US2362884A - Process of making aluminum coated paper - Google Patents

Description

Patented Nov. 14, 1944 PROCESS OF MAKIKC,

A; ALUMINUM COATED PER John W. lark, Westbrook. Maine, assignor to S. D. Warren Company, Boston, Mass., a corporation of Massachusetts No Drawing. Application April 9, 1941,

Serial No. saw-1s 1 Claim. 117-41) This invention relates to the provision of aluminum-coated web or sheet material, e. g., re-

generated cellulose fllms, resinous films, and fibrous webs such as paper or textile webs. In particular it pertains to aluminum-coated'sheet material in which the aluminum coating is deposited from an aqueous medium. The finished product has properties of glossiness and reflection of visible and/or infra-red radiation not possessed by previous aluminum-coated sheet material of the same general class.

Aluminum powder, both from the standpoint of original color and especially color after aging, has long been recognized as superior to powders of zinc, tin or other white metal (exclusive ofthe expensive noble metals) for use in making metallic coated or so-called silver papers. A large volume of fine quality aluminum-coated paper, having a mat or dull surface, is made annually. Generally speaking this production falls into two classes, i. e., so-called lacquer-metallics and caseinmetallics. The term "lacquermetallics is used to designate paper which is coated with a suspension of metal powder suspended in a cellulosic or resinous lacquer or varnish. Such coatings dry smooth, with a good sheen, but with low specular reflection. Lacquer metallics have usually been considered to be the best grade of aluminum-coated papers on the market. A cheaper grade of paper, the socalled casein metallics has been made by coating a paper base with an aqueous suspension of aluminum powder and a water-dispersible adhesive, such as casein. Such coatings usually do not dry smooth and so have to be treated, as by calendering, to remove cockle from the sheet. Such papers have been usually considered somewhat inferior to lacquer-metallic coated paper in appearance.

In both lacquer-metallic and casein-metallic coatings the metal powder and the binder are so mixed that the layer of coating is of similar composition throughout. It may be that leafing may take place to some extent-that is, that there may be a tendency for some of the metal flakes to rise to the top of the coating before the coating sets, so that there is a slightly higher concentration of metal at the surface than be neath the surface-but in any case the difference is slight, and-for practical purposes the composition may be considered to be uniform throughout...

Some aluminum-coated papers of both the lacquer-metallic type and the casein-metallic type fairly Well approximate the appearance of dull or unpolished' aluminum foil. Prior to the present invention, however, no such aluminumcoated paper of either type has, so far as I know, even remotely approached polished aluminum foil in appearance. In fact the belief has been quite general among paper-coaters that it was impossible to make an aluminum-coated paper of either the lacquer-metallic or casein-metallic type which would resemble polished foil in appearance.

The desirability of making an aluminumcoated paper resembling polished foil in appearance has long been apparent. Various methods of producing such paper have been suggested. One more or less successful method comprises spreading over a paper base a resinous composition and, while the resinous coat is tacky, dusting it with aluminum powder. The excess powder is then removed by a brush or other polishing device, leaving only the metal particles which actually contact the resinous coat. The coating is then preferably cured as by heating. It is apparent that such a coating, in contradistinction to the more or less homogeneous coatings of the lacquer-metallic and casein-metallic types, is composed of two distinct layers, 1. e., a layer of resinous binding material next to the paper, with a thin surface layer of metal flakes lying thereon. The lower side of each flake is anchored in the resin while the top side is exposed substantially free from any covering material. As a result, when properly burnished such coated paper may exhibit a polished aluminum surface which approaches polished foil in appearance. Such paper reflects visible light very well, but owing to the comparative thinness of the metallic layer it would not be expected that the more penetrating infra-red rays would be reflected so well; actually the most attractive of the products of this type with which I am familiar is known to have poor heat reflective properties, i. e., to have a high emissivity value.

The product of the present invention is an aluminum-coated sheet material' which, contrary to prior aluminum-coated papers, has a low heat emissivity value, in the order of 0.2 or less, preferably in the order of about 0.1 or less. The product has a coating which is substantially uniform throughout, being deposited from a liquid slurry in which aluminum powder is thoroughly mixed with an aqueous binding material. More- 'over the product has to a considerable degree the appearance of polished aluminum foil. The degree of resemblance of the product to polished foil and its difference from prior aluminumcoated papers having coatings deposited from aqueous media are easily apparent to the eye. In some cases the product of the invention may hardly be distinguished from polished foil in appearance, while in all other cases the appearance of my product is much nearer to that of polished foil than has been attained by prior coated paper having water-deposited aluminum coatings.

There is not, to my knowledge, any absolute measure of "polished metallic appearance. However, polished metal and other very shiny surfaces are distinguished by certain optical characteristics which, while having no absolute correlation with polished metallic appearance, are nevertheless definitely not possessed by material having no shiny appearance. Of the measurements of 1) specular gloss, (2) luminous apparent reflectance and (3) apparent-reflectance contrast, each affords means of scientific definition of certain such optical characteristics which go with polished metallic appearance. Without at present going into detail as to the nature of the aforementioned measurements it will sufllce to state that my products have a specular gloss of at least 25, while prior casein-aluminum coated papers have not had a specular gloss in excess of about 10. The products of my invention likewise have a luminous apparent reflectance of not over 0.20, whereas' prior caseinmetallic" papers (aluminum) have shown values of about .35 to about .30. The apparent-reflectance contrast value of my products is 100 or more,

while that of prior casein-aluminum paper has been not over about 25.

The considerable difference in gloss properties between my product and prior casein-aluminum coated papers indicates that either the proportions of coating ingredients used for my product have not been used before or that the particular finishing methods I use have not been applied to previous casein-aluminum papers, or both.

To make the product of the present invention an aqueous suspension is prepared comprising aluminum powder and suitable water-dispersible adhesive, the proportion of adhesive to aluminum powder being notably less than in similar waterdispersible adhesive-aluminum compositions here, tofore used for paper coating whereby to provide an adhesive-aluminum coating which is definitely slack-sized as contrasted with prior art, conventional, casein-aluminum coatings; the suspension slurry is applied to a suitable flexible web such as a web of paper; the coated web is dried; and then the coating is smoothed by heavy rolling pressure. If desired, the smoothed coated surface may subsequently be burnished, bufled, or otherwise polished.

The base stock to which the aluminum coating is applied may be a paper, textile, regenerated cellulose web or the like. When paper is used as the base, it may be calendered or supercalendered; or preferably the base may be a paper or a textile web bearing on mineral coating (e. g., clay or other pigment and adhesive such as starch, casein or the like), which mineral-coated surface may, if desired, be calendered and/or supercalendered. Superior results are obtained by use of a base stock carrying a relatively heavy mineral coating, e. g., a coating of over 15 pounds per ream on one side. In some cases it is advantageous to have the surface of the base stock and/or the mineral coating colored or dyed to approximately the color of the aluminum coating applied thereto: such color in the underlying surface makes possible the use of a thin layer of metallic coating without danger of the base stock being discernible through the metallic coating when the product is complete.

The aqueous slurries of aluminum powder and adhesive may be applied to the base stock by any conventional coating means, such as a brush, knife-blade, roll, or spray-gun. A preferred method, is the well-known air-blade or airits surface a layer of brush" method disclosed by Lebel in U. S. Patent No. 1,980,923.

Adhesives suitable for use in preparing aqueous slurries of aluminum powder in accordance with the present invention include various materials soluble in water. or dilute alkali, including: carbohydrates, such as starch; proteins, such as glue, casein, and soy bean protein; and water-soluble resinous substances, such as polyvinyl alcohol, Petrex resin, and the like. Because of its efficiency and relatively low cost, casein is expedient to use. In using casein, care should be exercised to use no large excess of alkali for dissolving the casein, because a large excess of alkali may act adversely by dissolving part of the aluminum powder.

Any of the various aluminum or aluminum bronze powders commonly used in the manufacture of aluminum coated papers are usable in the carrying out of the invention. Commercial aluminum powders are supplied in various grades such as litho, standard," "extra brilliant, lining, extra fine lining, etc. Commercial aluminum powders are prepared by the use of hammermills, or similar devices, and during the process of manufacture they are coated with more or less grease of some kind. For my purpose, the less grease on the powder the better. In any case, however, it is advisable, before preparing the aqueous coating mixture, to wet the metal powder with a wetting agent, such as an alcohol, a sulphonated naphthalene derivative, a sulphate of a high aliphatic alcohol, or the lik such wetting materially assists in obtaining good dispersion of the powder in the aqueous mixture and insures contact being made between the metallic particles and the adhesive.

It will be readily understood that variations in particle size, surface area, and other properties of the aluminum powder particles, as well as in quantity of grease upon them, will cause variations in the adhesive requirement among different powders. The adhesive requirement may be understood to be the least quantity of adhesive which will bind a unit quantity of the metal powder satisfactorily in the use to which the coating is to be put. Since the adhesive requirement varies depending upon the properties of the aluminum powder used and also to some extent upon the particular base stock used, it is apparent that no fixed ratio of adhesive to metal powder can be set as optimum. Generally speaking, the best results in respect to finish and metallic appearance result when the ratio of adhesive to aluminum is kept as low as possible. I prefer to use not over 35 parts by weight of adhesive to 100 parts of aluminum powder and I usually obtain better results if I use less than 35 parts, say or parts, of adhesive to 100 parts of aluminum powder. Of course, it is to be understood that the finish, gloss, or metallic appearance of the product depend not alone on the adhesive-metal ratio, but also upon the finishing treatment given the coated web. For instance, a coating containing not over parts of adhesive to 100 of aluminum may be calendered moderately to show a specular gloss of say 25 and thus fall within the limits of my acceptable products, or it may be treated further according to my preferred buffing treatment (to be described hereinafter) to acquire a considerably higher gloss. On the other hand, a coating containing a higher ratio of adhesive to aluminum, say, or to 100, almost certainly will not, when merely calendered, take a gloss sufilcently high to fall within the scope of my invention; but such a coating when further treated according to my preferred bumng process may attain a gloss high enough to bring it within my product. Even 45 par of adhesive to 100 parts of aluminum is considerably less than it has been customary to use in previous casein-aluminum coatings';-however: for example, the coating of a typical, high grade. commercial silver" paper was found to contain "l2 parts of casein to 100 parts of aluminum powder. On the other hand, as previously stated, I prefer not to use more than 35 parts of adhesive to 100 of al' minum, i. e., not more than approximately half the quantity that has been used in typical prior aluminum coated paper. Usually I need to use less than 35 parts, say from 20 to 30 parts of adhesive. I have successfully used as little as 15 parts of casein, or correspondingly small amounts of other water-dispersible adhesives, to 100 parts of aluminum powder.

The paper coated with aqueous aluminum pow- "der. slurry as hereinbefore described may be dried in any-usual manner, as by currents of hot air, on sticks, or against internally heated drying surfaces. The dried paper is then subjected to relatively heavy pressure to smooth the coating. The smoothing means may suitably be a calender stack of all metal rollspor a supercalender comprising alternate metal and softer rolls, or a friction-calender having a driven friction-roll running at a speed different from the paper speed, or any combination of such calenders, depending upon the exact type of finish desired. The heavy rolling pressure exerted by the calender acts to align the particles and press them together so that they present substantially a single plane surface more or less closely resembling that of sheet metal. Calendering the sheet while the coated surface is moist is advantageous in order to increase the flatness of surface; very heavymoistening of the sheet is permissible. The use of polished calender rolls is advantageous to increase the smoothness and gloss of the calendered surface. The method and apparatus disclosed for supercalenderi ng varnishable printing paper in Cates U. S. Patent No. 2,088,893 are very suitable for use to make the present product. Heavy calendering of some sort is a practical essential in the production of my improved product.

If desired, the calendered sheet may be further treated to increase the polish and specular reflection of the surface. Polishing or burnishing devices, such as brushes or cloth buffing rolls, may be used, or methods such as those commonly employed in polishing metal foil may be resorted to. In carrying out the buffing operation I have found it advantageous to use a polishing agent on the buiflng roll or rolls: the polishing agent should, naturally, be in an extremely finely divided state. For this purpose fine silicious powder, emery powder, polishing rouge, tripoli, and finely divided chromium oxide have been used with success, particularly when mixed with a small quantity of waxy, resinous, or other suitable binder. The process of buffing is generally described and claimed in my co-pending application filed July 12, 1939, Serial No; 284,088, entitled Paper with improved surface." Apparently a real grinding or polishing action is obtained when such polishing agents are employed in the buffing. Whatever the finishing process employed, however, the finished product exhibits an appearance more like that of true metal foil the lower limits acceptable for than is obtained by applying the same finishing operations to previous papers having coatings deposited from aqueous suspensions of aluminum and adhesive in which the conventional ratios of adhesive to metal powder have been adhered to.

The effect of polishing ordinary casein metallics" with a bufllng roll and abrasive is usually to raise the gloss appreciably; but no great improvement in actual metallic appearance results. In the case of my product, however, the effect of the bumng treatment is striking; the buffed product closely resembles highly polished metal foil in appearance, and its specular reflection is so greatly increased that, generally speaking, the gloss of the product can not be measured on the Bausch & Lomb Glossmeter, a standard instrument for measuring gloss of customary glossy papers. Hence, it is apparent that results are attained by buffing my product which have not been even closely approached by previous caseinaluminum coated papers.

The following examples, in which the parts" .are parts by weight, are given to show some specific coating compositions suitable for use according to the invention:

Example 1 v Parts Aluminum powder, litho grade 100 Butanol (wetting agent) 20 Casein dissolved by alkali (adhesive) 20 Water 200 Example 2 Parts Aluminum powder, extra fine litho grade--- 100 Aerosol, a sulphonate of a dioctyl malic ester (wetting agent) Soy bean protein dissolved by alkali (adhesive) Water- 250 Example 3 Parts Aluminum powder, extra brilliant litho grade 100 Turkey-red oil (wetting agent) 5 Oxidized corn starch (adhesive) 35 Water -4 250 v Example 4 Parts Aluminum powder, lining grade 100 Butanol 20 Polyvinyl alcohol (adhesive) 15 Water s 1'75 grades of Following is a description of the preferred manner of carrying out the process of the invention:

For the base there is selected a mineral-coated paper of which the uncoated body stock amounts to about 50 pounds per ream (of 500 sheets cut 25 x 38 inches) and which has on one side a mineral coating of from 13 to 20 pounds, the mineral coating comprising inorganic pigment such as clay, calcium carbonate and the like. and adhesive such as casein, starch and the like. To the coated side of the base paper is applied, by means of an air-brus coating machine, from 5 pounds to 1 pound, or even less, dry weight, of an aqueous aluminum coating composition similar to preceding Example 1, or any other of the examples previouslycited. The coated paper is dried by any conventional method, preferably by the passage thereover of hot air currents. The coated and dried sheet is then-preferably after being moistened (e. g., steamed) heavily .on the coated side-run through a stack of calendar rolls in bufling roll by a suppo ing the coated sheet have highly polished surfaces; the other rolls of the stack may be either metal or fiber rolls, but in any case relatively hard rolls are preferred. The calendaring preferably is conducted under heavy pressure, suitably the heaviest pressure which paper will bear.

If the treatment be stopped at this point, i. e., after the calendering operation, it will be found that the product is of excellent appearance, and has a gloss considerably higher than that hitherto known in aluminum coated papers iioated from aqueous suspension, and higher than that of dull or unpolished aluminum foil. If a higher gloss is desired, the calendered paper may be further treated by subjecting it to the action of one or more bumng rolls used in conjunction with suitable fine abrasive powder as previously described. Such bufling 'rolls havenot prior to my invention, to my knowledge, been .used in the treatment of paper. Such rolls have a soft yieldable surface, such as fabric. The bufilng roll or rolls revolve-at high speed while contacting the sheet, which be held against the g roll. As a result of the buffing action the luster of the aluminumcoated-surface is increased until it practically equals or even surpasses that of usual grades of polished aluminum foil.

The calendered, or calendered and buffed, product may, if desired, be embossed by known means, to provide a desired embossed effect.

It should be noted that in case colored coating's are desired, colored aluminum powders may be employed in the coating compositions. Or one may use uncolored aluminum powder, but color the coating compositions by dyeing the menstrua, e. g., by dissolving in them suitable dyestuffs. Or, a colored transparent coating may, if desired, be applied over the aluminum surface to change the color thereof.

As has been shown, the invention results in marked improvements in paper coated with aluminum powder deposited from aqueous suspensions. Improvements result in part from the novel composition of the coatings, and in part from finishing treatments not before given alumnum-coated papers. Coatings containing the relatively low ratio of adhesive to aluminum powder shown, that is, "slack-sized coatings, are found to respond to calendering and especially to bufling operations in a manner quite different from previous aqueous coated aluminum papers. The preferred form of the invention comprising a flexible base having a mineral-coating and a top coating of aluminum and adhesive, suitably calendred and buffed, is so similar in appearance to metal foil that it may be considered as an imitation foil.

In an effort to provide exact measurement of the optical characteristics of my calendered (and of my calendered and buffed) metal-coated products, I have caused the latter to be studied from the standpoints of (1) specular gloss and (2) luminous apparent reflectance, each of which measurements affords means of scientific definition of the optical characteristics, as will be explained in the following:

A distinguishing quality possessed by polished metal and other very shiny surfaces is a relatively high specular gloss. The specular gloss of the coated a surface may be readily determined by finding the apparent reflectance in the direction of mirwhich hard or metal rolls contact the coated side of the paper.

instrument aperture type of surface being ror reflection through an which is adapted to the studied. The photometric device used, of course, is calibrated against a permanent standard of apparent reflectance. Several reflectometers are available for making such measurements. One suitable reflectometer for measuring specular gloss is described briefly in the Journal of the Optical Society of America, volume 27, page 225. The sample in question may be illuminated at an angle of 45 to the normal and viewed at an angle of minus 45 to the normal using a reflectometer similar to that mentioned which has circular apertures about 18.

If ordinary casein metallics," comprising paper coated with'aluminum powder and a watersoluble adhesive in the heretofore conventional ratio of adhesive to metal, are measured under the conditions mentioned above, they are found to have a specular gloss of about 10 on a scale where a perfectly reflecting surface would show 100. On the other hand, paper coated according to the present disclosure and properly supercalendered will, under the same conditions, exhibit a specular gloss of at least 25, and preferably a specular gloss of 30 and very advantageously about 35 or better. In case the product of this invention is bufled after being supercalendered, it will be found to have a specular gloss of at least 45 and preferably has a specular gloss lying between 45 and or even higher.

In speciflc tests, specular gloss measurements showed the following values:

comprising aluminum powder 7.6 6. Another casein metallic" paper of commerce comprising aluminum powder 7.0 '7. A "lacquer metallic" paper of commerce comprising aluminum powder 5.3 8.Polished white Vitrolite 5.4

Each of the above values represents the mean of readings taken from a plurality of settings over a surface. It is found in the case of very shiny surfaces, i. e., surfaces havinga specular gloss of over 50, that considerable diflerences in measured specular gloss can exist without being particularly noticeable to the eye. On the other hand, in the case of less shiny surfaces, small differences in specular gloss attract quick attention. For example, a sample of the bufl'ed product of the present invention was found to have a specular gloss of 55, while a sample of very high grade polished aluminum foil had a specular gloss of nevertheless, the difference between the two from merely visual inspection was very slight. On the other hand, however, the visible difierence between sheets having specular gloss values of 10 and of 25 was very marked indeed.

The measurement of the luminous apparent reflectance is another test by which the products of this invention may be distinguished, and which affords a numerical definition of metallic appearance." In making this test the surface is illuminated at an angle of 45 from the normal. and is viewed at 0 from the normal, measurement being made with a Hunter multi iving an aperture sum of purpose reflectometer" (J. Opt. Soc.Am., vol. 27,

p. 225, 1937). The scale of measurement is calibrated between a value for the luminous apparent reflectance of a perfectly reflecting mirror, and a value of "100 for a prepared magnesium oxide surface.

On this scale, a very flnely polished aluminum foil was found to have a value of .05 and an unpolished or duil" aluminum foil was found to have a value of about .68. The supercalendered and bufled metallic-coated paper of this invention has avalue lying between .10 and .05 or less-e. g., about .08--while the calendered but Lnbufled paper product has a value lying between .20 and about .15, or even less. Tested in this same manner, standard caseinmetallic" papers (aluminum) showed values of about .35 to about .30, while a lacquer metallic" paper of commerce showed a value of about .24.

In determining numerical values of apparent reflectance contrast, Cr, the formula employed which formula recites the ratio of specular gloss to luminous apparent reflectance of the subject tometer. previously referred to, or a McNicholas goniophotometer (see Jour. of Research of the U. S. Bureau of Standards, vol. 13, page 211), employing a photox" cell and approximately the same total source andreceptor spreads as are employed with the Hunter instrument (e. g., about 18 in both the plane of the measuring light beams and the plane perpendicular thereto). Since the r" numerical values are large numbers, it may be preferable to cite the values in terms of logmCz.

In specific tests apparent reflectance contrast values were found in accordance with the data of the following table, in which samples 1 through '7 were the same as samples 1 through '1 of the data above relative to specular gross measurements, is another sample of the same material as 3 but made at a different time, and "11" is the calendered and buffed product of the present invention closely similar to sample 2 above but differing from the latter inthat sample 11 was machine buffed whereas sample 2 was hand buffed:

Apparent refleitance contrast, Log Cl Sample PM Plane of e of b E am 3 beams n- Mean Parallel Perpm' Mean parallel dicular in grain 850 550 700 2. 93 2. 74 2. 84 1, 050 330 090 3. 02 2. 52 2. 77 140 125 133 2. 2. 10 2. 12 10 72 41 1. 00 1.86 l. 43 16. 6 13. 4 15 1. 22 1. 13 1. 18 16. 1 14. 7 15. 4 1. 21 1. 17 l. 19 13.7 11. 2 l2. 5 1. l4 1. 05 1. 10 300 310 336 2. 56 2. 40 2. 52 900 340 665 3. 00 2. 53 2. 76

In general, the calendered but. unbufled products of the present invention show average apparent reflectance contrast values upward from a minimum of 100 (logmC==2.00), while the calendered and buffed products show average apparent reflectance contrast values upward from a minimum of 500 (10B1oCr=2.70).

It has been found that the products of the present invention have sufficiently low heat ray emissivity values to make them useful as heatinsulating materials. Emissivity may be defined as the ratio between the quantity of heat radiated from the surface of a given body and the quantity which would be radiated by a theoretical, perfect black body of the same dimensions and at thesame temperature, the emissivity of the black body being designated as unity. Polished aluminum foil is the. accepted standardof excellence for reflective insulation. Any material having an emissivity of not over 0.15 may be considered excellent for insulation while anything with an emissivity of not over 0.2 is useful in insulating. Emissivities of over 0.2 in general are of little interest for reflective insulation.

The following table shows the normal emissivities of a number of materials, as determined with the surface at a temperature of 212 F., and the thermoplle at normal room temperature:

Generally speaking my calendered product, when measured under the conditions specif d above, has an emissivity value of 0.2 or less, usually about 0.15 or less, and my buffed product has an emissivity value of about 0.1 or less.

Prior aluminum coated papers in which the metallic coating was substantially uniform throughout, e. g., coatings of either the caseinmetallic or lacquer-metallic type, have surely not approached the low emissivity value of 0.2, much less the still lower value of 0.1. In fact it is to be doubted that any paper coated with aluminum powder has, prior to my invention, had the low emissivity value of 0.1.

It appears, therefore, that my product provides an entirely new kind of heat-insulating p per. It also provides a decorative paper having ardegree of gloss not previously believed attainable in a casein-aluminum coated sheet. While very low emissivity and very high specular gloss are not necessarily correlated, they both may exist in the same sheet of my product.

While the invention has been described in relation to a coated paper it should be apparent that the coating described may equally well be applied to various other sheet materials without important changes in procedure, and will in those cases also result in aluminum-coated products having gloss and emissivity values-not before attained by water-deposited aluminum coatings.

While the present invention is particularly com cemed with the provision of metallic coated sheet material (e. g., metal foil-simulating metallic coated paper) wherein the metal is aluminum, it is fully within the scope of the invention to substitute, wholly or in part, finely divided copper, or finely divided copper alloy, in flake form, for the aluminum powder above described. In such event, one adheres to the concept of employing a less than conventional adhesive-to-metal powder weight ratio.

This application is a continuation-in-part, of co-pending Serial No. 284,089, filed July 12, 1939.

6 ascaaes I claim:

Process 01' making an aluminum coated paper product simulating aluminum foil, which comprises coating a paper base with a mineral coatin: composition and calenderinathe coated paper base, applvinl to and spreading over the mineral coated and calendered paper base a layer or aqueous "aluminum powder-containing coating composition containing substantially Ireaseless aluminum powder. a wetting scent for dilpersinx the aluminum powder and a water-dispersible protein adhesive, the veil'ht ratio of adhesive to aluminum beina' from 15 to 35 parts adhesive to 100 parts aluminum, the layer of aluminum coatin: composition being oi a thickness to add, when dried, from 1 to not more than 5 pounds of the