. 4 %

c , 1 . . . .

Aerospace Symposium January 25,1993

Table of Contents

A Survey of Stress in Nickel Deposits ................................................................................................................. 1 Roger W. Derby, Quality Control Instruments, Oak Ridge, TN Paper Not Available in Time For Publication

Ion Beam Enhanced Deposition of Hard Chrome Coatings ............................................................................... 2 Dr. Amold Deutchman and Robert J. Partyka, E3eamAlloy Corporation, Dublin, OH

A Non-Chromium, Non-Heavy Metal Deoxidizer System for Aluminum and Its Alloys .................................... 9 Phillip Johnson, Parker+A", Madison Heights, MI

Reduction of Cadmium Plating on Aircraft Wheels and Brakes ...................................................................... 26 Crittenden Ohlemadrer and Mark W. Gilbert, Aircraft Braking System Corporation, Akron, OH

Brush and Flow Selective Sulfuric Acid Anodizing .......................................................................................... 29 Joseph Nonis, SlFCO Selecbve Plating, Cleveland, OW

Hazardous Minimization: Saving Time, Money and the Environment ..._ ........................................................ 39 Mary Beth Fennel1 and Mark Roberts, Naval Aviation Depot, MCAS Cheny Point, NC

Low VOC Organic Finishing ............................................................................................................................... 47 Ed B m n f o q Crown Metro Aerospace, Atlanta, GA Paper Not Available in Time Far Publicatim

Regulatory Considerations in Choosing a Cleaning Solvent or Process ..................................... .................. 48 Steve Risotto, Center for Emissions Control, Washington, DC

1 ,I ,l-Trichloroethane Vapor Degreasing Elimination in Aerospace Repair Applications .......................................................................................................... .. .............. !i7 Glenn Travis. Environmental Management Services, Jennings, OK, and Cynthia Bosteq American Airlines, Tulsa, OK

Organic and Ionic Soil Removal by Vapor Degreasing with a Heterogeneous Azeotrope of Perchloroethylene and Water .......................................................................................................................... 63 Eric Maim, Vulcan Chemicals, Wichita, KS

I Paper Not Available In Time For Publication I

A Survey of Stress in Nickel Deposits Roger W. Derby, Quality Control Instmments. Oak Ridge, TN

Theinfluenceof stressinelectmdepositswasfmtobservedinthe 187Os.Aclual measurements of the effect, using the curve strip techniqce, were made in the early 1 9 0 0 s . T h e s p i r a l c ~ ~ c ~ ~ ~ w ~ ~ ~ ~ b y B r e n n e r a n d S e n d e r o f f at the National Bureau of Standards in late 1940s. Although their approach represented a distinct improvement over the curve strip, it still presents theoretical and practical problems. This presentation will focus on sources and remedies for deposit stress, the development of the spiral contractmeter, and new approaches that may form the basis of stress measurements in the 21st Century.

For copies of this paper, please contact the author directly:

Mr. Roger Derby Quality Control Instruments Post Office Box 5345 Oak Ridge, TN 37831 61 5/483-6498

1

, .. . . ~ . . . .

. . .

ION BEAM ENHANCED DEPOSITION OF HARD CHROME COATINGS

Arnold H. Deutchman and Robert J. Partyka BeamAlloy Corporation

Abstract

An advanced ion beam enhanced deposition process has been developed which is capable of depositing hard chrome coatings on most metallic, ceramic, plastic, and glass substrate materials. The hard chrome coatings are alloy& directly into the substrate surface thereby eliminating the need for the predeposition of interlayers and optimizing coating adhesion. The coatings are hydrogen-free and fully dense with a measured hardness exceeding 1100

(0.0004 inches) thick can be deposited on components that are manipulated during processing to allow selective area coating and provide for a uniform coating thickness.

Knoop (5 grams). coatings up to 10 microns

Background

The deposition of metallic coatings such as chromium, nickel, and cadmium is u s s l y achieved by wet chemical plating processes, and is done on a wide variety of engineered components and tools. There are however other techniques capable of depositing adherent metallic coatings that show much potential for use in the industrial marketplaq if certain technological and economic hurdles can be overcome. Vacuum coating techniques, which were initially developed for the deposition of thin fi lm in the electronic and optical industries, represent one class of deposition processes that are potentially attractive for the coating of mechanical components. The driving forces for investigating alternative deposition technologies include the need for better coated component performance and the increasing pressures on reducing the environmental effects of chemical plating processes. Both of these issues can be addressed by vacuum coating techniques and form the basis for increasing interest in developing these processes.

There are five general classes of vacuum coating processes that can potentially be used for the deposition of chromium and other metallic coatings. All involve the production of energetic metal atoms or ions and the transport of those energetic metallic particles to the surface to be coated, either by electric fields or by using a plasma of energetic inert particles. The properties of the metallic coatings produced will be determined by the nature of the deposition process chosen. Thus not all of the processes can be used to deposit chromium (or other metallic) coatings that will have the desired engineering properties. The five general classes of vacuum deposition processes include vacuum evaporation (VE); sputtering (SP); ion plating (IP); direct ion deposition (DD); and ion beam enhanced deposition (II3ED).IJ The key factors that differentiate the individual'processes are the energy of arrival of the netallic atom at the surface to be coated, and the degree of control provided by each prmss over this critical energy. This arrival energy will determine to a great extent coating morphology, residual stresses, physical properties, and adhesion.'-s

Figure 1 lists the relative energy ranges of the metallic atom that reach the surface of a component treated with all five of the general classes of vacuum deposition processes. Vacuum evaporation is the lowest in energy of all of the vacuum deposition processes. The metallic atoms that reach the surface to be coated by vacuum evaporation have energies in the 0.1 to 1.0 electron volt (ev) range. At these energies the atoms come to rest on, and can move around the surface being coated. They begin to agglomerate at nucleation sites and a film coating grows. Coating adhesion is determined solely by van der Waals attraction and is usually not strong. Industrial applications of vacuum evaporation include web coating of thin mylar films with aluminum; fabrication of thin film electronic

4

2

components such as resistors, capacitors, and strain gages; and coating of optical elements such as lenses and mirrors with aluminum, gold, and silver.6

IO - 1 -

5 0.1 - 0.01 .

1000 1

100 a

o.Ooo1 J . 1 i

VE SP IP DD BED

Figure 1. The arrival energy ranges of coating particles found with the five general classes of coating deposition processes.

.-

Sputtering @C, RF, or magnetron) is a higher energy process that utilizes a glow discharge to produce and transport coating atoms to the surface to be coated. Atomic particle arrival energies up to the 5 keV range are achievable, although the mean energies of the arriving atoms are two orders of magnitude lower. Atoms that arrive at the surface to be coated at these energies can react with the surface itself to form actual chemical bonds thereby improving adhesion. Pure metallic coatings as well as refractory nitride, oxide, and carbide coatings can be deposited with glow discharge based sputter coating processes. Sputter deposition is used primarily in the electrical component (integrated circuit) industry for the deposition of insulating and conductive thin films, and in the optical industry for the deposition of optical thin film coatings.'J It is also used routinely for the deposition of decorative chromium coatings on plastic components for automobile and consumer products.'

Ion plating is a vacuum evaporation process assisted by a glow discharge. The ion plating process can be implemented in one of two major forms. Simple ion plating involves evaporation of coating materials (usually a metal like aluminum or gold) into a DC- or RF- supported glow discharge. The atoms of the metal coating material are transported to the surface to be coated, and a coating is grown. In reactive ion

platmg, a reactive gas is added to the glow discharge. This produces a chemical reaction between the evaporated metal and the reactive gas. Compounds such as metallic nitrides can be formed by the chemical reactions and will condense on the surface to be coated. Simple ion plating is used primarily for the deposition of metallic coatings such as gold, silver, and aluminum. In one widely used ion plating process,, aluminum coatings are deposited as an alternative to cadmium plating for corrosion resistance.1%12 Reactive ion plating, also termed physical vapor deposition (PVD), is widely used to deposit hard coatings of titanium and chromium, nitride and carbonitride compounds. These Coatings are routinely deposited on a wide variety of metallic tools and engineered components for increased wear-re~istance.~'

Since ion plating is a sputter-assisted deposition process, the distribution of coating particle energies is approximately the same as is found with glow discharge sputtering. Coating adhesion is determined primarily by chemical reaction at the surface being coated and is improved at elevated temperatures by thermal diffusion. One of the amactive feahnes of ion plating is the increased rate of coating deposition. In general, most metals can be evaporated much faster than they can be sputtered. Thus much more material is liberated and available for coating growth and thicker coatings can be deposited at faster rates than can be deposited with simple sputtering.

Direct deposition of chromium and other metals has also been attempted in order to increase the energy of arrival of the coating species at the surface to be coated. The metallic species is first vaporized and ionized in an ion source, and then extracted at high energy by a high electrical potential.1c16 Metallic particles with energies up to approximately 10 keV have been used to build coatings on the surface of various materials. Improvements in adhesion have been noted because of actual penetration of the accelerated atoms into the surface to be coated. The flux of atomic particles that can be produced economically with this type of deposition technology is low however, which limits the rate at which coatings can be produced. Thus even though direct deposition technology produces adherent coatings, and allows accurate control over the energy of the arriving particles, it is not a viable deposition technology for industrial scale-up.

2 3

Ion beam enhanced deposition is a relatively new deposition process which combines many of the attractive fcatures of the other classes of vacuum deposition techniques with the ability to deposit metallic coatings relatively rapidly, and with a high degree of control over the properties of the deposited coating.*' The BED process is a simultaneous thin film deposition with illumination of the growing film by a second, independently controlhble, beam of energetic particles. A high vacuum environment is required for propagation of the secondary ion beam, which limits the thin film deposition technique used in the IBED process to either vacuum mporation or ion beam sputtering. Since the secondary ion beam

. is independently controllable, the 'energy of the particles in the beam can be varied over a wide range and chosen within ahery narrow window. This allows the energetics of deposition to be varied during the coating cycle and allows optimization of coating properties such as interfacial adhesion, density, morphology, and internal stresses. The wide energy range and controllable flexibility of the BED process enable the deposition of chromium coatings on the substrate materials reported in this work.

IBED Process

Ion beam enhanced deposition (IBED) is carried out in a high vacuum, at pressures of lxlW Torr or below. With proper choice of deposition parameters the temperature rise in components being processed can be held below 50 degrees Centigrade (92 degrees Fahrenheit). A general diagram of the implementation of the IBED process is seen in Figure 2. The surface to be coated is first illuminated with a flux of high energy inert gas ions that is initially used to remove surface oxides and other contaminants. This high energy flux is maintained, and once the surface is cleaned, a flux of pure chromium atoms is then directed simultaneously at the surface to be coated. The high energy inert gas ions are used to mix the initial few atomic layers of chromium into the surface being coated. This forms an alloyed bond layer in the surface that promotes adhesion of the chromium layer and is the mechanism that allows chromium coatings to be applied to virtually any substrate material without the need of an intermediate bonding layer of copper or nickel. Once the bond layer is formed properly, a chromium coating is then allowed to grow out from this bond layer. The high energy

chromium Atom

-2 Figure 2. Schematic diagram of the implementation of the BED process. Growth of a pure chromium coating is shown. Other metallic coatings, such as titanium, aluminum, copper, silver, and gold can also be deposited with the BED process.

inert gas ion flux is then used to control the morphology of the chromium coating that is being grown from the d a c e . This allows control over the grain structure of the coating as well as coating density and residual stresses. Chromium coatings in the thickness range of 5-10 microns (0.0002 - 0.0004 inches) can be grown with the techniques on most metallic, glass, ceramic, and plastic surfaces.

IBED Chromium Coatings - Characterization

Chromium coatings have been deposited directly on a wide range of metal alloy materials. Figure 3 is a scanning electron micrograph (SEM) of a typical pure chromium coating deposited on a metailic surface, in this case, copper. Coupons cut from copper (C11000, electrolytic tough pitch) sheet were buf€ed and polished to a mirror finish and then cleaned in acetone and methanol solvents to remove residual bdling compound. The copper was then coated with a 5 micron (0.0002 inch) thick layer of chromium using the BED process described above. The original polish on the copper coupon was maintained, and as can be seen in the SEM image, the chromium coating is smooth, continuous, and pinhole-free. Due to the nature of the IBED process

3

the chromium atoms that nucleate on the surface haw a high degree of initial adatom

coating and the top surface of the copper coupon. Therefore when the chromiumxopper interface is i n

Figure 3. SEM image of a 5 micron (0.0002 inch) thick coating of pure chromium on a polished copper substrate. (SEM: AMR, Model 1000).

mobility and the coatings that subsequently grow replicate the finish of the surface being coated. Thus chromium coatings deposited on polished surfaces with the IBED process do not require postcoating polishmg operations to restcre the original surface finish.

The interfacial region between the BED chromiun; coating and the copper substrate was examined by SEM and is pictured in Figure 4. An BED chromium coated copper coupon was sectioned, mounted and polished using standard preparation techniques, and then imaged. The sample was not acid etched prior to imaging. The chromium coating is the uniform structure on top of the copper base material. The softer copper base material shows remnants of embedded polishing media and residual polishing marks and streaks. The chromiumapper interface appears very uniform with no evidence of delmnation. The chromium coating itself shows no evidence of voids or cracks, and is smooth and featureless at this magrufcation (SOOOX). The brighter color of the uppermost regon of the chromum coating is due to a focuslng artifact from the SEM The chromium layer is much harder than the copper substrate, which after grinding and polishing leads to a slight sample height variation between the uppermost layers of the chromum

4

Figure 4. SEM image of the interfacial region at an IBED chromium coated copper surface. (SEM: AMR, Model 1000).

focus, the uppermost region of the chromium coating is out of focus and appears less resolved and brighter on the SEM image.

Further characterization of the BED chromium coating was done by X-ray photoelectron spectroscopy ( X P S ) in order to assess the chemical purity of the BED chromium. The XPS spectra of both the chromium layer and the copper substrate are displayed in Figures 5 and 6 , respectively. The spectrum from the chromium layer clearly shows strong K-alpha (5.29 kev) and K k t a (5.92 keV) X- ray peaks characteristic of pure chromium metal. Spectral lines from other elements including aluminum, silicon, copper, oxygen, and carbon present in trace amounts were also detected. The aluminum and silicon are residual contaminants left from the sample polishing media (alumina and silica). Likewise, the presence of copper is due to smearing of the copper substrate into the exposed face of the chromium coating during sample sectioning and polishmg. Trace amounts of carbon and oxygen are also present and are most likely remnants of sample polishing and mounting rather than contaminants introduced into the BED chromium coating dunng deposition.

5

The spectrum from the copper substrate clearly shows the strong K-alpha (8.05 kev) and IC-beta (8.86 kev) lines characteristic of pure copper. Also detected were peaks from aluminum, silicon, iron, and chromium, all in trace amounts, and all probably remnants of the sample preparation procedure.

Twe AI - 6061 c u - 110 Fe - 1018

Figure 5. X-ray photoelectron spectrum from the IBED chromium layer. (Princeton GanunaTech - IMIX Spectrometer).

Hardness Kl Hardness Kl 91.3 1121.7

101.7 1085.1 326.1 1046.5

Figure 6. X-ray photoelectron spectrum from the copper substrate. (Princeton GammaTech - IMlX Spectrometer).

IBED Chromium Coatings - Hardness

The mechanical properties of BED chromium coatings must be optimum in order for the process to be considered viable for use on engineered components. Since chromium is often specified for use as a hard, wear-resistant coating, the mechanical

hardness of B E D chromium coatings was investigated. The IBED process was used to deposit pure chromium coatings on a variety of engineered metal alloy materials. The microhardness of the deposited coatings was then measured.

A set of samples was prepared from aluminum (6061), copper (1 lo), carbon steel (1018), and alloy steel (4140) stock. Cylindrical coupons, 31.75 mm (1.25 inches) in diameter, and 19 mm (0.75 inches) long, were cut from rod stock. One flat face on each coupon was machined smooth and then lapped progressively with silicon carbide polishing media to a 15 micron (600 microinch) finish. Selected coupons were then lapped further with diamond paste to a 1 micron (40 microinch) finish. Ail sample coupons were then cleaned in acetone and methanol prior to BED coating. Microhardness measurements of the surfaces of selected uncoated samples were made using a h o o p microhardness instrument (Torsion Balance Company: Kentron Microtester) at an indenting force of 5 grams.

Four prepared samples of each alloy material were then IBED coated with chromium. All sixteen sample coupons were coated at the same time with chromium to a thickness of 5 microns (0,0002 inch). The BED chromium layer was deposited directly on the prepared surface of all sixteen samples without the use of any intermediate adhesion layers of copper or nickel. The temperature rise in the coupons during BED coating did not exceed 50 degrees Centigrade (92 degrees Fahrenheit). The d a c e finish on all of the coupons BED coated was not degraded. The 1 micron (40 microinch) finish on the polished samples remained shiny and optically reflective. The BED chromium coating adhered tightly to the surfaces of all four alloy materials coated. Microhardness measurements were made on all coated samples and are listed in Table 1.

Table 1 Knoop Hardness Measurements

I Material I MaterialBase I Chromium I

~~ I Fe-4140 I 380.6 I 1008.5

5

The Knoop hardness measurements of the base material surface and of the BED chromium coating are all averages of three separate measurements on each sample coupon. The Knoop hardness measurements are presented graphically in Figure 7.

1000

800

600

400

200

0

Figure 7. h o o p hardness measurements of the bulk material and the BED chromium coating on each base material. The loading force used for these measurements was 5 grams.

Results

Hard chromium coatings were deposited successfully on all of the aluminum, copper, and steel alloy materials investigated and tested in this experiment. No spalling or delamination of the coatings was found on any of the coated samples. Scanning electron microscopy showed the chromium coatings to be smooth, uniform, void-free, pinhole-free, and well bonded to the substrate. X-ray photoelectron spectroscopy showed the chromium coatings to be very pure and contamination-free. Knoop-scale hardness measurements of the IBED chromium coatings yielded an average that approached 1100 Knoop ( 5 gram). A number of individual hardness values exceeded 1200 Knoop (5 gram). In addition, the IBED chromium coating hardness was found to be essentially independent of the hardness of the base material on which the coating is deposited.

Conclusions

A new, versatile ion beam based deposition technique has been developed which is capable of depositing hard chromium coatings on a wide variety of materials. The BED technique differs from all other conventional vacuum deposition techniques in two key ways. First, the energy of amval of the chromium atoms at the surface to be coated is much higher with the B E D technique than is achievable with conventional vacuum deposition techniques. Second, since the IBED technique uses two independently controllable ion fluxes, the deposition conditions can be changed at different stages of the coating cycle. This provides the ability to produce adherent chromium coatings directly on the surface of a wide variety of materials, as well as the ability to control the morphology of the chromium coating produced. Chromium coatings with h o o p hardnesses exceeding 1100 (5g) were deposited successfully on aluminum, copper, and steel alloys. The new BED technique shows much promise for use as a potential alternative to electrodeposition for depositing hard chromium coatings on engineered components of all types for improved part perfOrmanCe.

References

1. RF. Bunshah, ed., Vacuum Deposition Processes, 1, Noyes Publications, Park Ridge, NJ (1982).

2. K.K. Schuegraf, ed., Handbook of Thin-Film Deposition Processes and Techniques, 1, Noyes Publications, Park Ridge, NJ (1988).

3. C. Weissmantel, J. Vac.Sci.Tech., 18(2), 179-185 (1981).

4, T. Taka@, J.Vac.Sci.Tech. A, 2(2), 382-388 (1984).

5 . D.M. Mattox,J.Vac.Sci.Tech. A, 7(3), 1105-1114 (1989).

6. S. Schiller, U. Heisig, and S. Panzer, Elecfron Beam Technology, 234, John Wiley & Sons, New York, NY (1982).

7. J.A. Thomton and V.L. Hedgcoth, J.Vac.ScI.Tech., 13, 117 (1976).

6

8. L. Maissel and R. Glang, eds.,Handbook of Thin Film Technology, 19-1, McGraw-Hill, New York, N Y (1 970).

9. L. Hughes, R. Lucariello, and P. Blum, Proceedings 20th Technical Conference. Society of Vacuum Coaters, Atlanta, Ga (April 1977), p. 15.

10. K.E. Sturbe and L.E. McCrary, J. Vac.sCi. Tech., 11,362 (1974).

11. 63( lo), 42 (1976).

A.W. Moms, Plating and Surface Finishing,

12. D.E. Muehlberger, Plating and Surface Finishing, 65,20 (1978).

13. J.E. Sundgren and H.T.G. Hentzell, J. Vac.Sci.Tech.A, 4(5), 2259-2279 (1986).

14. T. Takagi, I. Yamada, and A. S a d , J. VacSci. Tech., 12(6). 1128-1 134 (1975).

15. J.Vac.Sci.Tech., 13(2), 591-595 (1976).

J. Amano, P. Bryce,'and R.P.W. Lawson,

16. J. Amano, Thin Solid Films, 92, 115-122 (1982).

17. J.K. Hirvonen, Materials Science Reports, 6, 215-274 (1991).

. . . . . . . . . .

. . . .... . . . .

. .

. . . . . . . . .

. .. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . ' :. . . . . . - - . .

"A Non-Chromium, Non-Heavy Metal Deoxidizer System for Aluminum and its Alloys I'

by Philip M. Johnson, CEF

Research Chemist Parker+Amchem

ABSTRACT

This paper will describe a two-step, non-chrome, non-heavy metal deoxidizer system now commercially available which, when used prior to chrome conversion coating of 2024 T-3 and other aerospace alloys, meets all the requirements of military and commercial aerospace specifications. The first step of the deoxidizer system has cleaning capabilities, and when it is used following a suitable aqueous degreaser, the traditional phosphated-silicated alkaline cleaner is not necessary.

Data will be presented on bare and painted corrosion resistance tests as well as paint adhesion tests. This deoxidizer system can also be used prior to anodizing, welding/brazing, or other processes which must meet military or commercial aerospace specifications.

i

9

A Non-Chromium, Non-Heavy Metal Deoxidizer Bystem for Alum~num and its Alloys

by Philip M. Johnson, CEF Parker + Amchem, Madison Heights, MI

Backcrround In spite of the importance

of deoxidizing and desmutting in aluminum finishing, literature on the subject is somewhat sparse. Even some of the best books on aluminum finishing devote 'little space to deoxidizing or desmutting. For example, the 1987 edition of Wernick, Pinner, and Sheas- by1 (a two volume set, 1325 total pages) devotes 1 1/2 pages to desmutting, which is an improvement over the earlier editions of Wernick and Pinner2 . Brace and Sheasby3 devote 5 pages. Very few papers have been published which are strictly devoted to deoxidizing, although some, such as Ketcham and Brown4, and Mohler5, included deoxi-

* dizing variables in their studies, and in 1975 Smith6 wrote an excellent article on deoxidizing.

Deoxidizing has not drawn the attention that some of us would like to have seen because deoxidizing/desmutting of the majority of wrought aluminum alloys is not difficult. Many shops use commodity baths based on nitric acid, nitric + sulfuric acids, nitric acid with fluoride, or proprietary baths based on ferric sulfate with nitric or sulfuric acid, or sulfuric acid/hydrogen peroxide mixtures. These solutions are quite adequate for most aluminum alloys.

However, the aerospace industry has a unique problem in having to process highly alloyed materials.

In order to achieve the necessarymechanicalproperties to allow aluminum to be used in c r i t i c a l a e r o s p a c e applications, aluminum must be alloyed with other elements. Unfortunately for the manufac- turers and users of aerospace equipment (and fortunately for chemical suppliers and waste treatmentequipmentsuppliers), the majority of aerospace aluminum parts are necessarily made with alloys containing copper.

As most aluminum finishers in the aerospace industry know, the wrought aluminum alloys which present the biggest deoxidizing/desmutting problems are those which contain copper. The reasons for this fact are related to the two metals' position in the electromotive series. Aluminum and copper are quite far apart, and aluminum is on the anodic side, while copper is cathodic; this means basically that there is a strong tendency for galvanic corrosion to occur, and aluminum will have a tendency to dissolve. ,Metallurgically, copper and aluminum do not mix together very well, and alumi- num/copper alloys tend to have c o p p e r a n d c o p p e r intermetallics dispersed as

1

10

.'

L . I

aggregates within the aluminum. Areas of high copper concentration become excellent sites for galvanic and pitting corrosion to occur.

The information we have presented thus far is a review of facts known since alumi- num/copper alloys were devel- oped during World War 11. Aluminum finishers of the post World War I1 era soon found that addition of Chromic acid or other chromates to their deoxidizer tanks prevented pitting corrosion and enabled them to meet military and aerospace specifications for corrosion resistance of fin- ished aluminum parts, when used in conjunction with good quality chromate conversion coatings or anodize.

Through the 1950s and -60s improvements were made in chromated deoxidizers, and the technology peaked following a patent by Dollman7 in 1970 involving the addition of a ferricyanide salt to precipi- tate dissolved copper and other alloying elements which affected the successful opera- tion of the bath at levels as low as 200 ppm. This invention resulted in an increase in deoxidizer bath lives of three to four times over previously available technology. Not only did this result in saving of recharging costs, but through the 1970s and beyond when chromium-containing sludges were identified as hazardous waste, waste treatment cost savings were also realized. Unfortunately we now know that ferricyanide has joined chromium as being extremely hazardous and is quite

difficult to treat prior to disposal.

Until now no one has been able to successfully replace chromateddeoxidizertechnology for critical applications. The only available non-chromate deoxidizers which came close to their performance were those based on ferric sulfate, which were patented in the 1960s 6 t

However, for critical 8

applications such as chromate conversion coating, iron- containing baths are inconsistent and potentially risky, considering that free iron is a contaminant in a chromating bath (our company recommends 10 ppm maximum free iron), and even minor dragout problems can result in loss of that bath in a short time.

One unique chrome-free deoxidizer which is worthy of mention was developed by Bati- uk9. His idea was to follow a non-chrome, iron based deox- idizing solution with a solu- tion which would complete the deoxidizing; that is, oxidize any residual alloying elements on the surface to their more soluble state. Batiuk's system had some disadvantages, however. First, his system used salts of nitrites, which can contribute to the formation of carcinogens. Second, the data which he presented in his patent, in which he followed his deoxidizer with chromate conversion coating, indicated that he was able to pass the salt spray requirements of Mil- C-5541, (168 hours), but not the requirements of Mil-C- 81706, (336 hours). However, Batiuk's concept of deoxidizing in two steps was recognized as

2

11

highly feasible and shown to be practical by McMillen''' '', whose work was the basis for the presently described system.

The system described by McMillen was quite successful in the laboratory, but when it was tried in field tests in cooperation with the Boeing Co., a number of changes were found to be necessary by Carl- sonl2,13 , due to failing intergranular attack and end grain pitting tests. Carlson's optimization of the process produced results which showed that the performance of a chromated deoxidizer could be matched by a two-step, non- chromated system.

Deoxidizinc~ vs. Desmuttinq Before we present a de-

scription of our new system, it will be helpful to distinguish the difference between deoxidizing and desmutting. The following definitions describe the difference between the two:

Deoxidizing is defined as: the removal of oxides and other inorganics which would interfere with normal finishing procedures.

Desmutting is defined as: the removal of pretreat- ment residues without significant attack on the surface of aluminum.

For relatively pure alloys where oxides of aluminum or magnesium are the materials which must be removed, mineral acids such as nitric or sulfu- ric will dissolve them. When the surface inorganics are not very soluble in mineral acids,

as with 2xxx series alloys, the addition of fluoride and an oxidizer such as hexavalent chromium must also be present. There is a perceptible but controllable attack on the aluminum surface by deoxidiz- ers, and some include this parameter, usually termed "etch rate", in their processing control procedures. We use the term vaetchlt with caution, since we do not want to confuse the term with the etch of alkaline etchants, where the etch rate, or chemical attack, is at least ten times that of a deoxidizer. Deoxidizing is necessary almost anywhere, any time that surface impurities might interfere with further processing.

Actually, desmutting is a special type of deoxidizing, as pretreatment residues (smut) are "other inorganics" on the surf ace. However, most solutions termed tldesmutters*t normally have a much lower etch rate than those solutions we call ndeoxidizersaf. Desmutters normally follow an etchant (alkaline 01 acid) and remove those reaction products or alloying constituents which are insoluble in the etchant.

Deoxidizing is an integral part of a number of operations involving aluminum for Aerospace. A list of typical operations where deoxidizing is needed is shown in F i g u r e s 3 and 414. The most critical operation in terms of imparting satisfactory corrosion resistance and paint adhesion properties is conversion coat- ing, and most of our discus- sions from here on will involve conversion coating.

3

12

The Deoxidizina system Now let us look at our new

deoxidizing system, Note that I have used the word Igsystemgg. The reason for this, which we must emphasize, is that the deoxidizing is done in two steps: an acid etch/cleaner, which removes aluminum and magnesium oxides but leaves a smut of most of the other alloying elements, and a desmut solution, which completes the oxidation of the other alloying elements to their more soluble ionic states. 1

It might be helpful at this point to look at what a deoxidizer does to an aluminum alloy surface. Table 2 repre- sents the results of a surface analysis performed by Auger Electron Spectroscopy by Dr. Jack Kramer of our Analytical Department'' . Sample 1 rep- resents a #'Cleaned Only" sur- face; a bare 2024-T3 panel was immersed in a silicated\phos- phated cleaner, rinsed with deionized water, and air-dried. The magnesium, aluminum, and oxygen, represented here were actually oxides of these metallic elements. The oxide layer was quantified by depth profile as being about 900 A thick, and masks any copper which might be present on the metal surface. The other elements represent residuals from the cleaner. Sample 2 data represents another bare 2024-T3 panel, cleaned, then deoxidized in our non-chrome system, rinsed in deionized water, and air dried, and Sample 3 was similarly processed except that the deoxidizer in this case was our best chromated product.

Note the similarity be- tween the results of the two deoxidizers. The magnesium is entirely gone. There is still aluminum oxide on the surface, but the thickness is only 60 A, It may be surprising to some that we now have some copper on the surface. The depth profile shows that the copper layer is about 120 A. It is important to note that analyses over several points on the panel indicate that the copper is present over the surface of the panel in the same 120 A thick layer. Removing all of the copper is not possible, but it is only when the copper on the surface is present in clusters or agglomerates that potential galvanic and pitting corrosion sites are available.

Now let us look at the deoxidizer solutions in more detail:

1. Acid etch/clean- a fluoride-containing acid solution which attacks the aluminum surface at a highly controllable rate. The optimum etch rate has been found to be equivalent to the etch rate of chromated deoxidizers. As the name implies, this solution has a bui 1 t - in acid stable surfactant which gives the solution cleaning capability. Other than aluminum, which dissolves in the solution at a slow rate, and trace quantities of alloying elements, there are no heavy elements in this bath, It is important to note, as seen in Table 2, that copper is hardly soluble at all in this solution. The data in Table 2 was generated by processing only bare 2024-T3 panels through this solution16. -

4

13

2. Desmut - As we have said, the alloying elements and intermetallics in 2024-T3 and other aerospace alloys are not very soluble in the acid etch/cleaner. Copper is oxidized to the +1 state and exists as CuzO on the surface and some iron oxidizes to Fe+’ and exists as FeO. The insolu- ble smut products are removed in a desmutting solution, based on nitric acid with an oxidizer, which converts the insoluble reaction products to their more soluble oxidation states, such Cu+2 and Fe+3, thereby removing them from the surface. As also seen in Table 2, the buildup of dissolved aluminum or copper in this bath is quite small. Remember that this was a panel study where dragout was minimal, and if actual parts with higher dragout rates were run, buildup would undoubtedly be much slower.

Control of both of these baths is quite simple, involving testing which is no more difficult than control of chromated deoxidizers. We should mention that the fluoride- content of the acid cleaner bath is quite low, and can be adjusted to modify the etch rate, as necessary.

The desmut solution is controlled with a simple acid titration and an iodometric oxidizer titration. Tempera- ture control is not necessary on this bath. Both tanks are mainly replenished with two packages (a third package, an optional surfactant package, is available to improve cleaning capability of the acid

etch/cleaner) . Details are shown in F i g u r e 5.

Bath life of both solu- tions has proven to be extra- ordinary. One would particu- larly wonder about the life of the acid etch/cleaner since we have already acknowledged that there is a slow but definite dissolution of aluminum in this bath. In the bath described in Table 2 a sludge began to form at an aluminum level of 1200 ppm, but no panel failures were experienced when panels were further processed and tested. The sludge was identified as aluminum fluoride. Since the bath will operate successfully even when saturated with alu- minum fluoride, the sludge can be continuously. filtered or removed on a periodic basis.

F i g u r e 6 shows the bare panel tests which all pass using this deoxidizer system. With a good quality conversion coating on 2024-T3 panels we have not only been able to consistently achieve passing 336 hours neutral salt spray, but passing results of over 1500 hours exposure (of bare panels) are common. Since most aircraft are painted, it is also important to consider properties of painted surfaces. A list of paint adhesion and corrosion resistance tests run is shown in F i g u r e 7 and shows passing results in all cases. These results were obtained with a typical aerospace paint system consisting of an epoxy primer and an epoxy polyamine topcoat specified in Boeing Specification BMS 10- i l l 7 .

When operating a chromated deoxidizer, it is very

5

14

I .

important to know the copper content of that bath to successfully process parts through that bath. With this in mind, we tried to get an understanding how this system would operate with dissolved copper in the baths even though we could not get copper to d i sso lve in the acid etch/cleaner bath by running 2024 parts. The acid cleaner bath was doped with copper sulfate such that the copper level was 450 ppm. While processing panels through this bath the smut was a distinct metallic copper color. The copper level of the desmut bath rose to 260 ppm before it could no longer remove this copper smut. However, panels processed through this system and then conversion coated still passed 336 hours salt spray as well as all painted adhesion and corrosion resis- tance tests. Incidentally, the copper content of the etch/ cleaner bath actually dropped slightly while parts were being processed.

A one-year production trial of this process was conducted at Sure Power Indus- tries, Tualatin, OR, with consistently passing results and no bath failures. The aluminum level of that 1200 gallon production bath rose to about 1100 ppm, then leveled off (see Figure 8). A small amount of aluminum fluoride sludge formed near the end of the trial. No panel or part failures were experienced throughout the trial. A metal analysis of the production baths is shown in Table 4, and again shows very little heavy metal buildup in either bath.

The parts processed through this bath were a mixture of 5052, 6061, and 2024 alloys.

A list of some of the aerospace companies who have successfully tested this pro- cess is shown in Figure 9. Recently, The Boeing Co. issued a Process Specification Departure for its BAC 5765 Specification for Deoxidizing of Aluminum. This PSD allows the use of this system for heavy duty deoxidizing prior to conversion coating, anodizing, and many other operations.

Ask0 Processing, Inc., of Seattle, WA, has confirmed the viability of using this deoxi- dizing system prior to welding. They have reported that parts deoxidized in this system maintain low surface resistance longer than parts deoxidized with chromated deoxidizers.

Before we conclude our presentation, let's take a look at where we see conversion coating processing is going in the foreseeable future. In doing so, we will also address the concerns of some about the use of a two step deoxidizer system. Figure 10 shows the procedure which we have recommended for the past 40 or so years. Note the solvent vapor degrease, alkaline cleaner containing phosphates, chromated deoxidizer and conversion coating baths. Every one of these processes t o d a y r e p r e s e n t s a n environmental problem.

In the very near future our recommended process for commercial and military aero- space will be the process shown

6 I i !

15 1

in F i g u r e 2 2 . Here we have an environmentally safe aqueous degreaser, a two-step environmentally safe deoxidiz- er, and an environmentally safe conversion coatingolook at the number of steps: except for a rinse after the degreaser, the number of steps is the same.These two charts do not show the waste treatment fa- cilities required for each process. The process of the 90's will require a much smal- ler waste treatment facility and drastically less sludge to dispose of, and the sludge will no longer be extremely hazardous- a sizable premium for an extra rinse tank,

Conclusion A non-chromium, non-heavy

meta1,non-ferricyanide deoxi- dizing system is now commer- cially available which equals, and in some cases surpasses, the performance of chromated deoxidizers. Laboratory and production line test results have confirmed that the system contributes to the conformance of a good quality chromate conversion coating to the applicable military and aero- space specifications, even on alloys which present finishing difficulties, It has been approved for use by the Boeing Company and other aerospace companies.

7

16

1. Wernick, S., R.Pinner, and P. G. Sheasby, "The Surface Treat- ment and Finishing of Aluminum and its Alloys", 5th Ed., ASM International, Materials Park, Ohio, and Finishing Publications Ltd., Teddington, Middlesex, England, 1987; pp.181-2.

2. Wernick and Pinner, @@The Surface Treatment of Aluminum", 4th Ed., Robert Draper Ltd., Teddington, Middlesex, England, 1972.

3. Brace and Sheasby, "The Technology of Anodizing Aluminumtt, Technicopy Limited, Stonehouse, Glos., England, 1979; pp. 52-3, 55, 291-2

4. Ketcham, S. J. and Brown, S. R., Vhromating High Strength Aluminum Alloystt, Metal Finishing, Nov. 1976; 7 4 , p.37.

5. Mohler, J. B., "Deoxidizing and Brightening Aluminum Alloystt, Metal Finishing , O c t . 1972, 7 0 , p.43.

6. Smith, S. V. , ttAluminum Deoxidizing and Desmuttingtt, P l a t i n g and Surface Finishing, Sept. 1973, 62, p. 870.

7. Dollman, D. Y., U. S o Patent 3,549,540, 1970.

8. Smith, H. V., U. S. Patent 3,275,562, 1963,

9. Batiuk, W., U. S. Patent 4,451,304, 1984.

10. McMillen, M. W., Ambler, Pa., "Nan-Chrome Deoxidizer for Aerospacett, Parker + Amchem Laboratory Report 88-621-1; (May 1989).

11. McMillen, M. W., U. S. Patent 5,052,421, 1991.

12. Carlson, L. R, , Madison Heights , MI , "Interim Status Report #1 on TD-3056-A/B and TD-3057-B/C at Askott, O c t . 26, 1989.

13. Carlson, L., Madison Heights, MI, Parker + Amchem Memorandum, Jan. 29, 1990.

14. The Boeing Co. , Seattle, WA, Boeing Process S p e c i f i c a t i o n BAC 5765, R e v , M , Cleaning and Deoxidiz ing Aluminum Alloys , June 8 , 1984.

15. Kramer, J. A. , Madison Heights, MI, Parker + Amchem Laboratory Report 395005-B, O c t . 1991.

16. Johnson, P. M. , Madison Heights, MI, #*Loading Studies of the TD-3056-A/B - TD-3057-B/C Deoxidizer System - Part Itt, Parker + Amchem Laboratory Report 500001-A, Feb. 24, 1992.

17. The Boeing Co. , Seattle, WA, Boeing Material S p e c i f i c a t i o n BMS 10-11, R e v , U, Chemical and Solvent Res i s tan t F in i sh , O c t . 28, 1991.

8

17

FIGURE 1 Deoxidizer Chronology

~-

1950- Present 7

1950- Chromated Deoxidizers Formulated

1963- Ferric Sulfate Deoxidizer Patented

1 970- Ferricyanide in Chromated Deoxidizers Patented

1984- Batiuk Non-Chrome 2-Step Process Patented

1 1991- McMilten "Chrome 2-Step Process Patented 1

FIGURE 2 Deoxidizing vs. Desmutting

Deoxidizing- The removal of oxi other inorganics which would

I interfere with normal finishing I p roced ures.

Desmutting- The removal of pretreatment residues without significant attack on aluminum.

18

. .

FIGURE 3 DEOXIDIZER APPLICATIONS*

Heavy Duty Deoxidizing e

Removal of Heavy Oxides

Removal of Corrosion Products

Removal of Heat Treat Discoloration

After Abrasive Cieaning

After Shotpeening

to Penetrant inspection

From Boeing BAC 5765

FIGURE 4 DEOXIDIZER APPLICATIONS*

CONTINUED

Ught Duty Deoxidizing Oxides, Corrosion, Etc. Not Present

Surfaces for Resistance

of Surfaces for Fusion Welding

Preparation of Surfaces for Brazing

Preparation of Surfaces for Adhesive Bonding

Removal of Foreign Metal Contamination

Removal of Flux

'Skin Quality' Deoxidizing - Prior To Clear Conversion Coat

From Boeing BAC 5765

19

TABLE 1. Auger Analysis Cleaned and Deoxidized 2024-T3 Panels

Per Cent Atomic Concentration

f Sample Number

1

1

\ Depth, A SI C Ca 0 Cr F cu Mg Ai

0 2.7 4.3 6.1 31.8 - 32.3 22.8

30 2.3 2.0 1.7 36.7 - 33.0 24.3

Sample Identification 0 #1 =Cleaned Only 0 #2=Non-Cr Deoxidized 0 t3-Chromate Deoxidized

/

~

a- present but not quantified

\ Acid

Cleaner Desmut

TABLE 2. Deoxidizer Loading Aluminum and Copper Levels

Aluminum 1200 ppm

\ Copper <5 PPm

Baths Loaded with Bare 2024-T3 18 Sq. FtJGal. Processed 20

TABLE 3. Deoxidizer Loading Acid Etch/Cleaner Bath Doped with CuSO,

Copper Level

Processed Bare 2024-13 Panels Panels Conversion Coated after Deoxidizing All Passed 336 Hours Salt Spray, Paint Adhesion Tests

FIGURE 5. Process Control Deoxidizer System

EtchKIeaner

Fluoride Probe

- Acid Titration - Oxidizer Titration

21

FIGURE 6. Unpainted Performance Tests 2024-T3, 6061 -16, 7075T6 Panels (All Passing)*

Uniform Appearance

* All Panels Non-Cr Deoxidized + Chromate Conversion Coated

FIGURE 7. Painted Performance Tests Bare 2024-T3 Panels* (All Passing)

‘Primed Only and Top-Coated Panels Following Non-Cr Deoxidize + Chromate Conversion Coat BMS 10-1 1 U Paints 22

FIGURE 8. Acid Etch/Cleaner AI u mi n u m D iss o I u t i o n

AI Content 1,400

1,200

1,000

800

600

400

200

0 2 4 6 8 10 12 14 16 18 20 0

1,000 Square Feet Processed

I

. Production Bath

TABLE 4. Deoxidizer Loading Metals in Solution (ppm)

ICB Analysis

9 2

Chromium I 5 7

Production Bath- 21,350 Sq.'Ft. Processed Mixed Alloys: 5052, 6061, 2024 23

. . . . .

FIGURE 9 Non-Chrome Deoxidizer

Cooperative Evaluators

FIGURE 10 Chromate Conversion Coat Process

1950- 1993

24

,

FIGURE 11 Non-Chromium Conversion Coat Process

1993 and Beyond

25

REDUCTION OF CADMIUM PLATING ON AIRCRAFT WHEELS AND BRAKES Crittenden J. Ohlemacher and Mark W. Gilbert

Aircraft Braking Systems Corp., Akron Ohio 44306

ABSTRACT:

Cadlpium had been the standard for protection of steel parts on aircraft wheels and brakes. However, environmental concerns have led to the elimination of cadmium plating from these products. This paper will cover the first phase, now complete, of that effort. Cadmium was eliminated from parts that were plated in-house. 9 t h e r than searching for a universal replacement for cadmium plating, each part was reevaluated. This evaluation included performance requirements, available alternate materials and coatings, and estimated cost impact. Several examples will be discussed, &th solutions that include alternate coatings and base metal changes.

INTRODUCTION:

Electroplated cadmium had been the standard for protection of steel parts on aircraft wheels and brakes. Cadmium has a uniqu3 combination of properties that have led to its use in many critical applications. The advantages of cadmium include'*2: 1) sacrificial protection of steel, 2) corrosion products with low volume, 3) inherent lubricity that reduces galling, 4) galvanic compatibility with aluminum, 5 ) ductility, 6) solderability, 7) low electrical resistance even with chromate conversion coatings, and 8) bondability with adhesives.

Cadmium electroplating has a number of disadvantages, including? 1) hydrogen embrittlement during electroplating, 2) liquidhlid metal embrittlement of high strength steel - limiting use to temperatures below 232°C (450"F), 3) liquidlsolid metal embrittlement of titanium, 4) toxicity of cadmium vapor and cadmium compounds, and 5) toxicity and environmental concerns related to the commonly used cyanide based plating baths. The health and environmental concerns have become increasingly important. These concerns, in turn, are driving the cost of cadmium electroplating rapidly higher. This combination of disadvantages has initiated the elimination of cadmium plating from aircraft wheel and brake products.

TEXT:

Approach to Conversion Because cadmium has a number of unique

advantages, a replacement that had all of the same advantages would be extremely difficult to find. Rather than attempting to find a single material with identical properties, each application was evaluated to determine the appropriate requirements. For example, if a particular part does not experience galling conditions in service, the excellent galling resistance of cadmium is not necessary in a replacement. Two specific examples will be examined.

Example: Balance Weights The conversion of balance weights used on the

wheels was relatively simple. The significant requirements for the balance weights are that they not corrode at a significant rate and that they not cause galvanic corrosion of the aluminum wheel.

Over the years, three different systems had evolved. One system consisted of steel balance weights electroplated with cadmium. A layer of zinc chromate paste or primer is applied between the wheel and the weight as a galvanic barrier. A second system consists of lead weights with zinc chromate paste or primer between the wheel and the weight. The third system consists of stainless steel balance weights with zinc chromate paste or primer between the wheel and the weight.

The cadmium electroplated steel balance weights were converted to 300 series stainless steel with zinc chromate paste or primer between the wheel and the weight. This change had two added benefits. First, the coating step was completely eliminated because passivation of the stainless steel balance weights is unnecessary for this application. Second, field repairs to damaged plating are no longer necessary.

Example: Wheel Keys Wheel keys are used to transmit force from the

brake disks to the wheel. Historically, the keys have been made from high strength steel with cadmium electroplating. A coating of zinc chromate paint was applied to the surfaces of the key that

1

26

contact the anodized aluminum wheel. The paint acts as a galvanic barrier and helps to minimize fretting.

.. . . . . i

Figure 1: ‘Typical wheel keys.

Two other approaches had been developed for special cases. In one c~se, a foreign air force had been experiencing problems with cadmium liquid metal embrittlement of keys on a fighter aircraft. To eliminate this problem, the keys on this wheel had been converted to zinc phosphate plus a dry film lubricant of molybdenum disulfide in a phenolic binder. The other approach, used superalloys as the key material on programs where temperature and strength requirements were known to be severe.

Evaluatioeof alternatives for protecting the keys focused on galvanic compatibility with the aluminum, corrosion protection of the steel key, galling resistance, and abrasion resistance. An additional criteria was the ability to resist the elevated temperatures currently seen and projected for the near future.

Electroplating with another metal or alloy was considered. Nickel plating was eliminated from consideration because of galling problems. Based on the desire for galvanic compatibility, examination was focused on zinc and zinc alloys. Based on available information, zinc-nickel5 appeared to be

promising. However, a drawback to this system is the possibility of liquidkolid metal embrittlement6. Zinc is limted to temperatures below 260°C (500°F)’.

Vapor deposited aluminum coatings were considered. The galvanic compatibility is very good and the aluminum is sacrificial to steel. The major drawback, which halted further exploration of this approach, was the cost to import and implement a coating process technology< significantly different from those in use at that time.

The replacement of high strength steel coated with cadmium electroplating by superalloys was ruled out because of the high material and

A variety of organic coatings were considered andlor tested. A frequent problem was the inability to tolerate the elevated temperatures that the key might see. Another problem was the adverse effect on dimensional tolerances of the increased coating thickness required for most systems.

The final choice to replace cadmium electroplating on high strength steel keys was a heat cured dry film lubricant (molybdenum disulfide in a phenolic base) applied over a zinc phosphate conversion coating. The system is an improved version of the one that was being successfully used on the wheel keys of a fighter aircrafi. The dry film lubricant has excellent galling and fretting resistance. The coating acts as a bamer for both corrosion of the steel key and galvanic corrosion of the a l u ” wheel. The coating system is capable of withstanding temperatures of approximately 340°C (650°F). The dry film lubricant exhibits good resistance to synthetic hydraulic fluid.

Seved air drying dry tilm lubricants were considered; however the resistance to spilld synthetic hydraulic fluid was not adequate.

The use of the heat cured dry film lubricant without additional surface treatment was considered as a cost reduction. However, the corrosion resistance as measured by neutral salt spray resistance (ASTM B 117) was only about 48 hours, compared to a specification requirement for cadmium plate of 96 hours. The addition of a zinc phosphate conversion coating prior to application of the dry film lubricant increased the corrosion resistance to greater than 96 hours. The zinc phosphate pretreatment also resulted in greater durability of the coating after exposure to 316°C (600°F) for 1 hour than the dry film lubricant without the zinc phosphate.

machining costs.

2

27

. . . . . . . .. ~ ._..

Figure 2: Dry film lubricant without zinc phosphate pretreatment. Neutral salt spray (ASTM B 117) for 96 hours.

Figure 3: Dry film lubricant phosphate pretreatment. Neutral (ASTM B117) for 96 hours.

with dnc salt spray

The conversion to zinc phosphate plus dry film lubricant resulted in some added benefits. The processing time was reduced by elimination of the separate hydrogen embrittlement relief step after phosphatine. The curing time and temperature of the dry film lubricant is adequate for the hydrogen embrittlement relief of the coated steel. This result was confirmed using sustained load tests per the MIL-STD 1312/5, 4.1.1 Torque Method. The coating is field repairable using an air drying dry film lubricant. Also the galvanic barrier of epoxy primer is no longer necessary.

SUMMARY:

Successfull replacment of cadmium plating on parts which were being plated in-house has been acheived. Outside vendors have obtained some benefit for parts of these types which they make. The next step in the process will be to eliminate cadmium plating from parts which are made only by outside vendors.

REFERENCES:

1. Cook A. R., Proceedings from the workshop "Alternatives for Cadmium Electroplating in Metal Finishing", 1977, 241-57.

2. Lucas, J. M. and Hague, J. M., Proceedings from the workshop "Alternatives for Cadmium Electroplating in Metal Finishing", 1977, 465-78.

3. Federal Specification QQ-P-416F, "Plating, Cadmium (Electrodeposited)", October 1, 1991.

4. Lowenheim F. A., "Electroplating", McGraw- Hill, Inc., New Yo&, 1977, 182-7,

5. Zaki, N. and Budman, E., Products Finishing, 2 (I), 1991, 46-51.

6. Lynn, J. C., Warke, W. R., and Gordon, P., Materials Science and Engineering, B, 1975, 51- 62.

7. Military Specification MIL-S-5002D, "Surface Treatments and Inorganic Coatings for Metal Surfaces of Weapons Systems", November 30, 1989.

3

28

BRUSH AND FLOW SELECTIVE SULFURIC ACID ANODIZING by Joseph C. Norrls

SIFCO Selective Plating, Divlslon of SIFCO Industries 5708 Schaaf Road, Cleveland, OH 44131

Abstract

The brush and flow selective sulfuric acid anodizing processes are reviewed. The areas covered include: the equipment, supplies, and procedures used; the effect of operating conditions on the process; and the properties of the coatings. Examp$= of applications are given.

Introduction

Anodizing is the formation of an oxide film on aluminum using reverse current (part is positive) and a suitable electrolyte. As the coating is formed, three processes occur simultaneously. They are:

1.

2.

Consumption of aluminum to form AI2O3

Formation of AI2O3 at the aluminum surface.

3. Dissolution of some of the AI2O3 by the anodiiing solution.

The anodizing process, therefore, is more complex than the single build-up process that occurs in electroplating.

There are four principal types of anodizing: they are phosphoric, chromic, sulfuric, and hard coat. The four types of anodizing differ markedly in the typical thickness of the coatings and in the purposes of the coatings. See Table 1.

Most anodizing is done by immersing the part in a tank of solution. The anodizing, however, may be done using brush or flow techniques similar to those used in selective plating operations. The brush and flow techniques have a number of advantages Over tank anoduing. Some of them are:

1. They are portable.

2.

3.

4.

5.

They do not necessarily require the disassembly of a unit.

They reduce the amount of masking required.

They permit anodizing of parts too large for existing tanks.

They reduce downtime and production delays.

This paper will deal with brush and flow selective suffiric acid anodizing.

Brush Anodlzlng

Brush anodized coatings are applied using a hand-held tool. The tools are often identical tothose used in brush electroplating.

Table 1 Four Types of AnodWng

Type Typical Purpose of thickness of

anodizing pm (in.)

Phosphoric 0.25 Improve (O.ooOOl0) strength of

adhesive bonds

Chromic 2 5 Corrosion (0.OOOl) protection

sulfuric 12.5 Corrosion (0.OOOS) protection with

some wear resistance

Hard Coat 50 Wear resistance (0.002) and often

corrosion protection

1

29

Figure 1 shows a drawing of a typical tool that can be used for either brush electroplating or sulfuric acid anodizing. Figure 2 shows the tool being used to process test panels. The operator will place the tool on the part which will complete the DC circuit necessary for anodizing. As soon as the tool touches the part, the operator will move the tool on the part or ‘brush’ the surface being anodized.

Flow Anodlzing

In flow anodizing, the anodizing solution is pumped rapidly through a gap between the part and the cathode. Figure 3 shows a sketch of a simple focture to do an inside diameter.

Brush Versus Flow AnodIring

The same sulfuric acid anodizing solution is used with the brush and flow methods. Both methods are equally effective, that is, the same operating conditions may be used, and the coatings are identical when using a certain set of operating conditions. This is prwided that asufficient exchange rate (about 2500times

Electricat f i t t i n g Ci t tin9

I J

Fig. 1 Drawing of brush anodizing tool.

x.

per minute) is employed in flow anodizing. The exchange rate is the volume of solution that flows between the anode and the cathode per minute divided by the volume between the anode and the cathode. For example, if the anodizing flow rate is 15,000 cm3/minute and the anode-to-cathode gap volume is 6 cm3. the exchange rate is 15,OOO divided by 6, or ZSOO times per minute.

Fig. 2 Brush anodizing.

Cathode Soiu t ion

O u t c e t j Negative (-)

Polar; t y

+>

isit ive P o l a r i t y ‘t Solut ion

I n l e t

Fig. 3 Flow Anodizing Fixture for sulfuric acid anodizing.

2

30

There are a number of factors that enter into determining which method is the best for an application. Some of them are given below.

Factor

Type of application

Complexity of part andlor accessibility of area requiring anodizing

Necessity for a masking fixture

Number of Parts

Comments

Brush anodizing is usually more convenient for touch up of anodized coatings since they are usually one of a kind and the area'is small and accessible.

Brush anodizing generally is more /convenient for a simple part and an area that can be easily masked around with tape. An example is a bore in a plate. Flow anodizing is generally more convenient when it is difficult to mask around the area, such as when there are complex surfaces nearby, and when the surface to be anodized is not very accessible.

A masking fixture may not be required for brush anodizing. Flow anodizing requires that a chamber be built up around the area to be anodized.

Flow anodizing is generally preferred for a large number of parts since masking can be rapidly performed using masking fixtures, and since flow plating is less labor intensive.

important Sulfurlc Acid Anodlzlng Operatlng Conditlons

Operating conditions that affect the anodizing process and the thickness, structure, and properties of the coating are:

1. Alloy being anodized. 2. Temperature. 3. Current Density.

. .. , . . .

Alloy Being Anodized

Each aluminum alloy responds differently to the anodizing process. Some differences are:

1. Certain aluminum alloys bum more easily than others; these require lower current densities.

2. m e factor (coating efficiency) varies slightly depending on the alloy.

Table 2 shows how the aluminum alloy affects the anodiing process'

Table 2 Effect of Aluminum Alloy on Sulfuric Acid Anodizing Aluminum Current Density* Factor* Alloy Ncm2 (Mn.2)

For M a x

3003-Hl4 0.39 (25)

5052-H32 0.78 (5.0)

6061-T6 0.47 (3.0)

7075T6 0.39 (25)

Corrosion Avg Protection

0.08 (03)

0.19 (1.25)

0.39 (25 )

(1 -5) 0.23

0.19 (1.25)

0.1 6 (1 -0)

0.03 0.49 (0.2) (0.008)

0.03 0.49 (0.2) (0.008)

*Coatings will bum above the maximum current density. Use the average current density when anodizing for wear resistance or in dimensional restoration. Use a current density of 0.03 Ncm2 (0.2 Min.2) when anodizing for corrosion protection.

**The mAh for 1 pm on 1 cm2 (Ah for O.OOO1 in. on 1 in.2) at 0.03 Ncm2 (0.2 Win..).

3

31

Temperature 2. Repairing a damaged anodized coating to restore corrosion protection.

The proper temperature range for sulfuric acid anodizing is 18 to 30 OC (65 to 85 OF). In wear resistance applications, 18 to 24 OC (65 to 75 OF) is preferred. In corrosion protection applications, 24 to 30 OC (75 to 85 OF) is preferred.

Heat is developed during the anodizing process. The amount of heat developed is dependent upon the thickness of the anodized coating and the size of the area to be anodized. The higher the thickness and the larger the area, the more the solution will be heated. Excessive oveheating of the solution is prevented by:

1. Starting out with an adequate amount of solution.

2. Cooling the solution, such as by running it through cooling coils or using an immersion cooler.

3. Selective anodizing an area for dimensional reasons. An example is a bore that has been machined oversize and anodizing will be used to get it back to size. In this type of application, the maximum thickness of coating that can be applied depends on the alloy. The thickest coatings that have been applied to date on various alloys is as follows:

2024-T3 - 94 pm (0.0037 in.)

6061 -T6 - 67 pm (0.0027 in.)

7075-T6 - 97 pm (0.0038 in.)

A356-T6 - 94 pm (0.0037 in.)

Equipment and material requirements will vary depending upon the type of application.

Equipment and Materials Current Density

Power Packs Sulfuric acid anodizing is best controlled

by current density. When corrosion protection is desired, a current density of 0.03Ncm2 (0.2 ampfin.2) is recommended. Higher current densities are used in wear resistance or dimensional restoration applications. There is a current density above which buming of the coating will result even with an excellent set-up. This is called the maximum current density. Firty percent of the maximum current density is called the average current density: it is recommended that this be used as a working current density in wear resistance and dimensional restoration applications. See Table 2 for more information.

Applications for Selectlve Anodizing

The three primary applications for brush and/or flow sulfuric acid anodizing are:

1. Selective anodizing, for corrosion and/or wear resistance purposes, an area on a part that has not been anodized before.

A thirty volt dc output rectifier is adequate for applications 1 and 2 listed above. A voltage output of 45 volts, however, is required for application 3 listed above, since higher current densities are used and higher thicknesses of coating are applied.

A special power pack particularly suited for sulfuric acid anodizing has been developed. The power pack is intended for anodizing areas up to approximately 322 cm2 (50 in.2) in corrosion resistance applications and 39 cm2 (6 in..) in wear resistance or dimensional applications. Specifications on it are as follows:

Max Voltage Under Load 45 Max Amperage Output 10 Resolution of Meters

Ammeter 0.01 Ah Meter 0.001

A special feature of the power pack is the high resolution of the meters. The resolution of the ammeter allows current density to be

4

32

controlled adequately for areas down to 1.6 cm2 (0.25 in2).

Coating thickness is controlled by ampere- hours. The number of Ah required for a job is precalculated by multiplying three numbers: the factor for the alloy to be anodized, the area to be anodized in CW (in.2) and the thickness of the coating in terms of pm (0.0001 in.). This number of Ah, as measured on the power pack, is then passed while anodizing. The resolution of the Ah meter on an 1.6 cm2 (0.25 in..) area is sufficient to control thickness to within 2 1 1 %

A second feature of the power pack is the voltage and current regulation controls. This allows the power pack to automatically assure that a preselected current, (current density), and vottage are not exceeded. This eliminates the need to make repeated voltage adjustments during an anodizing operation.

Immersion Probe Solution C%ler

An immersion probe solution cooler has been developed to maintain the anodizing solution in the proper temperature range in production applications or when doing large areas. The unit has a refrigeration capacity of 941 watts (3200 B N / H R ) at 6 OC (42 OF).

Therefore, it has ample cooling capability for any anodizing job that can be done with the previously mentioned special power pack.

The solution cooler consists of a refrigeration cabinet, a stainless steel immersion cooling probe, a stainless steel immersion temperature sensor, and a PVC tank to contain the anodizing solution. The refrigeration cabinet includes an analog temperature control calibrated in OC. Solution temperature can be maintained to within 20.5 OC of the desired temperature.

are often modified to allow pumping solution through them, particularly when the hardest, most wear resistant coatings must be obtained.

When special tools are made, the cathode material may be graphite, stainless steel, or platinumclad niobium.

Standard or special tools should cover all, or as much as possible, of the area to be anodized. This is particularly important when the hardest, most wear resistant coating must be applied. Also, the coatings are applied as fast as possible when the tool covers the entire area Less than full coverage increases the required coating time and the amount of solution degradation of the coating. For example, if the tool covers only 25% of the area, the required coating time and the amount of degradation of the coating increases by a factor of four.

m e anodizing tool covers should be polyester, that is, Dacron@ batting, Dacr n

' tubegauze, Dacron felt, or white Scotch-Brite . Cotton materials should not be used.

Pumps

%

Standard, commercially available plating solution pumps are suitable for use in sulfuric acid anodizing provided that they are large enough for the size of the area to be anodized. General recommendations are as follows:

Size of Area to be Anodized cm* (in.21 Solution SUPP~ Method

Up to 6.5 (Up to 1)

6.5 to 65 (I to 10)

65 to 196 (10 to 30)

Dip for solution. Pump not necessary. Pump using small submersible pump. Pump using large submersible pump.

Brush Anodizing Tools Solutions

Standard, commercially available brush plating tools may be used for anodizing. The need for solution-fed lools to maintain proper temperature, however, is greater in anodizing as compared to electroplating; this should be considered in selecting tools. Standard tools

Two solutions used in all applications are:

1. Electrocleaning solution. 2. Sulfuric acid anodizing solution.

5

The electrocleaning solution is used to electroclean aluminum surfaces prior to masking and then again just prior to anodizing. The operation is carried out at 10 to 15 volts, forward current and is continued until the subsequent rinse water does not break on the surface.

The same sutfuric acid anodizing solution is used for all types of anodizing applications and when using either the brush or flow method.

Some other solutions that may or may not be required are:

1. Anodize strip solution.

2 Protective coating strip solution.

3. Seal solution.

4. Dye solution.

m e anodize strip solution is used to remove thin, partially damaged anodized coating immediately adjacent to the area where wear or physical damage has penetrated to the base material. If the thin coating is not removed, the thin area will not pass current, and it will remain unchanged while anodizing. The end result would be a full thickness of repair coating in the area where penetration occurred into the base material, and a thin original coating around it. This type of repair is generally not attractive and is suspect as far as wear resistance and corrosion protection.

The stripping is done after solvent cleaning the repair and adjoining areas, and after masking for anodiiing. Care is exercised in masking to assure sharply defined edges will be maintained throughout the masking, stripping, and anodizing process.

m e solution is applied to the area to be stripped using a Dacron pad. Rubbing the pad over the area is necessary for stripping to take place. No current is used in the operation. Stripping is continued until there is an obvious change in color that indicates stripping has been completed.

The protective coating strip solution is used to remove a film that has formed on the aluminum while using the anodize strip solution. The film protects the aluminum from excessive attack during the stripping operation. It is removed by applying the protective coating strip solution on the surface using a Dacron pad and no current. Some gassing occurs on the surface as the film is being removed. The operation is continued until the gassing stops and there is a change in color on the surface. These changes indicate the film has been removed.

Two room temperature seal solutions have been developed for use after anodizing. The use of either one will allow passing the salt spray requirements (336 hr) of MIL-A-8625E. The first solution provides for a greater tolerance for errors in processing while anodizing, rinsing, and sealing. However, it imparts a light yellow color to the coating which might be undesirable for appearance reasons. When the light yellow color cannot be tolerated, the other solution should be used.

m e seal solutions can be used by one of three methods: by dipping the part in the solution, by making a dam and then pouring the solution in the dam, or by swabbing. When swabbing, use polyester covers and keep the surface wet with solution. Sealing conditions are as follows:

Seal Temp.OC (OF) Time Solution -

1 15.5 to 18 (60 to 75) 2 min

2 30 to 32 (85 to 90) 10 min

Dyeing occasionally must be done for optical reflectivity reasons or to color match an existing dyed coating. Bmsh-on Mack, red, blue, etc. organic dyes are available for this purpose. A black, inorganic, electrolytic dye has also been developed for applications such as aerospace hardware.

6

34

Testlng of Coatlngs material and room temperature seal combinations tested were:

Corrosion Resistance Testing

Corrosion resistance tests per ASTM 61 17 for 336 hr have been run on 7.6 cm (3 in.) by 7.6 cm (3 in.) flat areas after brush sulfuric acid anodizing 12.5 pm (0.0005 in.) thick. The base

A test program was set up by an aircraft manufacturer to determine if a brush sulfuric acid anodized coating would crack after a bearing was press fit into the bore. The part was cast A356 and it was not a fatigue sensitiie part. The bores were 2.29 cm (0.900 in.) diameters, 0.86 cm (0.337 in.) long. Details on brush sulfuric acid anodizing the part are given in Table 3. It should be noted how quickly the anodizing operation was carried out.

Table 3 Data on Sulfuric Acid Anodizlng Test Parts

/

Part Thickness of Coating Anodizing Time (in.) Minutes

7 - pm - 1 15 (0.001) 4.5 2 15 (0,001) 4.5 3 50 (0.002) 9 4 46 (0.001 8) 9

After anodizing, the parts were Super Penetrant inspected. No cracks were found in the coatings. Four bushings were then manufactured out of 155 stainless steel to represent bearings. The bushings were then pressed into the anodized holes. The interference fits varied from 43 to 53 pm (0.0017 to 0.0021 in.). The bushings were removed immediately on two parts: the two parts passed a subsequent Super Penetrant inspection. A third part was cut with the bushing in for metallographic inspection. There was no evidence of cracking in the coating. The coating was intact and coating was present all around the internal diameter. The fourth part was salt spray tested and it passed the test. Based on these results, it was recommended that the process be approved as a method for salvaging undersize holes and other features on parts not sensitiie to fatigue.

1. 2024, Seal 1 2 2024,Seal2 3. 6061, Seal 1 4. 7075, Seal 1 5. 7075, Seal 2

There was no evidence of corrosion on any of the above panels.

Corrosion resistance tests were also run on simulated spot repairs. Four panels were tank sulfuric acid anodized and sodium dichromate sealed; two panels were 3003-Hl4 alloy and two were 7075-T6 alloy. An area was sanded into the base material, exposing approximately 1.3 cm2 (0.2 in.*) of aluminum on each panel. In another area on each panel, the tank anodizing was stripped exposing aluminum on a 4.5 cm* (0.7 in.*) area The panels were then masked exposing only the two simulated spot repair areas on each panel. The panels were then brush sulfuric acid anodized to a 12.7 pm (0.0005 in.) thickness at 0.03 A/cW (0.2 MM). The panels were then unmasked and sealed with one of two seals. The combination of base material and the seals used were as follows:

1. 3003,Seall 2. 3003,Seal2 3. 7075,Seall 4. 7075, Seal 2

The panels were then salt spray tested per ASTM 61 17 for 336 hr. No corrosion was noted

on any of the panels.

Selective Anodizing Applications

Selective Anodizing a Part

Figure 4 shows an aluminum piston head that goes into a marine, gas fueled, combustion engine. A sulfuric acid anodized coating 7.5 pm (0.0003 in.) thick is desired on the wrist pin bore for improved wear resistance. The surface is very selective, that is, no coating is desired on other areas and none is permitted in the piston ring areas. The area is also relatively

7

35

Fig. 4 Aluminum Piston Head

inaccessible. These factors make this a good Support Fitting selective anodizing application. An investiga- tion is currently under way to determine Dimensional Restoration whether the brush or flow method is the best way to proceed. The production rate will be approximately 300 pieces per hour.

Sutfuric Acid Anodized Coating Repair

Fig. 5 Brush Anodizing a Main Gear Box

Figure 6 shows an A356 alloy generator frame with numerous, close tolerance inside

in.) bore that is the second bore from the back, I diameters. A 5.199 to 5.202 cm (2047 to 2048

Figure 5 shows a Main Gear Box Support Fitting that is present on a CH-124 Sea King Helicopter used by the Maritime Command of the Canadian Forces. A bore in the part is being brush sulfuric acid anodized during a training session. This bore must be frequently repaired on the aircraft when damage occurs while installing or removing barrel nuts. Since the aircraft spends 90% of its flight time over sat water, it operates in a corrosive environment and the coating must be immediately repaired. When the area being repaired is less than 25% of the total area, the preferred repair method is brush anodizing. When the repair area is more than 25% of the total area, the entire bore is stripped and reanodiied. In this case, the flow method is more practical since access to the bore is very limited. tt takes approximately 4 hours to clean, mask, anodize, and unmask. The alternative is to replace the part, which costs approximately $5,OOO, and requires a replacement time of 150 hours. Fig. 6 Generator Frame

.

8

36

is periodically mismachined. Selective anodizing has proven to be a very successful method of repairing the bore. The bore is anodized to the middle of the tolerance and no subsequent machining or grinding operation is required.

The complexity of the part made it unlikely that it could be successfully masked with tape and paint. The area was also relatively inaccessible. For these reasons the flow approach was selected for anodizing. Figure 7 shows the flow foctures used. The rubber stopper shown at the bottom sealed off an adjacent smaller ID. A PVC cup with a rubber seal cemented to it (shown at the top) seals off an end face on the other side of the ID. The part is oriented vertically and anodizing solution enters the chamber from the bottom and exits through holes in the OD of the graphite electrode.

Figure 8 shows the anodizing in progress. Details on the application are as follows:

Area - Build up Required -

Current Densrty - Current 3A Ah 0.79 Coating Time - 16 minutes

38.7 cm2 (6 in.2) SOpm (0.002 in.) on diameter. C.08 Ncm2 (0.5 Nin.2)

Fig. 8 Anodizing Generator Frame

Figure 9 shows a 7075T6 helicopter control link, that is not fatigue sensitive, requiring a build up on an inside diameter. Also shown are the masking fixture and the cathode used. The brush method was used in this application since the area was very accessible.

Fig. 7 Flow fixture for generator frame Fig. 9 Helicopter control link and tooling for job

9

37

Figure 10 shows the anodizing in progress. Auxiliary equipment is being used to rotate the cathode and to pump solution into the work area Details on the application are as follows:

Area - Build up Required -

Current Density - Current 1.96 A Ah 0.22 Coating Time 6.7 minutes

10.1 cm2 (1.57 in.2) 5Opm (0.002 in.) on diameter 0.19 Ncm2 (1.25 Win.2)

Fig. 10 Brush anodiuing helicopter control link

Summary

Brush and flow sulfuric acid-anodizing are effective, reliable methods of anodiiing a select area on a part or repairing a damaged tank anodized coating. The coatings applied by these methods meet the quality and performance requirements of MlLAS625E and AMS 24710 and 2472C. There are important applications for these processes in aerospace and other industries.

lo

38

i

HAZARDOUS MINIMIZATION - SAVING TIME, MONEY, AND THE ENVIRONMENT

Mary Beth Fennel1 and

James Mark Roberts

Naval Aviation Depot Cherry Point, North Carolina

a

39

..

ABSTRACT

For a hazardous minimization (HAZMIN) program to work it must involve everyone at the facility from the Commanding Officer to the shop artisans. With this commitment, a HAZMIN program can reduce environmental impact and reduce labor, turnaround time, and costs.

At the Naval Aviation Depot, (NAVAVNDEPOT) Cherry Point we are trying to look at all steps of the overhaul process from the HAZMIN perspective. In our plating facility several new initiatives are being implemented to reduce waste. Aqueous cleaners replace vapor degreasers, an atmospheric evaporator and cold vaporization units recycle clean water back to the process lines.

As with other environmental changes at the depot, hazardous minimization initiatives in the plating facility save time, money, and the environment.

BACKGROUND

Traditionally at NAVAVNDEPOT Cherry Point, as at many industrial facilities, materials and process engineers have focused their priorities on specifying the materials and processes at the leading edge of technology to obtain the highest quality parts. Normally, environmental concerns and the quantities of hazardous waste generated from a specific process were not looked upon as a primary concern. It was left to the hazardous waste personnel to find the cheapest way to get rid of our waste without violating the environmental regulations.

In January 1990, NAVAVNDEPOT Cherry Point's Materials Engineering and Technology Division embarked on a different approach. Recognizing that the engineers that write process specifications requiring hazardous materials are ultimately responsible for the resultant hazardous waste generated, we began to accept hazardous waste reduction as one of our primary responsibilities as materials engineers. Current environmental goals dictate review of parts processing to identify alternatives beneficial to the environment without compromising safety or quality.

Significant improvements in reducing hazardous materials used and hazardous waste generated cannot be realized without every functional area understanding and fulfilling their environmental responsibilities. We.have worked to form partnerships with production mechanics, facilities and equipment designers, facilities maintenance personnel and environmental engineers. Each of these groupa is responsible for a necessary step toward implementing process improvements to reduce hazardous waste. Each organization has its own set of priorities and its own agenda. It is imperative that open communication be encouraged to ensure facility-wide goals rank first on everyone's priority list.

One of the initial steps in a hazardous minimization program is education. Time and money invested to inform shop artisans and supervisors about methods to use less material and generate less

1

40

i

waste yields tremendous returns. Artisans need, and deserve, to know not only what materials are hazardous and their associated costs, but also how they can affect the amount of waste generated. It is amazing how motivated some people will become once they understand the problem, and know that they are part of the solution. Most likely, the entire workforce will not adopt hazardous minimization as their personal cause. However, several advocates scattered through the production department can make a tremendous difference in the way change is accepted.

. SPECIFIC HAZARDOUS MINIMIZATION EFFORTS

Due to its extremely high ozone depletion potential, and because of its skyrocketing cost, Freon 113 was one of the first materials targeted for elimination from our facility. The characteristics that make Freon 113 so difficult to replace in certain applications, are the same ones that make it so widely used for many functions for which it was never intended. We can no longer afford, economically or environmentally, to accept this practice.

Our approach was to first evaluate where Freon 113 was being used, in what quantity, and where it was authorized for use. We found Freon 113 was being used €or engine test cell leak check, hydraulic test stand cleaning, wiping of surfaces prior to bonding or sealing, bench cleaning during assembly and disassembly, and cleaning of electronics and instruments, among other applications. It was also used in our Naval Oil Analysis Program (NOAP), calibration and hydraulic laboratories. In each of these areas the material was also being used for

additional general purpose cleaning or drying, because it was available, cleaned well, and dried by itself. Of the 8000 plus gallons used in 1990 in our facility, only about 25 percent was being used for an briginally intended purpose. And for many of those applications, alternates were readily available. We set out to eliminate the unnecessary uses immediately, resulting in a rapid decrease in usage from 1990 to 1991. Further decreases are proving more difficult. Real reductions are outlined in Figure 1.

Freon 113 Usage NADEP Cherty Point

U 1990 1991 1992

Figure 1.

Our next challenge was to provide as many shops as possible with alternate materials or processes that would perform the functions for which they were dependent upon Freon 113. Our electric shop was cleaning motors by placing them in a Freon 113

2

41

ultrasonic unit for several hours after a presoak in varsol. The units were then dried in an oven for several hours to insure dryness. This application accounted for the use of 1750 gallons of Freon 113 per year. Since the drying step was already in place, the conversion from solvent ultrasonic to aqueous ultrasonic was simple. We were later able to outfit the shop with a small spray washing cabinet to take care of additional cleaning problems and found these motors got as clean in 15 minutes in the spray washer as in several hours in the solvent ultrasonic cleaning tank. The shop turnaround time, as well as the attitudes of the artisans both improved dramatically.

The change from Freon 113 to soap and water for leak check in our test cells appeared insurmountable at first. Mindsets against other materials were strongly formed. But, when the cost of Freon 113 surpassed $2,500 a drum, the supervisors and management decided it was time for a change. Since the engines were still warm when the leak check was performed, the detergent solution used the heat from the part to improve cleaning efficiency. A good sturdy spray bottle with the ability to provide a strong steady stream of solution made the difference for many artisans convinced that soap and water would never work.

Once some of these simple changes were put in place, we took away authorization for Freon 113 from all shops except those engaged in the following applications for which we knew of no "easy solution."

a. Oxygen/Nitrogen system internal cleaning.

b. Limited electronics cleaning

c. Limited instrument cleaning. d. NOAP Laboratory analysis.

applications.

At first we had a surge of requirements for Freon 113. We looked at each application and supplied information on alternates as necessary. In some cases we were forced to allow the shop to continue to use Freon 113 until new materials or equipment were in place. But the number of production advocates also rose and the majority took on the attitude that we simply have no choice but to move on to something else. The number of suggestions concerning alternate materials for Freon 113 also rose and in many instances the logical solution had been in the minds of the workforce all along. Only the incentive to change had been lacking.

As we have pointed out already, one of the big changes,as we move away from Freon 113, is the return of soap and water for cleaning. While this seems to be an obvious substitution, many of the areas doing wipe cleaning were built without access to water, especially hot water, or drains. This was due to the fact that the solvents used for cleaning did not require rinsing. This fact complicated the facilities and equipment changes required. We have substituted some petroleum/terpene based wipe cleaners in disassembly bench operations to maintain a waterless process. Some of these applications can be as easily handled with soap and water when rinsing and drying capability is available.

3

42

A comprehensive wipe cleaning comparison is scheduled to begin during this fiscal year. Limited experimental data is available concerning the comparison of benchtop solvent cleaners with newly developed, less hazardous alternatives. In fact, it is questionable as to what criteria was used to qualify the currently used w i p e s o l v e n t f o r p r e b o n d i n g / p a i n t i n g / s e a l i n g applications. We are particularly interested in comparing materials with flashpoints above 38 degrees C (100 degrees F), preferably above 60 degrees C (140 degrees F) . They must contain no ozone depleting substances, no substances from the Environmental Protection Agency’s top 17 list, and may not be photochemically reactive. We plan to compare residue left on the part, effect on drying time, odor, effect on bond strength, and potential for corrosion.

l,l,l-Trichloroethane (methyl chloroform) was the next material to be targeted for elimination at our facility. The predominant use of this material is in vapor degreasing. In 1990, our facility had 14 functional vapor degreasers and they were considered sacred by much of the workforce. Getting rid of this tool was as much a struggle of wills as a technological substitution. The replacement most often consisted of soap and water cleaning in some application: aqueous immersiontanks, aqueous ultrasonics or spray cleaning machines. Through instaLlation of this alternate cleaning equipment we reduced our average monthly consumption of l,l,l-Trichloroethane from about 1300 gallons per month in 1990 to 1991, to 890 gallons per month during the first half of this year, and only 400 gallons per month

during the most recent six month period. This reduction is presented in Figure 2.

Methyl Chloroform Usage - NADEP Cherry Point

1990 1991 1992 1993

Figure 2.

Our two largest degreasers were in the plating facility and the engine cleaning shop. Replacing the degreaser in the engine shop required a combination of initiatives including aqueous tank cleanifig, spray cabinet washing, and steam cleaning. The degreaser has now been of€ for several months and the alternate processes have proved to be suitable replacements. One of the biggest concerns with this substitution is drying capability. The parts end up clean without residue, but the parts are not dry. This presents some real and perceived problems with the process change. Drying capability is still being addressed at our facility.

4

43

We currently have three remaining waste stream increased from 8,420 degreasers. They are used to clean pounds in 1987 to 88,000 pounds in bearings, engine blades and vanes, 1992 as shown in Figure 3. and pneumatic components. Plans are in the works to shut down all of these units by the end of this calendar year.

Plastic Media Blast Waste Stream

Wet sodium bicarbonate blasting is being used increasingly at our facility to eliminate the need for costly and hazardous carbon removing compounds. We currently have two open blast units and a wet sodium bicarbonate glovebox. The glovebox application is preferred because it solves the problems of noise and mess often experienced with open blasters. For larger parts, the open blast unit effectively takes off soils and carbon build-up but requires engineering controls to improve working conditions.

a) U r - .J 0 n y.

0 a) U c a a) 3 0 5

Plastic media blasting is being used locally to the greatest extent possible to replace methylene chloride based chemical strippers. A tremendous savings is being realized, particularly on support equipment, where larger size media makes removal rates more attractive. All chemical stripping has been eliminated from our support equipment strip and clean area.

Although plastic media blasting reduces the chemical hazards workers are exposed to, and the hazardous air emissions from methylene chloride based materials, it is still a process that generates large quantities of dust, as shown in Figure 3. This dust is generally hazardous due to cadmium or lead content. As our plastic media blasting processing has increased, the amount of waste dust from this operation has also increased. This

Figure 3.

We are investigating various options for eliminating this waste stream. One approach being pursued is to use spent plastic media as a raw material in the manufacture of a new product. This would eliminate our plastic media waste stream, improve the manufactured product, and result in disposal and new material procurement cost savings.

PLATING SHOP S P E C I F I C EFFORTS

Vapor degreasing was eliminated from our electroplating facility approximately 12 weeks ago. Prior to the day that the degreasers were actually "locked out" , shop personnel were briefed on what to expect in using the new cleaning and degreasing methods. It is our

5

44

i feeling that this step w a s critical in assuring the successful transition from vapor degreasers.

As most of us in the electroplating field are well aware, vapor degreasers have gradually assumed bigger roles than just parts cleaners. In a sense they have become catch-alls €or shop procedures from de-maskants to parts dryers. We have known for some time that their cleaning function -could be replaced by aqueous media but it is in finding alternatives to their other functions that shop personnel have been frustrated.

Two different types of aqueous cleaners are used in the plating shop : a parts washer and an immersion soak tank. Utilization of the parts washers far exceeds that of the soak tank, mainly due to the nature of the workload in the facility. To date, no quality deficiencies have been detected in the end product that could be linked to problems with pre-plate cleaning. This was a major concern for both production and engineering personnel, €or it was not known how the viability of the cleaning media would deteriorate over time. As long as the cleaners are maintained according to manufacturers' specifications, we have observed no problems with continual usage of the parts washers and immersion tank.

Another major concern that had to be addressed was what material would replace l,l,l-Trichloroethane in its other varied roles besides degreaser. For example, l,l,l-Trichloroethane could be used to remove plater's wax from recesses even after it had been baked on. No other material was available to the shop that could

perform that function. The solution was found in a wipe-on/ wipe-of f application of a citrate based cleaner that was heretofore unknown to the shop artisans. In another instance, l,l,l-Trichloroethane was commonly used to remove latex and tape residue following electroless nickel plating. It was discovered that with an additional 10 to 15 minutes of labor, most parts could be adequately cleaned using abrasive pads and a wipe-on/ wipe-off cleaner.

In addition to eliminating the use of l,l,l-Trichloroethane, many approaches have peen attempted to reduce hazardous waste generation in the electroplating shop. Where rate of evaporation allows rinse waters to be recycled into their respective plating tanks. But this common sense approach cannot be applied to all process lines. Two of the newest technologies considered are an atmospheric evaporator and freeze vaporization units. The atmospheric evaporator was targeted for use on the cadmium plating line. The freeze vaporization units could be used for cleaning the rinse waters of cadmium, copper and silver plating, among others. Neither has as yet been installed in the shop, but both show promise to further reduce hazardous waste generation and edge us towards our goal of zero discharge. A freeze vaporization unit has been used with success to reduce the volume of spent electroless nickel sent for disposal.

6

45

To summarize, success of phasing out vapor degreasers in the plating shop rested upon (1) shop personnel being committed to making aqueous cleaners work and (2) engineering working to provide suitable alternatives to fulfill all the functions of l,l,l-Trichloroethane. Reduction of hazardous waste has hinged upon some "common sense" solutions and application of some new technologies.

SUMMARY

Traditional phiiosophies suggest that process changes due to environmental concerns will result in sacrificing part quality, turnaround time, or competitiveness. In our plating facility, as well as various other depot shops, we have found ways to turn environmental improvements into productivity improvements. This not only allows us to reap economic benefits, but in this time of environmental consciousness, it allows us the satisfaction of doing what is right.

We hope our hazardous minimization program will have a finite life. When every processing decision made at this depot takes environmental impact into account, we will no longer have a need for the HAZMIN program and we will have achieved our goals.

e

7

46

Paper Not Available In Time For Publication

Low VOC Organic Finishing Ed Branrfors, Crown Metro Aerospace, Atlanta, GA

New developem in aerospace coatings are achieving substantial reductions in solvent content Coatings developed for new aircraft manufacture and coatings used for refinishing at the airfine maintenance facility will be dis- cussed. Solvent reductions are the result of formulation technology, better applications methods and longer life finishes.

For copies of this paper, please contact the author directly:

Mr. Ed Brannfors Crown Metro Aerospace 400 1 Riverdale Court Atlanta, GA 30337 206/687-4277

47

. :.. , :...:. . . . ._....( . . . .

ABSTRACT

REGULATORY CONSIDERATIONS IN CHOOSING A CLEANING SOLVENT OR PROCESS

bY Steve Risotto

Center for Emissions Control

The pending phaseout of 1,l.l-trichioro- ethane (methyl chloroform or MCF) and chloro- fluorocarbon 113 (CFC-113) has resulted in a flurry of activity on alternative cleaning products, processes, and services. .While some of this activity may have introduced true innovation to the marketplace, much represents a repackaging of long-standing cleaning chemistries.

The choices for cleaning include the following four basic approaches: (1) other chlorinated solvents, (2) aqueous cleaning, (3) semi-aqueous cleaning, or (4) petroleum solvents, alcohols, ketones, and synthetic aliphatic hydrocarbons. In addition to the technical aspects, there are a variety of regulatory considerations that may impact the approach chosen. Recent activity implementing provisions of the federal Clean Air Act may help to make the choice easier.

OVERVIEW

Production Phaseout

On February 11, 1992, President George Bush announced that the U.S. production of the CFCs, halons, carbon tetrachloride, and MCF for emissive uses would be phased out by January 1, 19%. This announcement represents a significant acceleration of the production phase-out required by the federal Clean Air Act. Regulations to implement the 1995 phaseout will be issued by the Environmental Protection agency (EPA) by the end of the year under the authority of Section 606 of the Clean Air Act. The international treaty on stratospheric ozone protection (the Montreal Protocol on Substances That Deplete the Ozone

Layer) was revised in November 1992 to also require a phaseout of CFCs and MCF by the end of 1995.

The President's February announcement has provided considerable incentive for companies using CFC-113 and MCF to find cleaning altema- tives well before the end of 1995. While both the national and international phase4ut schedules allow production beyond 1995 for feedstock and "essential" uses, EPA staff have indicated that they believe that alternatives exist for all uses of these substances.

Excise Tax

Users of CFC-113-and MCF also will have to pay more for these solvents after January 1, 1993. Before adjourning in October, Congress passed the Energy Policy Act of 1992 (Public Law 102-486) which increases the federal excise tax on both of these solvents. As a consequence of this legislation, the per-pound tax rates for CFC-113 and MCF will increase according to the schedule in Table 1.

As before, the tax would be paid by the producer of the substance and passed along to customers in the form of higher prices. Users of either substance would be responsible for paying a "floor-stocks" tax, however, on any inventory held for sale or future use in manufacturing over 400 pounds at the end of the year. This floor-stocks tax is equal to the difference between the tax rate in the previous year and that in the current year. In 19sr3, this difference is $1344 for CFC-113 and 7.4 cents for MCF.

Labeling Requirements

Perhaps the least recognized of the strato-

TABLE 1. COMPARISON OF OLD AND NEW EXCISE TAX RATES

/

spheric ozone protection provisions of the Clean Air Act is the requirement for labeling of products containing or manufactured with CFCs and 1,l.l- trichloroethane. The statute requires that any of these products introduced into commerce bear after May 15, 1993 bear a label stating the following:

Warning: Manufactured with [insert name of substance], a substance which harms public health and the environment by destroying ozone in the upper atmosphere.

Under the labeling regulations proposed in May 1992 by EPA, the label need not be placed on each individual part or product, but instead could be placed on accompanying packaging or paperwork Under this proposal, any assembled products containing even one piece cleaned with CFC or l,l,l-trichloroethane would have had to bear a label. In that way, the labeling would have passed through the "stream of commerce" to the ultimate consumer.

At a recent conference, however, EPA staff reviewed several likely changes ta the labeling regulations for products containing and manufac- tured with l,l,l-trichloroethane and CFC-113.' Among the changes that EPA is considering, the following two proposals will be of particular interest to users of these two solvents: (1) an exemption from the labeiing requirements for those products manufacturedbefore May 15,1993;

and (2) a more narrow definition of stream of commerce, whereby only direct suppliers manufac- turing with one of those solvents would be required to label products g6ing to the next person in the commerce chain.

These changes are being contemplated primarily because the labeling requirement was developed under the original phase-out schedules of the Clean Air Act, and has become U M ~ C ~ S - sarily burdensome in light of the accelerated phase-out. It is important to note, however, that these revisions are only proposals at this time, and may change before the regulation is finalized in February. Although EPA is behind schedule in completing the regulations, the law specifies that the labeling requirements become effective on May 15, 1993.

.

ALTERNATIVES

Several alternatives exist for cleaning with CFC and MCF. All of the most likely choices for metal cleaning have been around in one form or another for some time. In all cases, however, the technology or chemistry has been refined to improve cleaning performance and/or to ease regulatory compliance.

EPA's Safe Alternatives Program

To assist companies in choosing an alternative, EPA was instructed by Congress under

- 2 -

49

. . . . . . . . . _ . .

. . . -

. .

the Clean Air Act, as amended, to evaluate alternatives to CFCs, MCF, halons, and carbon tetrachloride. Under this "safe alterrlatives" policy, EPA would prevent a switch to a substance that may present adverse effects to human health or the environment provided that an acceptable alternative was available for that specific use.

While the Agency has not yet issued a proposed regulation establishing a safe alternatives policy, they have conducted a thorough review of the alternatives for cleaning and other applications. As a result of this review, EPA staff have developed preliminary findings that the other chlorinated solvents (trichloroethylene, perchloroethylene, and methylene chloride), aqueous and semi-aqueous cleaning, and cleaning with petroleum solvents, ketones, and alcohols are likely to be viewed as acceptable? These findings are expected to be finalized before the end of 1992.

/

Other Chlorinated Solvents

While their chemical and physical properties vary, it often is possible to replace one of the halogenated solvents with another. In the 197Os, for example, many companies substituted MCF for trichloroethylene to comply with state VOC regulations. More recently, as a result of the pending phaseout of CFC-113 and MCF, there has been increasing interest in the use of trichloro- ethylene, perchloroethylene, and methylene chloride. The properties of trichloroethylene and methylene chloride make them reasonable choices as solvent replacements in certain applications. In addition, although the high boiling point of perchloroethylene has limited its use for cleaning, the development of azeotropic formulations may broaden its applicability.

Switching to one of these other halogenated solvents may require companies to apply for a modification to their existing operating permit. These three solvents generally are subject to relatively low occupational exposure limits and are regulated under state air toxics programs. In addition, degreasing with these solvents will be subject to a national emission standard under Section 112 of the federal Clean Air Act. While

trichloroethylene and perchloroethylene are regula- ted as VOCs, EPA is expected to list perchloro- ethylene as an exempt solvent in the near future?

Aqueous Cleaning

Aqueous cleaning has been widely used in industry for many years for the removal of salts, rust, scale, and other inorganic soils from ferrous metals, and has been more recently adapted to the removal of greases and oils traditionally reserved for vapor degrwing. These water-based cleaning systems usually include synthetic detergents and surfactants, along with other ' additives (e.g., sequestering agents, saponifiers, emulsifiers, chelators, stabilizers, and extenders) in combination with mechanical, electrical, or ultrasonic energy. The various ingredients aid in the cleaning process by reducing surface or interfacial tension, by forming emulsions, and by suspension or flotation of insoluble particles.

Water-based cleaners typically are not as forgiving as solvent cleaning, and generally have poor solvent loadings.' Cleaning performance depends on several factors including the level of contamination, the conkntration, pH (i.e., the degree of acidity or alkalinity), and operating temperature of the cleaning solution, the amount of mechanical agitation, and the ability to conduct effective rinsing. Good engineering and process control are more critical in preventing problems, and part design can have a significant influence on the level of cleanliness that can be achieved: Aqueous cleaners tend to have higher surface tension, and may not penetrate as readily into holes and capillary spaces. For complex parts, considerable engineering and experimentation may be required.

Type of Cleaners

Aqueous cleaning solutions typically are tailored to the requirements of the specifi~ cleaning application, and generally are classified according to pH. Acid cleaners contain inorganic (and sometimes organic) acids, acid salts, and wetting agents or detergents. While not commonly used, acidic aqueous cleaners are generally preferred for cleaning pigmented drawing compounds becauseof

3 -

50

the chemical inertness and strong adherence of these contaminants6

Alkaline cleaners @H>9) are the most common, and generally consist of the following three components: (1) alkaline salts, or builders; (2) organic and inorganic additives; and (3) surfactants. Alkaline cleaning baths generally are operated at temperatures of 130°F to 190°F to enhance cleaning, although some formulations can be operated at room temperature.

Cleaning Processes

The principle stages in aqueous cleaning are washing, rinsing, and drying. Like vapor degreasing, aqueous cleaning equipment can be characterized as in-line, for high-throughput cleaning requirements, and batch for low- throughput applications. Each can be further subdivided into immersion, spray, and ultrasonic equipment.

Immersion aqueous cleaning typically employs temperature and/or mechanical agitation to remove the soil from the immersed part. The soil is removed by convection currents in the solution created by heating coils and/or mechanical action. The simplest type of immersion cleaner consists of a single wash tank, but the demands of most cleaning jobs generally require more complex equipment.' In most cases, rinsing of the part is required to remove any remaining contaminants and residual cleaning solution. Depending on the application and plant water quality, rinsing can be conducted with tap distilled, or deionized water.

The addition of ultrasonic agitation to an immersion cleaning system can effectively clean complex parts and configurations that would otherwise provide a difficult cleaning challenge for aqueous systems. Ultrasonics also can be used to clean difficult-to-remove contaminants like carbon and buffing compounds. In addition, ultrasonic agitation can be used to great advantage in rinsing.

As an alternative to immersion, spray-wash systems use the mechanical energy associated with spraying the aqueous solution at medium to high pressures to clean the parts. Spray pressures can

vary from 2 to 400 pounds per square inch (psi), or 14 to 2758 Wa, or more. In general, the higher the spray pressure, the greater the efficiency of soil removal. Spray cleaners are prepared with low- foaming detergents (Le., nonionic surfactants) which are not as chemically active as those used in immersion cleaners, but are still effective because of the mechanical agitation.

A high-pressure (100 to 200 psi) spray system also can be used as an effective final rinse step, especially when filtered water is used. While optimization of the spray is important to the rinse effectiveness, spray rinsing uses less water and can provide cleaner surfaces since the final rinse water can be quite pure.

Drying

Drying can present a major challenge when switching to aqueous cleaning, particularly when dealing with complex parts in production processes. Solvent equipment currently in use has no provision for drying, since the thermodynamics of the solvents are favorable to spontaneous evaporation. The addition of drying equipment will increase the amounfof space required for the cleaning operation and the time required for cleaning, and will significantly increase its energy consumption.' The drying step may not be necessary if the parts are not needed immediately and a rust inhibitor is used, or if the subsequent painting, electroplating, or other process is water- based.

Available drying techniques include compressed air blowaff, warm air circulation, drying ovens, infrared heating lamps, cloth wipe, and centrifugal drying.' Mechanical removal of 90 percent or more of the water can be achieved with the use of compact turbine blowers with filtered output to remove potential contaminants.' The use of high-velocity forced air can reduce drying times significantly, and can minimize the potential for water spotting and staining? Subsequent evaporative drying can then be accomplished with drying ovens, infrared heating, or centrifugal drying. For parts with complex geometries, vacuum oven drying can be an effective method for completely removing water.'

- 4 -

51

In addition, several features can be included in the overall design of the aqueous cleaning system to reduce the drying times and/or energy consumption." These features include the following: (1) use of rinse agents that can speed drying, (2) improved surfactants to allow easier rinsing, (3) improved rinsing with ultrasonics and fdtration to reduce overall water flow, (4) use of multiple wash tanks to reduce soil loading in the rinse tank, and ( 5 ) use of recirculating hot air dryers.

Semi-Aqueous Cleaning

Cleaning systey using terpenes, dibasic esters, glycol ethers, n-methyl pyrrolidone, or other hydrocarbons, generaq in combination with surfactants, have been developed as alternatives to solvent degreasing. These hydrocarbon solvents are used in one of the following three cleaning processes: (1) emulsification in water, (2) application in concentrated form, followed by a water rinse, or (3) a combination of both. Because all of these techniques require water, the process generally is referred to as semi-aqueous.

The steps in a typical semi-aqueous cleaning process resemble those in aqueous processes - washing, rinsing, and drying. The only significant difference between the two cleaning processes is the primary cleaning tank. Because of the relatively strong solvency of the hydrocarbon solvents, semi-aqueous cleaning exhibits good cleaning ability for heavy grease, tar, waxes, and hard to remove soils. In addition, hydrocarbon formulations used for semi-aqueous cleaning generally maintain their effectiveness through significantly higher soil loading than aqueous cleaners?

Types of Cleaners

While this discussion will focus on cleaning systems that have been developed for use with terpenes, dibasic esters, glycol ethers, and n-methyl pyrrolidone, the information generally can be extended to any such systems. These four solvents exhibit varying degrees of flammability, and

especially if used in spray cleaning applications. 4 generally require that certain precautions be taken,

All four have low vapor pressures, although they are still considered photochemically reactive and may be subject to VOC regulations in urban areas. This low volatility &nimizes the potential for worker inhalation exposure. While their low volatility means that emissions will be considerably lower than for vapor degreasing operations, these solvents do not dry quickly and require a water rinse to remove residues. As a result, rinsing is a particularly important step in the semi-aqueous process."

Cleaning Processes

Semi-aqueous cleaning systems also can be configured as in-line or batch operations. Immersion equipment is the simplest and most common design that has been proposed. In these systems, the parts are dipped into the concentrated hydrocarbodsurfactant bath, an emulsion bath, or both. Additional energy (i.e., heating, ultrasonics, or spray under immersion) can be added to enhance the cleanliness, although the solvency of the hydrocarbodsurfactant often makes it unneces- sary.

As noted previbusly, caution must be taken in the use of sprays or in addition of heat, because of the solvents' combustibility. The use of high- pressure sprays below the surface of the liquid prevents the formation of an atomized solation which can ignite at temperatures below the flash poinL" If heating is desired, it is recommended that the solution temperature be kept well below the flash point of the cleaner to avoid flammability concerns.'*" In addition, because of the viscosity of the formulation, some ultrasonic equipment may not work properly with semi-aqueous cleaners.

The parts then are rinsed with clean water to remove residues left from the wash step. A water rinse may not be necessary if the operation includes an emulsion wash or rinse step and the application can tolerate some cleaner residue. For more demanding applications, a cascading rinse arrangement, or a final alcohol rinse, can be used to ensure the appropriate level of cleanliness. The rinse also may be used for the application of rust inhibitors or for other finishing processes.

- 5 -

52

Semi-aqueous cleaning processes generally require the addition of drying equipment to remove remaining water from the part for further processing or to prevent rusting. The available drying methods are the same as those for aqueous cleaning, and are descriied above. Heated or high velocity room temperature air are the most common techniques. As in aqueous cleaning, the drying step may not be necessary if the parts are not needed immediately, and a rust inhibitor is used, or if the subsequent process is water-based.

Petroleum Solvents, Ketones, Alcohols

Petroleum solvents, alcohols, and ketones presently are used in some sectors of manufac- turing and repair industries for cold cleaning applications. Petroleum solvents (mineral spirits, kerosene, Stoddard solvent) show good solvency for most contaminants, are compatible with most rubbers, plastics, and metals, and have low surface tension. Alcohols (e.g., ethanol, isopropanol, and glycol ethers) have been used in certain applications requiring their high polarity and effective solvent power, including their use in azeotropic mixtures with halogenated =bents for defluxing operations in the electronics industry. Ketones such as methyl ethyl ketone, and acetone are powerful solvents, but like alcohols are not compatible with many polymeric and elastomeric materials.

These solvents, however, have several limitations. Their flammability restricts their use in enclosed systems and in vapor degreasing applications. The common alcohols and ketones have flash points that are quite low, and are considered flammable. Flash points of the petroleum solvents generally are between 100°F and 140°F. In addition, these solvents are photochemically reactive, and are regulated as omne (smog) precursors in most urban areas of the country."

Cleaningwith these solvents is accomplished by immersion of the parts to be cleaned in one or more solvent baths. In most cases, the baths are operated at ambient temperature. For higher

flash-point solvents, hot cleaning may be used to improve cleaning efficiency, but temperatures must be maintained well below the flash point of the solvent. Spraying processes generally can not be used with these solvent because the fine droplets produced can ignite at temperatures below the flash point of the bulk fluid

Because the parts are immersed in a solvent bath, the solvent must be maintained as free of contamination as possible. If contaminant oils and greases are allowed to build up, they may redeposit on the parts. In multistage wash processes, fluid from one bath is periodically transferred to the preceding bath as its soil level builds up. Fresh solvent is added only to the final bath to ensure the highest cleanliness of parts, and spent solvent is removed only from the first stage. Reclamation of these solvents can be accomplished using filtration and distillation, but flammability concerns make distillation difficult.

Most of these solvents, particularly the petroleum solvents, exhibit slower drying times than halogenated solvents. As a result, operations requiring a clean part for the next step in the production process may need to add a drying step. Drying can be accomplished using forced air, but precautions must be taken to prevent combustion of the solvent. Solvent vapor recovery can be accomplished with carbon adsorption or conden- sation.

In addition to the these hydrocarbons, several other synthetic products are available that offer favorable properties for metal cleaning. Commercially available products include those containing p a r a f i c or oxygenated hydrocarbons or aliphatic esters. While their properties vary, depending on the exact formulation, these products generally have lower flammability than petroleum fractions. In addition, they have reactively low surface tension allowing for good surface wetting and penetration."

As with the hydrocarbons descriid above, the solvent bath must be kept relatively free of contaminants to ensure effective cleaning. While the lower volatility of these synthetic solvent formulations reduces their emissions of volatile

- 6 -

53

organic compounds (VOG), it likely will further increase their drying times. Unlike those hydrocarbons used in the semi-aqueous processes descriied below, however, these aliphatic solvents reportedly 'can be dried from the surface without leaving a residual This would eliminate the need for a water rinse an4 thus, simplify any wastewater treatment requirements.

REGULATORY CONSIDERATIONS

Air Toxics

Under Section 112 of the federal Clean Air Act, as amended, EPA is required to develop regulations controlling emissions from sources of about 190 substances. The Agency is expected to propose regulations on degreasingwith chlorinated solvents by the fall of 1993, and to finalize these regulations about a year later. EPA staff have indicated that these regulations will be based on readily availablecontrol technology (e.g., freeboard chillers, powered covers), and will allow for more than one approach to controlling emissions. As a result, these regulations likely will not have a significant impact on companies using trichloro- ethylene, perchloroethylene, or methylene chloride in states that already have active !'air toxics" programs.

The iist of 190 substances in Section 112 contains several other substances that may have application for cleaning. Currently, however, EPA is not planning to issue regulations specifically directed at cleaning uses of these substances. These situations likely will be affected, however, by the volatile organic compound (VOC) provisions of the federal Clean Air Act.

Volatile Organic Compounds (VOCs)

The Clean Air Act also requires that states develop more aggressive implementation plans to achieve the national ambient air quality standard for ozone (smog). As a result, industries in urban (ozone nonattainment) areas will be required to reduce their emissions of VOCs. This likely will affect many cleaning applications using solvents classified as VOCs, including semi-aqueous pro-

cesses. Of perhaps greatest significance, however, EPA recently announced its intention to exempt perchloroethylene from VOC regulation.'

Wastewater

Wastewater treatment and disposal also may be an important consideration in aqueous and semi-aqueous cleaning. Wastewater generated from these cleaning processes may contain detergents and surfactants that are not readily biodegradable, as well as oil and greases and other organic contaminants and dissolved or suspended metal derived from the cleaning process. In addition, alkaline cleaning systems may produce effluent that has unacceptable high pH.

Because semi-aqueous cleaning generally demands water rinsing, the techniques available for treating semi-aqueous process effluent are the same as those for aqueous cleaning. One additional problem that may be encountered with some semi-aqueous cleaners, however, is the inability to separate quickly and completely from the rinse water and the resulting potential for significant organic discharge. More recent, "second-generation" sed-aqueous formulations have been developed to separate rapidly, enabling the implementation of closed-loop recycling for rinse water.

As a result, this wastewater may require pretreatment prior to discharge to the sewer systems to meet local, state, or federal requirements. The level of contamination and the quantity of wastewater will depend on the specific cleaning applicatian, and generally will determine the type of wastewater treatment required. In response to concerns about the effluent from aqueous cleaning, several "closed-loop" water recycling systems have been developed. These systems minimize the discharge of process water and the need to treat discharge water, and concentrate the oils, metals, and other contaminants for disposal.

The application of aqueous cleaning systems to heavy oil and grease removal requires the inclusion of equipment to remove the high . volumes of oils that can accumulate. This removal

- 7 -

54

will not only serve to concentrate the oils for disposal, but reportedly can extend the effective life of the cleaning solution. Gravity separators are the most c m " m n devices employed to sepa- rate ncn-soluble oils, and can be included as an integral part of the wash solution holding tank?. I '

In these systems, the oil-laden overflow from the cleaning bath is directed to a decanting chamber where bulk separation of the oil and aqueous cleaner occurs." The oil is decanted from the top of the chamber, and the cleaner can be returned to the holding tank. The collected oil also may contain suspended soils, like metal fines and chips.

For additional separation of the non-soluble oils and greases, a coalescing medium made of polypropylene or monofilament line, blankets, or pillows can be used Coalescing is a simple, but effective, method for removing the oil droplets. The droplets accumulate on the media and then rise to the surface of the solution as they combine to form larger particles.

The removal of emulsified oils may require chemical treatment or microfiltration (0.2 micron pore size). Chemical treatment involves the addition of alum, femc chloride, polymers, and/or organic compounds to break the extiulsion and agglomerate the small oil droplets into larger ones. Microfiltration is accomplished by pumping the wastewater through a semi-permeable membrane that collects the emulsified oils and suspended soils, but allows the aqueous cleaner and dissolved contaminants to flow through as permeate. The wastewater flows through the membrane in a cross- flow pattern so that the concentrated oils and soils are prevented from building up on the membrane.

Carbon adsorption or ultrafiltration (c0.2 micron pore size) can be used to effectively remove other organic contaminants from rinse water for discharge or reuse, including the hydrocarbon chemicals and surfactants used in the cleaners and any finishing and pigment compounds used in processing. Both techniques require that the water be relatively free of suspended solids, oils, and greases, although ultrafiltration mem- branes do not clog as easily." Carbon adsorption provides cleaner water, but can not remove dissolved metals and other ions, while membrane

separation reportedly removes ions and organic material simultaneoudy.''

While suspended metals generally are removed during gravity separation and filtration, the treatment of any dissolved metals in the effluent requires precipitation or ion exchange. Precipitation, the most common method, involves the addition of alkaline reagents such as lime and sodium hydroxide to the effluent to form a metal precipitate that can then be removed for disposal. The precipitation reaction can be conducted in a mix tank or in the sedimentation device or clarifier which, while more expensive to purchase and operate, can speed the sedimentation of the metal precipitate." An ion-exchange system uses the reversible interchange of ions between an exchange resin and the effluent to remove the dissolved metal from the solution.

The cost of wastewater treatment can add significantly to the overall cost of installing an aqueous cleaning system," and small users may find it more economic to contract with a reputable waste treatment company than to treat the wastewater on-site. In such cases, several methods exist for optimizing the cleaning process to minimize the quantity of wastewater produced, including prompt removal of sludge and soils, routine monitoring of the cleaningsolution, regular maintenance of equipment, and the use of dimineralized, deionized, or softened water to minimize unnecessary loading of the cleaner."

As described above, some semi-aqueous systems include a two-stage rinse method consisting of both emulsion and water rinsing. In the emulsion section, the parts are given a preliminary rinse with an aqueous emulsion that contains a relatively high concentration of semi- aqueous cleaning agent dragged out from the wash tank. This emulsion rinse makes it possible to keep most of the cleaning agent and dissolved soils out of the water rinse baths.16 The emulsion rinse is periodically discharged to a decanter where it separates into an aqueous layer and an organic layer. The aqueous layer can be directly returned to the emulsion rinse tank without additional purification, while the organic layer, comprised of solvent and soil, can be collected for disposal.

55

REFERENCES

1.

2.

3.

4.

5.

6.

7.

8.

9.

Bonnelycke, N., 1992, EPA's Safe Altema- tives Program, Presentation to the Ozone Layer Protection Conference Addressing Metal Cleaning Alternatives, Sponsored by EPA Region V, Schaumburg, Illinois, October 27-28.

U.S. Environmental Protection Agency, 1992, Update on Proposed Labeling Rule, Office of Air and Radiation, October 5.

U.S. Environmental Protection Agency (EPA), 1992, Air Quality: Revision to Definition of Volatile 'Organic Compounds - Proposed Rule, Federal Register 57(207): 48490-48492.

Harmon, J.,1991, Innovations in Chemical/ Process Solutions to Metal Cleaning, in Proceedings of the International CFC and Halon Alternatives Conference, Baltimore, Maryland, December 3-5, p. 165-168.

U.S. Environmental Protection Agency (EPA), 1991, Eliminating CFC-113 and Methyl Chloroform in Precision Cleaning Operations, Office of Air and Radiation, EPA/400/1-911018, June, 1OOp.

Chiarella, WJ., 1990, Solvent Versus Aqueous Cleaning, American Machinist 134(6): 56-57 (June).

United Nations Environment Programme (UNEP), 1991, Solvents, Coatings, and Adhesives Technical Options Report, December, 324p.

Chiarella, WJ., 1991, Alternatives for Sohent Cleaning, Presentation at Ozone Layer Protection Conference, U.S. Environmental Protection Agency Region I, Danbury, Connecticut, September 16-17.

Rowny, MJ. and S.D. Temple, 1991, Impingement: The Key to Effective Aqueous Cleaning, in Proceedings of the International CFC and Halon Alternatives

10.

11.

12.

13.

14.

15.

16.

Conference, Baltimore, Maryland, December 3-5, p. 174-183.

Maltby, P., 1991, The Efficient Use of Aqueous Cleaning for Precision Compo- nents, in Proceedings of the International CFC and Halon Alternatives Conference, Baltimore, Maryland, December 3-5, p. 74-83.

U.S. Environmental Protection Agency (EPA), 1991, Alternatives for CFC-113 and Methyl Chloroform in Metal Cleaning, Office of Air and Radiation, EPA/400/ 1-91/019, June, 94p.

Miasek, P.G. and J.L. Schreiner, 1991, Use of Advanced Hydrocarbon Fluids for Preci- sion and Metal Cleaning, in Proceedings of the International CFC and Halon Altema- tives Conference, Baltimore, Maryland, December 3-5, p. 49-58.

Temple, S.D., 1990, A New Era for Aqueous Cleaning, Products Finishing 54(6): 76-83.

Weaver, TJ.M., 1'991, Recycling and Recovery of Cleaning Solutions, Presenta- tion at Ozone Layer Protection Conference, U.S. Environmental Protection Agency Region I, Danbury, Connecticut, September 16-17.

I

1 Hood, C.C., 1991, Water Recycle in Semi- Aqueous Terpene Cleaning Processes, Presentation at Ozone Layer Protection Conference, US. Environmental Protection Agency Region I, Danbury, Connecticut, September 16-17.

Fritz, H.L., 1991, Waste Water Management for Semi-Aqueous Cleaning Processes, in Proceedings of the International CFC and Halon Alternatives Conference, Baltimore, Maryland, December 3-5, p. 96-104.

I

- 9 -

56

. . . ... . . . . . . . . . . . . . . . . . . . . . . . . : :...:$$g,$;$:.; . . .

h

1,l.l -TRICHLOROETHANE VAPOR DEGREASING ELIMINATION IN AEROSPACE REPAIR APPLICATIONS

Abstract

The aerospace repair industry has long utilized vapor degreasing to accomplish a myriad of cleaning tasks. The unque properties of chlorinated solvents and vapor degreasing are unmatched in providing a dean, dry surface for inspection and metal finish processes. The regulatory phase-out of 1-1 -1 Trichloroethane (TCA) and proposed tighter emission controls on vapor degreasers, has forced users to search for altemative chemistries and reevaluate the vapor degreasing process.

This paper is a case study of the program at the American Airlines Tulsa Maintenance and Engineering Center to identify and implement chemicals and processes to replace TCA vapor degreasing. It chronicles the decision making factors and program revisions necessitated by a changing and uncertain regulatory atmosphere.

Background

Traditionally, vapor degreasing with chlorinated solvents has been specified by OEMs for a multitude of cleaning operations including pre/post inspection, metal spray operations, plating, and prep before $lodining and painting. The vapor degreasing process was fast, effective, and versatile, so much so that mechanics came to depend on vapor degreasing and expand its use to many convenience cleaning operations. TCA was the solvent of choice. It was very effective, inexpensive and relatively safe. Its use spread to many cold cleaning operations. At AA's Maintenance and Engineering Center bulk TCA was stored in a 6500 gallon and 2500 gallon storage tank and piped to fifty-fie locations, with no control or accountability.

Regulatory History

so safe that it was exempt from vapor degreaser regulations in many states. However, early predictions of the Montreal Protocol Revisions were that TCA would be included on the ozone depletion list with CFCs and that a production phase-out date would be set and an ozone

1-1-1 Trichloroethane (TCA) was considered

depletion tax implemented. In late 1989. in anticipation of the Montreal

Protocol Revisions and expected adoption into the 1990 Clean Air Act ammendments, the Environmental Engineering department undertook a survey of TCA vapor degreasing operations and cold cleaning applications. As a result, a waste minimization program was developed to continue to economically use TCA until an elimination plan and schedule could be developed.

Waste Minimization Program Outline

1. Where possible, eliminate and/consolidate vapor degreasers to reduce their numbers.

2. Retrofit vapor degreasers with equipment to meet anticipated regulatory requirements and to operate more efficiently.

a Projected capital expenditures: $1 10,oo.

b. Projected 1st year solvent savings: $1 17,000.

3. Provide training to all personnel utiliiing a vapor degreaser to improve operational efficiency, and to reinforce solvent conservation practices.

a Video presentation of proper vapor degreasing technique, solvent handling procedures.

degreaser tanks. b. Eliminate solvent removal from

c. Assign responsibility for individual degreaser equipment to eliminate spills, leaks and abuse.

4. Extend the useful life of TCA by providing weekly chemical analysis to ensure the full cleaning potential is utilized before disposal.

58

1,1,1- TRICHLOROETHANE VAPOR DEGREASING ELIMINATION

IN AEROSPACE REPAIR APPLICATIONS

GLENN TRAVIS ENVIRONMENTAL MANAGEMENT SERVICES

AND

CYNTHIA BOSTER AMERICAN AIRLINES

57

The aqueous dip process consisted of up to a 25 minute dwell time in a 25% concentration cleaning tank, a cold water rinse, and a final hot water rinse to facilitate flash drying. Shop air was available for further drying where necessary. In general, the process was found to work as well or better than vapor degreasing for removal of light oil and soils. Most OEM-approved products were evaluated; several are in use today, dictated by shop preference.

Minor problems were encountered as more aqueous systems were installed. Mechanics were skeptical of using water in place of solvent, and for some, the aqueous dip process did not perform as expected. Some alkaline cleaners showed temperature stability problems, leading to residue deposition on parts. Corrosion inhibitors had to be evaluated and utilbed to address increased rusting problems. Oil contamination limited bath life, requiring investigation into oil skimmers and filtering systems. Shop supervision was concemed over increased process time required for degreasing in an aqueous system.

The year 1992 proved to be a tuming point for the replacement program. President Bush announced an accelerated phaseat of TCA by 1995. Costs of specialty equipment required to clean large and unique parts matched or was higher than new state-of-the-art vapor degreasers now available. Furthermore, EPA was now indit ing that retrofit of existing vapor degreasing equipment would be an option to meet the MACT standard to be promulgated in 1994.

These developments dictated that vapor degreasing be reconsidered. Perchloroethylene was already in use in the plating operation and was a proven process. A comparison chart was developed to organize and compare the pros and cons of both processes (See Table 1). A consideration of altematives must take into account both equipment and chemistry. User safety, environmental and disposal issues must be deal! with up front. Economics must be a major fador in every decision to replace vapor degreasing. The most expensive "high tech" system may not always be the best long term performer. Aqueous cleaning at any cost is an attitude few companies can afford.

Current Program

Final recommendations for the elimination of

TCA vapor degreasing at the AA M&E center were based on the following guidelines.

1. The total number of degreasing areas should be reduced to a reasonable minimum. Remaining degreasing systems, whether aqueous or solvent, should be centrally located. This will require the elimination of some degreasers, sharing of degreasing systems between shops, or routing of parts to areas with degreasing capabilities.

2. Aqueous degreasing tanks should be installed only in shops already using other aqueous cleaning methods,since cold and hot water rinse tanks would be available. Heated. agitated and filtered dip tanks are recommended. Special applications may require special equipment.

3. Shops with no cleaning capabilities,other than vapor degreasing, should be considered for continued vapor degreasing using perchloroethylene. Moddication of existing degreasers will be required.

Vapor degreasing is no longer required for engine repair work by OEMs but is now included as an altemate to aqueous deaniwdegreasing. Other OEMs are evaluating aqueous and semi- aqueous altematives. There are still a few critical aircraft system and component applications where vapor degreasing replacements have not been approved by OEMs. These include the aircraft oxygen system, honeycomb structures and some bearings.

Based on the above guidelines, a total of seven vapor degreasers will remain, using perchlorethylene as solvent in the following areas: metal spray (2), landing gear teadown, blade and nozzle, plating (3) and composites.

TCA vapor degreasing operation. Recycling offers several advantages:

Solvent recycling is an important part of the

Eliminates a hazardous waste stream. Spent TCA is collected and shipped to an off-base RCRA permitted vendor. The solvent is distilled, dryed, and restabilized to meet prearranged acceptance standards, then retumed for reuse.

Reduces the amount of virgin solvent purchases. The Clean Air Act set a production phase out schedule of TCA. Recycled TCA does not incur the ozone

3

60

a. Specific Gravrty (oil loading)

b. Acid Acceptance

c. pH

d. Water Content

5. Establish a recycie program to distill, dry and restabilie spent TCA to meet performance specifications.

a. 30% savings over virgin material.

b. Recycled material does not incur ozone depletion tax.

c. Eliminate dispos'al costs.

6. Establish central person to inspect, test and monitor the program. Pursue additional reduction opportunities where possible.

/

Program results as of year end 1991:

Number of operating vapor degreasers reduced by 45% (capital expenditures $145,000)

Virgin TCA purchases reduced by 78%

Recycle 228,294 Ibs of TCA

Disposal costs for spent TCA: 0

Total program savings as of 12/31/91 : $508,104.00

Eliminatlon of Vapor Degreaslng

The Clean Air Act ammendments of 1990 set the production phase-out of TCA by 2002, and implemented a progressive ODP tax that would reach 31 c per pound by 1995.

TCA Phase-out Schedule - Baseline 1989

1993 90% 1994 85% 1995 70% 1996 50% 2000 20% 2002 -0-

TCA ODP Tax Schedule

1991 13.7t/lb 1993 16.7tAb 1994 30t/lb 1995 31 $Ab

Also, the Clean Air Act specified BACT (Best Available Control Technology) controls for vapor degreasing and ordered MACT (Maximum Achievable Control Technology) regulations promulgated by 1994. Early predictions of MACT controls indicated "0" emissions equipment would be mandated by the EPA. Engineering had previously explored this type of equipment. Costs ranged from $250,000 to $750,000 per unit depending on size. With sohrent costs to double and vapor degreasing equipment replacement prohibitive, the Chemical Review Board, consisting of representatives f rom environmental, safety, engineering lab, purchasing, and industrial engineering, launched a revised vapor degreasing survey. Individual vapor degreasers were surveyed by shop to determine:

1. Specific degreasing operations performed

2. Attemative cleaning /degreasing

3. Repair process madifiitions

processes approved by OEMs

4. Safety considerations of altematives

5. Attemative equipment specifications

6. Economic considerations

The revised vapor degreaser sutvey identified potential equipment and chemical altematives for all remaining TCA vapor degreasers. Cost estimates to eliminate vapor degreasing was 2.8 million dollars. An elimination schedule was developed that eliminated all TCA degreasers by 1997, five years ahead of final production phase out of TCA.

Alternative Processes and Chemicals

Aqueous cleaning processes were the first to be approved by OEMs to replace solvent vapor degreasing. Hydrocarbon and terpene cleaners found increased use in various cold cleaning and wipe on applications.

2

59

Floor space

Water usage

Energy requirements

Water treatment /

Disosal

Chemical costs

Safety

Regulatoty future

IAELEL

AOUEOUS 3 - 4 'xi O'xl2; tanks

500,000 gallon + yearly

4500 gallon to 140°F

solids, ph, metals, phosphates

sludge

$2.73/gallon

non toxic

strengthened treatment standards

Process time

Equipment costs

1 to 1 replacement

Drying required Yes

30 minutes

$30,000 tanks $250,000 + ne1 no

t

- 14x1 O'xl2' tank

process water - no discharge

225 gallon to 26OOF

none

recycle

$2.75/gallon

exposure limits, land ban

VOC exempt, no phase out, no excise tax

10 minutes

$40,000 retrofit $250,000 new tec.,

Yes

no

5

62

depletion tax. Congress has passed new legislation that will accelerate final phase out of TCA by year end 1995. The ozone depletion tax will be adjusted as follows:

present adjusted

1992 13.7e/lb - 1993 16.7t/lb 21.1 t/lb 1994 30.0t/lb 43.5t/lb 1995 31 .Ot/lb 53.5e/lb

Environmental and Public Relations Advantages. EPA mandates a waste reduction program be in place. Environmentally responsible corporations support and practice waste minimization.

A project to retrofit an existing vapor degreaser, utilizing perchloroethylene as the solvent, to meet anticipated MACT regulations is currently being planned. The retrofiit will utilize extended freeboard, secondary freeboard chilling, reduced room draft, controlled hoist speed, and automatic power cover to control emissions. An extensive air monitoring protocol will be used to measure solvent emissions during operating and idle periods. Air emission data will evaluate control performance and determine further control requirements. Operator monitoring will be conducted to ensure exposure levels are well below OSHA regulations, and to develop additional operating procedure modifications if necessary.

Concluslons

.

The regulatory phase-out of 1,l.l- Trichloroethane has forced users to re-evaluate their cleaning processes and requirements. TCA vapor degreasing applications can be replaced with alternative chemistry that is non- ozone depleting. However, any program to replace TCA vapor degreasing must address all altematives. Vapor degreasing with one of the other chlorinated solvents is an option, and can be utilized safely if proper controls are maintained.

Process control is a prerequisite for any replacement program. A successful program will be a synergistic process of evaluating cleaning systems, chemicals, and policies that are effective, efficient, economically sound and environmentally responsible.

4

61

solvents. The cleaning system also contains 41 proprietary stabilizers for the perc fraction and proprietary corrosion inhibitors for the water fraction. The corrosion inhibitors are unique in that they provide both liquid phase and vapor phase protection to the degreaser and the parts being cleaned. A process patent has been granted.

Solvent Stability

For those people who have been vapor degreasing with MCF, it is safe to state that water is the root of all solvent stability problems. The presence of water via contamination, or condensation in a MCF degreaser has more often than not been the cause of the solvent going acid. An acid condition is quite serious in terms of degreaser corrosion, loss of solvent and down time.

In the perdwater azeotrope system, perc solvent compatibility with water is not a problem. The literature indicates that the reaction rate of perc in basic media is extremely slow.(l) This low reactivity or decomposition rate gives perc a half life of about lox6 years. Perc's resistance to hydrolysis is well known, with perc being used as a drying solvent in numerous applications. Perc which is not stabilized is susceptible to oxidation. There are a number of methods available regarding prevention of perc decomposition by oxidation.

Solvent and stabilizer package stability . in the presence of soils is also a key

performance issue. Several different tests have been completed to measure the durability of the azeotrope's proprietary stabilizer and inhibitor system. This includes short term endurance testing, long term testing in the laboratory and actual field testing.

Laboratory testing included reflux to failure testing, which has been a traditional method for evaluating MCF stability with different soils. Typically the chlorosolvent t o be evaluated was refluxed with water, soil (such as mineral oil), iron powder and aluminum shot. Under these conditions, stabilizer depletion normally occurs over a two week

period. With the perc and water azeotrope system, no significant depletion of the stabilizers and inhibitors were seen over two weeks.

Endurance tests were conducted using a small commercial vapor degreaser(21, a Branson 6400-R. After forty days of around the clock operation with soil loadings as high as 20 weight percent, no significant depletion of the stabilizer package occurred. Perc stability in trials with the azeotrope system has not been an issue.

Equipment Requirements

The perc/water system will function in modern vapor degreasing equipment with modifications. Stainless steel equipment is recommended. Most good quality equipment built in the last 10-15 years fits in this category.

More heat input is required because the latent heat of vaporization per pound of azeotrope is more than double that of conventional chlorinated solvents. Boil up rate should be maintained near that used with MCF or TCE or perc in order to produce sufficient quantities of clean vapor condensate "for spraying or dip tank use.

Limits on energy input can limit the type of degreasing equipment which will be capable of operating with the perc/water system. Degreasers which are steam heated and have 351 x102 kilograms per square meter (50 psig) steam available have operated well with the azeotrope. Electricalty heated units may require additional heat input. If energy input is limited, the degreaser's total parts throughput may be reduced.

The addition of supplemental heat may require more cooling capacity. However, this has not been found to be a significant factor in testing completed to dare. Water cooling has been shown to be sufficient in lab and field testing. In many cases the system will operate in a degreaser by first making adjustments in the degreaser vapor temperature and sump temperature limit settings to those commonly

2

64

Organic and Ionic Soil Removal by Vapor Degreasing with a Heterogeneous Azeotrope of Perchloroethylene and Water

Eric L. Mainz Vulcan Chemicals: Applications Research Group

6200 South Ridge Road Wichita, Kansas 6721 5

Introduction

Industries currently using methyl chloroform for vapor degreasing are facing a dilemma. Presidential action causing the acceleration of the methyl chloroform (MCF) phase out is causing unforeseen problems for operators of degreasers as well as associated government regulatory agencies. The Environmental Protection Agency (EPA) is currently developing the National Emissions Standard for Hazardous Air Pollutants (NESHAP) for vapor degreasing. Regulation is scheduled to be issued in late 1994. Rule making is also under consideration to define regulations needed to implement Clean Water Act legislation. All of these upcoming changes in regulations will have profound impact on all cleaning technologies available as options to the metal cleaning industries including vapor degreasing, aqueous and semi-aqueous technology. With all of these regulatory uncertainties, decisions on spending new capital to Upgrade cleaning systems are nearly impossible to make. However, the regulators wait for no one. End users must cease using methylchloroform in two years or less. The low cost approach with the lowest impact on pre and post cleaning operations and the fewest unknowns is to simply switch to an alternate chlorinated solvent. This approach has been recommended and is acceptable to the EPA if two criteria are met. Users must not over expose their workers to the alternate chlorosolvents which have much lower permissible exposures l im i ts than methylchloroform. Secondly, emissions must be controlled. Technology exists today which can meet both of these requirements.

This paper addresses a new vapor degreas ing t e c h n o l o g y based o n

perchloroethylene (perc). This is another chlorinated solvent based alternative to MCF designed to operate in modern vapor degreasing equipment. The system is based on a heter0geneOUS azeotrope between perc and water. Perc has been listed in preliminary Significant New Alternative Program (SNAP) drafts as an acceptable substitute for MCF. Perc, which has a long history of safe use in the dry cleaning industry and as a vapor degreasing solvent, is expected to receive an exemption from VOC reporting in 1993. Perc will continue to be regulated as a Hazardous Air Pollutant. The pedwater azeotrope is a novel system which combines the oil cleaning power of a chlorinated solvent with the polar soil cleaning power of w-ater. In fact the system can be described as a semi-aqueous vapor degreasing system with little or no waste water effluent.

Azeotrope Physical Properties To understand the azeotrope system,

first consider the physical properties, Table 1. The azeotrope has a boiling point of 88°C (1 9O"F), very near that of trichloroethylene (TCE), which is 87°C (188°F). This is slightly above that of MCF, at 74°C (165°F) but far below the 121 "C (250°F) boiling point of pure perc. This is a heterogeneous azeotrope. In liquid state, the perc and water do not mix and are only very slightly miscible. In fact the solubility of one component in the other is on the order of only .several hundred ppm at ambient conditions. When the system is boiled, a hOmOgeneOUS vapor cloud forms. The vapor composition is approximately 1 6 weight percent water and 84 weight percent perc. Latent heat of vaporization for the azeotrope is approximately 109x1 O3 joules/kilogram (228 btu per pound) (calculated). This is about 2.5 times higher than MCF or the other chlorinated

63

I

the azeotrope (3) or pure perc. The cleaning performance of the azeotrope system for ionic soils is impressive. In vapor plus dip cleaning tests, the azeotrope system removed 99.9 percent of the sodium chloride from the sample plate while only 11 percent was removed by the pure perc system.

The cleaning of aqueous milling fluids is another demonstration of the enhanced cleaning ability of the pedwater system. Five common water based milling fluids were selected for the cleaning tests. The fluids were Cimstar 40 (41, Boelube (51, E2000 (61, 4250 (7 ) and Trimsol (8;. Cleaning tests were conducted by preweighing dean stainless steel coupons followed by dipping in the milling fluid. After draining, the coupons were cleaned in the perdwater system. Perfarmance was compared with identical cleaning completed in pure solvent degreasing, using perc in one case and TCE in another.

A cleaning process cycle of 60 seconds in vapor, 15 seconds in the boiling sump and 45 seconds in the vapor was used with each solvent system. A Branson 8400-R was used for all tests. Overall performance of the perdwater system was superior. This was especially true with the Cimstar 40 and 4250 milling fluids. The Trimsol, Boelube and B- 2000 were cleaned equally well by all systems. This demonstrates an important point. The perc/water system will offer advantages when solvent soluble and water soluble soils must be removed.

Limited outside testing has been completed. Soil cleaning tests were completed by a major degreaser manufacturer to determine the perc/water systems ability to remove an array of soils. The test included standard milling oils, honey oils, common water emulsion fluids such as Trimsol as well as aged soils. Due to the presence of perc, all of the organic soils were easily removed. The pedwater system was also successful in removing buffing compounds.

In separate testing we have successfully removed high melting point waxes commonly used in the plating industry. The system has proven to be quite versatile.

Field Testing

The best measure of any cleaning system's performance is in actual field testing. We have been operating the azeotrope system at a plating operation in Wichita for over six months. The plating shop has three degreasers, two dedicated to aluminum parts and one dedicated to cleaning steel parts prior to zinc or chrome plating. The pedwater system being evaluated in the field tests is targeted for steel cleaning. The corrosion inhibitor package is not compatible with aluminum since it functions at a pH of 10. A second generation corrosion inhibitor system designed for ferrous and white metals is under development. Typical part configurations seen in the zinc plate shop range from pipe fittings to computer cabinet frames. This is a job shop, cleaning and plating parts with a range of configuration and soils. Normally the pans are not excessively soiled, being well drained during transport from the parent source.

The facility is degreasing with a Baron- BIakeslee open top unit, Model No. 3648. This is a stainless steel unit with steam heat through a carbon steam tube bundle. The unit open top area is .91 meters by 2.41 meters (3 by 8 feet]. To control emissions the degreaser is fitted with a one piece mylar roll cover. Worker exposures are minimized using a lip vent exhaust system. .

TCE was used in the unit prior to our tests with the perc/water system. This has provided an opportunity to compare performance and emissions between the two systems.

Very minor modifications were needed to complete the conversion from TCE to the pedwater system. Steam pressure was increased from 105 to 351 kilograms per square meter (15 to 35 psig). The water separator water exit line was repiped back to the boiling sump of the degreaser. This line now has a valve to direct the condensed water from the azeotrope vapor back to the storage tank. This modification allows recovery of the water during the degreaser boil down prior to routine degreaser clean out maintenance.

4

66

used with a pure perc system. Secondly, since water is a co-solvent in the cleaning system, the water separator water drain is repiped back to the degreaser boiling sump. There is no waste water stream from the water separator which must be handled as a hazardous waste.

Degreaser operator exposure must be minimized. MCF has a permissible exposure level of 350 ppm. The permissible exposure level for perc is low at 25 ppm. Technology is available from degreaser manufacturers which can practically eliminate any exposure problems while greatly minimizing solvent consumption.

Cleaning Process Description

The perdwater ' functions in the degreaser as follows. Stabilized perc is charged to the sump of the degreaser. The sump is filled with perc to completely cover the heating coils. Additional perc is added equal to two to three times the perc loss rate expected for the degreaser between solvent additions. This will assure that the heating coils are continuously covered with perc during operation.

Deionized water containing corrosion inhibitor is added to the sump. The total charge of water to the sump is not extremely critical with equal volumes of perc and water in the system being a good target value. The total charge of perc and water to the degreaser must include enough of the solvents to fill dip tanks, spray storage tanks and transfer lines.

DegreaSer start up is similar to that when using pure chlorinated solvent. The azeotrope vapor is formed at the interface between the perc and water layers in the sump.

Once formed, the azeotrope vapors flash through the water layer to form the vapor cloud in the degreaser. Heat up time may be extended due to the higher heat of vaporization

' for the azeotrope.

After t h e degreaser reaches temperature and the vapor cloud stabilizes, parts cleaning can begin. When cold parts are

placed in the degreaser, the perc and water condense and immediately separate into two liquid phases. Typically the perc will sheet from the parts. The condensed water phase will drain from the parts with some beading of water occurring.

Perc and water which condense on the cooling coils will be diverted to the spray storage tank in a typical degreaser (if so fitted). The water will remain on top of the perc phase and will overflow to the degreaser. Both the perc and water overflows can be sent straight to the boiling sump, to a spray reservoir(s1 or dip tank(s1, depending on the degreaser configuration.

A number of different equipment configurations can be utilized to take advantage of the perc/water system. Vapor with spray cleaning is commonly used in industry. This can include spraying of condensate perc and/or condensate water. The desired configuration depends on the cleaning requirements.

The option of using a water or perc dip tank to aid cleaning can also be incorporated into equipment design. This flexibility is one of the primary performance advantages of the perc/water degreasing system. Parts can easily be cleaned with both distilled perc and distilled water via a user selected combination of vapor cleaning, spraying and dipping.

Cleaning Performance

One unique advantage of the azeotrope system is the inclusion of water in the cleaning '

process. Polar soils are removed. As an example a very simple test comparing the cleaning power of the azeotrope versus TCE or perc, degreasing was conducted using sodium chloride deposited onto metal plates. Approximately 0.5 grams of salt in aqueous solution were placed in a one inch diameter well which had been milled in a 3 2 cm (1/8 inch) by 3.81 cm (1.5 inch) by 10.2 cm (4 inch) stainless steel coupon. The water was evaporated from the coupon in a drying oven.

Cleaning tests were conducted by vapor only cleaning and vapor plus dip cleaning in glass bench top vapor degreasers containing

3

65

degreaser will typically contain more than 16 weight percent water which is the azeotrope composition. To remove the water as azeotrope vapor, pure perc is returned to the boiling sump during the boil down while all water is retained in the storage tank. This is easily done by using the spray system to return perc to the degreaser and sump. Once the water has been removed the degreaser is operating on pure perc solvent.

By monitoring the vapor temperature it is easy to determine when all water has been removed. Pure perc has a boiling point of 121°C (250°F) far above the pedwater azeotrope boiling point of 88°C (1 90°F). The remaining perc is boiled over as condensate leaving a solvent and soil heel which is be removed from the degreaser by draining. The heal is processed on site in a batch still with most of perc being recovered. A small quantity of perc remains in the soil heel. This is disposed of by a certified hazardous waste hauler. With this system, all of the water is recycled as well as nearly all of the perc from the degreaser.

Conclusion

The perdwater azeotrope cleaning system offer unique cleaning capability to those desiring to continue to use vapor degreasing cleaning. The system designed for steel and stainless steels is being proven in a field trial program. A commercial product is expected to be offered in mid-1993. Current work is concentrating on development of a system compatible with aluminum as well as steel substrates. We are seeing progress in this area with field testing of the new system planned for early 1993.

(1) Jeffers, P.M.; Ward, L.M.; Woytowitch, L.M.; Wolfe, N.L.: Homogeneous Hydrolysis Rate Constants for Selected Chlorinated Methanes, Ethanes, Ethenes, and Propanes. Environmental Science and Technoloav. 1989, 23, 965-969.

(2) Branson B400R, Danbury, CT

(3) C h e m c i a 1 s , Wichita, KS (4) Cimstar" 40, Cincinnati Milacron,

(5) Boelube 70105, Orelube Corp.,

(6) 8-2000, Blasser Swisslube Inc.,

(7) Houghton-Grind@ 4250, E.F. Houghton,

(8) Tr im@ Sol, Master Chemical,

Hydroper", Vulcan

Cincinnati, OH.

Plainview, NY.

Goshen, NY.

Valley Forge, PA.

Perrysburg, OH.

6

68

TCE consumption was monitored for a period of ten weeks. Next the perc/water system was operated for six weeks without changing any of the procedures used by the operators. Solvent consumption was found to identical for both the trichloroethylene and perdwater systems.

The operators were then trained to vapor degrease parts using proper techniques. This included using the degreaser covers when not cleaning and also turning off the lip vent when the cover was in place. Drag out was minimized by longer free board hold times and tipping of the baskets prior to removal. A 70 percent decrease in solvent consumption was documented once the new techniques were adopted.

Degreaser Monitoring and Control

The perdwater system is more complex than a pure chlorinated solvent degreasing system. It is necessary to monitor the perc and water levels in the boiling sump. A slot type sight glass can be installed for a direct visible reading of the perc and water levels. A simple sample thief or profile sampler can also be used. This can consist of a half inch tube equal to the depth of the degreaser fitted with a valve having a remote actuator. The sample thief is passed into the degreaser prior to start up with the valve open and fills with a vertical cut of the perc and water. The perc and water are transferred to a graduated cylinder which has been calibrated for the specific degreaser. Volume of liquid in the cylinder can then be converted to inches of perc and water in the degreaser. Necessary make up volumes of the solvents can then be determined.

The concentration of the corrosion inhibitor in the water phase should also be monitored. This can be accomplished by catching a sample of the water and perc condensing from the cooling coils. A portion of the water collected is then titrated to a specific color end point using a titration kit provided with the initial charge of the pedwater system. The titration is simple, being very similar to that commonly used with MCF.

Training to monitor and control the degreaser was given to the shop foreman and the chemist responsible for monitoring the plating baths. They have been controlling the unit for four months without any problems. Sump perc and water levels are measured every Monday with additions made the same day. The record has shown that solvent and water consumption are about equal on a volume basis.

Cleaning Performance

Operators a t the plating shop are satisfied with the cleaning provided by the perciwater system. Organic soil removal has been equal to that seen with the TCE use previously. Some water spotting of parts is seen when a vapor only cleaning process is used. The water spotting is greatly reduced when the parts are sprayed with condensate perc from the storage tank.

Tests were also completed in which perc heated to 102°C (21 5°F) was sprayed on the parts after they reached the vapor temperature. The hot perc, heated above the vapor temperature, was very effective in removing beaded water from the parts. The hot perc spray was not installed on a permanent basis since in this plating operation, water spotting was not a problem. Standard procedures in the plating operation included an initial water rinse in the plating line. The water spots were easily removed and did not affect the plating quality.

Degreaser Clean Out Procedure

The perclwater system has no aqueous effluent for disposal. Water and corrosion inhibitors used in the system can be recycled back to the degreaser following the degreaser clean out procedure.

Degreaser cleaning at the plating shop is done on a monthly basis. The process involves a boil down to remove perc and water followed by draining the remaining heel.

The boil down procedure is completed by sending all condensate to the storage tank instead of back to the degreaser sump. The

5

67