reds, manganese violet and manganese blue. These variations are due to differences in the particle size and tinting strength of the pigments, as explained below.
Pigment load is sometimes apparent in the full shade color of the raw paint as it is squeezed from the tube between two paints made with the same pigment, the darker paint contains more pigment. Pigment load is decreased by adding fillers or brighteners to the vehicle, which both dilutes and lightens the pigment. Cobalt or ultramarine blues appear whiter, phthalos or cadmiums appear lighter, and quinacridones or perylenes appear less intense. A tinting test is the easiest way to compare the pigment load of two or more paints made with the same pigment.
Particle Size. Color in turn depends on the several material properties of a pigment. First among these is the pigment particle size, measured both as (1) the median or average particle size, and (2) the range or distribution of particle sizes. Most pigments are manufactured in a range of particle sizes intended for different applications (coloring plastics, ceramics, cements, cosmetics, paints, inks, etc.).
First, here is the basic information you should know. The table gives the average particle size for common pigments used in watercolor paints in microns (micrometers). A micron (abbreviated "µm") is one millionth of a meter, or 0.000036 inches. As reference, a human hair is about 100µm in diameter; a red blood cell, 5µm; a wavelength of light, about 0.5µm; the average virus, about 0.1µm.
Now there are many pigment properties determined by particle size. Of these the most important for artists are the light scattering properties of the pigment, including the pigment lightfastness, color, opacity and tinting strength.
Across different watercolor pigments, smaller particle sizes in general characterize pigments that are:
higher in tinting strength, because the smaller particle sizes produce a greater surface area in the same weight (mass) of pigment, producing more intense color in the same volume of water
more transparent, because the smaller, more intense pigment particles leave more gaps for the paper to show through, and
more staining, because the smaller pigment particles more easily penetrate into the spaces between paper fibers.
Within the same watercolor pigment, smaller particle sizes (down to a limiting size of approximately 0.5µm or the wavelength of light) tend characterize pigments that are:
weakly colored (less saturated and lighter valued), because the increase in surface area produced by the smaller particle sizes increases the total surface scattering from the same quantity of pigment. In most mineral pigments this offsets the increase in tinting strengh.
less lightfast than larger particles of the same pigment, especially in most synthetic organic pigments; in mineral pigments this effect is less pronounced.
In many pigments, smaller particle size also reduces the pigment permanence (resistance to damage from heat, water, acids, alkalis and other agents).
In oils and acrylics, smaller particle sizes make the pigments more transparent and up to a point more saturated, as the particles are entirely embedded in the dried paint vehicle, which reduces light scattering at the particle surface. In watercolors, pigments in smaller particle sizes are more transparent because the same number of particles cover less of the paper surface area, allowing more of the paper (or paint layer underneath) to show through. Even opaque paints can be made more transparent by diluting the color, which creates more visible spaces between the pigment particles applied to paper. (See this discussion of the luminosity myth for the differences between oil and watercolor paint layers.)
All these characteristics are related to the most fundamental pigment attribute affected by particle size: pigment surface area. If we start with a fixed quantity or weight of pigment, then divide this pigment into smaller and smaller particle sizes, the smaller particle sizes create a larger total surface area, as illustrated below with a single cubic crystal of pigment.
effect of particle size on pigment surface area
As the example shows, dividing a single cubic pigment crystal into 27 cubes of only 1/3 width increases the surface area of the pigment by 3 times, and a similar increase occurs if we reduce the particle size by as much again. A crude rule of thumb is that surface area increases X times if the average particle size is reduced by 1/X.
This larger surface area means that the ratio of vehicle to pigment must be greater more finely divided pigments. That is, more water or vehicle is required to "wet" or completely cover all sides of every pigment particle, or conversely that the same weight of pigment mixed in the same weight of water produces a thicker paste as the pigment particles get smaller. This accounts in part for the variations in pigment load across paints made with different kinds of pigment. It also accounts for the tendency of some pigments to sink in water (because of large particle size) and of others to diffuse or spread out in water and to backrun as water dries (through small particle size).
As for water, so also for light: smaller particles present a larger surface area for light to strike, which means that the light scattering effects of the pigment surface, and the pigment eroding effects of light absorption, are also increased.
However, particle size is not a simple attribute. Most pigment particles clump or flocculate into larger chunks called aggregates (clusters of particles) or agglomerates (clumps of aggregates), as shown below. In many cases this clumping can be controlled by manufacturing methods and chemical additives, and the largest clumps can be reduced by the paint manufacturer during the milling (mixing) of pigment and vehicle.
particle size is not a simple attribute
(left) particles of quinacridone pigment clumping into aggregates and agglomerates (manufacturer photo); (right) particle size variation in viridian pigment (from Newman 1997)
Pigment particles also come in a range of sizes, like the rocks, stones, gravel and sand found in mountain stream beds. In the sample of viridian shown above, the largest particles are visible to the naked eye but the smallest particles are too small to see with a microscope. A very large particle size variation is typical of low cost cobalt, manganese and chromium crystal pigments, but all pigments show a range of particle sizes that is described by the particle size distribution.
particle size distribution for typical titanium dioxide
The diagram shows a typical titanium dioxide pigment, which is usually manufactured so that the particle size distribution is as small as practical and so that the largest proportion of the total particles falls within the size range that produces an opaque or hiding pigment layer. Each bar represents the percentage of all particles that fall within a narrow size range; this domed or bell shaped curve is typical of all pigments used in printing and painting media.
Finally, many generic pigments can take on two or more crystal or particle forms depending on how they are manufactured, as shown at right for iron blue (PB27). Because these show a similar variation in the particle distribution, the striking difference in the form of the pigment will have the dominant effect on the light absorbing and light scattering properties of the pigment, which can significantly affect its color.
Lightfastness. Beautiful color is worthless without the permenancy to make it endure. So the third important pigment attribute is lightfastness, the ability of the pigment to remain unchanged under prolonged light exposure.
Lightfastness is fundamentally determined by the molecular structure of the dye or pigment: some molecules break apart, or change form, under the continuous pressure of light, which destroys their color. The effects are different depending on the type of pigment: all pigments that fade completely are organic, while most inorganic pigments either gray or darken. Most completely permanent pigments are inorganic, especially the synthetics, while the fastest fading pigments are organic.
The logic and methods for testing pigment lightfastness are presented on the page your own lightfastness tests. But lightfastness must be mentioned in any discussion of pigment attributes valuable to artists.
Tinting Strength. The tinting strength of a pigment is its colorant power in relation to its mass. As tinting strength goes up, the quantity (weight or volume) of pigment needed to produce a required color intensity goes down. Tinting strength also indicates how much a pigment will dominate the color of a mixture with other pigments, and for this reason some watercolorists incorrectly refer to it as mixing strength.
Tinting strength is defined by a tinting test, which measures the minimum amount of pigment required to impart a perceptible color to a specific amount of clear liquid or white paint, or the relative color intensity of a specific amount of pigment mixed with a specific amount of clear liquid or white paint.
In general, pigments of smaller particle size have a higher tinting strength than coarse pigments, and synthetic organic pigments have a higher tinting strength than mineral pigments. Phthalocyanine blue, for example, has a tinting strength about 40 times greater than ultramarine blue, and twice that of prussian blue.
Refractive Index. Next is what watercolorists call the transparency of the pigment after it has dried on the paper. This means that the paint allows everything behind it whether the paper, ink or charcoal marks on the paper, or an earlier layer of paint to show through completely.
In fact, watercolorists are actually describing the hiding power of a pigment, or the thickness of a paint film required to completely mask a black and white pattern on a painted surface. The transparency of a pigment is degree to which the paint layer disappears when applied over a black surface.
As many watercolorists dislike hiding power and do not paint on black surfaces, I define watercolor transparency in the guide to watercolor pigments as a lack of hiding, and opacity as substantial or complete hiding in masstone applications.
Paint transparency depends primarily on the average size of the pigment particles and the thickness of the paint layer that is, the number of pigment particles covering the paper. Despite common belief, it does not much depend on whether the pigment particles themselves are transparent or semiopaque. The reason is that by the time they dry, most watercolor pigments are opaque, so only spacing will allow the background to show through.
Here's why. Pigment "transparency" or hiding power actually depends on the ratio between the refractive index of the pigment and the medium around it, the RI ratio. Pigments appear transparent when this ratio is close to 1.0, as it is when pigments are enclosed in an acrylic or oil layer. But watercolor paints do not form a paint layer that completely surrounds the pigment particles as oils or acrylics do, but instead strews the almost naked pigment particles across the cellulose fibers and crannies of the paper. Their RI ratio is determined by the refractive index between pigment and surrounding air. As this table shows, nearly all watercolor pigments are opaque in that situation.
The following example shows the appearance of various white pigments with different ratios of refraction when mixed in the same acrylic vehicle, which has a refractive index about the same as the gum arabic in watercolors.
paint hiding power and pigment refractive index
equal quantities by volume of five white pigments in an acrylic vehicle; pigment refractive index in black, ratio of pigment refractive index to vehicle refractive index (1.5) in red; anatase and rutile are different crystal forms of titanium dioxide (PW6) Source: DuPont Coating Chemicals
This image is useful because it shows the the visual effect of the number in red, the RI ratio between pigment and surround. A pigment appears significantly cloudy when the RI ratio is around 1.33, and almost completely opaque when it is above 1.5. But 1.5 is the RI ratio of watercolor pigments in air, which means those pigments will appear somewhat cloudy or opaque, depending on how much gum arabic remains with the pigments on the surface of the paper. Pigments with refractive index of 2.0 or higher will appear cloudy or opaque whether they are enclosed in vehicle or not.
This cloudiness is caused by light scattering at the pigment particle surface: light is reflected back to the viewer without being partially absorbed or "colored" by the pigment itself. This scattered or "white" reflectance dilutes or desaturates the reflected light and causes a drying shift in the pigment color: many watercolors become lighter and less saturated as they dry, especially those with particle sizes below 1.0µm.
two crystal forms of the same prussian blue pigment
from Barbara Berrie (1997)
Typically, the most "transparent" watercolors are made with pigments of very small particle sizes and/or very high tinting strength, which means the pigment particles can be thinly spread on the paper and still provide an intense color. The "opaque" pigments have larger particle sizes and/or duller color, which means the tinting strength of the pigment must be compensated by increasing the pigment load of the paint. In short, "transparency" occurs between and not through the pigment particles. (For more discussion of paint transparency, see the section on the luminosity myth.)
Specific Gravity. All pigments have a characteristic specific gravity, which is simply the weight of the pigment in water that is, the ratio of the weight of the pigment to the weight of the water it displaces in solution. This ratio is constant regardless of particle size, and is the sixth basic attribute of pigments.
The specific gravity of water is 1.0; anything that floats has a specific gravity less than 1. Heavy pigments have a specific gravity much greater than 1, which means they weigh more than the volume of water they displace, and therefore sink in solution.
Watercolorists become aware of specific gravity when they discover that dilute solutions of some pigments in particular the cobalts, chromiums and cadmiums need to be stirred each time the brush is charged with more paint. Many watercolorists categorize these paints as sedimentary, which seems to mean that the pigment is very dense or very heavy, or both.
However, the smallest particles of heavy pigments can remain suspended in solution, like motes of dust in the air, because of the continuous jostling of water molecules. So "sedimentary" is not a fundamental pigment attribute but depends on the combined effects of specific gravity and particle size, as shown in the diagram.
the variety in "sedimentary" paints
If we locate pigments in terms of their approximate particle size and specific gravity, then the domain of pigments divides into two groups:
the mostly synthetic organic pigments (but including iron blue and carbon black) that have both small particle size and low specific gravity. These pigments are close to the weight of water, so they tend to float in suspension longer; they also are very small, which means the movement of water molecules can jostle them afloat longer.
the "sedimentary" pigments that are either very heavy (high specific gravity) or very large. Both attributes reduce the time they can remain in suspension through buoyancy or the jostling effect of molecular impacts. In addition, all these pigments have relatively low tinting strength, which means the pigment load (pigment density) must be increased to produce a satisfactory color concentration.
Note that some generic "earth" pigments (yellow ochre, venetian red) are "sedimentary" while others (raw sienna) are not. Burnt sienna and raw umber in particular can be either "sedimentary" or "transparent" depending on brand; the "sedimentary" paints contain pigments with larger particle size or lower tinting strength.
These differences in combined specific gravity and particle size can be revealed with a sedimentation test.
Staining (Resistance to Lifting). Peculiar to watercolorists is a concern with the staining or resistance to lifting of a paint. However staining is not an inherent property of a pigment but arises through the combined effects of several watercolor pigment, paint and paper attributes:
Pigments with a small average particle size can more easily flow or creep into the tiny spaces between paper fibers, where they "stain" because they are embedded too deeply to swab away; extremely small pigment particles can also display an electrostatic cling that makes them stain plastic palettes and brush hairs.
These small or finely divided pigments are often milled with a dispersant (such as ox gall) which speeds up the mixing of dry pigment and vehicle during milling; this dispersant in turn increases the capillary movement of pigment particles into the paper fibers. Paints made with a high proportion of humectant also tend to penetrate the paper more deeply.
Paints made with a high proportion of gum arabic binder tend to hold the paint on the surface of the paper and to surround both pigment particles and paper fibers in a gummy coating, which makes the pigment easier to remove because it redissolves in water.
A heavier surface and/or internal sizing seals the cellulose fibers from contact with the pigment and closes off more of the spaces between fibers, keeping the pigment particles on the surface.
A smoother paper finish (hot pressed rather than rough) is produced by calendaring the paper, or pressing it between iron rollers; this compresses the spaces between the paper fibers and reduces the channels along which pigment particles can penetrate deep into the paper. But this also makes hot pressed papers less absorbent, so they often receive a lighter coat of surface sizing which allows more staining; and hot pressed papers still have a surface texture that is many times larger than finely divided pigment particles, so staining can occur. The result is that the staining can be aggressive on hot pressed papers, and scraping to remove staining paints can be more obtrusive on the smooth finish.
Whether staining is desired or considered an annoyance, the painter can decrease it in several ways: by using paints with coarser pigment particles or with less dispersant and/or more gum arabic in the vehicle; by adding gum arabic to the dissolved paint; by using papers with a smoother finish; or by using papers with a heavier coating of surface sizing, including papers precoated with a gum arabic wash.
Dispersability. A covert but important pigment attribute is its dispersability, which is simply a measure of how easily the pigment powder can be thoroughly and evenly wetted in the paint vehicle or in water. Unfortunately, I can't give you specifics about pigment dispersability because information is difficult to acquire and because it depends on the methods of pigment manufacture and chemical properties.
There are however two visible consequences of pigment dispersability: the watercolor paint's activity wet in wet and its expected shelf life. To ship the pigment to the paint manufacturer, the pigment manufacturer may add one or several chemicals that prevent the pigment from clumping, forming a skin or settling in solution; these manufacturer additives remain with the pigment all the way to the finished product. To turn pigment into paint, the paint manufacturers must mill the pigment with the vehicle, and to accelerate or extend this wetting process they may add a surfactant or dispersant to the paint mixture, which acts exactly like dishwashing soap to loosen, break apart and dissolve the greasy particles from your dinner plates.
Pigments that are prone to clumping because of their electrostatic cling to one another (phthalos, iron blue), or that are difficult to dissolve because of their tiny particle size (earth pigments, carbon black), end up with much more surfactant in the finished paint. In watercolors, the main culprits are generally the pigments with tiny particle sizes the phthalos, carbon blacks, iron blue, many earth pigments, alizarin crimson, and many of the semitransparent synthetic organics and larger soft pigments, such as ultramarine blue or the cadmiums, which will cake together under milling pressure. Cheaper paint brands try to save money by rushing the milling process, and may add surfactants to almost every color they make.
Just as a drop of soap in a greasy sink causes the grease on the surface to shoot outward, a drop of paint containing surfactant shoots outward from the point of contact with the wet paper. This is not inherently a pigment attribute, but results from manufacturer additives intended to stabilize the pigment as it is transported and ease the milling of the pigment into paint.
basic paint tests
With the pigment attributes in mind, there are a few traditional methods for testing pigment tinting strength, color appearance, particle size and the presence of paint additives. These tests require no special materials or tools, with the exception of a high school microscope.
Draw Down Sample. The tinting test does not give a reliable sense of the appearance of the paint on paper. For that, the draw down sample is effective. Used by printers to test inks, the draw down is a very quick, accurate and reliable way to assess paint color appearance you can make color samples of an entire paint line in 20 or 30 minutes.
Place a small daub of the paint (about the size of a lentil) on a hot pressed (smooth) watercolor paper or heavy (2 ply) white bristol paper placed on a flat surface, and smear it out with the edge of a large palette knife or putty knife, using a quick, downward swiping movement applied with firm pressure. The stroke must thin out the paint to an even, matte finish; dark or glossy patches indicate you should apply more pressure with the knife.
Pigment texture or granularity (but not flocculation) can be evaluated with this method, but unfortunately the paint sample is so thin that it is not a reliable test of pigment transparency all pigments will appear semitransparent. (Note: this palette knife drawdown is separate from the liquid drawdown method used to prepare standardized lightfastness samples, in which a paint solution is smeared over a tilted piece of white filter paper.)
draw down comparison of eight brands of cobalt blue
left to right: Rowney Artists, M. Graham, DaVinci, MaimeriBlu, Holbein, Schmincke, Daniel Smith, Winsor & Newton
The photo shows draw down samples for the same paints shown in the tinting test illustration. Note that it is somewhat easier to see the differences in paint lightness, chroma, and hue across all paints. The Rowney Artists and Winsor & Newton paints appears rather dull, while the DaVinci paint seems artificially brightened. Holbein is again the weakest, and the MaimeriBlu is so dark it is possibly mixed with a darker cobalt pigment, such as cobalt blue deep. The test also reveals the partial separation of pigment and vehicle in the Schmincke paint.
These palette knife draw downs are effective for color comparisons at a standard paint thickness, though they cannot show the range of color appearance from masstone to tints. For that, you need to make paint swatches.
Tinting Test. This is the single most powerful and revealing test you can make of your watercolor paints. It shows the tinting strength of the paint, which is a measure both of the pigment concentration and of the pigment quality in a paint.
In the traditional tinting test, you mix a small amount of the paints you want to test in equal, large quantities of white paint, usually titanium dioxide PW6. This is the only feasible method for oil or acrylic paints, but for watercolorists, mixing in water is quicker and cheaper. I show both methods.
Solution Tinting Test. To prepare a tinting solution, thoroughly dissolve a precisely measured, small quantity of paint in a large volume of water. I use about 1/8th teaspoon of paint, scraped level with a palette knife, in two quarts of water but the right proportions may depend on the intensity of the paints you are testing. If too much paint is dissolved in the water, the water becomes opaque; if too much water is used, differences among the paints will be too small to see.
For light valued yellow or earth pigments, the water can be tinted with a very small quantity of phthalocyanine green; tinting strength is shown by the amount of change from this base color and the hue (bluish or yellowish) of the resulting green mixture stronger paints will have a yellower hue.
All paints should be measured and diluted in exactly the same way in identical clear glass or plastic containers. (Gallon or half gallon juice containers, with the labels removed, work very well.) Then visually assess the color brightness, clarity and intensity of the solutions side by side. It is best to view the samples against a brightly lit white background (a sheet of white paper receiving direct sunlight, or a white sheet hung over a brightly lit window, are ideal). Make sure that one sample does not cast colored light into another.
Paints with higher tinting strength (better quality pigment, higher concentrations of pigment) will produce a darker, more intense color. The best solutions will have a deep, nearly transparent color without any milky cloudiness. (This can be assessed by placing a spoon in the mixture.)
The tinting test is inconvenient to do on all the paints offered by several different brands, but by focusing on a handful of the most expensive pigments a selection from among cadmium red, cadmium orange, cobalt blue, cobalt violet, dioxazine violet, viridian, phthalo green YS, pyrrole orange or red, benzimida orange, quinacridone violet, indanthrone blue, etc. you can get a fair idea of the quality standards of the paint brands.
tinting test solutions of eight brands of cobalt blue
left to right: Rowney Artists, M. Graham, DaVinci, MaimeriBlu, Holbein, Schmincke, Daniel Smith, Winsor & Newton
The photo shows a comparison of cobalt blue across eight common brands of watercolors in a half gallon glass vase of water. Note the variations in color lightness, intensity and the opacity or milkiness of the mixture. Paints with too little color are, almost always, paints that reduce the pigment load to lower manufacturing costs.
For the expensive cobalt paints, an especially strong color may indicate the addition of cheaper pigments (this is usually phthalocyanine blue). Other tests are required to identify these additives but sometimes the hue of the mixture will be different.
Paintout Tinting Test. For the paintout method, mix 10 parts titanium white paint with 1 part of the paint(s) to be tested. The exact proportions do not matter so long as you measure them consistently with clean implements. If you use kitchen measuring spoons, 1-1/4 teaspoons of white and 1/8 teaspoon of paint works out right. (One teaspoon is just under 5 milliliters, so you can get 9 paintouts from 4 15ml. tubes of titanium white.) You can measure in smaller quantities but this makes it harder to keep everything consistent no sense doing the test if you can't trust the results.
Squeeze the paint into the measuring spoon, scrape it off smooth and flat, and make sure there is no excess on the bottom. Then scoop it out of the spoon into a mixing dish with a moist, synthetic bristle 1/2" flat brush. Do the same for the white, and mix with the brush until the color is homogenous and unchanging. Then paint it out on white paper. You want a thick coat of raw paint, so no paper shows through; do not dilute with water. For reference, also paint out a sample of the raw paint.
tinting test paintouts of 12 brands of cadmium red
(top, left to right): Art Spectrum, M. Graham, Daniel Smith, Maimeri, Rowney Artists, DaVinci; (bottom, left to right): Sennelier, Winsor & Newton, Holbein, Utrecht, Rembrandt, Cotman
These tinting paintouts show 10:1 mixtures of titanium white and cadmium red (red medium or red light, depending on brand), with samples of the pure paint color for comparison. This test divides the paints into three groups by CIELAB lightness (italicized number):
the darkest four (Art Spectrum, Rowney Artists, M. Graham, Daniel Smith)
the middle four (Winsor & Newton, Holbein, Maimeri, Utrecht)
the lightest four (Sennelier, Rembrandt, Cotman, DaVinci)
A tinting test will not be accurate if there is a visible separation of vehicle and pigment, as this will make the pigment sample too concentrated; this happened with my tubes of Art Spectrum and Blockx cadmium red.
The digital image does not capture the visible hue differences across paints, which affects the lightness as well. Even so, the test does separate the very strong from the very weak (or "student" grade).
Sedimentation Test. A sedimentation test provides some insight into the average particle size of pigments, as well as the presence of unannounced colorants.
Pigments each have a slightly different specific gravity, which is the ratio of the weight of the pigment to the weight of the water it displaces in solution. All pigments have a specific gravity greater than 1, so they sink rather than float, but differences in their specific gravity cause the pigment particles to sink at noticeably different rates the heavy particles drop to the bottom of the container, while the light particles remain in solution. For pigments of similar or the same specific gravity, differences in the pigment particle size distribution has a significant effect: the largest particles will settle out within a few minutes, while the smallest particles may remain suspended indefinitely.
To make a sedimentation sample, dissolve a leveled 1/4 teaspoon of paint in a drinking glass of water try to use a glass with a flat bottom and straight sides. It is important to stir the solution repeatedly until you are sure all the paint has completely dissolved. Then let the paint samples sit entirely undisturbed for several days. (I set the samples on a north facing windowsill, where they are out of the way but easy to inspect each day.)
Let tinting solutions stand until you can see through the upper half of the glass in most of the samples (which happens in from three to seven days, depending on the particle size and specific gravity of the pigment). Then look for (1) how far down in the solution the pigment particles have settled (which indicates the relative average size of the pigment particles); (2) the distinctness of the boundary between heavy and light pigment particles (which shows the variation in size of the pigment particles); and (3) the amount of cloudiness in the clear part of the solution (which can indicate the presence of other ingredients).
sedimentation comparison of six brands of cadmium red
left to right: control sample (cadmium orange #2), Daniel Smith cadmium red scarlet, DaVinci cadmium red, Holbein cadmium red light, M. Graham cadmium red light, Rowney Artists cadmium red pale, Winsor & Newton cadmium red
The illustration shows sedimentation samples for six cadmium red paints that present a similar scarlet red hue in drawdown samples and tinting solutions. As a control sample (far left), I thoroughly mixed a pure, high quality but relatively coarse cadmium orange #2 pigment (from Kremer Pigments) in an equal quantity of gum arabic solution. After four days, note that this cadmium has settled completely out of solution (white arrow), the boundary between this layer and the clear layer is distinct, and the solution of lighter particles is nearly transparent. That is, this cadmium sample is made of coarse grains of similar size and without additives.
The DaVinci paint has also completely settled out, indicating a comparably coarse pigment; the greater coloration of the solution results from a higher proportion of small pigment particles (uneven particle sizes) and more additives in the paint vehicle. The Daniel Smith and Rowney Artists paints have settled to nearly the same halfway level, with a vague boundary and opaque solution. The Winsor & Newton has settled out slightly farther, and the solution is slightly cloudier, indicating it has a larger average particle size and is either mixed from a light and dark shade of cadmium selenosulfide (darker shades contain more selenium and are often more coarsely ground, making the pigment particles heavier) or contains a lighter filler which clouds the rest of the liquid.
M. Graham and Holbein have settled out of solution much less, indicating that these cadmiums are made using a much smaller particle size. They should be left for a longer period to assess the variations in particle size and the presence of vehicle additives.
In general, within the same type of pigment, smaller particle size should produce a higher tinting strength, but the sedimentation test is not entirely consistent with the tinting test paintouts (above). The ranking here is (1) M. Graham, (2) Holbein, (3) Daniel Smith, (4) Rowney Artists, (5) Winsor & Newton, (6) DaVinci; in the tinting test it was (1) Rowney Artists, (2) M. Graham, (3) Daniel Smith, (4) Winsor & Newton, (5) Holbein, (6) DaVinci.
Combining the two ranks gives: (3) M. Graham, (5) Rowney Artists, (6) Daniel Smith, (7) Holbein, (9) Winsor & Newton, (12) DaVinci.
Be sure to do the sedimentation tests in clear glass containers. In some paints the brighteners or fillers may settle out first, forming a whitish layer on the bottom that is only visible from underneath (example at right).
Dissolving Test. This very revealing test is simple to do.
Set out separate small, white porcelain or plastic containers for each paint to test. Fill each container with 2 to 3 tablespoons of water. Place the containers where they can remain undisturbed for a few days out of sunlight or drafts.
Squeeze into the first container a 2 cm (1 inch) length of paint directly from the tube; let the paint drop into the water without disturbing the container. Watch the paint for a few minutes to observe the immediate reaction. Many paints will start to diffuse pigment into the water, as a thin film on the surface or as clouds of paint through the solution sometimes both. Note your observations, then proceed to the next paint. Once all samples are prepared, cover with foil or a book and let sit for two days. Then uncover and examine again.
dissolving test of single pigment viridian paints
left to right: M. Graham, Winsor & Newton, Rowney Artists, Holbein; (top) after 2 minutes, (bottom) after 2 days
The four samples above were prepared from supposedly single pigment paints, but displayed marked differences across the four brands.
I chose viridian (PG18) because the pigment is a chemically simple, completely inert mineral crystal that has a distinctive color and a high specific gravity. Its behavior in solution should be no different from a very fine sand.
The M. Graham paint shows how a geniunely pure paint should behave. After two days the water has mostly evaporated but the paint has slumped in place.
All the other paints diffused promptly when dropped in water, indicating the presence of dispersants in the paint formulation. The Winsor & Newton and Daler Rowney mixtures had a distinct whitish and cloudy color, suggesting a filler or brightener with a lighter specific gravity predominates in the mixture.
The Holbein did not diffuse nearly as rapidly, indicating that it contained very little dispersant. But the paint did gradually discharge a cloud of very bright, pure color, too bright to be pure viridian. It is phthalocyanine green (PG7), added in small quantity to boost the color of the mineral pigment.
Drying and Rewetting Test. A second test that can help you examine the various ingredients in a paint is the drying and rewetting test. This can reveal large differences across paint brands and important differences across the major pigment families (cadmiums, cobalts, ultramarine, chromiums, phthalocyanines, quinacridones, azos and earth pigments) within a single brand.
These tests are easiest to perform on dark paints. You will need several identical, flat bottomed porcelain, plastic or china containers or mixing cups. China saucers, condiment or soufflé dishes, or your standard mixing cups can all serve the purpose. (A large sheet of glass, laid perfectly flat, also serves well.) You'll also need a large (at least 11"x15"), heavy weight (600GSM or higher) sheet of cold pressed or hot pressed watercolor paper. (Block watercolor paper is not heavy enough and will warp.)
For all paints you want to test, measure out an equal quantity of paint (I use a level 1/4 teaspoon) into a large, flat bottom mixing cup or large mixing well, and dissolve completely in about 2 teaspoons of water. Use an acrylic brush to disperse the paint and to clean out the residue paint in the measuring spoon. Set the paint samples aside on a level surface, cover with cheesecloth to protect them against dust, and allow them to dry out completely depending on heat and humidity, this may take 4 to 10 days. Do not move or stir the paints as they dry.
drying comparison of nine brands of cobalt blue
left to right: (top) Rowney Artists, Holbein, Utrecht; (middle), MaimeriBlu, Daniel Smith, M. Graham, (bottom) Winsor & Newton, Old Holland (cobalt blue deep), Schmincke
This first part of the test gives you some insight into the paint vehicle. The photo shows a drying test applied to nine brands of cobalt blue. There are two obvious differences among the various brands: (1) some show slight to considerable cracking and peeling of the dried paint layer, while the rest (Holbein, Utrecht, Daniel Smith, M. Graham, Old Holland) do not; (2) most show obvious discoloration or separation of the pigment, while some (Holbein, M. Graham, Old Holland) do not.
The cracking is generally caused by a low proportion of humectant glucose or honey to gum arabic in the vehicle. It is also typical of synthetic (glycol) binders, which is likely the case with the Schmincke paint. A glassy surface (Old Holland, Holbein) indicates a relatively high proportion of gum arabic and glycerin to pigment. (We already confirmed this for Holbein in the tinting test.) The discolorations reflect separation of the vehicle ingredients, in particular separation of the dextrin, glycerin and gum arabic.
Next, rewet the dried samples with 1 teaspoon clear water, and thoroughly dissolve the pigment in the saucers by mixing for a minute or two with a wet acrylic brush. As you rewet the samples, you will notice that some paints dissolve evenly and quickly, while other paints tend to break up into coarse particles, which only slowly dissolve. Paints with a relatively high proportion of binder, plasticizer and humectant in the vehicle will redissolve smoothly and easily. However, some pigments, such as viridian (PG18), always redissolve slowly and leave grainy clumps.
Next, premoisten an area 2" to 3" square on the heavy (600GSM or higher) watercolor paper, and pour (do not paint) about half the redissolved paint mixture into the square. (If you dried out the paint samples on a sheet of glass, rewet and pour off the paints one at a time, starting at the bottom edge, wiping up excess paint as you go.) The paper should be resting on a perfectly level surface where it will not be disturbed. Let the paint dry completely, which may take 2 or 3 days.
rewetting comparison of nine brands of cobalt blue
left to right: (top) Rowney Artists, Holbein, Utrecht; (middle), MaimeriBlu, Daniel Smith, M. Graham, (bottom) Winsor & Newton, Old Holland (cobalt blue deep), Schmincke. (Image contrast increased and saturation reduced by 10% to enhance surface variations.)
This last test reveals variations in the "dry" contents of the paint, including pigment, brighteners, and filler. The dried paints have settled out in layers, with the lightest particles on the surface, and the paint layer is much thicker than you can usually obtain with a brush. As a result, the paints will display much larger variations in color. In these samples, the Rowney Artists appears much duller and lighter than the other paints, with a surface like housepaint, indicating significant levels of brightener and/or filler. Several of the paints (Holbein, Maimeriblu, Daniel Smith, Winsor & Newton, Old Holland) show a flocculated or grainy texture that is actually caused by uncoated paper fibers. The Daniel Smith and Maimeriblu paints show extensive bronzing along the lower edges, and the Utrecht paint presents a peculiar splotching. Note the large variations in color and lightness, which are much more noticeable in the original samples.
Finally, rub each sample several times with a cotton swab or a finger wrapped in a paper towel. Some of the paints show some pigment rub off this way (M. Graham especially), while others (Winsor & Newton, Old Holland, Daniel Smith) yield almost none.
The optimal paint will show the least ruboff, discoloration, splotching, bronzing, flaking or cracking across all tests, and will rewet quickly and smoothly. Against those criteria the Holbein, Old Holland and M. Graham paints perform relatively well.
Black Field Test. As watercolor paints are designed to be applied to white paper, and most pigments are rather dark, the "white" brighteners or extenders in paints are typically invisible. But if a dark pigmented paint is applied to a black or very dark gray background, the pigment becomes largely invisible, and any additives are displayed prominently. In effect, you can "x ray" the paint to see its hidden backbone structure.
The substrate you use can be (1) a sheet of smooth, matte, high density black bristol board (to prevent cockling), (2) a sheet of clear acetate which you view over black paper, or (3) a large piece of black acrylic (available from specialty plastics stores or some hardware or hobby stores). The acetate or black acrylic are especially useful as a single piece can be swabbed and repainted many times and it holds absolutely everything on the surface, which produces strong contrasts among the surface deposits and allows you to assess pigment texture by rubbing it with your fingers.
For the most dramatic results, paints you test should contain a single pigment that is (1) relatively expensive, (2) dark valued, and (3) finely divided (small particle size). The quinacridone pigments are ideal as a paint quality test, as these pigments are expensive, dark, and have a very small particle size; the best pigments will also not show the separate presence of a whitish or opaque laking substrate. The similarly dark and expensive perylenes, pyrrols, indanthrone blue, cobalt blue and dioxazine violet may all provide useful tests across brands or within a single brand.
To look for the presence of cost cutting fillers or extenders within a single brand, you need a comparison sample that shows the backbone composition of binder and humectants in pigments less likely to be adulterated. These should be pigments that are finely divided, dark but inexpensive. Iron [prussian] blue, ultramarine blue or carbon black, all relatively cheap, are probably useful for this, but best is phthalo blue GS or phthalo green BS, commonly used in printing inks.
To prepare the acrylic test, wipe down the black acrylic sheet with water and a paper towel. Squeeze out a small quantity of paint (approximately the size of a lentil or green pea), then pour over it one teaspoon of distilled water. (Impurities in tap or spring water may obscure the paint ingredients.) Mix paint and water on the sheet using a clean acrylic brush, then spread the mixture around to produce a large puddle. Repeat for other paints. Let the sheet sit completely undisturbed for one or two days, until all the water has evaporated from all the samples. Samples must be poured on black bristol instead of mixed on it, as this can scuff the surface.
It's important to use diluted paint samples in this test and to let the deposits spread out on the substrate. Do not brush or fuss with the samples once you have mixed or poured them out. Different brands contain different vehicle mixtures, and excessive gum arabic, glycol or honey can obscure the results, especially if concentrated in a small puddle.
black acrylic comparison of nine brands of quinacridone paint
(top, left to right): M. Graham quinacridone rose (PV19), Winsor & Newton quinacridone magenta (PR122), Daniel Smith quinacridone magenta (PR202), Holbein rose violet (PR122), Schmincke purple magenta (PR122); (bottom, left to right): Old Holland magenta (PR122), Rowney Artists quinacridone magenta (PR122), MaimeriBlu verzino violet (PR122), Sennelier quinacridone purple (PR122)
black paper comparison of eleven brands of quinacridone paint
(top, left to right): M. Graham quinacridone rose (PV19), Winsor & Newton quinacridone magenta (PR122), Daniel Smith quinacridone magenta (PR202), Holbein rose violet (PR122), Schmincke purple magenta (PR122); (center): Da Vinci alizarin crimson hue, Da Vinci quinacridone carmine; (bottom, left to right): Old Holland magenta (PR122), Rowney Artists quinacridone magenta (PR122), MaimeriBlu verzino violet (PR122), Sennelier quinacridone purple (PR122)
The illustration shows black field samples for common quinacridone paints on acrylic (top) and heavy black bristol (matt board, bottom). The "white" additives in the Rowney Artists and Sennelier paints are glaringly obvious brilliant when viewed in sunlight. On acrylic, the Sennelier ingredients were grainy, like superfine sugar crystals, and rubbed off when touched. The M. Graham and Da Vinci paints seem to use identical pigments. Winsor & Newton and Holbein paints had some opaque whitening at one side of the dried sample, but not nearly as obvious as in the Rowney Artist and Sennelier paints. The Daniel Smith, Old Holland, Schmincke and MaimeriBlu paints showed the least opaque discoloration and had the darkest dried color, suggesting that very little additives or fillers, and high quality pigments, have been used in these paints.
Microscopic Examination. Even the sedimentation and drying tests are not conclusive for pigments mixed with extenders or brighteners of similar specific gravity or very fine particle size; the movement of water molecules keeps these very fine particles from sinking. Usually a centrifuge is required to separate out fillers and pigments into distinct layers.
The water tinting test will neutralize the effect of brighteners so that the actual pigment coloration can be compared from one brand to another. However, I've found a microscopic analysis is often useful. If you have access to a standard high school microscope or materials inspection microscope (with a magnifying power of about 100x to 500x), use it to inspect the draw down samples of relatively expensive, dark mineral pigments such as cobalt blue or cadmium red, or dark synthetic organic pigments such as perinone orange or pyrrole red. (Microscopic inspection seems to me less reliable with strongly staining pigments such as phthalo green or phthalo blue, as the proportion of gum vehicle to pigment is so large.) Pure pigments in gum vehicle will coat the paper fibers in a smooth, transparent coat, like oil or varnish; the largest pigment particles will appear as dark, intensely colored and barely visible flecks, and the cellulose fibers will appear transparent.
Paints made with large quantities of fillers and/or brighteners will show one or both of two obvious features:
a frothy, opaque, slightly whitish surface texture, strikingly like sprayed stucco or the sugar coating on "frosted" breakfast cereals.
a litter of colorless, semitransparent chunks, which may have the jagged appearance of salt crystals or the worn, rounded appearance of river glass.
Either is a clear indication of additives that are not pigment particles and not gum vehicle, and therefore are something else.
Although these traditional tests can provide you with interesting comparisons of paint brands or paint colors, in my experience making side by side paint wheels with your favorite brand and a comparison brand is often the single best paint test you can use. You're actually handling the paints, mixing them with many other colors from the same manufacturer, and doing this in a highly consistent way. You see the paint quality across many pigments, in repeated mixing and brushing out and mixed to make a complete range of colors; and you can compare the results visually across all hues of the color wheel. If either brand of paint has important flaws, you will notice them.
Keep in mind that the methods described above are not the only way to "test" watercolor paints. You can study your paints as you use them, provided you take the time to look and treat routine tasks as an opportunity to look and discover.
Some of the routine examinations I make as I work:
Each morning the dissolved paint in mixing cups or paint wells has had time to settle out of solution. The quantity of sludge and the clarity of the liquid the next morning tell me about the specific gravity and particle size of the pigment; I can evaluate variations in color by tipping the mixing cup to one side to reveal the sludge at the bottom.
I take a few seconds to look at the residue of a wash mixture that has completely evaporated in a porcelain dish, as this reveals pigment texture quite clearly.
I pay attention to how quickly a pigment seems to settle onto the paper, and how smoothly it can be feathered or graded by gentle brushing.
When using tube paints, I always watch for the clear vehicle that may have separated from coarse or very heavy pigments. When this happens I make a note of the color, darkness, cloudiness, odor and taste (no swallowing!) of the vehicle.
As I paint I note how quickly a paint rinses from the brush and how easily it is displaced by water in diffusion or backruns.
When mixing paints, I pay attention to the relative proportions of each required for the mixture of a specific hue, and from that make a casual judgment of their relative tinting strengths.
One practice that has been useful to me is making test paintings on medium to small sized watercolor blocks, scraps of paper, or failed paintings. The guiding principle is simply to play with the paint rather than direct it, and allow backruns, puddling, diffusion, random color mixtures and other accidents to reveal the effects possible with a paint. Three examples are shown below.
test paintings that explore pigment attributes
These paintings let me explore radiating backruns, color mixtures using diffusion, and pigment effects on backrun edges. In each case I did not set out with a specific goal, but had a general pattern in mind backruns in a wash, bicolor stripes, parallel layers of contrasting texture that I developed and elaborated as the water evaporated. Most watercolors by Gerhard Richter are essentially "play paintings" of this type.
There are very simple design choices that can improve the esthetic quality of these paintings a geometric pattern, or repeated shapes, or symmetry around an axis, or basic color contrasts. But they should not get complicated, otherwise you are focused on the image rather than the image creating pigments.
I also use these paintings as "warm up" or "get in the mood" exercises. I normally clean my tools when I start each morning, so I sometimes make these paintings with excess paint mixtures or the paint left in mixing cups that I want to rinse and reuse. This limits my experiments to paints I am actually using, and adds a recreational benefit to clean up chores.
making paint swatches
After you have tried these basic paint quality tests, it is useful to make paint swatches to show the paint color appearance on watercolor paper and the paint handling attributes with brush and water. I describe here the paint test swatches I used to make the pigment ratings in the guide to watercolor pigments.
My approach puts many paint swatches on a single sheet. In addition to their value as test materials, I keep these finished pages in a portfolio where they provide a reference to the colors and handling attributes of various paint brands.
You may decide to use different testing methods. Before you get rolling, I suggest you try your method with a few paints you already know well staining or nonstaining, granulating or smooth, transparent or opaque. Your method should display clearly the differences among these paints, show the paint color in both masstone and tint, and create a large enough color swatch to be evaluated from the same viewing distance as an actual painting.
Whatever method you use, practice it before starting. Comparisons across different paint samples will not be reliable if you change your procedures as you go along.
a sludge of brightener at the bottom of a sedimentation test
(Lukas "cobalt violet deep")
Prepare test papers. You will first need to create the sheet of watercolor on which to make the test swatches:
Choose the brand, weight and finish of watercolor paper to use in the tests. (If feasible, you should choose the type of paper you use in your actual painting.) Apparently trivial differences among papers can have a significant effect on the appearance of paints. I used 300 GSM Arches cold pressed watercolor paper in a 12" x 16" block (see figure at right).
Prepare a separate sheet for each major color category: red, yellow/orange, green, blue, violet and earth. Placing similar colors on the same page assists visual comparisons across colors and brands. (You probably will not fill the sheet, but the empty spaces can be used to test new paints as you buy them.)
Rule each sheet with pencil guidelines to space the swatches evenly. On a 12x16" sheet, using the swatch layout shown below, I can get twelve 1" wide swatches spaced 3/16" apart, in three rows of swatches 3" high with 5/8" spacing between them, with 1" of margin on all sides.
Draw a fat black indelible line across the paper, or two narrow black lines with a thin space between them, about 1/4" below where the swatch brushstroke will start, with a Sharpie felt pen (or other indelible ink pen). (The black lines are used to judge the paint's opacity.)
The finished swatch, once painted, will look like the example below.
layout for a single paint swatch
Here are the test methods I used to display these paint differences for evaluation:
Buy a small notebook to use to record your observations as you paint the swatches. Some of the things you will notice as you work are: stuck caps, air bubbles or pure vehicle in the paint, differences in paint texture (syrupy, gummy, clayey or hard), paint that dissolves slowly or easily, paint that is smooth or gritty (you will hear this), paint that is difficult to rinse off the brush, paint smells, and paint separation (sedimentation).
Lay out the paints you will test, left to right, in the order they will appear on the paper. Any rearranging you want to do, do it now.
Write above or below each blank swatch area the manufacturer's name, the paint name, and the pigment color index name(s). Do this for all the paints you are going to test at one time.
Squeeze a standard quantity of tube paint into the paint well of a flat plastic palette or onto a flat mixing sheet. I extruded about 5mm of paint from a 7mm wide tube nozzle (the size used on 14ml Winsor & Newton paints) or an equivalent amount (narrower nozzles will require you to extrude a longer bead of paint). Cut the paint by swiping the nozzle flat against the edge of the mixing well or mixing area. Repeat for the other paints.
Important! Look at the label on the tube, and the information written on the test sheet, and make sure that they match. Squeeze out the paints in exactly the same order that they will appear on the sheet.
What to look for: What is the texture of the paint out of the tube? Do you notice any air bubbles or pure vehicle (a pale or dark, thick liquid) squeezes out as well? Does the paint retain its shape or settle into a puddle after it is squeezed out?
Important! If you encounter pure vehicle as you squeeze out the paint, squeeze this mess out onto a paper towel until you get pure paint again. Use this paint for the test sample. Do not test paint that is mixed with pure vehicle.
Important! Don't try to adjust the size of the sample to compensate for apparent differences in pigment concentration or thickness across manufacturers. Try to make all your samples as similar as possible you want to reveal differences in consistency or concentration across paints, not disguise them.
Mix the paint sample thoroughly with fourteen drops (1/4 teaspoon) of pure water roughly the amount that can be carried to the mixing area by a fully charged 1" acrylic flat watercolor brush (Daniel Smith). It's more accurate and faster to squeeze out the drops of pure water from an eyedropper, available in most pharmacies, or to measure with a small kitchen measuring spoon, but I found the brush was a sufficiently accurate measuring tool.
Important! Take care to use a constant amount of water, and to completely dissolve the paint including any raw paint stuck on the brush bristles.
What to look for: What is the texture of the paint smooth or gritty, dense or liquid? Does the paint dissolve easily or only with patient effort? Is the paint mixture thick (lots of gum arabic) or thin (like water)?
Charge the brush with the paint mixture and apply it to the paper in a single stroke, about 6 cm (2-1/4" inches) long. Paint the stroke in a slow, even movement, about two seconds from start to finish. The swatch will vary from masstone at the start of application to a medium tone at the end.
What to look for: What is the texture of the paint thick, watery, granulating, smooth? Did the paint flow smoothly out of the brush? As you reach the end of the swatch, does the paint continue to flow freely, or seem to gum up or dry out at the end? What is the immediate color appearance of the wet paint, and does this change as the swatch dries?
Thoroughly rinse and drain the 1" brush (this gives the end of the swatch time to dry, approximately 15 seconds). With a 1/2" acrylic flat watercolor brush (Daniel Smith), apply a stroke of clear water across the bottom end of the swatch. To ensure a constant amount of water, rinse the brush, hold until it stops dripping, then lightly touch it to the edge of the rinse container.
What to look for: This stroke partly dissolves the paint swatch, as the water gradually diffuses up the swatch, producing a blossom or backrun. Does the paint dissolve immediately, or only after a time? Is the dissolved area limited to the stroke of water, or does it expand up the swatch? Is the backrun or blossom barely visible, or very obvious? (A greater amount of gum arabic in the vehicle, and smaller particle sizes, increase the tendency of the paint to blossom or backrun.)
This is a good point to stop and write down your observations, if any, in your notebook. Then continue:
Rinse the 1/2" brush, and hold until it stops dripping. Apply a second line of clear water horizontally (left to right) at the bottom of the test swatch area, as long as the swatch is wide. Wick off any remaining water with a paper towel, then tap the brush three times vertically in the test mixture to pick up a small quantity of paint. Touch the brush to the left end of water stroke to release paint into the wet area.
What to look for: This stroke is used to judge diffusion wet in wet, and also the color of the pigment in dilution. Does the paint diffuse in the wet area rapidly or slowly? What is the texture of the paint as it diffuses? Does a thin film of paint shoot out across the surface of the water (indicating a dispersant, such as ox gall, has been used in manufacture)? Does the pigment separate into two colors as it diffuses (indicating the paint is a mixture of different pigments, or contains impure pigment)?
Saturate the 1" brush with clear water, and make a small puddle of water on the palette by pressing the brush firmly against it. Pick up the test mixture by daubing it with a 1/8" acrylic flat brush (Daniel Smith), and mix this with the pure water. Use the same brush to apply the diluted mixture as a line between the two test swatches.
What to look for: This stroke is used to judge the hue and chroma of the paint tint. Is the hue or brilliance of the color different from the main swatch? Do you see more or less paint texture?
This completes the application of a test swatch. Go on to the next paint once you have finished. When all swatches are finished, let the sheet thoroughly dry (overnight is best), then proceed with the final two tests:
Lightly rub the large swatch, just below the black lines, five times with a cotton swab (Q-Tip) soaked in clear water. Use moderate pressure and rub in a single direction. Blot the wet area dry with a clean paper towel.
What to look for: This tests the paint's staining and lifting behavior. Does the swab remove a small or a large amount of paint? How much color is left unremoved on the paper? Did swabbing seem to drive the paint further into the paper? Did more paint come off when you blotted the paint with the towel?
Draw another indelible line over the top of the first test swatch, parallel to black test lines already drawn on the paper.
What to look for: This line helps to show how much the paint has masked or changed the black color of the lines on the paper. The degree of difference between the two lines shows the relative opacity of the paint.
Photos of four paint swatches, with an explanation of how they are interpreted, are presented at the bottom of this page.
I found, by making duplicate swatches, that this method displayed the paint attributes quite reliably.
The major sources of error (random variation) will come from variations in your behavior while making the test swatches: in the amount of the test sample squeezed from the tube, in the amount of water used to dilute it, and in the speed and timing of the steps in making the swatches.
To reduce these variations, make several practice swatches to decide on the exact methods you want to use, and get familiar with them. If you can make three swatches from the same tube of paint that appear identical to you, you can assume your methods are sufficiently consistent to be trustworthy.
a page of test swatches
placing similar colors on the same page assists visual comparisons and use as a paint reference