technique

For more information email: rellesh@yahoo.com

    Watercolor paintings with surprising designs, color arrays, and surface textures can be achieved using a new saltwater technique (see the test images on this page and the paintings in the galleries).

Brief Description of the Method
(a) Dissolve a salt in water.
(b) Add watercolor pigment to the solution.
(c) Paint a paper surface with the mixture.
(d) Allow it to dry.
(e) Coat with a clear shellac.
    With certain choices and amounts of materials, painted areas can be obtained, for example, that consist of darker, pigment-rich "spots" distributed throughout a lighter-pigmented field. The "spots" actually turn out to be crystalline salt particles that are well cemented to the paper, thus giving it a uniquely granulated, rough-feeling surface. With other mixtures, especially those containing more than one pigment, surprising and visually exciting results are obtained. These are described in more detail below.

Detailed Description of the Method
Salts and Solutions:
    The grade of common salt (sodium chloride) used was “Coarse Kosher Salt” from Morton International; solutions of this grade were clear. This was preferred over Morton’s “Iodized Salt” which has 3 major impurities; solutions of this grade were cloudy. (Crystal growth is normally better when fewer impurities are present.) The sodium chloride solutions that were used were about 90-95% saturated. (Results were similar, although less pronounced, when lower concentrations were used.) Typically, solutions were prepared from about 1.3 oz. of sodium chloride and 4.0 oz. of water. The copper sulfate used was a commercial grade. Similarly, the most often used copper sulfate solution was about 90-95% saturated. It was prepared from about 0.8 oz. of copper sulfate and 4.0 oz. of water.
    The watercolor pigments used were obtained from several manufacturers: Winsor-Newton, DaVinci, Rembrandt, and Holbein. Most often, the pigments were squeezed out of their tubes and allowed to dry (“harden-up”) over time in air at ambient temperatures on plastic palettes. Wetted brushes were then used to pick up some of the pigment and mix it with salt solutions.
    Salt areas are coated with a clear shellac such as that produced by William Zinsser & Co., Inc., Somerset, NJ. It is available as a water/alcohol solution so, technically, it is another water medium. (See "Other Observations" below for more information about the coating.)

Test Example 1:
    Using a wetted #4-Round watercolor brush, some pigment was combined with about 0.1 oz. of an almost saturated solution (see above) of sodium chloride in water. The amount of pigment (or the number of pigment-wetted brushfulls) used varied with the intensity of the color desired on the final painted surface. The resulting mixture was then thickly coated (puddled) onto test squares (1.5” x 1.5”) on Strathmore 140-lb. cold-press watercolor paper and allowed to dry slowly at about 75% relative humidity. Table 1 lists pigments used, whether they were transparent or opaque, and the result obtained for each. Some representative digital scans (about 1.5x actual size) are also included below.
    Each transparent pigment example gave the deepest (highest pigmented) color at the exact randomly dispersed spots where the cemented sodium chloride crystals appeared and a lighter colored (lower pigmented) overall background field upon which the crystals had grown.
    On the other hand, each opaque pigment example resulted in a test square having relatively uniform overall color with randomly dispersed, cemented sodium chloride crystals that were either the same color, lighter, or almost colorless.

Table 1
Pigment #	Description	T/O	Mfr	Results
1	Permanent Red Deep	T	R	See note 1
2	Permanent Red	T	DV	See note 1
3	Scarlet Lake	T	WN	See note 1
4	Permanent Alizarin Crimson	T	WN	See note 1
5	Quinacridone Red	T	WN	See note 1
6	Permanent Rose	T	WN	See note 1
7	Quinacridone Magenta	T	WN	See note 1
8	Perylene Maroon	T	WN	See note 1
9	Thioindigo Violet	T	WN	See note 1
10	Winsor Violet	T	WN	See note 1
11	Perm. Magenta (Quinacridone)	O	DV	See note 2
12	Winsor Blue (Green Shade)	T	WN	See note 1
13	Winsor Blue (Red Shade)	T	WN	See note 1
14	Winsor Green (Blue Shade)	O	WN	See note 2
15	Winsor Green (Yellow Shade)	T	WN	See note 1
16	Winsor Yellow Deep	T	WN	See note 1
17	Benzimida Orange	T	DV	See note 1
18	Winsor Lemon	T	WN	See note 1
19	Lemon Yellow	O	WN	See note 2
20	Acrylide Yellow	T	DV	See note 1

Note 1: Deeply colored crystals and lighter colored background.
Note 2: Relatively uniform color throughout with the same or lighter colored crystals.
T=Transparent; O=Opaque. The assignments come mostly from the manufacturers designations
or from Hilary Page’s book: “Guide to Watercolor Paints”, Watson-Guptill Publications, New York, 1996.
Manufacturer (Mfr): DV=DaVinci; WN=Winsor-Newton; R=Rembrandt.

Pigment 4 and sodium chloride	Pigment 7 and sodium chloride	Pigment 9 and sodium chloride
Pigment 12 and sodium chloride	Pigment 13 and sodium chloride	Pigment 16 and sodium chloride
Pigment 11 and sodium chloride	Pigment 14 and sodium chloride	Pigment 19 and sodium chloride

Test Example 2:

    Using a wetted #4-Round watercolor brush, some pigment was combined with about 0.1 oz. of an almost saturated solution of copper sulfate in water. The amount of pigment (or the number of pigment-wetted brushfulls) used varied with the intensity of the color desired on the final painted surface. The resulting mixture was then thickly coated (puddled) onto test squares (1.5” x 1.5”) on Strathmore 140-lb. cold-press watercolor paper and allowed to dry slowly at about 75% relative humidity. Table 2 lists the pigments used. The corresponding digital scans (about 2x actual size) of each test square are also included below.
    Each transparent pigment combination resulted in deeply colored, cemented crystals of copper sulfate. The inherent light blue color of the copper sulfate crystals themselves was clearly overwhelmed by each pigment color. Also, as in the case of the sodium chloride examples, a much lighter colored overall background field surrounded the crystals.
    The designs obtained (see scans) are a consequence of how much the wet paper buckled and how the aqueous mixtures flowed during drying.

Table 2
Pigment #	Description	T/O	Mfr
4	Permanent Alizarin Crimson	T	WN
7	Quinacridone Magenta	T	WN
9	Thioindigo Violet	T	WN
12	Winsor Blue (Green Shade)	T	WN

T=Transparent; O=Opaque. The assignments come mostly from the manufacturers designations or from Hilary Page’s book: “Guide to Watercolor Paints”, Watson-Guptill Publications, New York, 1996.
WN=Winsor-Newton

Pigment 4 and copper sulfate	Pigment 7 and copper sulfate
Pigment 9 and copper sulfate	Pigment 12 and copper sulfate

Test Example 3:
Using a wetted #4-Round watercolor brush, some of each of two pigments, one opaque (in this example, a nickel titanate-based Lemon Yellow was used) and one transparent, was combined with about 0.1 oz. of an almost saturated solution of sodium chloride in water. The amount of each pigment (or the number of pigment-wetted brushfulls) used varied with the intensity and shade of the color desired on the final painted surface. The resulting mixture was then thickly coated (puddled) onto test squares (1.5” x 1.5”) on Strathmore 140-lb. cold-press watercolor paper and allowed to dry slowly at about 75% relative humidity. Table 3 lists the pigments used. A few representative digital scans (about 1.5x actual size) are also included below.
Each example resulted in cemented crystals of sodium chloride that were quite deeply-colored with the transparent pigment’s color and a yellow-colored overall background field, from the opaque pigment #19, upon which the crystals had grown. This separation of color upon drying is remarkable. It also occurs in numerous other cases when a transparent and an opaque pigment are combined.

Table 3
Pigment #	Description (All Transparent)	Mfr	Opaque Pigment*
1	Permanent Red Deep	R	19
2	Permanent Red	DV	19
3	Scarlet Lake	WN	19
4	Permanent Alizarin Crimson	WN	19
5	Quinacridone Red	WN	19
6	Permanent Rose	WN	19
7	Quinacridone Magenta	WN	19
8	Perylene Maroon	WN	19
9	Thioindigo Violet	WN	19
10	Winsor Violet	WN	19
12	Winsor Blue (Green Shade)	WN	19
13	Winsor Blue (Red Shade)	WN	19
15	Winsor Green (Yellow Shade)	WN	19
16	Winsor Yellow Deep	WN	19
17	Benzimida Orange	DV	19
21	Carmine	H	19
22	Alizarin Crimson	WN	19
23	Vermillion	R	19
24	Permanent Sap Green	WN	19
25	Payne's Gray	WN	19

* Lemon Yellow (Winsor-Newton; nickel titanate based).
DV=DaVinci; WN=Winsor-Newton; R=Rembrandt; H=Holbein.

Pigments 1 and 19 plus sodium chloride	Pigments 5 and 19 plus sodium chloride	Pigments 8 and 19 plus sodium chloride
Pigments 10 and 19 plus sodium chloride	Pigments 12 and 19 plus sodium chloride	Pigments 25 and 19 plus sodium chloride

Discussion

Background:
    Traditionally, watercolorists have achieved an interesting effect by sprinkling salt (sodium chloride) granules on wet areas of watercolor pigment on paper. The effect is described in numerous “How-To” watercolor books and in U. S. Patent #5,362,518: “Method for Watercolor Painting Using Rock Salt”, Rodney J. Johnson, Nov. 8, 1994. Visually, “star-like” light to white (the original paper surface) areas begin to appear and grow from the spot where the individual salt granules lay on the wet surface. This is the result of small amounts of salt dissolving from each crystal creating a salt concentration gradient that forces the hydrophilic pigment molecules away from the crystal toward lower salt concentrations. (With many other types of organic molecules, this phenomenon is often referred to as the “salting-out effect.”) The ultimate size of the light to white areas depends on the wetness of the paper and the rate of drying. Eventually, when the surface is dry, there is almost no adhesion of these salt granules to the paper and they are easily brushed away.

Summary:
    A new method has been developed which produces novel designs, color arrays, and textures in watercolor paintings. The process involves first completely dissolving a salt (e.g.: sodium chloride, copper sulfate, etc.) in water, combining the resulting solution with watercolor pigments, painting a watercolor paper surface, allowing it to dry, and then protecting it with a coating of clear shellac.
    In the case of mostly transparent (i.e.: mostly soluble in the aqueous salt solution) pigments, the resulting solutions were spread on watercolor paper and allowed to dry. As the water evaporated, crystals of the salt began to grow from the paper surface. At the same time, the pigment color tended to concentrate at each crystal. When all of the water had evaporated, the painted area showed dark-colored, pigment-rich crystalline spots distributed throughout a relatively light-colored, pigment-poorer field. Furthermore, the crystalline particles were well cemented to the paper resulting in a unique granulated, rough-feeling textured surface.
    With mostly opaque (i.e.: well-dispersed but mostly insoluble in the aqueous salt solution) pigments, the resulting solutions/suspensions were spread on watercolor paper and allowed to dry. As the water evaporated, crystals of the salt, as above, began to grow on the paper surface but, in these cases, little differential distributions of the pigments occurred. The resulting painted areas showed relatively uniform color plus salt crystals; the latter were either about the same color as the rest of the painted area, lighter, or, in some cases, almost colorless. Once again, these crystalline particles were well cemented to the paper thus giving it a unique granulated, rough-feeling textured surface.
    With mixtures of more than one pigment, unexpected and striking results were obtained.   For example, when a highly transparent (mostly soluble in salt solution) and a highly opaque (mostly insoluble in salt solution) pigment were used together and the resulting mixture was spread on watercolor paper and allowed to dry, the transparent pigment tended to concentrate at the growing salt crystals while the opaque pigment produced a continuous background field of color.
    In all cases, the size and shape of the final crystals depended on the material used and on several other parameters, namely drying rates, impurity levels, presence of seed crystals, pigments, initial concentration, how undisturbed the surfaces were during drying, etc. With sodium chloride, most of the crystals were cubic and smaller than 1 mm in largest dimension, although some of up to about 3 mm have been observed. With copper sulfate, the crystals appeared to be elongated, pointed shapes in the 1-20 mm range. Other salts gave other shapes and size ranges. In general, the height that the crystals grew out above the paper surface ranged from about 0.1 mm to about 1 mm, thus producing a painted area with a distinct textural feeling.

Other Observations:
    All paintings produced by this overall technique are matted and framed the same way as any other watercolor renderings.
    As mentioned at the outset, salt areas are coated with a clear shellac that is typically available in a water/alcohol solution and, technically, is just another water medium. The reason for the coating is two-fold: (1) to minimize moisture absorption by the salt and (2) to protect it from accidental abrasion (even though it is already fairly well cemented to the paper through this method). About 20 different coatings were tried before settling on the shellac. First, it worked about the best in its salt-protecting capacity, probably because its chemical makeup allows it to bond most effectively with the polar salt particle surfaces. Second, being available in a water/alcohol mixture means that the solvent evaporates quickly and leaves no odor. Most of the other coating materials that worked almost as well: polyurethanes, acrylic lacquers, varnishes, etc., are sold in petroleum distillate mixtures that are comparatively quite noxious during coating and leave long-lasting objectionable odors. A few latex mixtures were also tried but these caused significant dissolving and running of the previously dried watercolor paints.
    Many other salts were explored. They included ammonium chloride, calcium acetate, calcium chloride, potassium chloride, potassium acid phthalate, potassium sulfate, potassium citrate, sodium bromide, sodium sulfate, sodium citrate, sodium benzoate, strontium chloride, sodium carbonate, aluminum sulfate, and magnesium sulfate. Initial results with these salts were less interesting than those with sodium chloride and copper sulfate.
    Pigments that watercolorists have traditionally called “transparent” were almost always found to be quite soluble in water and saltwater. Pigments that have traditionally been designated as “opaque” were almost always found to be quite insoluble (although usually well-dispersed) in water or saltwater.   Pigment formulations known traditionally as “semi-transparent” or “semi-opaque” were found to have soluble as well as insoluble components in water or saltwater. That is, part settled out of a solution upon standing while the solution itself remained quite rich in color.
    In general, the more soluble the pigment was in the saltwater, the larger the differential distribution of pigment was between the crystals and their surrounding areas.
    The largest crystals in the dried painted areas were generally obtained by (a) using the purest possible crystalline materials, (b) using the highest possible aqueous concentration of the crystalline materials, and (c) using a relatively thick layer (a puddle) of the dissolved crystalline material/water/pigment on the painted surface. It has also been noted that the size of the crystals can be affected by the particular choice of the watercolor paper used.
    If one allows the pigment/saltwater mixture to stand too long on a palette, some water will evaporate and some salt crystals will begin to form. If you then paint the mixture on paper, these preformed crystals will act as seeds that will initiate the growth of many crystals, all of which will be tiny, instead of the fewer larger ones normally produced.
    One should try not to move the painted paper until it is dry. Such movement will also tend to create too many seeds that will initiate the growth of too many tiny crystals.
    I hope that anyone interested in trying this method will enjoy it as much as I have. I would be most interested in hearing and seeing what others can achieve using it. Email address: rellesh@yahoo.com

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Random Technical Musings:
    It is possible that the above-described cemented crystals began their growth from the lowest level to which the solutions had penetrated into the paper, namely from several layers deep into the cellulose fibers. During this early growth, it is also possible that either strong ion-molecule coordination was established between the crystalline component (for example: sodium chloride) and the numerous hydroxyl groups of the cellulose chains/fibers, or that some of the chains/fibers actually became encapsulated by the crystals as they began to grew up and out of the paper. In any event, something special must have occurred to explain why the salt crystals end up so well cemented to the dried paper.
    The tendency of transparent (mostly soluble) pigments to concentrate at the salt crystals is perhaps more complex. During the drying, the last areas with visible wetness were minute “puddles” surrounding each of the growing crystals. These puddles were much darker (more pigment-rich) than their surroundings and, of course, saturated in salt. The puzzle then is why is the transparent (soluble) organic pigment drawn there since, as mentioned in the Background section, salt tends to drive organic pigments away when placed on a film of pigment and water.
    The tendency of opaque (mostly insoluble) pigments to not concentrate at growing crystals is probably because, as insoluble particles, they simply settle out on the paper early and then never move.
    Again, I'd welcome your communications about any aspect of this method. Email address: rellesh@yahoo.com