WO1997013005A1

WO1997013005A1 - Metal cleaning process with improved draining uniformity

Info

Publication number: WO1997013005A1
Application number: PCT/US1995/012361
Authority: WO
Inventors: Thomas H. Fick; Timm L. Kelly
Original assignee: Henkel Corporation
Priority date: 1995-10-06
Filing date: 1995-10-06
Publication date: 1997-04-10

Abstract

In a process of cleaning aluminum cans by contact with a principal aqueous liquid cleaning composition containing acid and surfactants and subsequently rinsing the cleaned cans with an aqueous first postcleaning rinsing composition, the uniformity of draining (freedom from 'water breaks') is improved by including aluminum cations in the first postcleaning rinsing composition.

Description

Description METAL CLEANING PROCESS WITH IMPROVED DRAINING

UNIFORMITY

BACKGROUND OF THE INVENTION Field of the Invention

This invention relates to an improvement in a conventional process for acid clean¬ ing and, optionally, subsequent surface conditioning, of aluminum surfaces, particularly those of aluminum beverage containers which are processed at high production speeds. Discussion of Related Art

Aluminum cans are commonly used as containers for a wide variety of products. After their manufacture, the aluminum cans are typically washed with acidic cleaners to remove aluminum fines and other contaminants therefrom. After the principal acid cleaning of the containers, it is conventional to rinse the cleaned surfaces with an aque¬ ous composition that is largely tap water, although it often contains other dissolved and possibly suspended materials as a result of drag-out of liquid on the surfaces ofthe con¬ tainers from the preceding acid cleaning step and/or overflow from subsequent rinse stages. More particularly, in what is believed to be the most widely used current com¬ mercial practice, at least for large scale operations, aluminum cans are typically subjected to a succession of at least six cleaning and rinsing operations as described in Table A below. It is preferable to include another stage, usually called "Prerinse", before any of the stages shown in Table A; when used, this stage is usually at ambient temperature (i.e., 20 - 25 °C) and is most preferably supplied with overflow from Stage 3 as shown in

Table A, next most preferably supplied with overflow from Stage 1 as shown in Table A, and may also be tap water. Any of the rinsing operations shown as numbered stages in Table 1 may consist of two or preferably three sub-stages, which in consecutive order of their use are usually named "drag-out", "recirculating", and "exit" or "fresh water" sub-stages; if only two sub-stages are used, the name "drag-out" is omitted. Most preferably, when such sub-stages are used, a blow-off follows each stage, but in practice such blow-offs are often omitted. Also, any ofthe stages numbered 1 and 4 - 6 in Table A may be omitted in certain operations. TABLE A

Stage Action on Surface During Stage Number

1 Aqueous Acid Precleaning

2 Aqueous Acid and Surfactant Cleaning

3 Tap Water Rinse

4 Mild Acid Postcleaning, Conversion Coating, or Tap Water Rinse

5 Tap Water Rinse

6 Deionized ("DI") Water Rinse

It has been frequently observed that after Stage 3 as shown in Table A, drainage of rinse liquid from the cans is somewhat non-uniform, leading to the phenomenon gen¬ erally known in the art as "water breaks", i.e., areas where the liquid film on the contain¬ er surface after draining and/or drying in the air is visually obviously thicker in some patchy areas ofthe surface than in immediately neighboring areas. Inasmuch as any non- uniformity of surface conditioning at any stage in the cleaning process is often perceived by users as likely to lead to suboptimal final appearance of the container surface after subsequent processing, it would be commercially advantageous to avoid or at least mini¬ mize any appearance of water breaks at any stage ofthe processing. DESCRIPTION OF THE INVENTION

Object ofthe Invention

A major object of this invention is to provide a process for cleaning and subse¬ quent processing of aluminum cans or similar containers which will minimize the appear¬ ance of water breaks at any stage of processing and will not cause any substantial adverse effect on any stage of processing or conventional use ofthe cleaned containers. In partic¬ ular, the beneficial effects of can mobility enhancing, dome staining reducing, and/or sludge minimizing processes such as are described in U. S. Patents 5,080,814, 5,064,500, and 5,378,379 and in pending U. S. Applications Serial Numbers 08/109,791 and 08/ 362,687, the entire specifications of all of which, to the extent not inconsistent with any explicit statement herein, are hereby incoφorated herein by reference, should not be ad- versely affected.

General Principles of Description

Other than in the claims and the operating examples, or where otherwise ex¬ pressly indicated, all numbers expressing quantities of ingredients or reaction conditions used herein are to be understood as modified in all instances by the term "about" in de¬ scribing the broadest scope ofthe invention. Practice within the numerical limits given, however, is generally preferred. Also, unless expressly stated to the contrary: percents, "parts of, and ratio values are by weight; the term "polymer" includes "oligomer", "co¬ polymer", "terpolymer", and the like; the description of a group or class of materials as suitable or preferred for a given purpose in connection with the invention implies that mixtures of any two or more ofthe members ofthe group or class are equally suitable or preferred; description of electrically neutral constituents in chemical terms refers to the constituents at the time of addition to any combination specified in the description, and does not necessarily preclude chemical interactions among the constituents of a mixture once mixed; specification of materials in ionic form implies the presence of sufficient counterions to produce electrical neutrality for the composition as a whole (any counter¬ ions thus implicitly specified should preferably be selected from among other constitu¬ ents explicitly specified in ionic form, to the extent possible; otherwise such counterions may be freely selected, except for avoiding counterions that act adversely to the objects of the invention); and the term "mole" and its variations may be applied to elemental, ionic, and any other chemical species defined by number and type of atoms present, as well as to compounds with well defined molecules. Summary ofthe Invention

It has been found that, surprisingly and unexpectedly, substantial concentrations of aluminum cations have a water break reducing effect when contained in the first aque¬ ous rinse (such as Stage 3 in Table A) that is used after principal acid cleaning of con¬ tainers with an acid and surfactant composition, compared with the use of "cleaner" tap water rinses in this processing step. Detailed Description ofthe Invention and Preferred Embodiments This invention is particularly useful when the acid and surfactant cleaning compo¬ sition used in a process according to the invention comprises alkoxylated, especially ethoxylated, alkyl phenol nonionic surfactant molecules and/or fluoride ions, preferably both. More particularly, the concentration of alkoxylated alkyl phenol surfactant molecules in the acid and surfactant cleaner that is used (for example, in Stage 2 in Table A above) immediately prior to a rinsing process according to this invention, or in a process according to the invention including such a rinsing process step, preferably is at least, with increasing preference in the order given, 0 02, 0 05, 0 10, 0.15, 0 20, 0.25,

0.30, 0.35, 0.40, 0.45, 0.50, 0 55, or 0.60 grams per liter (hereinafter usually abbreviated as "g/L") and independently preferably is not more than, with increasing preference in the order given, 8, 6, 5, 4.0, 3.0, 2.2, 1.9, 1.7, 1.6, 1 4, 1.30, 1.20, 1 10, 1.00, 0 95, 0.90, 0.85, 0.80, 0.75, or 0 70 g/L. Independently of concentration, these alkoxylated alkyl phenol surfactant molecules preferably are selected from molecules in which the alkyl moieties have an average number of carbon atoms that is at least, with increasing preference in the order given, 3, 5, 6, 7, 8, or 9 and independently preferably is not more than, with increasing preference in the order given, 20, 16, 14, 12, 11, or 10, and, inde¬ pendently, the average number of alkoxyl units per molecules is at least, with increasing preference in the order given, 2, 3, 4, 5, 6, 7, 8, or 9 and independently preferably is not more than, with increasing preference in the order given, 20, 18, 16, 15, 14, 13, 12, 11, and 10.

When the principal acid and surfactant containing cleaner composition used be¬ fore a characteristic rinse according to this invention contains alkoxylated alkyl phenol surfactants as described above, it preferably also contains alkoxylated, more preferably ethoxylated or both ethoxylated and propoxylated, aliphatic alcohol(s) nonionic surfact¬ ants) The base alcohols from which these nonionic surfactants are normally made preferably have at least, with increasing preference in the order given, 8, 10, or 12 carbon atoms per molecule and independently preferably have not more than, with increasing preference in the order given, 22, 20, 18, 16, or 14 carbon atoms per molecule Inde¬ pendently, the concentration ofthe alkoxylated aliphatic alcohol(s) nonionic surfactant molecules preferably is such that the ratio of the total concentration of the alkoxylated alkyl phenol nonionic molecules to the total concentration of the alkoxylated aliphatic alcohol nonionic surfactant molecules in the principal acid and surfactant cleaner is not less than, with increasing preference in the order given, 0 4 1 0, 0 6 1 0, 0 8 1 0. 1 0 1 0,

1.1 : 1.0, 1.15.1.0, 1.20.1 0, 1.25 1 0, 1 30.1 0, or 1.35 1 0 and independently preferably is not more than, with increasing preference in the order given. 5 1 0, 3 1 0. 2 5 1 0, 2 0 1 0, 1 8 1 0, 1 7 1 0, 1 65 1 0, 1 60 1 0, 1 55 1 0, 1 50 1 0, or 1 45 1 0

Primarily for reasons of economy and independently of the other preferences stated above, much ofthe acidity ofthe pπncipal acid and surfactant cleaner composition used in connection with this invention is preferably derived from sulfuric acid However, as already noted above, the presence of fluoride ions in the principal acid and surfactant cleaning composition is also preferred, and the source of these ions often is hydrofluoric acid. Therefore, a mixture of sulfuric and hydrofluoric acids is generally preferred to provide the total and free acidity and fluoride content in this principal cleaning composi¬ tion It is well known in the art that fluoride ions can form complexes with undissociat- ed hydrofluoric acid and with numerous metal cations, including those of aluminum, which are always present to some extent in cleaning compositions of this type when aluminum substrates are being cleaned Therefore, the preferred fluoride content for these cleaning compositions is normally specified by use of a fluoride sensitive electrode rather than by any measurement ofthe total concentration of fluorine atoms in the com¬ positions, because the part of the fluorine atoms bound into some complexes present in the composition is not effectively available to promote the controlled dissolution ofthe substrates being cleaned as is desired

The effective fluoride activity for purposes of this description is measured by use of a fluoride sensitive electrode as described in U. S Patent 3,431 , 182 and commercially available from Orion Instruments Fluoride activity was specifically measured relative to a 120E Activity Standard Solution commercially available from the Parker Amchem ("PAM") Division of Henkel Coφoration by a procedure described in detail in PAM Technical Process Bulletin No 968, Revision of April 19, 1989 The Orion Fluoride Ion Electrode and the reference electrode provided with the Orion instrument are both im¬ mersed in the noted Standard Solution and the millivolt meter reading is adjusted to 0 with a Standard Knob on the instrument, after waiting if necessary for any drift in readings T e electrodes are then rinsed with deionized or distilled water, dried, and im¬ mersed in the sample to be measured, which should be brought to the same temperature as the noted Standard Solution had when it was used to set the meter reading to 0 The reading of the electrodes immersed in the sample is taken directly from the millivolt (hereinafter often abbreviated "mv" or "mV") meter on the instrument With this instru- ment, lower positive mv readings indicate higher fluoride activity, and negative mv read¬ ings indicate still higher fluoride activity than any positive readings, with negative read¬ ings of high absolute value indicating high fluoride activity. The principal acid and surfactant cleaning compositions used in connection with this invention preferably have a fluoride activity measured by this method that is not more than, with increasing preference in the order given, 50, 40, 30, 20, 15, 10, 5, or 1 mv and independently preferably is not less than, with increasing preference in the order given, -15, -10, -7, -4, -2, or - 1 mv.

The acidity ofthe principal acid and surfactant cleaning composition used in con- nection with this invention preferably is measured and controlled by a "Free Acid points" value measured by titration of a sample ofthe composition in a manner well known per se in the art. For this description, the Free Acid points are defined as the number of mil¬ liliters (hereinafter usually abbreviated as "mL") ofOΛ N strong alkali solution required to titrate a 10 mL sample ofthe composition being measured, after dilution with deion- ized water and addition of a large excess of a neutral fluoride such as sodium or potas¬ sium fluoride, to an end point of pH 8.2 - 8.5, as with phenolphthalein indicator. For a principal acid and surfactant cleaner used in connection with this invention, the Free Acid points value so measured preferably is at least, with increasing preference in the order given, 2, 4, 5, 6.0, 6.5, 7.0, 7.5, or 8.0 and independently preferably is not more than, with increasing preference in the order given, 18, 16, 15, 14.5, 14.0, 13.5, 13.0,

12.5, or 12.0. Higher Free Acid values within these ranges are preferred when the time of contact between the principal cleaning compositions is very short, and lower values are preferred when the time of contact is relatively long. Also, high Free Acid and high concentrations of effective fluoride (which as noted above correspond to lower values of millivolt readings with a fluoride sensitive electrode as described above) both promote more rapid dissolution ofthe substrates being cleaned, so that, unless the time of contact between the principal cleaner and the substrate being cleaned is very short, the highest parts of these preferred ranges for Free Acid and fluoride activity normally would not be preferred together. Contact between the substrate to be cleaned and the principal acid and surfactant containing cleaner composition used in connection with the invention is preferably ac¬ complished initially by spraying for a time interval that preferably is at least, with in- creasing preference in the order given, 10, 15, 20, 25, or 30 seconds (hereinafter usually abbreviated as "sec") and independently preferably is not more than, with increasing preference in the order given, 180, 150, 120, 90, 80, 70, or 65 sec. In addition to the greater economy of operating with shoπer contact times, the drainage uniformity itself 5 is often adversely affected by longer contact times The temperature of the principal acid and surfactant containing cleaner used in connection with the invention preferably is at least, with increasing preference in the order given, 25, 30, 35, 40, 45, 50, 55, 57, or 59 ^CC and independently preferably is not more than, with increasing preference in the order given, 90, 85, 80, 75, 70, 67, 65, 63, or 61 °C. io After contact by spraying as described immediately above has been accomp¬ lished, the substrates preferably are allowed to stand in ambient air for a time interval that is at least, with increasing preference in the order given, 0.2, 0.5, 1.0, 1.5, 2.0, 2.5, or 3.0 sec and independently preferably is not more than 60, 45, 30, 25, 20, 17, or 15 sec. Independently, before being contacted with a rinsing composition according to this in-

15 vention, but after any period of standing in ambient air that is used as described immedi¬ ately above, the surfaces ofthe cleaned substrates preferably are contacted with laterally moving forced air ("blown-off"), in order to reduce the volume of liquid retained on their surfaces, particularly their dome surfaces, for a period of time that preferably is at least, with increasing preference in the order given, 0.02, 0.05, 0.07, 0.10, 0.13, 0.15, 0.18, or

20 0.20 sec and independently preferably is not more than, with increasing preference in the order given, 6, 5, 4, 3.0, 2.5, or 2.0 sec. Also independently, any such blow-off is preferably followed by another interval of standing in ambient air, with the same time preferences as are described above for the first period of standing in ambient air, before contact with a rinsing composition according to this invention. 5 Before being contacted with the principal acid and surfactant containing cleaner composition in connection with this invention, the substrates to be cleaned are preferably contacted with an acidic precleaner that does not contain surfactants added specifically to assist in cleaning¹. Most preferably, this precleaner consists essentially of water and

^•'The lubricants used to form cans usually contain surfactant, and if they do. some surfactant will usually accumulate in the solution used for prewashing, even though no surfactant is added specifically to aid cleaning. Furtheπnore, water overflowed from the liquid nnsmg composition used in Stage 3, which after contmued use usually contains some surfactant earned over from the cleaning soluUon in Stage 2. is often supplied to the solution used in prewashing Stage 1. and. in some commercial operations. sulfuric acid and independently preferably has a pH value that is at least, with increasing preference in the order given, 1 1, 1.3, 1.5, 1.7, 1 8, or 1 9 and independently preferably is not more than, with increasing preference in the order given, 4, 3.5, 3.2, 2 9, 2.7, 2.5, 2.3, or 2.1.

5 Contact between the substrate to be cleaned and the precleaner composition used in connection with the invention is preferably accomplished initially by spraying for a time interval that preferably is at least, with increasing preference in the order given, 10, 15, 20, 25, or 29 sec and independently preferably is not more than, with increasing pref¬ erence in the order given, 90, 80, 70, 60, 50, 45, 40, 35, or 31 sec. The temperature, dur- o ing spraying contact with the substrates being cleaned, of the precleaner composition used in connection with the invention preferably is at least, with increasing preference in the order given, 25, 30, 35, 40, 45, 48, 50, 52, 53, or 54 °C and independently pref¬ erably is not more than, with increasing preference in the order given, 90, 85, 80, 75, 70, 66, 62, 59, 57, or 55 °C. 5 After any precleaning as described above has been accomplished, the substrates preferably are allowed to stand in ambient air for a time interval that is at least, with increasing preference in the order given, 0.2, 0.5, 1.0, 1.5, 2.0, 2.5, or 3 0 sec and inde¬ pendently preferably is not more than 60, 45, 30, 25, 20, 17, or 15 sec. Independently, before being contacted with a rinsing composition according to this invention, but after 0 any period of standing in ambient air that is used as described immediately above, the surfaces ofthe cleaned substrates preferably are blown-off, in order to reduce the volume of liquid retained on their surfaces, particularly their dome surfaces, for a period of time that preferably is at least, with increasing preference in the order given, 0.02, 0.05, 0.07, 0.10, 0.13, 0.15, 0.18, or 0.20 sec and independently preferably is not more than, with in- 5 creasing preference in the order given, 6, 5, 4, 3.0, 2.5, or 2.0 sec. Also independently, any such blow-off is preferably followed by another interval of standing in ambient air,

cleanmg solution which has been discarded from Stage 2 in order to reduce the contanunaUon level in Stage 2 is added directly to Stage 1 to help mamtam the desired pH These expedients provide additional possible sources of surfactant in the prewashmg solution which are not intended to be excluded by the language "not specifically added to aid in cleanmg" However, the total surfactant concentrauon from all sources in the solution used in acid prewashing Stage 1 preferably does not exceed, with increasing preference in the order given. 50, 40, 30, 20, 15, 10, 7, or 5 % of the concentration of surfactant m the principal acid and surfactant cleaner solution used in Stage 2. with the same time preferences as are described above for the first period of standing in ambient air, before contact with the principal acid and surfactant containing cleaner.

In the liquid rinsing compositions which are used in the first contact ofthe sub¬ strates being cleaned with any liquid phase after the principal acid and surfactant cleaner composition, and which are the characterizing feature of this invention, the concentration of dissolved aluminum cations preferably is at least, with increasing preference in the order given, 0.0001, 0.0002, 0.0005, 0.001, 0.002, 0.003, 0.004, 0.005, 0.006, 0.007, or 0.008 moles per liter (hereinafter usually abbreviated "Λ-f ) and independently preferably is not more than, with increasing preference in the order given, 0.1, 0.08, 0.06, 0.05, 0.04, 0.03, 0.02, or 0.01 M. Also, independently, the pH value of these first post-cleaning li¬ quid rinsing compositions preferably is at least, with increasing preference in the order given, 1.5, 1.9, 2.3, 2.7, 2.9, 3.1, 3.3, or 3.4 and independently preferably is not more than, with increasing preference in the order given, 6.5, 6.1, 5.9, 5.8, 5.7, or 5.6. No other constituents except water (and counterions) are necessary in these rinsing composi- tions, but some other constituents may be included without adverse effects, as noted fur¬ ther below.

Contact between the substrate being cleaned and the first liquid rinsing composi¬ tion used according to the invention after the principal acid and surfactant containing cleaner composition has been used is preferably accomplished initially by spraying for a time interval that preferably is at least, with increasing preference in the order given, 0.5, 1.0, 1.5, 2.0, 2.5, 3.0, 3.5, 4.0, 4.5, or 4.9 sec and independently preferably is not more than, with increasing preference in the order given, 45, 35, 30, 25, 20, 17, or 16 sec. The temperature ofthe first liquid rinsing composition according to the invention that is used after the principal acid and surfactant containing cleaner used in connection with the invention preferably is at least, with increasing preference in the order given, 15, 18,

20, 22, 23, 24, 25, or 26 °C and independently, primarily for reasons of economy, prefer¬ ably is not more than, with increasing preference in the order given, 70, 53, 47, 42, 38, 35, or 33 °C.

After contact by spraying with the first liquid rinsing composition subsequent to principal acid and surfactant containing composition cleaning as described above has been accomplished, the substrates preferably are allowed to stand in ambient air for a time interval that is at least, with increasing preference in the order given, 0.2, 0.5, 1.0, 1.5, 2.0, 2.5, or 3.0 sec and independently preferably is not more than 60, 45, 30, 25, 20, 17, or 15 sec. Independently, before being contacted with a second liquid rinsing compo¬ sition, but after any period of standing in ambient air that is used as described immedi¬ ately above, the surfaces ofthe cleaned and rinsed substrates preferably are blown-off, in order to reduce the volume of liquid retained on their surfaces, particularly their dome surfaces, for a period of time that preferably is at least, with increasing preference in the order given, 0.02, 0.05, 0.07, 0.10, 0.13, 0.15, 0.18, or 0.20 sec and independently prefer¬ ably is not more than, with increasing preference in the order given, 6, 5, 4, 3.0, 2.5, or 2.0 sec. Also independently, any such blow-off is preferably followed by another inter- val of standing in ambient air, with the same time preferences as are described above for the first period of standing in ambient air, before contact with a second liquid rinsing composition.

It is customary in the art and preferred in a process according to this invention to form the first liquid rinsing composition used after principal cleaning of the substrates being cleaned by inputting only tap water and, optionally, overflow from the second li¬ quid rinsing composition, into the first liquid rinsing composition and allowing all other constituents to be supplied by carryover, also called "drag-out", of constituents ofthe principal cleaning composition that adhere to the surfaces of the substrates as they are transported from the area where they are contacted with the principal acid and surfactant containing cleaning composition to the area where they are contacted with the first subse¬ quent liquid rinsing composition. After the principal cleaning composition has been used to clean aluminum substrates, it will normally contain a sufficient amount of dissolved aluminum cations that the preferred levels of aluminum cations in the first liquid rinsing composition can be achieved by allowing more carryover of principal cleaning composi- tion constituents than is now usual, and/or by reducing the amount of tap water added to already existing carryover from the principal acid and surfactant cleaner composition. This procedure, of course, means that some ofthe contaminating soil(s) that are removed from the cleaned metal substrates and transferred to the principal cleaning composition will also become constituents ofthe first liquid rinsing composition used after principal cleaning. Suφrisingly and unexpectedly, it has been found that such "contamination" ofthe first rinsing composition has no deleterious effect on the degree of cleanliness ulti¬ mately achieved by the entire cleaning process, provided that subsequent "uncontamin- ated" tap water and deionized water rinses as shown in Table A above are used.

For a fuller appreciation ofthe invention, reference may be made to the following examples, which are intended to be merely descriptive, illustrative, and not limiting as to the scope ofthe invention. Examples and Comparison Examples

GROUP 1 Materials Used

The sulfuric acid used was a technical grade, approximately 50 % H₂SO₄ in tap water. (Each lot was assayed to determine percent sulfuric acid before use in the labora- tory, in order to assure the reliability of the significant figures given below for H₂SO₄ concentration.

Dissolved aluminum cations were provided by technical alum (i.e., variably hy¬ drated aluminum sulfate), which has an average molecular weight of 631.34 and contains 8.55 % of aluminum atoms, with two such atoms per molecule, Ammonium bifluoride, technical grade, > 97 %, typically 98.3 %, of NH₄HF₂, with the balance predominantly NH₄F, was used.

Ammonium hydroxide, 26° Baume, technical grade, was used when needed to adjust free acid and/or pH values. (This material is also referred to as "aqueous ammon¬ ia".) Triton™ N-l 01 is commercially available from the Industrial Chemicals Division of Union Carbide Chemicals and Plastics Company Inc. in Danbury, Connecticut and is reported by its supplier to be a nonionic surfactant consisting of polyethoxylated nonyl¬ phenol with an average of 9.5 moles of ethylene oxide per molecule.

Surfonic™ LF-17 is commercially available from Huntsman Coφoration in Houston, Texas, and is reported by its supplier to be a non-ionic surfactant that consists of ethoxylated and propoxylated linear primary 12 - 14 carbon number alcohol mole¬ cules.

The following lubricants, believed to be typical of those generally used commer¬ cially in the manufacture of aluminum cans by drawing and ironing of aluminum sheet, were used:

Atochem™ SDO5L54N2J, a metal working lubricant used as a coolant; Ato- guard™ C-l 01 A Modified, a metal working lubricant applied to aluminum prior to the cupping operation; and Atoguard™ 1-102, circulated through the tool pack in the bodymaker; all of these are commercially available from the Metalprep De¬ partment of Elf Atochem North America in Cornwells Heights, Pennsylvania. DTI™ 5600-M4, applied to aluminum prior to the cupping operation; and DTI™ 5600- WB, a metal working coolant circulated through the tool pack in the body¬ maker; both commercially available from Diversified Technology, Inc. in San Antonio, Texas.

Daralube™ 17A2, circulated through the tool pack in the bodymaker and com¬ mercially available from the Grace Can Forming Fluids Division of W. R. Grace. Mobil™ DTE 24, a hydraulic lubricant applied to the aluminum prior to the cup¬ ping operation; and Mobil™ SHC 629, a gear and bearing lubricant; both com¬ mercially available from Mobil Oil Coφoration in Fairfax, Virginia. All other materials identified by chemical name below were reagent grade mater¬ ials. Precleaning Acid Composition: This consisted of a solution of sulfuric acid in water with a pH of 2.

Principal Cleaner Composition: The principal cleaning compositions used for all these examples and comparison examples had a Free Acid value of 12 points, a Total Acid val¬ ue of 36 points, and a Fluoride Activity of 0 mV as measured with a fluoride sensitive electrode versus a fluoride sensitive electrode in contact with the standardizing fluoride containing aqueous composition as already described above. A base stock consisting of 15.1 g/l of ammonium bifluoride, 8.9 g/l of a solution of 52 % hydrofluoric acid in water, 6.25 g/l of 66° Be sulfuric acid, and 142 g/l of alum as described above, with the balance being water, was prepared. Two hundred-fifty grams ofthe stock solution were mixed with tap water to give one liter of cleaner composition. The Free Acid, Total Acid and Fluoride Activity ofthe cleaner composition were checked. Additions of appropriate in¬ gredients were made to the stock composition to adjust its composition so that the desired values of Free Acid, Total Acid, and Fluoride Activity were obtained when it was diluted as described above. It addition to the four components listed above, ammonium hy- droxide was added if the Free Acid value was otherwise too high. Surfactant was then added as follows: 0.79 g/l of Triton™ N-101 plus 0.57 g/l of Surfonic™ LF-17. All of the cleaning compositions also contained 3.13 g/l of total lubricants except where oth- erwise indicated. The compositions of the lubricant mixtures will be given below.

The composition of cleaning composition described immediately above is be¬ lieved, on the basis of analysis of used cleaning compositions, to simulate closely the steady-state composition in many commercial can cleaning lines that utilize acid-surfact- ant as their principal cleaner. Such cleaners when freshly made usually contain only wat¬ er, sulfuric and hydrofluoric acids, and surfactant(s). However, as the cleaning composi¬ tions are used, aluminum dissolves into them from the surfaces ofthe cans treated, there¬ by reducing the initial acidity so that hydrogen difluoride ions are formed in solution, and lubricants from the can surfaces are dissolved into and/or suspended in the cleaning compositions, thereby cleaning the cans.

Modified First Liquid Rinsing Compositions: Three types of modified rinse composi¬ tions were prepared to simulate compositions that would arise in commercial practice when making the first liquid rinsing composition by mixing freshly input tap water with dried and/or liquid materials carried over from a used principal acid and surfactant con- taining cleaner composition by adherence on the surfaces ofthe substrates being cleaned.

These types were designated below as "Inorganic Contamination", "Organic Contamina¬ tion", and "Combined Contamination".

Rinsing compositions described as "Inorganic Contamination" contained a per¬ centage as indicated in the accompanying Tables of the aluminum cations and fluoride and sulfate anions in the cleaning composition. Rinsing compositions described as "Or¬ ganic Contamination" normally contained a percentage as indicated in the accompanying Tables of the surfactants and lubricants in the principal acid and surfactant containing cleaning composition. In some instances as specifically noted in the Tables, the Organic Contamination rinsing compositions contained a percentage of either the surfactant(s) only or the lubricants only ofthe cleaning composition. Rinsing compositions described as "Combined Contamination" contained a percentage as indicated in the Tables of both types of ingredients specified separately for Inorganic and Organic Contamination rins¬ ing compositions. Apparatus and Procedure All cans were prepared on a laboratory carousel can washer which has been designed so that, in most respects , it closely simulates commercial large scale operations. Each run used fourteen cans. Four were used for water break evalua¬ tion, and the remaining ten were dried and used for coefficient of static friction (herein¬ after usually abbreviated as "COF") and interior brightness (hereinafter usually abbrevi¬ ated as "INT") testing. The procedure used to prepare cans is given in Table 1

Table 1 : CAN PROCESSING SEQUENCE AND CONDITIONS

Stage Process Temperature, Time in Seconds for: °C

Spray Dwell Blow- Off

1 Precleaning 54 30 10 30

2 Principal Cleaning 60 60 10 30

3 First Post-Cleaning Rinse See Tables Below 30 10 30

5 Tap Water Rinse Ambient 30 0 0

6 Deionized Water Rinse Ambient 90 0 30

Dry Oven Drying 150 - 300 -

Test Procedures Evaluation of Waterbreak: The water break characteristics of cans were evaluated by visually rating the amount of waterbreak on each ofthe four major surfaces ofthe can: interior dome and sidewall and exterior dome and sidewall. A score of 2 is assigned to a completely waterbreak free surface, zero to a completely waterbroken surface and inter¬ mediate values to extents of waterbreaks in between Values were converted to Percent¬ age Waterbreak Free by dividing the sum ofthe values determined for each area of inter¬ est on four cans by 8 and multiplying the result by 100. Coefficient of Friction ofthe Exterior Sidewalls ("COF") The cans were evaluated for

² The time periods for rinsing, standing, and blowing-off operations are higher in the laboratory apparatus, because it has only a single spray chamber, which must be used for all stages ofthe process. As a result, longer draining, rinsing, and blowing-off times are required in the laboratory apparatus to avoid contamination. In commercial scale apparatus, there are separate chambers for each spraying and blowing-off step, so that much shorter times can be used. Extensive experience, however, has established that this difference between laboratory and commercial practice does not normally affect the results achieved. this property with a laboratory static friction tester. This device measures the static fric¬ tion associated with the outside sidewall surface characteristics of aluminum cans. This is done by using a ramp which is raised through an arc of 90° by using a constant speed motor, a spool and a cable attached to the free swinging end of the ramp. A cradle at- tached to the bottom of the ramp is used to hold two cans on their sides in horizontal position approximately 13 millimeters apart, with their domes facing the fixed end ofthe ramp and restrained from sliding along the ramp as it is raised by the cradle. A third can is laid on its side upon the first two cans, with the dome of the third can facing the free swinging end ofthe ramp, and the edges of all three cans are aligned so that they are even with each other. The cradle does not restrain the movement ofthe third can.

As the ramp begins to move through its arc, a timer is automatically actuated. When the ramp first reaches an angle at which the third can slides freely from the two lower cans, a photoelectric switch shuts off the timer. The elapsed time, recorded in sec¬ onds, is commonly referred to as "slip time". The coefficient of static friction is equal to the tangent ofthe angle swept by the ramp at the time the can begins to move. This angle in degrees with the particular apparatus used is equal to [4.84 + (2.79-t)], where t is the slip time.

Reflectivity of Metal Containers: This property was measured by use of a device consist¬ ing of a power stabilized high intensity lamp, a fiber optic bundle conveying the light from this lamp to the can surface, a photocell onto which the light reflected from the can impinged, and an International Microtronics Inc. Model 350 amplifier by which the pho¬ tocell current output was amplified and converted to a digital readout; the number dis¬ played was recorded as the brightness ofthe surface. The instrument was calibrated be¬ fore testing by measuring the reflectivity of a back silvered plane mirror and adjusting the readout to a measured reflectivity of 440. The reflectivities of ten cans were mea¬ sured and averaged to yield the results reported in the tables.

Surface Carbon Analvsis Surface carbon analyses were performed using a Carbon Di¬ oxide Coulometer, Model 5010, made by Coulometrics Inc. of Wheatridge, CO, with a digital readout of micrograms of carbon. After the cans were evaluated for water break, they were put into the drying oven for five minutes at 150 °C. All cutting and handling tools were wiped off with isopropyl alcohol before being used in subsequent operations and between cleaner formulations. The can was held with pliers while the dome was re- moved with scissors. A paper cutter was used to cut the sidewall into strips approximate¬ ly 4 x 0.37 inches in size. Tongs were used for handling the strips of the can sidewall. The tongs were flamed over a Fisher burner to red heat before every run to remove any carbonaceous matter. Four strips were weighed on an analytical balance and loaded in

5 the carbon analyzer. The strips were placed in the cool zone of the carbon analyzer for two minutes. After two minutes, the strips were pushed into the hot zone (450°C) of the oven and the counter reset to zero. The sample was tested for five minutes before a read¬ ing was recorded. A second series of four strips from the first can was analyzed similar¬ ly. The complete process was repeated using the second can. o The sidewalls of five untested cans were individually measured and weighed on an analytical balance to determine how much each square foot of sidewall weighed. The total surface area ofthe sidewall for each can was divided by its weight and rationalized to obtain the grams per square foot of sidewall. The five results averaged 14.71 grams per square foot. The carbon analyses results were normalized using the following formu- 5 la: μg of carbon per square foot = [(14.71 ÷ sample weight) * total μg of carbon] The median values for the four samples are reported. The median was reported because it is a better measure of location than the mean for small samples which have broad ranges.³ 0 2.2.5. Analvsis of Results The RS/1 Explore® Version 2.0 module of RS/1 , Release 4.3, commercially available from BBN Software Products, Bolt Beranek and Newman Inc., Cambridge, MA (version dated 1990), was used to analyze the data from each experi¬ ment. The Correlate and Rank Correlate procedures of RS/1 were used to test for corre¬ lation between some sets of pairs of measurements. The value ofthe correlation coeff - 5 ient, R, is a measure ofthe extent of linear association between two variables. It ranges from -1.0 to 1.0. When it is near zero, there is no linear relationship between the varia¬ bles. Values of R close to -1.0 indicate that the variables are negatively correlated, so that as the values of one variable increase the values of the other variable decrease. When the value of R is near 1.0, the values are positively correlated, so that when the

³Davies, O.L. and Goldsmith, P.L., ed., Statistical Methods in Research and Production, Fourth Edition (Longman Group Limited, London, 1980), pp 35 and 49 values of one variable increase so do those ofthe other variable.

Another measurement of linear correlation is the coefficient of determination, R². It measures the proportion of the variation in the second variable that is "explained by" the first variable. It ranges from 0 to 1.0, with values near zero implying little or no cor- relation and values near 1.0 implying strong correlation.

The correlation coefficient measures only the amount of linear correlation be¬ tween variables. It may not be sensitive to certain types of non-linear correlation. Val¬ ues of R far from 0 mean that the variables are linearly correlated. Values of R near 0 mean that the variables are not correlated or there is some non-linear association present. Two variables that demonstrate associated increasing or decreasing trends, but that do not fall near a straight line when plotted against one another, may be non-linearly correlated. The Spearman's Rho (hereinafter usually abbreviated simply as "Rho") statis¬ tic is a measure of association designed to measure such monotonic trends. The values of Rho fall between -1.0 and 1.0. If Rho is near zero, the ranks of the samples are not correlated. If Rho is near 1.0, the ranks are in similar order and are, therefore, correlated and dependent. If Rho is close to -1.0, the ranks are inversely correlated and dependent. Determination of Cleaner Solution: Free Acid. Total Acid. Reaction Product and Fluoride Activitv: The test procedures described in Parker Amchem Technical Process Bulletin No. 1580 of January 3, 1994 for Ridoline® 123 Cleaner were used. Results

Subgroup 1.1 This subgroup was designed to determine how the waterbreak observed on cans after cleaning was affected by the conditions in Stage 3, the rinse immediately after the cleaner. The following factors were used: pH (3.5 vs 5.5); inorganic contamin¬ ated Stage 3 composition (none vs 10 % of Stage 2); surfactant contamination Stage 3 composition (none vs 10 % of Stage 2); lubricant contamination Stage 3 composition

(none vs 10 % of Stage 2); and the nature of the lubricant component in the Organic Contamination (hydraulic oil, Mobil DTE-24, vs synthetic can making lubricant, Daralube (DL) 17A2 designated as Fraction DL17A2). The following results were mea¬ sured: conductivity of the Stage 3 rinsing composition; interior reflectivity; percent of the area ofthe exterior sidewall which was waterbreak free ("% WBF ESW"); percent waterbreak free on the interior dome ("% WBF ID"); percent of the area ofthe interior sidewalls which remained waterbreak free ("% WBF ISW"); and coefficient of static friction on the exterior sidewalls ("COF")

The compositions ofthe Stage 3 rinsing compositions and the results of the mea¬ surements are reported in Table 2. The Explore® module of RS/1 was used to determine the main effect of each of the factors on each of the responses. Only one of the re- sponses, exterior sidewall waterbreak, had any factors which had an effect on it at the 95 % confidence level: "Inorganic Contamination" had a positive effect on % WBF ESW at the 95 % confidence level. As the amount of Inorganic Contamination in Stage 3 in¬ creased, the percentage ofthe exterior sidewalls of the cans which were waterbreak free increased. Subgroup 1.2. This subgroup was also designed to determine how the waterbreak ob¬ served on cans after cleaning was affected by the conditions in Stage 3, as in Subgroup 1.1, but the lubricant contamination in this subgroup was an Atochem bio-stable lubricant system. It consisted of 27.6 parts by weight of Atoguard™ C-l 01 Modified cupper lubri¬ cant and 72.4 parts by weight of Atoguard™ 1-102 bodymaker coolant. The cleaner con- tained 2.5 g/L of lubricant. The same results as for Subgroup 1.1 were measured and were analyzed by the same means.

The compositions ofthe stage 3 rinsing compositions and the results ofthe mea¬ surements are reported in Table 3. Three ofthe results, exterior sidewall waterbreak, in¬ terior dome waterbreak and interior sidewall waterbreak, had factors which had an effect on them at the 95 % confidence level: Inorganic Contamination had a positive effect on all three results. As the amount of Inorganic Contamination in the Stage 3 rinsing composition increased, the percentage of each of the three surfaces of the cans which were waterbreak free increased. Inorganic Contamination was the only factor to have a significant effect on waterbreak at the 95 % confidence level. Subgroup 1.3. In Subgroup 1.1 above, it was found that the waterbreak observed on the exterior sidewalls ofthe cans after cleaning decreased as the amount of Inorganic Con¬ tamination in the Stage 3 rinsing composition increased. The Inorganic Contamination consisted of aluminum, sulfate and fluoride. The conductivity and ionic strength of this rinsing composition increased as the concentration of Inorganic Contamination increased. The improvement in waterbreak resistance which was observed when the concentration of Inorganic Contamination in the Stage 3 rinsing composition increased might have been caused by the increase in conductivity and ionic strength, or it might have been caused Table 2 EFFECT OF FIRST RINSING COMPOSITION CHARACTERISTICS, SUBGROUP 11

Run Directly Controlled Characteristics of First Rinsing Composition Results

H pH InCom, as % SuCom, as % Total Fraction of DL Conductivity of RiCo, μ % Waterbreak Free on: COF INT of Cleaner of Cleaner Lub, g/L 17A2 in Lub Siemens

ESW ID ISW

1 55 10.0 100 0313 0.5 3700 100 100 100 1316 283

2 35 00 100 0313 10 430 79 100 100 1385 286

3 35 00 00 0000 00 400 79 100 100 1431 279

4 55 00 100 0313 00 270 90 100 100 1449 283

5 55 00 100 0313 05 260 81 100 100 1438 287

6 35 00 100 0313 05 440 75 100 100 1352 288

7 55 100 100 0313 00 3850 100 100 100 I 510 291

8 55 00 00 0000 00 250 41 100 100 1260 288

9 35 100 00 0000 00 3400 89 100 100 1446 283

10 55 00 100 0313 10 300 69 100 100 1369 287

11 35 100 100 0313 05 3300 83 100 100 1451 290

12 35 100 100 0313 10 3300 80 100 100 1341 291

13 55 100 00 0000 00 3900 96 100 1 0 1294 288

14 55 100 100 0313 10 4000 99 100 100 1354 294

15 35 00 100 0313 00 360 86 100 100 1330 286

16 35 100 100 0313 00 3300 98 100 100 1308 290

Abbrcv iations for Table 2 InCoi n = Inorganic Contamination, SuCom = Surfactant Contamination; Lub = Lubncant; gL = grams per liter, RiCo = First Rinsing Composition

Table 3 EFFECT OF FIRST RINSING COMPOSITION CHARACTERISTICS, SUBGROUP I 2

Run Directly Controlled Characteristics of First Rinsing Composition Results

PH InCom, as SuCom, as % ClOl 1102, g/L Conductivity of RiCo, μ % Waterbreak Free on: COF HNT

% of of Cleaner MOD, g/L Siemens Cleaner ESW ID ISW

1 3 50 00 00 oooo oooo 330 0 93 18 1 048 283

2 3 50 00 100 0086 0 226 390 1 68 5 1 073 286

3 3 50 100 00 0000 0000 3400 31 100 96 1 053 279

4 3 50 100 100 0086 0 226 3450 55 100 100 1 304 283

5 5 50 00 00 0000 0000 300 0 86 19 0 970 287 ro 6 5 50 00 100 0086 0 226 275 4 78 0 1 178 288 O

7 5 50 100 00 0000 0000 4050 95 100 100 1 042 291

8 5 50 100 100 0086 0 226 4000 94 100 100 1 048 288

Abbreviations for Table 3 InCom = Inorganic Contamination, SuCom = Surfactant Contamination, ClOl MOD = Atoguard™ ClOl A Modified Metal Working Lubncant, g/L = grams per liter, 1102 = Atoguard^{1 M} I- 101 Metal Workmg Coolant; RiCo = First Rinsing Composition

by the addition of specific ions such as aluminum or fluoride to the Stage 3 rinsing com¬ position. This Subgroup 1.3 was designed to determine which factor(s) were actually controlling for this result. The following factors were used: pH (3.5 vs 5.5); temperature (27 - 58 °C); total concentration of Inorganic Contamination (0 vs 0.004 M); the nature ofthe Inorganic Contamination (aluminum sulfate vs sodium sulfate); the Organic Con¬ tamination (none vs 10 % of that in the principal cleaner); and the calcium concentration. The Organic Contamination consisted of 0.08 g/L of Triton N101, 0.056 g/L of Surfon¬ ic™ LF-17, 0.156 g/L of Daralube™ 17A2 and 0.156 g/L of Mobil™ DTE24. The fol¬ lowing results were measured: conductivity ofthe Stage 3 rinsing composition; percent ofthe area of the exterior sidewall which was waterbreak free; percent waterbreak free on the interior dome; percent ofthe area ofthe interior sidewalls which remained water- break free; and surface carbon which remained on the cans after cleaning and drying.

The compositions ofthe Stage 3 rinsing compositions and the results were mea¬ sured and analyzed in the same manner as for the previous Subgroups. The results are shown in Table 4. Two of the results, % WBF ESW and surface carbon, had factors which had an effect at the 95 % confidence level: The aluminum concentration and the Organic Contamination in the Stage 3 rinsing composition had a positive effect on % WBF ESW at the 95 % confidence level. As the concentration of aluminum or the Or¬ ganic Contamination in Stage 3 increased, the percentage of the exterior sidewalls ofthe cans which were waterbreak free increased. The concentration of sodium in the Stage 3 rinsing composition did not have a significant effect on % WBF ESW. Therefore, the positive effect of the Inorganic Contamination in the Stage 3 rinsing composition on % WBF ESW observed in Subgroups 1.1 and 1.3 was probably due to the increase in alum¬ inum concentration as the result of increased Inorganic Contamination. It is surprising that the positive effect of increased aluminum concentration on %

WBF ESW is not reflected by a reduction in surface carbon as the aluminum concentra¬ tion in the Stage 3 rinsing composition increases. While the main effect of aluminum concentration on surface carbon is not significant at the 95 % confidence level, the aver¬ age effect, the mid point ofthe range, is positive. The positive effect of the Organic Con- tamination level on % WBF ESW is reflected in its negative effect on surface carbon. The higher the concentration of Organic Contamination the lower the surface carbon. The sodium concentration in the Stage 3 rinsing composition also has a negative effect Table 4 EFFECT OF FIRST RINSING COMPOSITION CHARACTERISTICS, SUBGROUP 1 3

Run Directly Controlled Characteristics of First Rinsing Composition Results

PH Temp, Alum Concen¬ Na-SO,, ToOr, CaCI-, Conductivity of RiCo, μ % Waterbreak Free on: Surface ° C tration, M M g Siemens Carbon,

ESW ID ISW μgire

1 5 5 26 7 0 0 0446 o ooo 1200 58 100 100 169

2 3 5 26 7 0 0 0000 oooo 1000 5 100 98 187

3 5 5 100 0 0 0 446 o ooo 1200 34 100 100 160

4 5 5 26 7 0 0 0000 0 368 1300 4 100 100 181

5 4 5 90 0001 0001 0 223 0 184 1700 35 100 100 168 ro 6 3 5 100 0 0004 0 446 0000 1700 31 100 100 169 ro

7 3 5 26 7 0004 0 0 446 0000 2100 86 100 100 169

8 5 5 100 0004 0 0000 0000 3000 94 100 100 190

9 5 5 100 0 0004 0446 0 368 2300 45 100 100 148

10 3 5 26 7 0 0004 0000 0 368 2000 34 100 100 160

1 1 4 5 90 0001 0001 0 223 0 184 1700 48 100 100 171

12 3 5 100 0 0 0446 0 368 1200 26 98 99 163

13 3 5 100 0004 0 0000 0 368 2800 53 100 100 190

14 3 5 100 0 0 0000 0000 1000 26 100 100 167

15 5 5 100 0 0 0000 0 368 1400 9 100 99 191

Table continued on next page ...

Run Directly Controlled Characteristics of First Rinsing Composition Results pH Temp, Alum Concen¬ Na.SO„ ToOr, CaCI-, Conductivity of RiCo, μ % Waterbreak Free on: Surface " C tration, M M g/L g/L Siemens Carbon,

ESW ID ISW μg/ft³

16 3 5 267 0 0 0446 0.368 1400 65 100 100 163

17 5 5 26.7 0.004 0 0446 0 368 3500 100 100 100 169

18 4 5 90 0001 0.001 0.223 0.184 1800 48 100 100 171

19 5 5 267 0 0004 0.000 0000 2100 70 100 100 169

Abbreviations for Table 4 Temp = Temperature; ToOr = Total Organic Concentration; g/L = grams per liter, RiCo = First Rinsing Composition

ro

on surface carbon. It, however, does not have a significant effect on % WBF ESW Subgroups 1 4. 1.5. and 1 6 These Subgroups examined additional variables to buttress or refute the tentative conclusion from Subgroup 1 3 that the concentration of aluminum cations in the Stage 3 rinsing composition was the primary factor affecting the degree of waterbreak observed after completion of Stage 3 The independent variables in these ex¬ periments were the pH and aluminum concentration ofthe Stage 3 rinsing composition. The measured results were conductivity, % WBF ESW, % WBF ED, % WBF ISW, and surface carbon. The Organic Contamination of the Stage 3 rinsing compositions re¬ mained constant at 10 % of that of stage 2 and the rinsing composition temperature was constant at 32.2 °C. The Organic Contamination in Stages 2 and 3 consisted of Mobil

DTE 24, cupper lubricant, and Daralube 17A2, body maker coolant, in Subgroup 1 4. The concentrations were the same as in Subgroup 1 3

DTI 5600 M4, cupper lubricant (28.6 parts by weight), Atochem SDO5L54N2J, bodymaker coolant (46.8 parts by weight), and Mobil 629 (24.6 parts by weight), hy- draulic lubricant, constituted the Organic Contamination in Subgroup 1.5. In Subgroup 1.6, the Organic Contamination consisted of DTI 5600 M4, cupper lubricant (28.8 parts by weight), DTI 5600 WB, bodymaker coolant (40.0 parts by weight), and Mobil 629, hydraulic lubricant (31.2 parts by weight).

The compositions ofthe Stage 3 rinse compositions and the measurements ofthe results for Subgroup 1 4 are shown in Table 5 Only one ofthe dependent variables, %

WBF ESW, had factors which had an effect at the 95 % confidence level. The aluminum concentration in the Stage 3 rinsing composition had a positive effect on % WBF ESW Neither dependent variable had a significant effect on surface carbon

The compositions of the Stage 3 rinsing compositions and the measurements of the results for Subgroup 1.5 are shown in Table 6 Only one ofthe dependent variables, % WBF ESW, had factors which had an effect at the 95 % confidence level The aluminum concentration in the Stage 3 rinsing composition had a positive effect on % WBF ESW Neither factor had a significant effect on surface carbon

The compositions of the Stage 3 rinsing compositions and the measurements of the results for Subgroup 1 6 are shown in Table 7 Each ofthe dependent variables had a single factor which had an effect at the 95 % confidence level The aluminum concen¬ tration in the Stage 3 rinsing composition had a positive effect on % WBF ESW, and the Table 5 EFFECT OF FIRST RINSING COMPOSITION CHARACTERISTICS, SUBGROUP 14

Run Directly Controlled Characteristics of First Rinsing Composition Results tt pH NrI<F*HF Concentration, g/L Alum Concentration, M Conductivity of RiCo, μ % Waterbreak Free on: Surface Car¬ Siemens bon, μg/fr¹

ESW ID ISW

1 45 oooo ooooo 280 51 100 100 192

2 35 0274 00040 2150 80 100 100 187

3 35 0000 ooooo 340 36 100 100 202

4 55 0274 00040 2900 89 100 100 198

5 55 0110 00016 1200 98 100 100 189

6 55 0274 00040 2950 100 100 100 198

7 55 0000 ro ooooo 260 35 93 95 194

8 35 0110 00016 1050 61 100 100 205

9 35 0274 00040 2200 56 100 100 200

10 45 0129 00024 1500 98 100 100 190

II 45 0274 00040 2500 95 100 100 184

Abbreviations for Table 5 gL = grams per liter, RiCo = First Rinsing Composition

Table 6. EFFECT OF FIRST RINSING COMPOSITION CHARACTERISTICS, SUBGROUP I 5

Run Directly Controlled Characteristics of First Rinsing Composition Results

#

PH NH,F^«HF Concentration, g/L Alum Concentration, M Conductivity of RiCo, μ % Waterbreak Free on: Surface Car¬ Siemens bon, μg/fr²

ESW ID ISW

1 4.5 oooo 0.0000 240 18 98 99 201

2 35 0274 00040 na 63 100 100 192

3 35 0000 ooooo 560 25 100 100 178

4 55 0274 00040 3300 100 100 100 207 ro 5 55 0.111 00016 1500 94 100 100 202

6 55 0274 00040 3400 100 100 100 219

7 55 0000 ooooo 270 20 66 98 205

8 35 0111 00016 1150 75 100 100 194

9 35 0274 00040 2300 70 100 100 201

10 45 0162 00024 1800 98 100 100 188

II 45 0274 00040 3000 96 100 100 211

Abbreviations for Table 6. g/L = grams per liter, RiCo = First Rinsing Composition.

Table 8

CORRELATION COEFFICIENTS BETWEEN SURFACE CARBON AND PERCENT OF EXTERIOR SIDEWALLS THAT ARE WATERBREAK-FREE

Subgroup # CorCoef (R) CoefDet (R²) Spearman's Rho

1.4 -0.101 0.010 0.004

1.5 -0.474 0.224 -0.388

1.6 0.393 0.155 0.447

1.7 -0.347 0.121 -0.391

Abbreviations for Table 8: CorCoef = Correlation Coefficient; CoefDet = Coefficient of Determination.

pH ofthe Stage 3 rinsing composition had a positive effect on surface carbon.

Correlation Between % Waterbreak Free ofthe Exterior Sidewall and Surface Carbon: The % WBF ESW and Surface Carbon data from Subgroups 1.4 - 1.7 were tested to de¬ termine whether any type of correlation existed between % WBF ESW observed and the amount of surface carbon found on the cans. The correlation coefficients, coefficients of determination and Spearman's Rho values from these calculations are reported in

Table 8.

The absolute values ofthe correlation coefficient and Spearman's Rho are closer to zero than to one in all cases. Therefore, there is very little correlation between % WBF ESW and Surface Carbon under the conditions of these examples. The small value of the coefficient of determination, R², indicates that very little of the variation in %

WBF ESW is "explained by" the variation in the quantity of surface carbon.

Although the absolute values of R and Rho are relatively small, the positive sign of both coefficients that was found for Subgroup 1.6 was unexpected. In general, it would be expected that as the quantity of surface carbon on a can increased, the percent of its exterior sidewall surface which remained waterbreak free would decrease. A nega¬ tive correlation between waterbreak and surface carbon is expected because it is general¬ ly assumed that a higher quantity of surface carbon implies more organic material on the surface and more organic material on the surface means that it will be more difficult for water to wet the surface. These results imply that all organic compounds are not equal with regard to their effect on waterbreak. In the case of this lubricant combination, DTI 5600-M4: Atochem SDO5L54N2J:Mobil 629, it appears that some ofthe organic com¬ ponents decrease waterbreak as the amount of them on the surface increases. This does not appear to be the case with the other lubricant combinations tested in these Subgroups. The surface carbon measurements were made on dried cans, while the waterbreak observations were made on wet cans prior to drying. Changes in the surface composition may occur during drying. Some organic compound(s) that promote waterbreak may be sufficiently volatile to evaporate during drying. Thus, although the surface carbon results on dried cans do not correlate with expectations regarding waterbreak on wet cans, they may not reflect the surface conditions on the wet cans.

The surface carbon measurements include the carbon present on both the interior and exterior sidewall surfaces. In the examples noted here, differences in the amount of waterbreak are in most cases only observed on the exterior sidewalls. Small differences in the surface carbon on the exterior sidewall surface may affect waterbreak but may not be large enough to significantly change the surface carbon measurement. Summary of Group 1 : The aluminum concentration in the rinsing composition following the cleaner is the only factor among the ones tested which consistently has an effect on the percent ofthe exterior sidewall surface of aluminum cans which are waterbreak free after principal cleaning with cleaning compositions which contain aluminum and organic soils. In all ofthe examples, as the concentration of aluminum in the rinse increased the percentage ofthe exterior sidewall which remained waterbreak free also increased. GROUP 2

In experiments which were conducted using the same general techniques as in Group 1 , but which utilized first rinsing compositions that were made by inputting only tap water and materials carried over from the principal cleaner into the first rinsing zone, instead of utilizing precisely controlled first rinsing compositions as in Group 1 , it was observed that:

( 1 ) The amount of waterbreak increased as the concentration of lubricant contamina- tion in the cleaner increased and as the cleaning time increased.

(2) Cans which were cleaned for 30 seconds had significantly less waterbreak than those cleaned for 180 seconds. (3) A higher concentration of free fluoride (more negative relative millivolts com¬ pared to a standardized fluoride sensitive electrode as described above) in the cleaner bath also had a detrimental effect on waterbreak.

(4) The spray pressure of the cleaner, the ratio of surfactants with alkylaryl hydro- phobe groups to surfactants with aromatic-free hydrophobe groups, and the total concentration of surfactant in the cleaner did not have a significant effect on waterbreak.

(5) When the contamination levels in the Stage 3 rinsing composition were increased by decreasing the quantity of water which was added to the rinse stage to dilute the cleaner composition being carried in from stage 2 by the cans, the following changes were observed in the Stage 3 rinsing composition, (i) the pH decreased; (ii) the concentration of dissolved aluminum increased; (iii) the concentration of Organic Contamination, which consists of both can forming lubricants and sur¬ factants from the cleaner, increased; (iv) the temperature increased; and (v) the amount of waterbreak observed on the cans during Stage 3 after the cessation of spraying decreased.

Claims

1. A process for cleaning and rinsing a soiled aluminum surface, said process comprising steps of:

(II) contacting the soiled aluminum surface with an aqueous liquid principal cleaner composition comprising acid and surfactant for a sufficient time at a sufficient temperature and while maintaining sufficient relative motion between the principal cleaner composition and the soiled aluminum surface to transfer at least part of the soil from the aluminum surface into solution, suspension, or both solution and suspension in the principal cleaner composition, so that the soiled aluminum surface is converted to a cleaned aluminum surface; and

(III) discontinuing contact between the cleaned aluminum surface and the principal cleaning composition, except for any ofthe principal cleaning composition that resists the force of natural gravity to remain in place as a surface layer on the cleaned aluminum surface after discontinuing contact between the bulk of the principal cleaning composition and any such surface layer, and thereafter rinsing the cleaned aluminum surface by contacting the cleaned aluminum surface with an aqueous liquid first postcleaning rinse composition comprising at least about 0.001 moles per liter ("A-f ) of aluminum cations, so that the cleaned aluminum surface is converted to a cleaned and first rinsed aluminum surface.

2. A process according to claim 1, wherein (i) the principal cleaner composition comprises from about 0.1 to about 8 g/L of alkoxylated alkyl phenol nonionic surfactant molecules and sufficient free fluoride to correspond to an electrical potential reading from a fluoride sensitive electrode in contact with the composition that is from about -15 to about 50 mV compared to a standardized fluoride sensitive electrode and (ii) the concentration of aluminum cations in the first postcleaning rinsing composition is from about 0.0002 to about 0.1 M.

3. A process according to claim 2, wherein (i) the principal cleaner composition comprises^*. from about 0.15 to about 6 g/L of alkoxylated alkyl phenol nonionic surfact¬ ant molecules; an amount of alkoxylated aliphatic alcohol nonionic surfactant molecules such that the ratio ofthe concentration ofthe alkoxylated alkyl phenol surfactant mole¬ cules to the alkoxylated aliphatic alcohol molecules is from about 0.4: 1.0 to about 5:1.0, from about 2 to about 18 points of Free Acid; and sufficient free fluoride to correspond to an electrical potential reading from a fluoride sensitive electrode in contact with the composition that is from about -10 to about 40 mV compared to a standardized fluoride sensitive electrode and (ii) the concentration of aluminum cations in the first postcleaning rinsing composition is from about 0.0005 to about 0.08 M.

4. A process according to claim 3, wherein (i) the principal cleaner composition comprises: from about 0.20 to about 5 g/L of alkoxylated alkyl phenol nonionic surfact¬ ant molecules in which the alkyl groups have from 5 to 20 carbon atoms; an amount of alkoxylated aliphatic alcohol nonionic surfactant molecules, with from 8 to 22 carbon atoms in the aliphatic alcohol residue moiety thereof, such that the ratio ofthe concentra¬ tion of the alkoxylated alkyl phenol surfactant molecules to the alkoxylated aliphatic alcohol molecules is from about 0.6.1.0 to about 3:1.0; from about 2 to about 18 points of Free Acid; and sufficient free fluoride to correspond to an electrical potential reading from a fluoride sensitive electrode in contact with the composition that is from about -7 to about 30 mV compared to a standardized fluoride sensitive electrode and (ii) the con¬ centration of aluminum cations in the first postcleaning rinsing composition is from about 0.001 to about 0.08 M.

5. A process according to claim 4, wherein (i) the principal cleaner composition comprises: from about 0.25 to about 4.0 g/L of alkoxylated alkyl phenol nonionic surfact- ant molecules in which the alkyl groups have from 6 to 16 carbon atoms; an amount of alkoxylated aliphatic alcohol nonionic surfactant molecules, with from 10 to 20 carbon atoms in the aliphatic alcohol residue moiety thereof, such that the ratio ofthe concentra¬ tion of the alkoxylated alkyl phenol surfactant molecules to the alkoxylated aliphatic alcohol molecules is from about 0.8:1.0 to about 2.5:1.0; from about 4 to about 16 points of Free Acid; and sufficient free fluoride to correspond to an electrical potential reading from a fluoride sensitive electrode in contact with the composition that is from about -7 to about 20 mV compared to a standardized fluoride sensitive electrode and (ii) the con¬ centration of aluminum cations in the first postcleaning rinsing composition is from about 0.002 to about 0.06 M.

6. A process according to claim 5, wherein (i) the principal cleaner composition comprises: from about 0.30 to about 3.0 g/L of ethoxylated alkyl phenol nonionic surfact- ant molecules in which the alkyl groups have from 7 to 15 carbon atoms; an amount of alkoxylated aliphatic alcohol nonionic surfactant molecules, with from 10 to 18 carbon atoms in the aliphatic alcohol residue moiety thereof, such that the ratio ofthe concentra¬ tion of the ethoxylated alkyl phenol surfactant molecules to the alkoxylated aliphatic alcohol molecules is from about 1.0: 1.0 to about 2.0: 1.0; from about 5 to about 15 points of Free Acid; and sufficient free fluoride to correspond to an electrical potential reading from a fluoride sensitive electrode in contact with the composition that is from about -1 to about 15 mV compared to a standardized fluoride sensitive electrode and (ii) the con¬ centration of aluminum cations in the first postcleaning rinsing composition is from about 0.003 to about 0.06 Ki.

7. A process according to claim 6, wherein (i) the principal cleaner composition comprises: from about 0.35 to about 2.2 g/L of ethoxylated alkyl phenol nonionic surfact¬ ant molecules in which the alkyl groups have from 8 to 13 carbon atoms; an amount of alkoxylated aliphatic alcohol nonionic surfactant molecules, with from 12 to 16 carbon atoms in the aliphatic alcohol residue moiety thereof, such that the ratio ofthe concentra¬ tion of the ethoxylated alkyl phenol surfactant molecules to the alkoxylated aliphatic alcohol molecules is from about 1.1 : 1.0 to about 1.8: 1.0; from about 5 to about 15 points of Free Acid; and sufficient free fluoride to correspond to an electrical potential reading from a fluoride sensitive electrode in contact with the composition that is from about -4 to about 10 mV compared to a standardized fluoride sensitive electrode and (ii) the con¬ centration of aluminum cations in the first postcleaning rinsing composition is from about 0.004 to about 0.06 Ki.

8. A process according to claim 7, wherein (i) the principal cleaner composition comprises: from about 0.55 to about 0.75 g/L of ethoxylated alkyl phenol nonionic sur- factant molecules in which the alkyl groups have from 8 to 11 carbon atoms each; an amount of ethoxylated and propoxylated aliphatic alcohol nonionic surfactant molecules, with from 12 to 14 carbon atoms in the aliphatic alcohol residue moiety thereof, such that the ratio ofthe concentration ofthe ethoxylated alkyl phenol surfactant molecules to the ethoxylated and propoxylated aliphatic alcohol molecules is from about 1.3 : 1.0 to about 1.5: 1.0; from about 6.5 to about 15 points of Free Acid; and sufficient free fluoride to correspond to an electrical potential reading from a fluoride sensitive electrode in contact with the composition that is from about -4 to about 5 mV compared to a standardized fluoride sensitive electrode and (ii) the concentration of aluminum cations in the first postcleaning rinsing composition is from about 0.007 to about 0.06 Ki and the pH ofthe first postcleaning rinsing composition is from about 3.4 to 5.6. s 9. A process according to any one of claims 1 - 8, wherein the first postcleaning rinsing composition is prepared by mixing tap water with:

(A) at least a portion of (i) a part ofthe principal cleaning composition that resists the force of natural gravity to remain in place as a surface layer on the aluminum surface being cleaned, or on other aluminum surfaces previously cleaned with the o same principal cleaning composition, after discontinuance of contact between such surface layer and any other part of the principal cleaning composition, (ii) a concentrated liquid composition or a solid composition formed by drying a portion of a surface Iayer of principal cleaning composition as recited in part (i), or (iii) both parts (i) and (ii); and, optionally, 5 (B) overflowed liquid from another aqueous rinse solution used in a distinct process step carried out subsequent to step (III).

10. A process according to any one of claims 1 - 9, said process comprising an additional step (I), performed before step (II), of contacting the soiled aluminum surface with an aqueous acidic precleaner composition that does not contain surfactants added 0 specifically to aid in cleaning.