Reusing Aluminium Lithographic Plates
by Nik Semenoff


Aluminium is the metal of choice amongst North American printmakers and commercial offset printers. For waterless lithography, this turns out to be a very lucky decision because of the strong bond between aluminium and silicon molecules. While most art printmakers who use metal, are using ball grained plates because of the necessary texture for drawing and for the amount of water they can hold in traditional gum-etch processes, the grain is not as important for waterless techniques. In fact, much better fine detail is possible from smoother plates, as has been discovered by commercial offset printers. For years I have been using quite smooth commercial plates for my waterless process, doing toner washes and flats with Sumi ink, because of the superior quality of the image -- not to mention the saving in plate costs. Because of the many plates I use in the multicolor editions I print, I have been looking for some way to reuse plates, in the hope to save money. I had found that all plates could be stripped of the silicone coating and reused, but the need to use a stronger silicone stripper, made this technique less desirable. I looked for a way also to reuse the many ball grain plates I had on hand, trying sandblasting and graining with levigators or a woodworkers palm sander. None of these seemed a good approach because of various problems that arose. Finally, after much experimentation and many attempts, I have found a simple and effective method. I have included a simple explanation of the technique in the previously updated paper on my waterless process.

Problems to overcome

Because I have found that the smoother the plate, the better the image in most cases, I looked at the back of plates as a source of metal. This would also overcome the deeper grain that had to be eliminated by working down the surface. When one tries to work with the backside, one soon finds that water is rejected because of dirt, oxides and other things that accumulate on used plates. Trying to rid these with counter etches did little good, and regaining with silicone carbide was very slow, as well as more damaging to the thin plate. Using gentler kitchen abrasives alone, similar to the technique of preparing intaglio plates, was not successful, because of the larger size of litho plates and the contaminates. The main problem was to find a method of removing the debris so that water would take to the surface.

Success at last

Since I am in the practice of using sodium silicate or sodium metasilicate on the aluminium in areas stripped of silicone and redrawn, I had these chemicals on hand. They were used to form a thin layer of aluminium silicate, to which the caulking silicone bonded much better to form the ink rejection layer. These chemicals are quite alkaline, therefore the metasilicate can be found in many industrial cleaning compounds since the use of phosphates has been curtailed. Bondex is one company that has sold it as a TSP substitute and is available at Home Depot outlets here. I first remove all dried ink with hydrocarbons and/or lacquer thinner to make sure the surface would be as clean as possible to accept the water based solutions. I found that by applying a dilute solution of either silicate, which had a pH over 10, the surface of the metal could be quickly moistened very effectively. If I added some scouring powder like Comet or Old Dutch Cleanser, the preparation of the metal was improved. Using a special constructed pad, I could quickly abrade the entire surface and be ready for imaging. All I had to do was to flush off the silicate solution and scouring powder, then dry the plate for use.

Amongst the many used plates I have accumulated, there were some with surface blemishes that got there from chemicals used in processing traditional litho plates. To improve the surface of these plates and maybe get some grain for use with Omnichrom pencils and my toner chalks; I used 3M scouring pads. They come in different grits that are bonded to some sort of felt-like tough Nylon fibers and are very effective for my purpose. Using my special pad as a backup for the 3M material, I could get a satisfactory grain for use with drawing materials. After flooding the back of a plate with the silicate solution, I would simply rub the surface with the pad, trying to keep a random pattern to the grain; or sometimes a grain parallel to one of the edges. Because the grain is not anywhere as pronounced as in ball graining, the pattern doesn't become a problem with any of the drawing materials I have tried. These would be considered very light grained plates by any standards.

Another approach I have tried with success is the use of pumice, silicone carbide or aluminium oxide in place of kitchen scouring powders. These coarser materials are very effective in putting a bit of a grain on the surface, and also in helping to remove the more difficult blemishes on the metal. They also are useful in dislodging silicone, when you decide to reuse the plates again.

Drawing on reused plates

Because of the smoother surface, some drawing techniques are better -- some are worse, depending on the material. I often use mechanical pens (Rapidograph) filled with Sumi ink to get line images or to touch up areas. These work much better on the smoother surface. Omnichrom and other drawing pencils will take to the slightly grained plates, and not show any of the grain direction. There are a number of different water-soluble drawing pencils that work in my waterless process, other than Staedtler Omnichrom. All the other brands have to be heated to make them bond to the metal surface and prevent them being dissolved by the odorless paint thinner in the diluted silicone.

Toner washes on very smooth surfaces are different than on grained ones. They tend to reticulate more on smoother surfaces as the electrical characteristic of the particles comes into play. On coarser grained surfaces, the toner particles descend into the grain and are held until the water evaporates; on smooth surfaces they are freer to move. Toner chalk drawings can be effective if the grain is coarse enough, but this requires that the coarser 3M pad be used.

The technique of transferring toner images drawn on Mylar (see toner paper) gives mixed results on very smooth surfaces; I would suggest abrading the metal with 3M scouring pads to get some grain to hold the toner. Toner chalk drawings on frosted Mylar are good on the more textured surface, but the best images are toner washes on smooth Mylar. Just make sure enough pressure is used to temporarily bond the toner particles to the plate, until heat sets them properly.

One of the more exciting transfer techniques is to use toner chalks on newsprint, just as if you were sketching with charcoal. The loose toner can be smudged or removed with Kneadable erasers to get interesting effects. While not 100% of the toner gets transferred like with Mylar images, it is only those particles caught in the newsprint fibers ; it does not seem to effect the image.

Sumi ink brushwork is better on smooth plates and very effective. Since I use a fair amount of flats in my editioning, this technique is an important part of my imaging. I sometimes outline the areas with Sumi ink with different sizes of mechanical pen, before filling in with a brush. I use water soluble glue that is diluted with water and some Sumi ink, as the black color is only needed for visual considerations.

Using water-soluble block printing inks, I have found that it is possible to transfer images much like in traditional lithography. I have used it by picking up textures from wood and other materials and transferring to the smoother plates. After the ink is dried by heat, it will mask out the silicone coating and later be removed with water to reveal the metal for printing. Pick an ink that has a long enough open time to allow the transfer, or add glycerin, glycol or other retarders to give you the needed interval.

I have found another perfect material for imaging and transfers – molasses.  This sugar based product dries to a silicone resistant film quite fast, but the speed can be controlled by the retarders one uses.  One of the more interesting techniques is based on the monoprinting method of rolling out ink on a slab and drawing on a piece of paper placed on the slab. The monoprint image is the ink transferred to the paper by the pencil pressure, producing some interesting textures, depending on the paper and ink consistency.  You can manipulate the molasses “ink” consistency to get the image you want.

The basic molasses ink is made up of common blackstrap molasses. While honey and syrups work, I prefer the character of molasses.  I start with one part each of molasses, Golden Acrylic Retarder, DOT 3 brake fluid and dry pigment Prussian blue.  Do not use carbon black if you want to use this formula as an image protector in the event of needing to add more silicone to a plate – the carbon has an affinity to silicone and hard to remove.  If you want to thicker ink for rolling out, you can add more pigment, but I suggest magnesium carbonate or calcium carbonate as it is cheaper and Prussian blue will make it easily visible. Depending on how fast you are in drawing, maybe you will need to add more of either retarding materials, or glycerin from the drug store.

After you do your drawing, place it face down unto a clean plate and pass it through the press. The molasses should leave a good film on the surface that can be dried with a heat gun.  Additions can be made to the plate before silicone is applied.  This material can also be used like a stamp pad by  selecting interesting shapes to press to the inked slab first, then unto to the plate.  Some of my students have diluted it with water for drawing with a brush.

I have discovered a way to make my own positive waterless plates that relies on the smoother surfaces of the back of plates. This is discussed in another paper.

Siliconing the plate

With a smoother surface, the silicone layer does not have to be as thick. This means a more dilute solution can be used, which also allows the plate to cure more quickly. Because there is a thin layer of aluminum silicate to which the silicone rubber bonds, the plates are just as effective as the others. If you want to double coat the plate, special precautions may have to be taken. Since any hydrocarbon will affect the silicone coating, even the gentle odorless thinner can strip off your first layer on the smoother plates, especially if the second coat goes on too soon after. To prevent this, I do not use foam rubber pads because of their tendency to act like an eraser and grab at the silicone. Instead I use a folded facial tissue, which is gentler on the previous layer. I also buff with less pressure. Once the silicone has cured, it holds up just as well as on ball grained plates.

Reclaiming the plate again and again

The purpose of all this is to make use of the plate as many times as possible. Because of the very thin silicone layer and the aluminium silicate underneath, it is easy to reclaim the plates, as the layer seems quite susceptible to removal with weak acids. I use a counter etch made up of phosphoric acid, with a very small component of hydrofluoric acid. About four ounces phosphoric acids are added to a gallon of water. To this is added about half a teaspoon or less of ammonium bifluoride crystals, to help destroy the silicone layer. This should result in a solution with a pH of around 1.5, depending on the concentration of the acids you use. Even weaker solutions have worked for me and I would recommend you could dilute this formulation even more. I have used a number of aluminium cleaning solutions in place of pure phosphoric acid as these contain the acid as their main ingredient. The very small amount of hydrofluoric acid in solution should not be a concern, as long as gloves and other standard precautions are taken. I like to add some red food coloring to this solution to indicate the danger of the chemicals being used.

While it is cheaper to use technical grade acids instead of special commercial products, they usually cause beading marks as the silicone rejects the waterbased stripping solution. Some commercial products contain surfactants that prevent this and give cleaner looking metal surfaces. Try various products meant for aluminium and even cedar siding as these contain phosphoric acid and the surfactant.

I have found on the market some effective products to strip of silicone. When I first started to do the research I found a number of products sold in automotive supplies shops, these chemicals are meant to be used as a cleaner for magnesium wheels, and contain a small amount of hydrofluoric acid in the form of ammonium bifluoride. These have disappeared from the shelves of stores and I looked elsewhere.  Craftspeople working with glass use a solution of hydrofluoric acid to etch designs on their products. The one I have run into consistently is “Armour Etch”, available in craft stores and from glass workers. Use only as little as possible since this can be a dangerous material to handle.

Because of the concern in using any solution with fluoride, I have asked our chemistry department to analysis the commercial cleaning products and the resulting diluted one made from these. While the original products contained less than 1% HF, the diluted stripping solution tested at .36%. This would depend somewhat on the particular product you use, as the content of HF varied from .9% to undetectable amount.  While this information was dealing with the aluminium wheel cleaning materials, the same ratio of HF in any stripping solution will recycle the plate. 

Because of concern over this use of HF in any form, the manufacturers of automotive products have been reducing or eliminating it from their wheel cleaners. For individual printmakers this convenient method of stripping silicone may become harder to acquire. If ammonium bifluoride in crystal form is available to them  at teaching institutions, I would recommend that you make up your own stripper because of the high cost of buying the commercial cleaners. I have found that it takes an extremely small amount of the crystals to make a useful stripping solution, as long as the silicone coating is thin, according to the best practice of waterless lithography.

While the analysis shows there is very little HF in the working solution, it does not mean that one should be careless when using it. Protective gloves and eye shields should be used to prevent any of the solution contacting the skin. If some does splash on your bare skin, quickly remove it with plenty of water and application of calcium chloride to neutralize any HF remaining. Calcium chloride can be purchased as crystals used to dehumidify wet areas, or as a replacement for common salt to keep driveways free of ice. Make up a solution and keep on hand just in case. The other acids you may have in the solution are just as dangerous, so use the silicone stripping material with care.

After the ink is removed, I generally use a solvent or a bit of lacquer thinner to get rid of any ink that resists removal. This prevents the previous image becoming lightly etched into the metal because of the resist factor of the remaining ink. With ink removed shortly after printing the edition, the image should be clean. The silicone stripper etch is poured on the plate and quickly spread to even out the etching action. Using one of the special pads made just for this part of the process, the chemicals are moved around as you see the silicone disappear and the entire surface takes water. If you want a more abrasive attack on the surface, I would suggest you use aluminium oxide instead of the other common abrasives. The bifluoride will be depleted if you use abrasive containing silica, like pumice or silicone carbide; and the acids will be neutralized if kitchen scouring powders are used, as these commonly contain calcium carbonate. I have found the use of aluminium oxide not necessary if other processes are done properly.

Once the surface is taking the stripping solution all over, this indicates that the silicone layer is removed. Flush the surface with plain water and then squeegee off the excess. By applying sodium metasilicate in dilute form, a new layer of aluminium silicate is produced for the next image, as well as neutralizing any acid left in the pores of the metal. If there seems to be some water rejection, use Comet with the silicate solution or a green 3M scouring pad to abrade the surface of the plate. Flush off with water while running the palm of your hand over the surface to feel any slipperiness as this may leave dry crystal and prevent silicone from sticking. Squeegee off the water and dry; you have recycled your plate for reuse.

Simple Wash/Deck

I have recently found a perfect product to make a good silicone stripper. In Canada there is a cedar deck restorer sold under the name of "Simple Wash" or Simple Deck". Made up of phosphoric acid, it also contains a surfactant that prevents beading of the solution on the rejecting silicone layer and produces a new surface without blemishes. It can be diluted by a factor of 4-5 and still be effective. To this diluted solution, add about a teaspoon of ammonium bifluoride crystals to a gallon of liquid. You will have to experiment on the amount of “Armour Etch” that is needed to effectively remove silicone.

Using other aluminium

At the University of Saskatchewan, we have for years been using common industrial 16-gauge aluminium for intaglio, which we buy in 24 x 32-inch size. I have experimented with these plates for waterless and have had excellent results. The first time you use one of the sides, use the same process as for the back of old litho plates. After the oxide etc., is removed, the surface is just as easy to work as the litho plates. Because of the thicker metal, there are some adjustments that have to be made for most techniques. First, the greater mass of metal makes it more difficult to set any toner images with heat, using a paint-stripping gun. While it is possible, it takes more time and there is danger of permanently damaging the plate through warpage if one is not careful. I would suggest using the wick flow method on heavier metal sheets as it does just a good a job of bonding as heat. In case your toner is tending towards an A type material, then modify the white gas with a little acetone to get the proper adhesion (see my toner paper).

The other problem that arises is in rolling up the plate. It is necessary to roll completely off the surface of a plate with any waterless process, so this means the thickness becomes a minor problem. By beveling the edges to 45 degrees, like in intaglio, the edge is not a likely to damage your roller. You could buy thinner gauges of aluminium, which would reduce this problem.

I would suggest that the scraper bars used for small images on thick plates be as wide as the plate. This would prevent stretching of the metal in the overused portion in the center, which will "bowl" the plate in time. If most of your images are to be about the same size, then cut your metal to appropriate dimensions.

Using thicker plates such as this can save students a great deal of money over the length of their studies, as the plates could be resurfaced enough times to complete all their yearly assignments. The initial cost of the industrial aluminium is less than for ball grained plates, making the investment a good one. Even printmakers could take advantage of using common industrial aluminum sheeting from metal suppliers. I have not tried any other metal as the price and characteristics of aluminium are perfect for this waterless process. If surplus Dow metal for making photo-engravings can be obtained cheaply from scrap metal suppliers, this harder metal could be an interesting choice. While it contains some magnesium in the alloy, this metal should take to silicone as well as the aluminium.

Where to get the materials

Hydrochloric acid can be bought in most hardware stores or swimming pool suppliers as muriatic acid. The phosphoric acid is the main component of aluminium cleaners on the market, but make sure, as some contain enzymes, which are useless in this operation. Ammonium bifluoride is the active ingredient in such compounds as Armour Etch and Jack Frost, a glass frosting chemical used by glass workers. In this form it is very expensive, so I would suggest you obtain it in pure crystal form from a chemical supplier. A 500 gram bottle will last you a very long time in the concentrations I recommend.

Sodium silicate is available from pottery suppliers as silicate of soda. It is sometimes referred to as waterglass, which is an impure mixture of sodium and potassium silicates. Sodium metasilicate is available from garage suppliers as it is used to remove grease and oil from the garage floor because of its strong emulsifying action. I have also seen it in hardware stores to remove oil stains from driveways. Home Depot sells it as a TSP substitute in a product make by Bondex.

A stripping method using solvents

When using the smooth backs of plates, I have found a reasonable method of removing the silicone without the need of acids. While this procedure requires a hydrocarbon, turpentine or citrus thinner, it can remove enough silicone to allow a grit containing 3M pad to produce a reusable plate.

If you are reclaiming a number of plates, then I would moisten the surface of two plates and stack them face to face so that the solvent can start softening the silicone. After a while, take a piece of steel wool and start scrubbing the surface. The silicone should peel off and produce a polished finish that you can see. Never let the solvent dry as the silicone can be rebonded to the metal. Wash the surface with a strong detergent to get rid of the water rejection, then flush with water. Apply sodium silicate or metasilicate with a pad and a green 3M pad to get the all important layer of aluminium silicate on the surface.

This method takes more care as any deposit of silicone left on the plate can be a problem. It is up to the printmaker to decide which of the two methods that works best for them. Both require the use of undesirable materials, but a well ventilated area makes solvent use acceptable.

Making the special pads

The best material for working the surface of plates is the tough carpeting fabrics available these days. I have used both looped and fuzzy types, and think both work well. They can be obtained as scrap from your local carpet store, as only a couple of square feet are needed. I suggest you use pieces of scrap 2 x 4 lumber for a backing to the carpet. Use common contact adhesive to bond the carpet to a piece about 5 inches long. Trim the excess fabric and pound the edges with a hammer to make sure there is good contact between the two elements. Make a number of these, as you should not use only one for all the operations. This will neutralize any chemical from the previous chemical application for which the pad was used. With a felt marker, indicate for which chemicals that pad is reserved. Remember that the sodium silicates are alkaline and the silicone stripper is acidic.


The price of plates can be a major component in small editions; any way this can be reduced means more editions can be printed by money-strapped printmakers. While in traditional lithography, the regraining of a metal plate requires special reciprocating tables, in waterless lithography I have not found the need for a coarse tooth on the surface. In fact, the smoother the plate, the better the detail. Commercial offset printers have found this to be true as they discovered better ways to control the fountain solution needed for ink rejection. Fine art printers have been accustomed to using more water to reject the heavier ink deposits they required for textured art papers and the paper to plate contact printing method. Today, printmakers can use pin registration, which allows them to apply more than one pass of ink to the paper, if sufficient ink is not deposited the first time. Offset proofing presses are common enough, which allow many passes over the paper to build up the ink density. I have not found these smoother plates required heavier inking or multiple passes through the press.

The time it takes to resurface a plate is much less than it takes to grain even a very small stone -- and there is no expense for graining abrasives. Plates can be reused a great number of times, until other damage makes printing or imaging difficult. Since we all go through the trouble to recycle aluminium soft drink and beer containers, just consider how many cans could be manufactured from one .012 of an inch thick litho plate. Since it is so easy to do, recycling printing plates should be a matter of course for most printmakers, as quality of editions should not suffer from the practice. With the exception of ammonium bifluoride crystals, all chemicals are easily obtainable and not very dangerous to use. Even the bifluoride is not dangerous in the concentrations needed to strip a properly applied silicone layer. While it would be admirable to have no dangerous materials in any printing technique, it is not always possible; but cautious and knowledgeable use of such materials is more important than their complete elimination.

Printmaking is an expensive pursuit and bringing down plate costs should bring it more in line with ones ability to produce more complicated editions. I hope this method will help printmakers produce these better editions as they worry less about putting on another color or texture.

This information is copyrighted material, but I urge printmakers to copy the paper to hand to others interested in the method. Publication in any commercial or educational journal is prohibited without permission only because I may have up to date changes and the only reason I reserve the copyrights.

Nik Semenoff
Department of Art and Art History
University of Saskatchewan
Saskatoon, SK. S7N 5A4

First published January 1999, updated March 2003