US3374155A - Modified oxide-coated aluminum and the method of modifying - Google Patents

Description

L. J. WEBER March 19, 1968 MODIFIED OXIDE-COATED ALUMINUM AND THE METHOD OF MODIFYING Filed Feb. 19, 1965 .0.... 002.2 20222. m .0. 0022.220 .002222.0 .00222002 .0 020.22 022 2222022020222 .20.0.20 m .20.002022.20.000 .20.002202 20 020.220 0022220 202.02.22 002002200 w 0.22020 ..2220 02.02202 20 ...2202 .0 20.20.60 0000002 .2.0-0. 22.2. 2.2022 w 4 Y 002.2 20222 B 002200220002 2 .222600222002 ...220 ..2 ...002022 0202002 000.22 022. 002 0022220 202 .0 000.070.-. |222222 .0022.220 .002222.0 .00222002 ...0 20.22 22 02.2.. 00.202022 020.22 022 02 -...0V.0.2.202. .2..2000 002002200 20.2 20.20222 .20.20200 02 .0 020.220 .0..22220 202 ...20200.-0 l.0222.0 I0222 20220 .0 .0 20.20.50 02.22.0.2202 20020 202. .0 02002200 0.22 20.20222 00 20.20000 0000002 0000002 -020 ...0 20.20.60 0000002 2.0-0222.. 22022 20.20000 0000002 002.2 20222. 2020 v. l. 022202220 20.20222 ....2020 00.-. 022200220 20.000

lrraRA/fys United States Patent O 3,374,155 MODIFIED OXIDE-COATED ALUMINUM AND I THE METHOD OF MODIFYING Ludwig J. Weber, 6016 Birch Crest Drive, Minneapolis, Minn. 55424 Continuation-impart of application Ser. No. 397,018,

Sept. 16, 1964. This application Feb. 19, 1965, Ser.

23 Claims. (Cl. 204-38) ABSTRACT OF THE DISCLOSURE A process of improving the corrosion resistance of aluminum articles that includes removing the impurities from the article surface, then chemically or electrolytically forming an artilicial aluminum oxide coating, treating the articially oxide coated article to a dilute aqueous solution of an inorganic base such as NaOH or KOH, and thereafter treating the article to an alkaline silicate solution. Advantageously intermediate the above mentioned treatments, the article is treated to one or more of aqueous solutions of (l) organic compounds having cations of various iron group metals and anions of acetates, citrates, oxalates, tartrates, (2) organic compounds of various alkali and alkali earth metals having anions of acetates, citrates, oxalates, (3) ammonium hydroxide, (4) ammonium compounds having an anion of such acetates, citrates, carbonates, and (5) various mixtures of the above.

This application is a continuation-in-part application of my application Ser. No. 397,018, iiled Sept. 16, 1964, and now abandoned.

This invention is directed to treating aluminum to form a modified oxide coating thereon that substantially increases the corrosion resistance above that obtained with conventional aluminum oxide coatings. More particularly this invention is directed `to forming an aluminum oxide coating on aluminum, then activating said coating and then subjecting the activated coating to further treatment to provide a highly corrosive resistant coating.

In order to `facilitate the description of the invention, the following definitions are applicable unless from the accompanying description the terminology is clearly used in another sense. Oxide coating refers to a coating that is substantially composed of aluminum oxide on the surfaces of aluminum or various alloys of aluminum. `Such oxide coatings may be produced by various methods that comprise a chemical or electro-chemical reaction between the aluminum surface and a solution of a chemically active substance or substances, but are not deemed to include an oxide coating on aluminum resulting from ordinary atmospheric oxidation. Theterm aluminum includes aluminum per se and its various alloys; while a modified oxide coating refers to coating formed by treating aluminum to iirst form an oxide coating and then further treating an oxide coated aluminum as set ,forth hereinafter to substantially increase the corrosion resistance over and above that obtained from only forming an aluminum oxide coating.

The protection of metals from corrosive influences is constantly a problem to the manufacturer as well as to the user. Various materials and procedures are being used at the present time but the search for improvement is going on constantly in order to increase the useful life of the products made from metals. Aluminum has many applications and its use could be greatly increased if the resistance to corrosive mediums could be improved.

Oxide coatings on aluminum and its alloy can be readily applied by chemical and electrochemical means. Even 3,374,155 Patented Mar. 19, 1968 exposure to the ordinary atmosphere produces a protective oxide lm. This oxide coating is actually integral with the metal so that it will not peel like paint or plastics which are so commonly used for the protection of metals. An oxide coating which may be formed on aluminum by an anodic process using, for example, sulfuric acid as the electrolyte has many advantages such as increased hardness, good appearance, non smu-dging, resistance to mild corrosive solutions and others. It has,ihowever, these disadvantages which seriously limit its usefulness, namely:

(1) The oxide coating is soluble in caustic solutions such as sodium hydroxide, sodium carbonate and sodium phosphates which are used in detergents for cleaning various types of equipment like cooking utensils, dairy equipment, clinical ware etc. Usually the cleaning in commercial installations is done by automatic machines in which the solutions are set at a temperature of to F.

(2) The oxide coating is subject to deterioration by hydration when exposed to moisture. This action increases as the temperature of exposure increases.

(3) The oxide coating is also attacked by comparatively weak acid solutions.

In accordance with this invention it has been found that by proper treatment of the aluminum oxide coating formed by chemical or electrochemical methods, these disadvantages ean be overcome, and thus allowing aluminum to be used in many applications where resistance to corrosion and deterioration is required.

One of the objects of this invention is to provide a new and novel process of treating aluminum to form a modied oxide coating of other stable compounds in order to obtain a substantial increase in corrosion resistance of the oxide coating obtained by present procedure. Another object of this invention is to provide a new and novel activation treatment of oxide coated aluminum prior to exposing said aluminum to a silicate treatment with or without subjecting the article being treated to intermediate treatment including various solutions of organic compounds of alkali, alkaline earth and iron group metals in the process of forming a modified oxide coating. A further object of this invention is to subject aluminum to a new and novel coating treatment that provides a coating very resistant to caustic and salt solutions as well as overcome the action of hydration.

An additional object of this invention is to provide a new and novel process for treating aluminum to obtain a substantial increase in corrosion resistance of the oxide coating obtained by present procedures that includes activating the oxide coated aluminum, treating the activtaed oxide coated aluminum to a solution of an ammonium compound which readily releases ammonia or ammonium hydroxide and then to an alkaline silicate treatment.

Other and further objects are those inherent in the invention herein illustrated, described and claimed and will be apparent as the description proceeds.

To the accomplishment of the foregoing and related ends, this invention then comprises the features hereinafter fully described and particularly pointed out in the claims, `the following description setting forth in detail certain illustrative embodiments of the invention, these being indicative, however, of but a few of the various ways in which the invention may be employed.

With reference to this disclosure an explanation of terms used is pertinent as follows:

1) Sulfuric acid electrolyte: sulfuric acid concentration of 15-l8.5% current density of about `1 2 amperes per square foot, and a temperature of electrolyte solution about 70i2 F.

(2) Chromic acid electrolyte: -chromic acid concentration of 10%, current density of one to three amperes per 3 square foot, 40 volts and a temperature of solution of about 65 F.

(3) Hard coat: sulfuric acid electrolyte having a sulfuric acid concentration of 7%, current density of 30 amperes per square foot, and the temperature of solution of about 257 F.-35 F.

(4) Alrok: trademark of the Aluminum Company of America, wherein a coating on aluminum is formed by a chemical treatment of (a) 2% Na2CO3 and 0.1% KzCrzOq, (b) 3% and 0.1% respectively or (c) 0.5 and 1.0% respectively in water at a temperature of 175 F.- 2l2 F. (ratio of carbonate to dichromate being from '20 to l to 30 to 1) depending on the product being treated; rinsing in water and which can be sealed in a solution of about 5% K2Cr2O7 at about 180 F. to 190 F. with an original pH of 4.5 to 5.0.

(5) Iridite: proprietary name for a powder of 61.7% ammonium phosphate, 22.9% ammonium fluoride and 15.4% potassium dichromate that is put in solution for chemically forming a phosphate conversion coating on aluminum.

(6) *FeNH4C an abbreviation used at various places in the specification to designate ferrie ammonium citrate.

(7) *NH4C an abbreviation used at various places in the specification to designate ammonium citrate.

(8) Liquid G.T.W.-an arbitrary non-proprietary designation used for a non-trademarked liquid sodium silicate having a ratio of percent Na2O2percent SiO of 1:3.25 which is produced by Lyons Chemical, Inc., of St. Paul and distributed by Geo. T. Walker Co., Inc., of Minneapolis, Minn.

The normal compositions of the aluminum alloys appear in the table below:

NORMAL COMPOSITION 1 Alloy Copper Manga- Magne- Chro- Zinc nese sium mium 1100 (99.00 percent minimum aluminum) 3003- 1 2 0. 8 5052 2. 5 .25 5357 1. 5457 (3) 30 1. 0 7075 1. 6 2. 5 3 5. 6 High Purity.-. 99. 99

l Percent of alloying elements-aluminum and normal impurities constitute remainder.

2 .29% maximum silicon and iron, 0.2% maximum copper.

3 .18% maximum silicon and iron, 0.2% maximum copper.

4 Aluminum.

. All of the below listed silicates are listed by their trademarks, the trademarks being those of Philadelphia Quartz The procedures hereinafter outlined and described consist of modifying aluminum oxide so as to change it from a product that is readily attacked to one that resists the attack of solutions that are very corrosive to aluminum as well as conventional oxide coatings that are presently employed.

First a brief description of a preferred procedure including a preferred form will be set forth in order to facilitate an understanding of the invention and then the invention will be more fully described including alternate procedures. This includes starting with a clean aluminum sheet or other aluminum articles which is provided with an aluminum oxide coating by an electrochemical procedure using sulfuric acid (l-18.5%) as an electrolyte and the aluminum sheet as the anode. A current density of 12 ampere per square foot with varying lengths of 4 v time is used in applying the coating. The sheet is then rinsed in water.

The next step which appears to be the most important for obtaining best results is to activate the oxide formed on the aluminum sheet. This is accomplished by heating the oxide coated sheet at about 212 F. for 5-15 minutes in a solution of about 0.2 to .55 gram sodium hydroxide or potassium hydroxide per liter of Water.

The activated sheet is then exposed to a solution of an organic compound of the iron group metals or to either ammonium hydroxide or ammonium citrate. Best results are obtained using about 5 grams of nickel acetate and 5 grams of ferric ammonium citrate in a liter of distilled water, heating for 15-30 minutes at 14C-165 F. and then rinsing the thus treated aluminum sheet in water; or using .5-5 cc. concentrated ammonium hydroxide per liter of water at about 200-212" F. for about 15-20 minutes, or using about 9-10 grams of ammonium citrate in a liter of water at about 200-212 F. for about 15 minutes.

The thus coated aluminum sheet is exposed to an organic solution of an alkaline or alkali earth metal. Preferably a solution of 10 grams per liter of magnesium acetate or potassium acetate at about 212 F. for 30i10 minutes is used and then the thus treated sheet is rinsed with water. Finally the above coated sheet is exposed to a solution of about 27 cc. potassium silicate, Kassil #1, in a liter of distilled water at about 212 F. for 30 minutes to form a spinel type compound, rinsed with water, dried and then buffed to remove any loosely adhering film.

Desirably all solutions are made up using water low in mineral constituents and near neutral; and preferably distilled or deionized water.

Finishing and treatment prior to the application of the oxide coating The finish or appearance of the oxide coated specimens depends to a great extent on the surface condition prior to the application of the oxide coating. In order to obtain best results, the aluminum should be substantially free of foreign material, including any scratches or metal defects, dirt, oil and grease and any substantial coating formed by air oxidation, both before applying the oxide coating and thereafter. The proper surface finish can be obtained by proper procedure in rolling the sheet, bufiing or other mechanical finishing methods.

The aluminum to be treated is cleaned, for example, by one or more of the steps as follows:

( l) Degrease in vapor degreaser,

(2) Degrease as in (l) followed by caustic etch,

(3) Degrease as in (1) followed by bright dip,

(4) Degrease as in (l) followed by caustic etch and bright dip.

Tests, examples of which are set forth hereinafter have shown that the condition of the sheet free from scratches and other defects with a finish as rolled buffed and etc., or the various chemical treatments of the above steps had no substantial effect on the corrosion resistance of the final (modified) oxide coating.

Part I.-Alamnum oxide produced by electrochemical means Aluminum oxide coatings can be produced by electrochemical methods using an oxidizing acid as electrolyte and making aluminum the anode. Acids which can beused are sulfuric, chromic, sulfamic and oxalic although mixtures of these as well as other acids and compounds` may also be added to these in order to obtain certain desired characteristics as hardness, color or texture. Sulfuric and chromic acids were selected for the majority of the specimens set forth herein as they are the QIles generally used in this country.

As an example of applying the oxide coating using the sulfuric acid, the electrolyte was as follows:

( 1) Concentration of sulfuric acid, percent 15-18.5

Procedure to make the oxide coating resistant t corrosive agents The aluminum oxide coating as produced by the above electrolytic process with or without being sealed in boiling water is not resistant to weak caustic solutions such as are used in detergents. This makes it impossible to use this coating on equipment subjected to many detergents such as cooking utensils, clinical ware, automobile trim and other applications unless special precautions are taken in the cleaning operations and selection of detergent. The oxide coating is also subject to hydration when exposed to water, especially at elevated temperatures, which again limits its use on aluminum articles. Such aluminum oxide coatings are also attacked by salt solutions of rather low concentration which prevents its use in many applications where it would come in contact with these solutions. Likewise forming a coating chemically on aluminum in accordance with the teachings of the prior art is subject to the same disadvantages.

In order to provide the above oxide coated aluminum with a modied coating of this invention it was treated as follows:

Procedure A.Activate the coating The oxide coating obtained by the anodic process is thoroughly rinsed in water and then activated by heating it in a solution of an inorganic base, for example, 0.2-0.55 gram sodium hydroxide or potassium hydroxide, per liter of water at boiling temperature (212 F.) for 5 to 15 minutes. In order to obtain best results there should be a constant stirring of the solution either from the heat applied .or by means of mechanical or air agitation.

That is, the aluminum oxide coating is immersed in a solution of an inorganic base of about .l-1.1 grams solute per liter of water; for example, about .15-l.1 gm./liter of NaOH or KOH (preferably about .2f- .55 gm./liter); or about .3-.6 gm./liter of Na3PO4 (preferably closer to .3 gm./liter); or about .26 gm./liter of Na2CO3 or about .4 gm./liter of NH4CO3. Sodium hydroxide and potassium hydroxide are preferred to the other bases mentioned. That is the activation treatment can be carried out at about 5-30 minutes at a temperature range of about room temperature to 250 F. with a base of the general type indicated of about .O-.11% concentration. However as indicated before, the activation step is to be carried out so that it does not too severely *attack lthe coating, an-d in particular, leaves a partial aluminum oxide coating covering the entire exposed surface.

Preferred procedure:

(l) Rinse the oxide from the anodzing process in water.

(2) Expose to a solution of about 0.2-.55 gram sodium hydroxide or potassium hydroxide per liter (.02-.055%) of water at boiling temperature for 5-15 minutes.

(3) Rinse with water.

Procedure B.-Expose to a solution of organic compounds of the iron group metals After operation A, the thus coated sheet is exposed to 0.5 to 2% solutions (desirably 510 grams/liter) or organic compounds of the iron group metals such as acetates, citrates, oxalate's, tartrates and others which hydrolyze on heating to form oxides such as chromium, nickel, iron, cobalt, and zinc. The specimen from Procedure A is exposed to the solution mentioned of this procedure at about 120 F. to 250 F. for a period of 6 15 to 30 minutes (desirably 140-180 F.) and then rinsed in water.

Alternately, the solution is desirably .5-l5.5 cc. concentrated ammonium hydroxide or 910 grams of ammonium citrate per liter of water at about room tem-perature to 250 F. for a period of 10 to 30 minute-s and then rinsed in water.

It is preferred that a mixture of about 5 grams of nickel acetate and 5 grams of ferrie ammonium citrate dissolved in one liter of distilled water or .5-5 cc. of concentrated ammonium hydroxide per liter or about 10 grams ammonium citrate, to 'be used to give the most consistent results. If the solutions of nickel acetate or ferric ammonium citrate yare used separately, a concentration of about 10 grams per liter is used. It is also preferable that the solution is at most only slightly acid or one which can be made nearly neutral by adding, for example, NaOH without precipitating the metallic ions. The more acid solutions do not produce as good results.

Preferred procedure:

1a) Expose to a Isolution of about equal parts by weight of nickel acetate (5 grams per liter) and ferric ammonium citrate (5 grams per liter) at 160 F.i5 F. for a period of 15 minutes. (Note: either of these salts by itself will produce good results but the mixture gives more consistently reproduceable results and increased corrosion resistance); or

(1b) Expose to a solution of ammonium hydroxide (.5-5 cc. concentrated ammonium hydroxide per liter) `at about 200 F. for a period of 15-20 minutes; or

(1c) Expose to a solution of ammonium citrate (9-1O grams per liter) at about 200 for 15 minutes.

Procedure C.-Exp0se t0 a solution of alkali o-r alkaline earth organic compounds The coated sheet from Procedure B is exposed to a .5 to 2% solution (desirability 5-10 grams per liter of distilled Water) of organic compounds of potassium, sodium magnesium, calcium, barium, or strontium, such as acetates, citrates, oxalates, tartrates and others at about 250 F. for about 15-45 minutes and desirably at 160- 212 F. However a near boiling temperature is preferred.

It has been found that about 10 grams per liter solution of magnesium acetate or potassium acetate at about 212 gives very good results. The coated sheet of Procedure B is preferably exposed to about 8 to 20 grams per liter solution of magnesium or potassium acetate at about 212 F. for 30i10 minutes and rinsed in water.

Preferred procedure:

(1) Expose to 8-10 grams per liter solution of magnesium acetate or potassium acetate at about 212 F. for about 20-40 minutes.

(2) Rinse in water.

Procedure D.Reactng the treated coating of Step C with silicate The coated sheet from Procedure C is next exposed to an aqueous solution of sodium or potassium silicate (14 cc. to 112 cc. liter of water) for a period of l5 minutes to an hour at about 200 to 250 F. It is preferred that Kassil #l silicate made and sold by Philadelphia Quartz Co. be used. The solution should be kept at about 212 F. and agitated either by heat or other methods. Silicates having a percent NazOzpercent Si02 ratio of about 113.20 to 1:3.75 or ya percent K20:percent Si03 of about 1:2.50; the potassium silicate solution being preferred to the sodium silicates.

Preferred procedure:

(1) The coated sheet from Procedure C is exposed t0 a solution consisting of 28 to 56 cc. of potassium silicate (Kassil #1) in a liter of water at about 212 F. for a period of 30 minutes.

(2) Rinse in Water, dry and buff to remove any loosely adhering lm.

7 Part ll.-Alternate method (A) of obtaining efject of iron group metal oxides in the aluminum oxide coating Some of the heavy metals such as vanadium, molybdenum, tungsten, colum'bium, chromium and manganese are amphoteric which makes it possible to add these oxides to the aluminum oxide coating during the anodizing process. For some applications, sufficient resistance to corrosion can be obtained by the method to be described and thus save the cost of Procedure B given in Part I. The results indicate that for maximum corrosion resistance, Procedure B is necessary.

The amphoteric metals mentioned above form vanadates, molybdates, tungstates, columbates, chromates and manganates or permanganates. By adding either the sodium salt or the acid (l-%) of these amphoteric metals to l5-'20% sulfuric acid, the negative ion containing the metal will goto the anode with the sulfate ion during the anodizing operation and deposit the metallic oxide in the aluminum oxide coating.

The procedure for making the coating corrosion resistant would be the same as that previously disclosed under Part I. However for some applications the addition of the amphoteric compounds to the sulfuric acid woul-d replace Procedure B.

Part IIL- Alternate method of introducing oxide in the aluminum oxide coating Many metals including the heavy metals, the alkali earth metals, magnes-ium and zinc can be introduced into aluminum as lan al-loying constituent. yFor example, by adding these elements to the aluminum as an alloyirlg constituent and choosing the proper anodizing conditions of acid composition and temperature, the desired oxides can be incorporated in the aluminum oxide during the anoidzing process. Typical alloys would be 1-2% magnesium and .5 to 1% of any one of the following: chromium, molybdenum, vanadium, iron, nickel, manganese, cobalt and columbium. Also more tha-n one heavy metal could be added to the alloy to obtain the desired :resul-t. There are limits as to the amount of metal that can be added depending on the solubility, the effect on the properties of :aluminum as well as the finish on the final product. lFor some applications sufiicient resist-ance to corrosion can be obtained by the method described in Part I (with or without Procedure B) with the indicated modification and thus save the cost of Procedure lB given in Part I. Results indicate that for maximum corrosion resistance Procedure B is necessary.

The alkali earth metals are very insoluble in aluminum and cause difficulties in the fabrication of the aluminum so that -th-is addition of these to aluminum is not recommended.

Magnesium and zinc are very soluble and there are many alloys of aluminum that contain these two metals. The results thus far indicate that w-ith the present alloys there is insufficient oxide formed to give maximum corrosion Iresistance and therefore it is not expected that the addition of these to the aluminum will replace Procedure C.

Part IV.--Oxide coating by chemical methods Aluminum oxide can be formed ras a protective coating by chemical methods. Essentially the solution used consists of a Ibase such as sodium carbonate and an oxidizing agent such as potassium dichromate. The coating is applied by immersing the aluminum in this solution at a temperature of `140 to 150 F. for various periods of time. Two of these conventional coatings used to considerable extent in industry are designated by the trade names Alrok and I-ridite. These coatings are soft as compared with those obtained by the anodic process and are not as resistant to corrosive mediums.

The corrosion resistance of these coatings (for example, Alrok and Iridite) can however be increased considerably by the same procedure outlined under Procedures A, B, C

The sheet of drawings summarizes the various steps of this invention.

General comments relative the examples All of the specimens of the examples were made of aluminum or an alloy thereof, the type of alloy being indicated for a given specimen when a record was kept of the alloy. Although not specifically indicated for each of the specimens that were subjected to an anodic treatment, each of such specimens was cleaned in a manner set forth under the heading Finishing and Treatment Prior to the Application of the Oxide Coating or as set forth in the various examples. Further even though not specifically indicated, after the anodic treatment, if used, the respective specimens were thoroughly rinsed in water. Likewise after each treatment in accordance with Procedure A, Procedure B, Procedure C and Procedure D respectively, where used, each of the specimens was thoroughly :rinsed in water.

Unless otherwise noted, the effectiveness of the successive treatments in the various solutions indicated was determined by immersing the treated specimen in a 1% NaOH solution at room temperature (corrosion test). The results of the corrosion test are based on visual observations by the naked eye. Other than for the indication of no attack and no noticeable attack, no observations were made other than for the period of elapsed time indicated.

In some cases two compounds were mixed to form a binary solution. lFor example with reference to Specimen 26 (Table 4) under Procedure B, ,NiAc3-i-FeNI-I4C (4.9 gms. each/liter) indicates that a quantity of nickel acetate and a quant-ity of ferrie ammonium citrate were each added to water to give the concentration per liter referred to; i.e. if one liter of water was used, 4.9 grams of nickel acetate and 4.9 grams of ferric ammonium citrate were added to one liter of water.

A rsum of the test results for the specimens of all examples which were not subjected to any of the treatment steps of Procedures A through D is found in Example 3.

EXAMPLE l In order to determine the effect of pretreatment on specimens before artificially producing an aluminum oxide coating on the specimens, various specimens were preltreated as indicated in Table 1.

TABLE 1 Specimen Alloy Pretreatment 5457 Degreased and then a caustic etch. 3003 Do. 5052 D0. 5457 Degreased and then a bright dip. 3003 Do. 5052 Do. 5457 Degrease, next a bright dip, and then a caustic etch.

Solution Time Temperature (min.)

(a) NaOH (.32 gms/liter) 10 Boiling. (b) NiAcq-l-*FeNILC (4.9 gms each/liter) 15 1GO-170 F. (c) MgAcg (9.8 gms/liter) 30 Boiling. (d) Kassil #1 (56 celliter) 30 Do.

To be mentioned is that after the anodic treatment and after the treatment in each of the above solutions each specimen was rinsed with water. The specimens were then subjected to the ycorrosion test referred to heretofore. The results are given in Table 2 below.

10 A and B; Table 5 the treatment of Procedures C and D; and Table 6 the results of the NaOH lcorrosion test.

5 TABLE 2 TABLE 3 Specimen t Elapled Results of Corrosion Test Specimen Ally Treatment une rs.

12-15 3003 Only chromic acid electrolyte for 40 min.

24 Slight edge attack and in scratch, no other 3003 Degrease, caustic etch, bright dip, 1 hr.

attack. sulfuric acid electrolyte, 24 Attack on one side, checks with 1, 5. 17 3003 Degrease, caustic etch, bright dip, 15 min. 24 Slight attack in scratches, 2 small pits each sulfuric acid electrolyte.

side no other attack. 18 5457 Degrease, 30 second caustic etch, water 24 A few small pits one side, 2 pits on other. rinse, 30 min. sulfuric acid electrolyte 24 Slight attack one scratch, no other attack. followed by 5 mln. chromic acid elec- 7 One side no stains, other large area of trolyte.

stains no attack. 19, 20 5457 Degrease, 30 second caustic etch, Water 17 No attack one side without stains, severe rinse 30 min. sulfuric acid electrolyte.

pitting and corrosion in stained areas. Same as specimen 18.

7 7 Stained areas, but no attack. Same as specimen 19.

17 4 small pits one side, 18 small pits other, Do.

some of these stained areas. Same as specimen .19 other than foaming 8 7 Stained, but not attacked. agent was added to electrolyte.

17 Medium pitting on one side, slight pitting Hard coat.

on other, severe attack in stained areas. No attack, but numerous stains. Severe small pitting.

Degrease, caustic etch and 45 min. sulfuric acid electrolyte.

2 min. caustic etch, 2 min. in HzSO4 etching solution and 45 min. sulfuric acid electrolyte.

401,402 3003 Iridite 3 min., dilute caustic etch 3 min.,

and 45 min. sulfuric acid electrolyte. The modified coated Specimens 6, 7 and 8 appeared 25 stained in areas, however this was undoubtedly caused by the bright dip as staining was observed on the specimens after ,the bright dip treatment. The caustic etch did not remove these stains entirely. From Specimens 1-11 3 TABLE 6 it can be seen that using only the caustic etch instead of 0 t Elapsed the bright dip or in combination therewith better results Speelman Corroslm Results were obtained. However good results can be obtained in 9 using a bright dip during pretreatment as shown by i5; 15' 16m" gi) No iigick Specimens 6, 7 and 8. Further as indicated by Specimens 35 18, 21 25 20 No noticeable attack. 16 and 17 of Example 2 very good results can be obtained 19' 20 gg Isigtttg-mg. when a bright dip is used in the pretreatment of speci- 23 20 No attack.

28 Considerable pitting. mf1s 24 2o No attack.

27 Attack in tlmell scratches on one side,

1 pi on o er. EXAMPLE 2 22 No attack.

7 Do. A further illustration of using various pretreatments 9% Smatlsl gsegt bottom @dse 011e Side. 281mm and types of `aluminum oxide coatings are shown by the 20 Same aset'hrs. 402 14% N o attack. specimens set forth in this eiiample, Table 3 setting furA 17 l Start of attack m Iarge-aa one side ther the pretreatment steps, if any, and the type of alu- 45 severe attacklarge ama on other. miiium oxide coating; Table 4 the treatment of Procedures TABLE 4 Procedure A Procedure B Specimen Solution Temp. Solution Time Temp. 12-17 NaOH (.30 gms/1).... 212 F NiAc3+FeNH4C (4.3.gmseach/liter) 15 min 1Z0-140 F. 18, 21, NaOH (.29 gms/1.). Boil NiAcs-i-*FeNHiC (4.6 gms. each/liter). 15 min.. 148 F. 19, 20, 22, 23 NaOH (.29 gms-71.)-- do. do 15 min 17e-180 F. 24 NaOH (.29 gms/1.).. do CrAc3-i*FeNH4C (4.6 gms. each/liter). 15 min.. 175 F.

NaOH (.32 gms/1. do. NiAca-I-*FeNH-iC (4.9 gms. each/liter) 15 min-- i60-170 F. NaOH (see Note 1 do NI-LOH (see Note 2) 15 min-- 200-212 F NaOH (.30 gms/l.) do. *NHiC (9 gms./liter) 30 min-, 20D-212 F. 402 NaOH (.30 gms/1.).. do *NHiC (9 gms/liter) 15 min 20D-212 F.

TABLE 5 Procedure C Procedure D Specimen Solution Time Temp. Solution Ttme Temp.

MgAci (8.9 gms./1.) min 212 F Liq. G.T.W. (28 cc./liter) 1 hr ltNigAcz (9.4 gms/1.) 30 ruin Boil.. MgAcz (9.4 gms/1.)

MgAc2 (9.8 gms/1.) N

NOTE 1.-NaOH of a concentration to require 2.1 cc. of 0.1 N HC1 NOTE 2.NH4OH of a concentration to require 4.80 cc. of 0.1 HC1 to neutralize a 25 cc. sample of solution. to neutralize a 5 cc. sample of solution.

With reference to the specimens included in Tables 4-6, attention is also directed to the specimens of other examples Where various other pretreatments also gave very 12 EXAMPLE 3 Interspersed with other specimens that were treated in accordance with this invention were various specimens that were not so treated. For these various specimens the 5 pretreatment, if any, is set forth in Table 7 while any gOOd results aS OHOWS- further treatment and the test results are set forth in Table 8.

TABLE 7 Specimen Alloy Pretreatment Artificial Oxide Coating Degree-se, bright dip 1 hr. sulfuric acid electrolyte.

5457 Caustic etch, bright dip. min. sulfuric acid electrolyte. 3003 Only 40 min. chromic acid electrolyte. 3003 Only Iridite dip. 5457 30 sec. caustic etch, water 30 minutes sulfuric acid electrolyte,

rinse. 5052 Degrease, l min. caustic etch, 2 min. Do.

bright dip. 3003 See Example See Example 20. 3003 do Do. 3003 do Do. 1100 1% caustic etch 45 min. sulfuric acid electrolyte. 1100 .do Do.

TABLE 8 Specimen Further Treatment Erllapsed Corrosion Test Results une 27 30 min. boiling water seal 15 min Noticeable attack.

l 4 hrs All coating dissolved. 28 Boiling water seal Oxide coating partially remove.

Very severe attack. 29 do 2 m1n Noticeable attack.

Very severe attack. 30 None Coating removed in minutes. 31.. 20 min. boiling Water se Very severe attack.

32.- do Severe attack.

217. 15 min. boiling water seal. 2 min- Do.

221 do hrs Noticeable attack.l

44 hrs Considerable attack, some coating still on specimenl 222 .do 2% mln. Coating dissolved..z

637 Boiling water seal 1% min. Very severe attack.

640 do 1 min DO.

1 Oxalic corrosion test (room temperature). 2 Oxalic corrosion test (boiling).

From the results shown in Examples 2 and 3, it is to be noted that a very wide variance of pretreatment is possible without having an adverse effect on the final product, it being noted all the specimens of |Example 2 (except 401,

402), including those subjected to a bright dip during the 50 pretreatment, withstood the corrosion test for at least 20 hours with no noticeable attack. These results are to be contrasted with those set forth in Example 3.

As will become more apparent hereinafter, treating specimens in accordance with this invention substantially enhances the corrosion resistance of aluminum over and above that obtained in treating specimens in accordance with this Example.

EXAMPLE 4 A series of specimens were provided with an articial oxide coating by various electrochemical and chemical procedures and then subjected to various treatments of one or more of Procedures A-D. The type of aluminum oxide coating is set forth in Table 9 for some of these specimens while the treatment intermediate the oxide coating of Table 9 and the corrosion test is set forth in Tables 10 and 11. With reference to Specimens 408, 409 they were given a Iridite coating by exposing to the solution for ve minutes and then rinsed in water.

TABLE 9 Elasped Specimen Alloy Oxlde Coating Corrosion Test Result 33 3003 Iridite dip 19 Considerable pitting but coating intact. 34 3003 Alrok #4 4 No attack.

30 Very severe attack in center areas. 35 3003 Hard C0at l5 No noticeable attack.

20 Severe pitting. 36 5052 ...do l5 IN o noticeable attack.

20 Slight pitting. 37 5457 do 8 Only attack on scratches.

15 Do. 20 Very slight pitting. 38 7075 do-. l5 No noticeable attack.

20 Slight pitting one side, none on other. 20 Medium amount of medium size pits.

4% No attack. 6 No attack one side, series of pits in a scratch on other.

TABLE Specimen Procedure A Procedure B 33 NaOH (.30 gms/1.). 10 min 212 F NiAcB-l-*FeNHtC (4.3 gms. each/liter) 15 min 12o-145 F. 34.... 2 gms/1.)- 10 min. Boil NiAcri-*FeNHiC (4.9 gms. each/liter) 15 min. 160-170" F. 35..-. 9 gms./l.) 10 min. None. 36.... rngs./l.) `10 min. 37.-.... gms./l.) 10 min. 38.- gms./l.) 10 min.-. 39 gms/1.). 10 min. 405. gms/l) min 406. gms/l.) 15 min 407... gms./l.) 15min-.. 408, 409 (NaOH-.32 gms/1. plus 15 min NILDE-7.5 cc. conc./ liter).

TABLE 11 Specimen Procedure C Procedure D Na2S103 (55 Cth/liter) The specimens of Tables 12-14 illustrate other variations in the method of forming an aluminum oxide coating on aluminum articles prior to subsequent treatment.

next given the treatment indicated under Further Treatment, rinsed and treated to the solutions of Tables 13 and 14 before the corrosion test.

TABLE 12 Specimen Alloy H2804 Further Treatment Elapsed Corrosion Test Eleetrolyte Time (hrs.)

40 1 hr Heated at 375 F. for 1 hour.-. 12 No attack.

15 Considerable shallow pitting on both sides.

3003 min Foaming agent added 27 Fentrhpits one side, attack in scratches on o er.

5457 30 mln 3 min. chromic acid electrolyte. 17 No attack.

24 Both sides numerous small pits and large area of attack like fingerprint.

3003 24 Medium small pits both sides, 2 large areas of attack like fingerprints.

5052 24 Severe attack that appeared like ingerprints, other, a iew small pits and attack in scratches.

5052 24 Few small pits one side, 3 pits other.

5457 24 Numerous small pits both sides, severe attack on bottom.

5457 do 24 Numerous small pits both sides.

3003 5 min. ehromie acid electrolyte- 20 Few small pits both sides.

3003 do 20 No attack one side, five small pits-on other.

5457 do 15 No noticeable attack.

20 Numerous small pits.

5052 do 20 Very slight pitting.

3003 lrntin. chromic acid electro- 6 No attack.

10 Severe pitting.

3003 do -do 10 No attack.

13 Medium general pitting one side, several general pitting other.

5005 do .do 16 N o attack.

20 Consider-able pitting.

Each of Specimens -52 of Tables 12-14 were cleaned by a pretreatment, then subjected to a sulfuric acid electrolyte for the time indicated in Table 12 and su-bsequently treated to solutions indicated in Tables 13 and 14 before the corrosion test; while Specimens 410, 411 and 412 were degreased, subjected to a dilute caustic etch,

Specimen TABLE 13 Procedure A With reference to Specimens 40, after the anodic treatment it was heated at 375 F. before being treated to the Procedure A solution; to Specimen 41 the foaming agent was added to the electrolyte solution; specimens, the chromic acid electrolyte treatment followed the sulfuric acid electrolyte treatment but was before the treatments set forth in Tables 13 and 14.

and for' the other Procedure B one NiAct+FeNllC (4.6 gms.each/liter) None (4. gms.each/1iter) 15 min.-- 120-140 F.

F. gms. each/litem.. 175 F.

TABLE 14 Specimen Procedure C Procedure D 4o..- MgAcz (8.9 gina/1.) 30 min 212 F Liq. Gfnw. 28 Cannery-.. oil Kassil #1 (28 ee./llter) 4312 44.. 45 l 46.- MgAcz (9 4 gms/l) 47.. MgAcz (9 4 gms/l) 30 mln d 48.. MgAcz (9 8 gms/l 30 ruin 2002l2 F do 49 30 min 20D-212 F do 212 Liq. G.T.W. (56 ec./liter) With reference to the specimens of Tables 12-14, attention is also directed to specimens of other examples as follows:

Specimen: Example 12-15, 18, 21, 24, 25 2 29, 30 3 122, 124 10 104, 107, 108 8 166-169, 172 12 GENERAL COMMENTS 1) Contrasting Specimens 33, 34 with the specimens of Table 3 and 30 of Example 3, it is to be noted that the corrosion resistance of a chemically produced aluminum oxide coating is substantially increased by providing the modified oxide coating of this invention.

(2) As indicated by Specimens 153-156 (Example 1l), 50 and 51, and Specimens 18, 21, 25 of Example 2, and 39 (Example 4), the regular sulfuric acid anodic treatment, the regular sulfuric acid anodic treatment followed by a chromic acid anodic treatment and the hard coat sulfuric acid treatment, each followed by steps that include an activation step and exposing the specimens to the silicate solution give good results, i.e. substantially increase the corrosive resistance above that of the untreated specimens. Also the chromium oxide in the modiment followed by the noted steps which included the NiA3, FeNH., citrate treatment. The manganese did not give the aforementioned effect (see Specimen While the 5457 and 7075 alloys did.

EXAMPLE 5 In order to ascertain which of Procedures A, B, C and D were necessary in order to substantially enhance the corrosion resistance, the series of specimens referred to in Table 15 were tested.

All the specimens of Table l5 were subjected to treatment in accordance with the following steps other than for the steps indicated as omitted in Table 15. The specimens were treated as follows (note adjustment of alkalinity of Example 6):

(a) Flash sulfuric acid electrolyte treatment of one minute.

(b) caustic etch-20 seconds.

(e) NaOH (.32 gm./liter) 10 minutes at boiling.

(d) Thoroughly rinse with water.

(e) NaOH (.32 gms./liter) 10 minutes at boiling.

(f) NiAcg-l-*FeNH4C (4.9 gms. each/liter) 30 minutes, 160 13d- 5 F.

(g) Mg. acetate (9.8 gms/liter) 30 minutes, 200 E+, but not boiling.

(h) Kassil #l (28 `cc./liter) 30 minutes, 200 but not boiling.

ed coating produced by chromic acid in the electrolyte or by the chromium in the starting alloy increases the corrosion resistance. However modifying the aluminum oxide by the steps which include the activation step and treatment in a silicate solution substantially increases the corrosion resistance above that obtained by only forming an unmodified oxide coating by utilizing chromic acid in the anodic electrolyte or an aluminum chromium alloy, see Specimens 28, 29, 31, 32 (Example 3) and 217 (Example 20) where the oxide coating even with a boiling. water seal was severely attacked or dissolved in minutes, and Specimens 18, 21 of Example 2 and 39 of Example 4 where the coating remained intact for at least 20 hours.

(3) As indicated by Specimens 50, 5l, 52 (Example 4) and 35-38 of Example 3, the CR2O3 introduced into the regular sulfuric acid anodic treatment, and the chromium in the 5052 alloy produced coatings by omitting -the NiAc3|FeNH4 citrate, yielded coating as resistant to corrosion as the .ones with the sulfuric acid anodic treat- As may be noted in particular from Specimens 53-57 (which were treated in freshly made solutions), the treatment in a basic solution of Procedure A and the silicate treatment of Procedure D are the two most necessary steps of this invention in order to enhance the corrosion resistance of an articially (chemically or anodically) oxide coated aluminum article. These results cross checked with the treatment of other specimens. Also note Example 3 where specimens were not treated with any of the solutions of Procedures A, B, C and D.

Also as indicated by the results from Table 15 and other examples, the presence of alloying elements such as manganese, chromium and magnesium form a modied oxide during the sulfuric acid anodic treatment and act substantially the same as when these same elements are deposited in the oxide chemically. In manyapplications, the resistance to corrosion obtained with only forming an oxide coating (electrically or chemically), and then utilizing steps of Procedures A and D would be ample.

Another procedure to obtain. a modified oxide coating is to add amphoteric compounds such as manganic, chromic, molybdic, vanadic and permaganic acids, etc. to the anodic solution. This would act in a manner similar to that of the preceding paragraph to increase the corrosion resistance after the specimen was activated with sodium hydroxide and then exposing them to the silicate solution within the approximate concentration and temperature ranges disclosed.

A series of specimens were first degreased, a ash sulfuric acid electrolyte for one minute, caustic etched, a 45-minute sulfuric acid electrolyte, thoroughly rinsed with water and then successively treated to the solutions indicated in Table 16, the' solutions being selected from the following:

(a) NaOH (.32,gm./llter) 10 min (b) NaOH (1.2 gms.lliter)+KAc (9.8 gms/ 30 min liter (c) NiAcVl-*FeNHrC (4.9 gms. each/liter).. min 160 F.

(d) *FeNH4C (9.8 gms/liter) 30 min. 200-212 F. (e) MgAc2 (9.8 gms./liter) (i) KAc (9.8 gms./liter) (g) Kassl #1 (28 cc.lliter) (h) NaOH (.3 gm./liter).. (l) KA Boiling.

TABLE 16 Time elapsed Corrosion Test Results B i s.)

No attack.

Only slight edge attack.

No attack.

Only attack on bottom edge.

No attack.

Pitting on bottom and side edge.

Above somewhat Worse, 6

small pits on side.

No attack.

Few very small pits one side, no attack on other.

No attack.

Only slight edge attack.

No attack.

2 small pits.

Only 2 very small pits.

Pitsllecame larger.

o. No attack. Attack along scratches and one side, small pit 4on both sides. No attack.

Pitting on bottom and side edge. g Abovel and 2 pits on one side.

A lo M A occa E Severe pitting on bottom and side edges, 2 medium large areas and 3 small pits on side but none on other. No attack.

Attack in scratches.

Verysevere attack on both sides.

No attack.

No attack on side, slight attack in scratches on TABLE lf3-Continued Speei- Procedures Time A en elapsed Corrosion 'lcst Results A B C D (hrs.)

11 Edge attack increase on side, medium attack in scratches and severe attack at one side on other.

418 k i g 14 N o attack. 419 A k i g 11% D0.

2 areas where coating dissolved one side and most of coating dissolved on other.

420... k c i g 7% No attack. l

8% Slight attack in scratches one side, no attack o er.

11 Slight attack in scratches one side, no attack other, plus medium small pits, numerous smaller pits other.

13 Same as above.

Prior to the pretreatment of specimens of Table 16 it was noticed Specimens 66-68, 70 had visually noticeable rmetal defects while Specimens 65, 69 and 7l, 72, 73 did l1101;.

Treatment with the basic solution 0i Procedure A followed by the silicate treatment of Procedure Din some tests give as good results as using the additional treatments of Procedures B andC; however, the results are not as consistent as when at least one of Procedures B and C are used. Further as exemplified by Specimens 67, 73 attempting to combine Procedures A and C does not give as good results as where thespecimens are separately treate-d to solutions of Procedures A and C.

Also asmay be noted, potassium acetate and mag/- nesiuin acetates intermediate thetreatments of Procedures A and D give good corrosion resistance and which is somewhat better than where only a ferrie ammonium citrate treatment is interjected between Procedures A and D.

EXAMPLE 6 Fresh solutions were madeup and a titration of a 25 cc. sample of each solution was made to determine the alkalinity before treating Specimens 53-57 as described in Example 5. The initial alkalinity being indicated in Table 17 by the amount of .5 normal HCl required'to neutralize the respective solution under the heading initial alkalinity, and after treating these specimens, the amount required to restore the solutions to the original alkalinity under the heading restore alkalinity. These specimens were pieces of strip aluminum, the area of each surface exposed to treatment being about 3 by 5" and the area subjected to the corrosion test being about 3" by 2" to 2% each/liter) MgAc2 (9.8 gms/liter) .1 cc. of .5 NHCl. 6 ce. of 1% NaOH. Kassil #l (28 cc./liter) 1.1 ccl of .5 NHCl... Left asis.

Samples (25 cc.) of solutions used for treating Specimens 75 and 80-90 of Example 7 plus' two others were titrated to neutral to ydetermine the original alkalinity, then the solutions used for treating specimens (area of each specimen ybeing about the same as given above) and again titrated, the alkalinity being expressed in .1 NHCl in Table 18.

TABLE 18 Solution Initial Number oi Specimens Restore Alkalinity Final Alkalinity Alkalinity NaOH 1.85 cc 74, 75, 86, 87 6 cc. 1% NaOH 1.9 NarCOq 84, 85, 89 and one.- Dilute 1-1.. 75 NH4CO3 80, 81, 90 and one..

aqP 4 g2, 83 (lgilutlety 1-llT...I 6 NiAc FeN H C- specimens. cc. a p MgAtZJ.F 4. do 10 cc. oi 1% NaOH. .15 Kassil #1 12.15 cc- 15 specimens Left as is Solutions that had their alkalinity restored as per the column under Table 18 were used for Specimens 76 and 77 (Example 7) and then their alkalinity restored as indicated in Table 19. 15

5 min. 1Z0-140 F.

NazCO3 (.26 gm /l1ter). NiAc3 (5 gms/liter) 15 min.- 1Z0-140 F. Na2S1O3 (liq. G.T.W. 28 c 1 hour 240-2l2 F.

While Specimen 92 was degreased, bright dipped and TABLE 19 subjected to a iO-minute sulfuric acid electrolyte and then 0 treated to solutions as follows:

Solution Specimens Restore Alkalinity Final Al- Na PO (.30 gm./l1ter) 5 min 1Z0-140 F. Tested kahmty Ilrdnae. (28.9 gms/haten) 145 m1n. g c2 9.5 gms. ier 5min i. III); 2070117 NaOH"" 1' 95 Na1s1o3 (uq. G.'1.W.2sec./1iter) 1 hour 212 F. N'Ac ii c. 76,77 zee. ofifyNaoH- Migliaia- 4 76,77 ecc. or 117g NaOH- .1 25 Speclmen 93 was treated the same as Specimen 91 Kassll #1 76 77 133 other than the Na2CO3 and NiAc3 solutions were each at room temperature. The results of the corrosion tests were as follows:

30 Specimen 91...- hours isi? gitaar. k

OlllS. g 8. 8.0 EXAMPLE 7 Specimen 92..-. 10 hours.. No attack except where specimen touched testing rack. 20 hours More numerous and larger pits than In order to determine the eifectiveness of varying the tiempl 11)1 also basic solution of Procedure A, the specimen of this exspecimen 93.--- ahours No attack.

3% hours Considerable attack.

ample that appear in Table 20 were all treated as per steps (a), (b), (c) and (d) of Example 5. The additional steps referred to in the table below are the same as the steps set forth in Example 5. With reference to the Also note that KOH can be used, see Example 17. As may be noted above the NaOH activation step gave the best results with NH4CO3 second best; whereas activation step of this example, the treatment was for 10 40 for Specimens 82-85, 88 and 89 it was noted that the minutes at boiling. respective activating solutions did attack the aluminum TABLE 20 Specimen Alloy Activation Step Additional NaOH Cor- Remarks Steps rosion Test 74 3003 NaOH, .32 gm./1 f, g, h 5 hrs Only edge attack. 16 hrs-. Only severe edge attack.

75 3003 NaOH, .32 gm./l l1 5 hrs Only edge attack.

16 hrs Severe edge attack, 13

medium to small pits one side, none other.

76...` 3003 NaaPO, .56 gm.ll. 1, g, h l? hrs... IS\T0 attack. k

412 hrs. evere attac 77 3003 NaOH, .32 gm./l. 4% hrs No attack. 78 3003 NazCOa, .25 gin/l.. Do.

Brown spots noticeable. Pitting started in brown 3003 NaOH, .32 gin/1 7 hrs.

24 hrs.- 3003 NHCOJ, .48 gnL/l i, g, h 15 min 2% hrs.

3003 NHiCOe, .48 gm.[l h 15 min 21/12 hrs..

3003 Na3PO4, 1.2 gms/l i, g, h 15 min 45 min 3003 NaaPO, 1.2 gmS./l. h 15 and 45 min. 3003 NazCOa, .50 gm./l f, g, h 15 and 45 mln. 3003 NazCOx, .50 g1I1./l h 15 and 45 mln. 86 HiP NaOH, .32gn1./1 !,g,h 5% hrs- 15 hrs.

y87 HiP NaOH, .32 gm./i h 1% hrs-...

3% hrs- 88 HiP Na3PO4,1.2 gms/1 i, h 1V hrs 11?) Naicoa, .so gm./1 f, n.-. ll/ hrs..

4% hrs k Severe attack in scratches,

slight pitting.

No attack. Very severe pitting and attack in scratches.

No attack. Severe attack.

Same as 82.

Edge attack, 2 medium pits ou both sides, edge attack.

. 10 large pits one side; fairly Severe attack one side, none other Severe attack one side, severe pitting on other.

. Severe attack.

oxide coating produced by the sulfuric acid electrolyte step. However as may be noted from specimens of Example 3 which had only a chemically or anodically aluminum oxide coating with or without being sealed in boiling water, the Na2CO3 and Na3PO4 activation plus further treatment that includes subjecting the specimen to a silicate treatment provided increased corrosion resistance. That is, although Specimens 82-85, 88 and 89 at the interval indicated were severely attacked, the modied coating had not been completely removed.

Also to be noted from Specimens 76 and 78, lower concentrations of Na2CO3 and Na3PO4 did give better results than higher concentrations of the respective activation solutions.

Further, the high purity specimens (HiP.) that were provided with the modied coating which included Steps (f), (g), and (h) had good corrosive resistance but not as good as the 3003 alloy given the same treatment. Also the high purity specimens that were treated as per the preceding sentence had better corrosion resistance than those in which Steps (f) and (g) has been omitted. This shows that the presence of the iron group metals such as manganese and chromium substantially increases the corrosion resistance. These iron group metals can be incorporated into the tinal product in three ways:

(1) Exposing the specimen to a solution such as NiAc3-l-FeNH4 citrate.

(2) As a constituent in the starting alloy.

(3) As an acid radical in the electrolyte used to provide the aluminum oxide coating in the specimen.

Regardless of the above manner of incorporating the iron group metals, the activation step plus the silicate treatment step substantially enhances the corrosion resistance over and above that obtained without said steps.

EXAMPLE 8 (c) NaOH (concentration of Table 21) for 10 minutes` at boiling.

(d) NiAc3-i-*FeNH4C (4.3 gms. each/liter) for 15 minutes at 175-180 F.

(e) MgAcZ (9.4 gms/liter) for 30 minutes at boiling. (f) Kassil #l (56 cc./liter) for 30 minutes at boiling.

TABLE 21 Specimen Alloy NaOH Elapsed Corrosion Test Results Cone. ime (gm/l.) (hrs.)

94 5 5457 1. 6 Coating dissolved in solution e.

96 1 3003 1.0 50 l small pit, no attack other side.

97 1 5457 1. 0 15 large areas of severe attack as if coating dissolved, other areas no attack.

5457 53 50 2 pits one side, 6

medium and few small pits on other.

3003 53 50 2 small pits one sidey 2 :medium pits on other.

5457 53 50 No attack.

5457 53 50 A few small pits ou both sides.

5457 .32 48 N o attack except at edges.

5457 25 3 Attack between splotches, otherwise OK.

5457 .20 3 Attack around edge and ottom.

3003 l 3 Very severe pitting attack over entire area.

3003 10 3 Similar but somewhat better than 106,107.

22 (1) Specimens 96, 97 Reddish color on lower part after treatment to solution (d). (2) Specimens 100, 101 Reddish color after treatment to solution (d). Subjected to a 3 minute chromic acid electrolyte just after the sulfuric acid electrolyte treatment.

Subjected to a 5 minute chromic acid electrolyte just after the sulfuric acid electrolyte treatment.

Not treated per Procedures B, C and D.

(3) Specimens 104, 107

(4) Specimen 108 (5) Specimens 94, 95

NaOH (.15 gm./liter) 5 mi'n 120l40 F. NiAci (8.9 gmJllter) 15 miu 1Z0-140 F. Na2S1O3 (lq G.T.W. 42 cc./1iter) 1 hour 212 F.

After 101/2 hours in the corrosion test both sides were attacked only in the scratches on the specimen.

For each of the specimens of Table 21, an observation was made at the end of the treatment in solution (c) to ascertain the effects of Varying the concentration 'of solution (c). For each of Specimens 102, 103 (.32 gm./liter) and 104 (.25 gm./liter) it was noted that the anodic aluminum oxide coating was lSeverely attacked by the activating solution while Specimen 105 (.20l gm./lter) showed considerable attack and 106-108 (.10 gm./liter) had no noticeable attack. With 1.6 gms/liter, the tanodic aluminum oxide coating was completely dissolved, i.e. Specimens 94, 95. This shows that the concentration of the activating solution and the time and temperature cannot be such that the aluminum oxide coating is too severely attacked; otherwise `the subsequent treatment steps will not provide best results. On the other hand, too low a concentration (considering time and temperature), apparently does not properly condition the aluminum oxide coating for further treatment. The aforegoing is to be taken in consideration with the corrosion test results after the specimens had been treated as per Procedures B, C and D which indicated the finally treated specimens, Specimens 96 and 98-103 had the best corrosion resistance. Also contrast Specimen with 96. Using NaOH for the activating solution of approximately .3-.55 gram/liter at boiling for 10 minutes has given best results while .l5-1.0 gram/liter normally give satisfactory results. Of course the thicker aluminum oxide coating can withstand 'a more severe attack by the actuating solution than a thinner coating.

As to the observations after the NiAc3+FeNH4 citrate treatment it was noted that each of Specimens 102-104 were discolored While there was no discoloration of Specimens 106108. In order to avoid this discoloration the NiAcBl-"eNHqE citrate solution should be kept at about i60-170 F.

With reference to the edge attack on Specimens 103, and 108 before the activation treatment, the edges were sponged with degreaser to attempt to remove fingerprints. On the basis of this, it appears that some of the edge attacks and attack in the form of splotches results from handling the specimens and points out that the specimens should be kept free of oil and grease during the treatment thereof.

A series of specimens were pretreated and subjected to a sulfuric acid electrolyte treatment for 40 minutes. They were then treated in a NaOH solution of .30 gm./liter for 5 minutes at various temperature ranges of 1Z0-140 F., -180 F. and violent boil respectively, next treated to a solution of NiAca-l-FeNHi citrate (4.3 gms. each/ liter) for 15 minutes at 1Z0-140 F. and then to liquid 23 GTW silicate (56 cc./liter) for one hour at boiling. The specimens were subjected to a corrosion test, Specimen 110 which was treated in a violent boil solution showed no attack at 4 hours, and 2 small pits together with some edge attack at 21 hours. Specimen 111 which was treated 5 attack in scratches, and after 81/2 hours numerous small pits and attack along scratches on both sides.

With reference to the specimens of Table 22, all of these specimens were rst degreased, given a 30-second caustic etch, rinsed in water, treated in a sulfuric acid electrolyte for 30 minutes, rinsed in water and then treated in the solutions as follows:

(a) NaOH (.29 gm./liter) min Boiling.

(b) NiAcrl-*FeNHiG (4.6 D each/litor) min 140-100 1". 0 (c) MgAcz (9.4 gms/liter) 30 In' Boiling.

(d) Silicate solution (5G (zc/liter) 30 min Do.

Afterwards these specimens were subjected to the corrosion test.

TABLE 22 Specimen Alloy Type Silicatc Elapsed Corrosion Test Results Time (hrs.)

5457 Slight pitting one side, no attack other. 3003 20 Few small pits. 5052 20 1 pit one side, 5 on other, scratches not pitted. 14 No attack. 15 Some pits noted. 115 5457 S-35 20 Numerous small pits.

14 No attack. 15 Some pits. 116 3003 S-35. 20 A few small pits, scratches pitted.

14 No attack. 15 Sorne pits. 117 5052 S-35 20 Severe pitting in scratches some attack on coating. 118 5457 Kassil #1 20 Fcwhsmall pits one side, none o er. 119 3003 20 1 pit one side, 2 on other, scratches not pitted. 120 5052 ...do 20 No pits one side, 3 other, scratches not pitted.

kept the 110 specimen surface free from bubbles and it better withstood attack. This shows that during the activation step, the solution should be either mechanically or otherwise agitated to keep bubbles from clinging to the article being treated to obtain best results.

EXAMPLE 9 NaOH (.15 gin/liter) NiAca-l-FeNH4C (4.3 g MgAc3 (9.4 gms/liter) The specimens then treated with liquid GTW, a sodium silicate showed less attack than those treated with Metso 99 silicate which in turn were better than those treated with G.D. silicate or sodium metasilicate. With reference to the specimen treated with liquid GTW silicate, after 7 hours in the corrosion test it had Slight pitting and slight As shown by Specimens i12-120, Kassil #l gave better results than N-silicate and S- silicate. Although some of the pits in these specimens may have occured at defects in the specimens, the overall performance of Kas- 0 sil #1 is somewhat better than that of N-silicate and considerably better than the S-35 silicate. In this connection it is noted that the respective ratios of percent and per-cent Na2O1SiO2 for Kassil #1, N-silicate and S-35 silicate are respectively 122.50; 1:3.22; and 113.75.

EXAMPLE 10 A series of specimens were treated in order to determine the effects of varying the concentration of the silicate solution, these specimens being set forth in Table 23. The specimens of Table 23 were treated the same as those of Table 22 except that only Kassil #1 of the concentrations of Table 23 were used and for specimens 124 and 122 a foaming agent was added to the electrolyte solution before the anodic treatment.

TABLE 23 Specimen Alloy Silicate Elapsed Corrosion Test Results Concentration Time (hrs.)

121 5052 14 cc./1iter 27 A few pits and attack in scratches,

7 pits other. 122 5457 27 A few pits and attack in scratches,

` other few medium size pits. 5052 28 cc./liter 27 5 pits one side, other 5 pits and attack in scratches. 5457 .do 27 Medium number of pits one side,

l numerous pits on other. 5052 5G cc./hter 27 No attack one side, few small pits on other. 5457 .do 27 A number of pits one side, 4 pits on scratches on other.

25 A concentration of 56 oc./liter of Kassil #1 was somewhat better than 14 cc. and 28 cc. per liter.

EXAMPLE 1 1 The specimens of this example are primarily directed to variations in Procedure B. However various specimens of the other examples show additional variations in` this procedure (in particular note Example 21 and the ones that follow), and some of the specimens of this example TABLE 24 Specimen 127 45min 7 131 40min 132 40min 133 40min 135 40 min Time Elapsed (hrs.)

Electrolytic Treatment Alloy Corrosion Test Results Negligible attack.

Localized attack at what appeared to be fingerprints on specimen and some small pits.

Only one side had attack in scratches and in one area that appeared touched by ngers.

Numerous small pits on both sides.

N o noticeable attack.

Considerable attack and pitting.

No attack.

Practically none.

Slight attack on both sides.

Medium pits both sides.

No attack.

Has consi erable attack.

No attack.

Considerable attack.

N o attack.

Practically none.

Slight pitting on one side, severe pitting on other. l

Numerous pits on one side, severe pitting on other.

No attack except edges where touched testing racks.

Some medium size pits also large areas no attack.

Few small pits one side and few pits and attack along scratches on other.

5457 40 min 13 Same except more attack along scratches.

3003 1 hr 13 Slight attack.

20 Numerous small pits both sides, large areas of no attack.

5457 1 hr 13 No attack.

2O Large areas no attack, large pits one side smaller pits on other.

5457 1 hr 13 No attack.

20 Large areas no attack, large pits one side; sliraller pits on other; 134 better than 3003 40 min. 20 4 pits one side, noue other, attack in scratches.

5457 40 min 20 Very few small pits except on sides and bottom.

5457 40 min 19 A few small pits and attack along scratches.

5457 1 hr 15 Few scattered pits on one side, numerous medium size pit-.s and attack in scratches.

5457 40 min- 20 Small pits more numerous than 143 but still very good.

3003 30 min. 20 Numerous shallow fine pits on both sides.

3003 30 min. 20 Medium number oi pits on both sides.

5052 30 min 11 One side severe attack along scratches other medium number of small pits.

5457 30 min- 20 Some attack on side otherwise no attack.

5457 30 min- 20 Only small area attack on one side.

5457 30 mn 11 Medium number of pits both sides.

5457 30 min 20 Very slight pitting, appeared associated with dent in sample.

3003` 30 min- 20 Same as 153.

5052 30 min. 20 o.

5052 30 min- 15 No noticeable attack.

20 Numerous small pits.

5052 30 min- 20 Slight pitting near bottom otherwise OK.

3003 30 miamA 2o D0.

5457 30 min- 20 Slight pitting, except where touched by fingers on one side.

5457 30 min 32 3 small pits one side, very few pits on other, very good condition.

5052 30 min 32 3 small pits each side, otherwise no attack.

3003 30 min. 32 No attack except where touched in corner.

45 min. 3 No attack.

4% 2 pits and attack in small scratch one side, no attack other. 6 Same on said one side, 1 pit other. 11 5 pits and edge attack one side, 3 pits 5 other 1 5 large 'pits and edge attack one side, 3

pits other.

TAB LE 25 Specimen Procedure A (Activation) Procedure B 120-140 F..... NiAe3 (5 gms/liter) 212 F NiAc3 (8.9 gms/liter).

NiAc3 (8.9 gms/liter). CrAc3 (8.9 gms./1iter).. CrAc3+K2C2O7 (4.5 gm

NaOH (.30 gms/l.) NaOH (.30 gms/l.) 10 min- NaOH (.30 gms/1.) 5 min.. NaOH (.30 gms /l l min- ZnAcg (4.1 gms/liter) NaOH (.30 gms /l min- NiAca-i-*FeNHiC (4.3 gms. each/liter) NaOH (.28 gms /l min NiAea-I-*FeNI-LC (4.6 gms. each/liter) NaOH (.28 gms /l 10 min NiAca-l-*FeNH4AcQ (4.6 gms. each/liter) NaOH (.28 gms/1.)... 10 min. NiAeg-i-*FeNlLC (4.6 gms. each/liter) NaOH (.28 gms/1.)... l0 min -.-do NaOH (.28 gms/1.)... 10 min. 10 min. 10 min. 10 min- 10 min. 156 10 rnin 157,158,159... NaOH (.28 gms/l 10 min- 160, 161, 162... NaOH (.28 gms/1.) 10 min. 404 NaOH (.30 gms./l.) 10 min Boil 15S-165 F.

TABLE 26 Specimen Procedure C Procedure D 127 None L'iq. G.T.W. (28 cc./liter) 212 F 128 M 212 F. Liq. G.I.W. (42 ze/liter) 212 F 129-. -..d Na2SiO2 (56 c2c/liter) 212 F 130.. G.T.W. (28 cc./liter). 212 F 131.. d 212 F 132 212 F 133...- 212 F 134..-. 212 F 135...- 212 F 136 212 F 137 212 F 138-.-- 212 F 139.-.. 212 F 141.. 212 F 142.. 212 F 143.- 212 F 144.. 212 F 145.. 212 F 146...- 212 F 147,148- B011, 149.... 0. 150.. 212 F. 151.- 212 F 152-- Boil. 153.- 212 F 154.. 212 F 155.. min 212 F 156 30 min. 212 F 157,158,159- None 30 n1in BoiL 160, 161, 162... MgAcg (9.4 gms/ iter) 30 min.-. 250 F. 404 MgAcz )9.8 gms. 30 min 20o-212 F.

COMMENTS acetate increases the corrosion resistance; while otherv tests indicate that varying the time of treatment in the NiAc3-l-PeNH4 citrate solution between 15 and 30 minutes, the SO-minute treatment increased the corrosion resistance of the final coating over that of the 15-minute treatment.

(4) A comparison of Specimens 157-159 (NiAc3 +FeNH4 citrate at 158-162 F.) versus 19-20 of -Example 2 (same solution at 178-180 F.) shows that the higher temperature, ie. 180 F. gives better results than the lower ternperatures. However increasing the temperature from 180 F. to boiling did not improve the corrosion resistance, compare Specimens 98, 99 (Example 8) with Specimens 100, 101 (Example 8); and that the coating was discolored, Specimens 100, 101 having a reddish color after being treated to the boiling solution of NiAc3+FeNH4 citrate. In view of the fact that for some applications, discoloration is undesirable, it is desirable that the solu- 29 tion of Procedure B be at a temperature of about 140- 180 F. and preferably about 15S-165 F.

Test such as for Specimens 12, 14, 15 (Example 2); 145 (Example 11) and 192 (Example 14) indicate that using ZnAc2 and MgAcZ, or NiAc3+FeNH4 citrate and BaAc2 in the solutions of Procedures B and C respectively, does produce accepted results but not as good as the NiAc-l-FeNI-L citrate and MgAc2 treatment. With reference to using MgAc2 in Procedure C also note Example 14. Further using NiAc3+FeNH4 citrate gives more consistent results than using either of these compounds by itself in Procedure B.

EXAMPLE 12 A series of specimens all were first degreased, given a caustic etch and then a Lt5minute sulfuric acid electrolyte treatment. Specimens 166, 168, 169 were additionally then subjected to a 3-minute chromic acid electrolyte while Specimens 167, 172 were subjected to a live minute chromic acid electrolyte. The specimen alloys were as follows:

After the electrolyte treatment each specimen was treated in NaOH solution of a concentration of .26 gm./ liter for minutes at boiling and just before the corrosion test to a silicate treatment of Kassil #l (28 cc./liter) at boiling for 30 minutes. Intermediate the activation treatment and the silicate treatment the specimens were treated as follows:

TABLE 27 Specimen Further Treatment; Time Temperature (min.)

163, 164, 165... NiAca-/I-lleNHr; citrate (4.9 i60-170 F.

gms. Than MgAc2 (9.8 gms/1 30 Boilmg. 30 Do.

30 Boiling.

EXAMPLE 13 The series of specimens of Tables 28 and 29 Were tested to in part determine the effect of Varying the temperature of the Procedure `C solution. All of these specimens were first pretreated and subjected to a 45-minute sulfuric acid electrolyte treatment, specimens 173 and 177 then being sealed in boiling water for 15 minutes. All of these specimens were then successively treated in solutions as follows:

NaOI-I (.32 gms/liter) 10 min-.- Boiling. NiAcg-I-*FeNHiC (4.9 gms. each liter) 15 mim--- See Table 28. MgAcg (9.8 gms/liter) 30 min Do. Kassil #l (56 cc./liter) 30 min-... Do.

Time Specimen Alloy Eapstd Corrosion Test Results v rs.

Medium pitting on both sides.

No attack except on bottom adge. Large areas of severe attack in of width of specimen, the other t no attack. Some areas of severe attack on 1X; oi

etmen, no attack on other Numerous pits 'one side, few pits o n other side near bottom and No attack one side, 2 small pits on other.

1 small pit one side, other one small pit and slight attack in scratches.

Medium pitting one side, slightly less on other.

It was noted a heavy gelatinous coating was on the specimens 1754177 which were heated in MgAc2 at 250 F. This coating was removed by steel wool on both sides of and one side each of 177 and 176. In other tests some gelatinous coating was noticed when `treating specimens to a `boiling solution of MgAc2 but not nearly as heavy as when heating at 250 F. These examples clearly indicate that the gelatinous product from the MgAc2 treatment affects the corrosion resistance and must be avoided to yield good results. However note Specimens 160, 161, 162, Example 1l, which show that good results were obtained when heating the MgAcz solution at 250 F. (the NaOH solution of Procedure A also being at 250 F).

With reference to Specimens 178 and 179, no gelatinous precipitate was noticed after the MgAc2 150-l60 F. treatment Ibut considerably for the 180L190 F. treatment.

Some of the pits of Specimen 182 are at the scratches or metal defects.

In View of results such as obtained for Specimens 180- 182 and Specimens of other Examples ZOO-212 F. for Procedure C is preferred.

EXAMPLE 14 In order to determine the effect of using various acetates for procedure C solutions, all the specimens of Table 30, which were a 3003 alloy were first subjected to the steps successively as follows: degreased, flash sulfuric acid electrolyte for one minute, and a 15-second caustic etch, a sulfuric acid electrolyte for 45 minutes, rinsed in water and then to an activating solution of NaOH of about .32 gram per liter of water at boiling for 10 min- 3l utes. The further treatment and the results of the 1.0% NaOH corrosion test are as follows:

sequent precipitating of the silica was noticed in the areas attacked.

TABLE 30 Specimen B C D Time Results 183 NiAc-*FeNHiC MgAcz. Kassil #1 1% hrs Severe attack similar to that as when Ac solution to acid. 184 NiAc-l-*FcNII4C KAc do 1% hrs No attack.

5% hrs 3 pits one side, 1 pit other, attack in scratches.

8 hrs 3 pits one side, 2 pits other, attack in scratches.

17 hrs 8 large pits, 15 small one side, 3 large pits, 3 small other, severe attack in 2 scratches.

185 NiAc-l-*FcNIhC CaAcz .do 1% hrs No attack.

5% hrs 5 pits and 2 scratches attacked one side, 2 pits other side.

8 hrs 13 pits and 3 scratches one side, 6

pits and one scratch other side.

17 hrs 20 pits one side, 15 pits other, severe attack in scratches both sides.

186 NiAc4-*FeNH4C BaAC2 do 1% hrs No attack.

5% hrs Only attack in scratches on one side.

8 hrs 4 small pits one side, severe attack in scratches both sides, severe attack in scratches and near edge.

17 hrs 10 pits one side, sever attack in scratches and near edge.

187 NiAc-l-*FeNH4C ZnAcz do 1% hrs No attack.

5% hrs Only small pits on one side.

8 hrs Number of small pits one side, 2 large areas of attack and a number of small pits on other side.

17 hrs Very severe attack and pitting on both sides.

188 NiAc-l-*FeNH4C MgAcz Boiling H2O seal 10 min. No attack.

17 min Very severe attack. 189 NiAc-I-"FeNHiC MgAcz No attack. 190.. NiAc-l-*FeNI'hC MgAcz Do. 191 NiAc-f-*FeNH4C KAc Do.

With reference to Table 30 the concentrations of the solutions and other conditions other than those given above are as follows:

Specimen 192 (5457 alloy) was degreased, caustic etched, bright dipped, subjected to the sulfuric acid electrolyte treatment for one hour and then treated to the solutions as follows:

Concentration Time Temperature NaOH (.30 gms/liter) 10 min 212 F. NiAcg-l-*FeNHiC (4.3 gms. cach/liter) min 12o-140 F BaAcg (8.1 gms/liter) min 212 F. Liq. G.T.W. (56 cc./liter) 1hr 212 F.

At 7 hours in the corrosion test, Specimen 192 had numerous small pits on both sides.

With reference to specimens 184-187 the KAc treatment gives the best results followed by CaAc2 which specimen had attack in the scratches but considerable parts unattacked. The ZnAc2 specimen had the poorest corrosion resistance of this group, there being considerable pitting and large areas attacked.

With reference to specimen 186 it is thought that BaSO4 could have precipitated and prevented the Kassil #1 properly reacting with the aluminum oxide. With reference to the relatively early attack in the scratches of these specimens and others, a possible explanation is that some acid from the electrolyte treatment was retained therein `and caused the precipitation of silica from the silicate treatment step.

Prior to carrying out Step C of the example, it was determined that each of the KAc, CaAc2 and BaAc2 solutions were slightly acid while the ZnAc2 solution was very acid. The areas of attack on the ZnAc2 specimen were typical of an acid condition wherein the sub- With reference to the severe attack of Specimen 183 in a short time, it was unexplained. In running another test with a specimen from the same sheet of metal as Specimen 183 it was noted the surface was dull before pretreatment which indicates that there was present a rather thick film of aluminum oxide from atmospheric oxidation. It appears probable that this thick lm was on Specimen 183, and not completely removed prior to the anodic treatment.

Also with reference to Specimen 183 the MgAc2 solution used had `been previously used. Accordingly before Specimens 189-191 were treated as indicated above, the MgAc2 solution that had been used for Specimen 183 was first titrated with suicient NaOH solution to make the solution basic and this titrate solution was used for Specimen 190, while freshly made MgAcz solutions were used for treating specimens 188 and 189. Also a new solution of NiAc3-l-FeNH4 citrate was used, which was basic for Specimens 188-191. As indicated in the results for the last mentioned specimen (contrast to Specimen 184); in order to obtain maximum corrosion resistance, it is important that the solutions of Steps B, C and D of this example be slightly basic or near basic as possible. In this connection it is to be mentioned that the titrated solution of MgAcz was used for Specimen 190 and the freshly made solution for Specimen 189 for the time period of the corrosion resistance test of the modified coating gave the same results. Further to be noted is that the KAC (Specimen 192) gave the same results as MgAc2 (Specimens 190 and 191).

Based on titrations of the NaOH activating solutions and the Procedure C solutions for Specimens 184486 and 188-191 (both before and after use), it was noted that in treating these specimens, these solutions became more acid. In this connection note Example 6.

With reference to Procedures B and C of this example, after using the respective solution (those given or solutions used in place thereof), the pH should be adjusted to bring it back to the pH of a freshly made solution, by, for example, adding NaOH, in order to obtain best results (also see Examples 6 and 15). However if the same solutions are used many times with readjustment of pH, it is to be understood that additional solute will have to ibe added since the concentration thereof is re- In order to determine the effect of reusing a solution, Specimens 193 and 194 were treated the same as Specimens 190 and 191 of the preceding example other than as noted.

TABLE 31 Specimen `Diierence Time Results (hrs.)

193 Usedsolution of 27 No attack.

MgAcz of 190 which was first brought back to original 'alkalinity by adding NaOH.

51 8 small pits one side;

8 very small and 5 small pits along scratch on other s1 e. 194 KAc cf solution 191 27 No attack.

brought back to original acidity and then 5 ce. 1% of NaOH added to one liter of solution.

`51 1 small pit one side;

11o attack on other.

'This Ishows that the acetate solutions may be reused if the alkalinity is controlled and still obtain very good results.

EXAMPLE 16 All the specimens of this example were a 3003 alloy and were successively subjected to a 1minute sulfuric acid electrolyte, a `15-second caustic etch, rinsed in water, a 45-minute sulfuric acid electrolyte and `rinsed in water. The subsequent treatment comprised subjecting the respective specimen to one or more of the following solutions for the time and temperature indicated. The various `solutions used were as follows:

Concentration Temperature 'Time (mini) (a) NaOH (.45 gms/liter) .1 Boiling (i).tNAc3+*FeNH4C (4.9 gm's. each] 160 F l5 l el (cit(9.8 gms. KAc-l-.l gms. Na0H)/ 200 butinot 30 er. g. (d1).t(9.8 gms. KAc-l-.32 gms. NaOH)/ ....do 30 1 er. (e) Do `Boiling (i) Do do.. 10 (g1)i (9.8.gms. MgAcz-l-.IB gms. NaOH)/ 200 30 r. o1 g. (h) Kassil #1 .(28 cc./liter) .do 30 The solutions used in treating the various specimens one side, attack along edge and top of other side, numerous small pits.

The appearance of Specimen 197 after the corrosion test was similar to that obtained when all the electrolyte solution had not been removed prior to further treatment, while the attack on Specimen 198 appeared similar to that where the yconcentration of NaOH of the activating step was too high or the exposure too lon-g. It is to be noted "that combining the NaOH with the acetate solution in an attempt to minimize the number of different solutions that a "specimen is treated with does not give as Igood results as where the specimen is-successively exposed to each individual solution, including an activating solution, separately (contrast Specimens 196 and 197). However the corrosion resistance of Specimen 197 is rgood and was substantially increased. so that it will be ample for many purposes.

EXAMPLE 17 All the specimens of this example were a 3003 alloy and were pretreated as set forth in -the yfirst sentence of Example 8.

The subsequent treatment comprised subjecting the respective specimens to one or more of `the following solutions for the time and temperature indicated below.

Concentration Temperature Time (min.)

(a) KOH, .32 gms/liter Boiling 10 (b) NaOH, .32 gms/liter d 10 (c) NiAc3+FeNH4 Citrate (4.9 gms. l5

each/liter) (d) FeNH4C2O4, 9.8 gms/liter plus 200 but not 15 boiling. (e) Do .,do 30 (f) FeNH4C2O1, 9.8 gms/liter plus added- .do 30 NaOH until turned alkaline. (g) Na2C2O4 NaOH added to H2G2O4 .do 30 until slightly alkaline. (h) KAG, .32 gms/liter plus per gal. of --.do 30 solution, 5 cc. of 1% NaOH/2 liters was added. (i) NaOH, .32 gms/liter do `30 (j) Kassil #1 (28 tac/liter) do 80 The solutions used and the results of the NaOH corrosion test are given below `in Table 33.

TAB LE 33 Specimen A B C D Time Results (hrs.)

199 a d i j 1 No attack.

2 No attack one side, 3 fairly large areas attacked on other side.

200 b e 1 j 1 No attack.

2 No attack one side,

other l fairly large and 3 smaller areas attacked.

201 b e j 1 No attack.

2 N o attack one side, very large area attacked on other.

a `c i 1 22 No attack, removed trom test. b c 1 j 22 Do. b i j l `No attack.

1% Severe attack. b g j 13 No attack.

22 No attack one side, -2

very small pits on other. 206 b h j 13 No attack.

22 Medium attack in scratch As may `be noted, KOH can be substituted for NaOH in the activating step and still obtain the same results. That is the NaOH and KOH solutions worked best for activating the previously produced aluminum oxide coating. However a too strongly alkaline solution, see Specimens 82-85 (Example 7), and Specimens 94, 95, 97 (Example 8a), apparently does overlyfseverely attack the aluminum oxide coating and thereby prevents -obtaining maximum corrosion resistance in accordance with this invention. Of course it is to be understood that with a stron-ger alkaline solution, a short time of treatment and/or a lower temperature would provide a degree of compensation for not having used a less basic activating solution.

In substituting FeNH4C2O4, which is very acid, in Step B of this example, Specimen A204 was severely attacked at 11/2 hours but not at 1 hour. This shows ferrie ammoniurn oxalate together with the activating and silicate NaOH Vused per Step C for these specimens. Further making the ferrie ammonium oxalate alkaline precipitated the iron and did not impart the desired corrosion resistance (see Specimen 204). It was noted after the respective oxalate treatment, Specimen 204 was a deep red; whereas for the acid oxalate, Specimens 199, 200 and 201 were a golden color. The aforegoing is in contrast to using sodium oxalate which was made by adding sodium hydroxide to oxalic acid until the resulting solution was slightly alkaline; and which imparted very good corrosion resistance, see Specimen 205.

From the above examples it is to be noted that by activating the chemically or electrolytically formed aluminum oxide coating with an inorganic base, provided the concentration, time and temperature are such that the oxide coating is not severely attacked; and a subsequent silicate treatment, the corrosion resistance is substantially enhanced. The corrosion resistance is substantially additionally enhanced by interjecting a treatment between the activating step and the silicate treatment step that comprises subjecting the specimen to an organic compound of a metal which deposits an oxide in the aluminum oxide coating. However as appears above, optimum results are not obtained if such a compound is more than slightly acid. That is, if such a compound is more than slightly acid and cannot be made nearly neutral or somewhat basic by adding, for example, NaOH, without precipitation, the said interjected treatment will produce adverse results. It is preferred that the pH of the solutions of Procedures B and in particular that of nickel acetate and ferrie ammonium citrate be about a pH of 5.3-6. Additionally, in order to obtain best results the article being coated with the modied Oxide of this invention should preferably have all the natural oxide coating cleaned oi, for example, used in a rolled condition, subject to a good etch and/or buff clean.

.It is considered that the activation step and the silicate treatment steps are the two most important steps of this invention in modifying an aluminum oxide coating regardless of chemically or anodically formed. Additionally to maximize the corrosion resistance a precipitate or an oxide of a heavy metal such as indicated in Procedure B and a precipitate of an oxide of an alkali or alkali earth metal of Procedure C should be included in the modified coating, the solutions utilized in forming these precipitates preferably being slightly basic or as close to being basic as practical without causing the solute to precipitate in the solution.

EXAMPLE 18 To deter-mine the effects of adding an oxidizing agent (nitric acid) to the solutions of Procedures B and C four specimens of 5457 alloy were treated. `Specimens 208 and 210 were degreased, caustic etched for one minute, treated to a Ibright dip and subjected to a sulfuric acid electrolyte for 40 minutes, While Specimens 207 and 209 were degreased and then subjected to a sulfuric acid electrolyte for 40 minutes without any intermediate treatment. Each of these specimens were treated in a NaOH (15 gms./ liter) for l minutes at 120140 F. and treated to the one or mo-re of the solutions as set forth in Table 34 under Procedures B and C and finally to solution (e) before the corrosion test. The various solutions used are as follows:

TABLE 34 Procedure Procedure Elapsed Corrosion Test Results Specimen B C Time (hrs.)

207 a c 9 No attack.

15 Small pits one side,

considerable attack in scratches and numerous pits on other.

No attack.

Very slight attack.

Numerous small pits one side, other numerous small pits in area that appeared like fingerprint.

Considerable attack.

Considerable attack on both sides.

Some attack.

209 a d f 210. b d S o. 9 Considerable attack on both sides.

As may be noted, the addition of nitric acid'decreases the resistance .to corrosion. This is consistent with the remarks made under Example 6.

EXAMPLE 19 NaOH (.32 gmsJliter) 30 min Boiling. KAG (9.8 gms/liter)- 30 min 20D-212 F. Kassil #l (28 cc./1iter) 30 min 20o-212 F.

and then subjected to the corrosion test. The results are given below in Table 35.

For these specimens one of the following electrolyte solutions was used, the particular solution for a given specimen being indicated in Table 35:

(a) 15% H2SO4-i-5% chromic acid,

(c) 15 H2SO4-i-hydrazine sulfate.

TABLE 35 Electro- Elasped lyte Time (hrs.)

Specimen Alloy Corrision Test Results Only attack in scratche both sides.

Sere attack on both Both sides attack in scratches, some pitting on polished side.

Severe attack on unpolished side; attack in scratches and along bottom oi polished side large area notattacked.

Attack in scratches o both sides slight pitting on polished side.

Moderate attack in scratches on unpolished side, attack in 2 scratches and iew scattered pits which started at defects on polished side.

No attack.

Unpolished side had severe attack on sides and bottom and small pits in central area; polished side had modcrate attack on sides and bottom and few small pits in central area.

No attack.

8 Unpolished side had severe attack on sides,

bottom and inner area; polished side had moderate attack on sides and bottom, few few small pits in central area.

It is believed that the severityof the attacky of the specimens of Table 35 was largely due to the washing Iand drying procedure .at the end of the anodic treatment since the specimens were stored with adjacent planar surfaces abutting one another. Thus ifV the specimens were not completely rinsed and dried, during storage the water would seep to the side and bottomfedges and thereby cause a condition that gave-the above results. It is believed that upon further investigation, the use of hydrazine sulfate and similar compounds in the electrolyte will give better results than -those indicated to date.

Although the tests for corrosion resistance given above are indicated by the resistance to a 1% NaOH solution, it is to be understood the modiiied coatings of this invention `also resist acid-attack. This is shown in the Example 20 as follows: i

EXAMPLE 2O Specimens 216-224 inclusive which were alloy 3003 were rst degreased, flash sulfuric acid electrolyte treated, rinsed in-water, subjected to a 20-second caustic etch, rinsed in water, dipped in acid fto remove smudge, rinsed in water, then subjected to a 40-minute sul-furic acid electrolyte and then thoroughly rinsed with water. Specimens 216, 220, 223 and 224 Iwere then subjected to the treatment of Steps (e), (i), (1g) and (h) for Table 15 of Example 5; while Specimens 218 and 219 were then subjected to the treatment of Steps (e) and (h) for Table of Example 5 and Specimens 217, 221 and 222wer'e thenonly given a boiling water seal for 15 minutes. The thus treated specimens were rinsed with water and vexposed to the corrosion medium as given in the Table 36.

TABLE' 36 Specimen Corrosion Medium Time Results 216 1% NaOH, room temp. 16 hrs.. No attack. 218 do 4 hrs Attack in scratches on each side.

16 hrs. Above attack but more severe, medium pitting on both sides.

217 do 2 min- Severe attack. 223 1% oxalic acid, room 44 hrs- No attack.

temp. 219 do 44 hrs D0. 221 -do 16 hrs. N oticeable attack.

44 hrs. Considerable attack,

some coating still on specimen.

220,224 1% oxalic acid, 212 F.. 11 min. No noticeable attack.

25 min Some gas being liberated at surface.

45 min- N oticeable attack but not all coating dissolved.

222 do 2% min- Coating dissolved.

As may be noted for Specimens 220, 222 and 224 the oxalic acid was at about boiling in order to increase the EXAMPLE 21 It has been found that various compounds that release ammonia in any aqueous solution duringV the time `that aluminum specimens are immersed therein enhances the resistance to corrosion. It is preferred that such ammoniumcompounds be applied as a Procedure B step, how.- ever, as may be noted hereinafter, an ammonium compound advantageously be used in a basic aqueous solution in the Procedure A step (see Example 26-Specimen 463). Illustrative of the manner of treatment and the very good results obtained are shown by Specimen 400 of Example 2 and Specimens 422-432 below.

ASpecimens 422-432 and 428e were first degreased and then pretreated as follows: Specimens 422, 423, 427-429, 428a, 431 and 432 which were a 3003 alloy, were successively subjected to a 5-minute dilute caustic etch, and a sulfuric acid electrolyte for 45 minutes; while Specimen 424 was subjected to a 5% caustic etch for 1 minute, a 1% caustic etch for 3 minutes and a sulfuric acid elec- 38 trolyte for 45 minutes; while Specimen 425, an 110.0" alloy, rwas subjected to a, 1% `caustic etch and a sulfuric acid electrolyte for 45 minutes; `and Specimen 426, an 1100 alloy was subjected to a 1% caustic etch, a 30 second bright dip and a sulfuric acid electrolyte for 45 minutes. Each of Specimens 422-432, and 428a were then rinsed in water, next successively treated to two or more of the solutions of Table 37 as indicated in Table 38 and finally to 's1-1.0% NaOH corrosion test. The results of the test are given in Table 38.

With reference to Table 37, wherethe concentration for NaOH is indicated in cc., this means the solution used required the given number of cc. of 0.1 NHCl to neutralize a 25 cc. of the NaOH solution which for solutions (g), (h) it indicates the cc. of 0.1 HC1 to neutralize a 5 cc. sample. l

TABLE 37 f Solution Time Temperature a) NaOH (.3 gms./liter) 15min Bolling. gb) NaOH (1.6 cc.) 15 min Do. (c) NaOH (2.10 cc.) 15 min Do. (d1) t(7.5 cc. conc. NHtOH-I- gms. NaOH)/ 30 min Do.

1 er. (e) Kassi1#1 (27 oc./1iter) 30-min 20o-212 F. (f) N H40, 9 gms.i-NaOH, 1.35 gms.)/1 15 min 20D-212 F. (g) NH4 H (4.80 cc.) 15 min 20G-212 F (h) NH4OH (2.60 cc.) 15 min 20D-212 F' (i)A (N 1%.;)2003 (9.0 gms./l.)+NaOH (1.5 15 min 20D-212 F.

gms.

(j) .Solution (i) with 7.5 ce. conc. NHiOH 30 min Boiling. a e

(kHtOH (7.5 cc. conc/1.) -l-Dequest 15 min 20G-212 F. (m) NHiOH (7.9 ce. conc/1.) 15 min 20G-212 F. (n) (NI-102 CO3 (9.8 gms/1.) -l-NHiOH 15 min 20G-212 F.

(4 cc. cone/1.).

TABLE 38 Specimen A B C D Elapsed Corrosion test Results time (hrs.

422 d e 20 No attack.

27 No attack one side, 2

very small pits on other.

60 5 small pits one side, 2

very small pits on other.

423 a I e 20 no attack.

24 Do. 48 1 small pit each side. 4 b h e 20 No attack. 425,426 e g e 20 Do. 7 a f e 24 Do.

a i e 17 No attack one side, l

very small pit other. 27 Same as 17 hrs. 428a a n e 12 No attack.

20 No attack one side,

large area of attack on other.

27 No attack.

30 Uniform attack.

40 No attack.

72 Slight attack in scratches one side, pitting starting on other.

To be noted is Specimen 429 wherein the solution for Procedure A contained NaOH, (NH4)2,CO3 and NH4OH which shows that a basic solution using a compound that releases'ammonia in solution together with NaOH in the same solution gives good results, as well as other specimens where such an ammonium compound were used only in the Procedure B solution. Further Specimens 422-432, 4280: show that very good results can be obtained by eliminating the Procedure C treatment if 'the aforementioned type of ammonium compound is used. Also attention is directed to Specimens 428, 428a of Example 21 Where ammonium carbonate is used.

EXAMPLE 22 The specimens of this example are primarily directed to show the use of various ammonium compounds in the Procedure B step. All these specimens were given a pretreatment that included degreasing a Icaustic etch, a 45- minute sulfuric acid electrolyte treatment, and rinsed in

Images