Corrosion Resistant Vacuum Brazing Sheet

1. Field of the Invention

This invention relates to brazed aluminum, and particularly to the corrosion properties of vacuum-brazed aluminum materials.

2. Description of the Prior Art

Aluminum has long been known for its resistance to corrosion, a property which renders it a favorable material of construction for a wide variety of purposes. With the relatively recent development of vacuum brazing technology for joining together aluminum parts, aluminum has gained wide acceptance in the manufacture of heat exchangers such as automotive radiators and evaporators for air conditioning units. Vacuum brazing has considerable advantages over the earlier-developed technique of flux brazing, since it avoids the need for a special brazing flux and for expensive cleaning operations to remove the flux after brazing, and it eliminates the possibility of flux contamination corroding the aluminum.

A problem which arises in all brazing operations is the intergranular penetration of core material by certain species in the brazing alloy, notably silicon. When the brazed product is subjected to a corrosive environment, the intergranular regions in the core, where silicon is present in high concentrations, are particularly susceptible to corrosion. In automotive radiators, for example, where the brazing alloy is present only on the exterior surface of the tubes, the salts and moisture from the road are sufficiently corrosive to cause attack from the outside, resulting ultimately in pinhole formation.

This problem has been addressed in the literature in a variety of ways. An early example is Miller, U.S. Pat. No. 2,821,014 (Jan. 28, 1958), where it is disclosed that intergranular corrosion problems in flux and dip brazing are alleviated by adding an interlayer between the structural member portion and the brazing layer. The interlayer is aluminum or an aluminum-base alloy, particularly certain magnesium-containing alloys, having a melting point greater than that of the structural alloy. The solution offered by Singleton et al., U.S. Pat. No. 3,788,824 (Jan. 29, 1974) and its divisional, U.S. Pat. No. 3,881,879 (May 6, 1975), is directed to vacuum brazing, and involves the addition of iron to either the core alloy or the cladding alloy as an alloying element, resulting in improvements in both corrosion resistance and sag resistance. Anthony et al., U.S. Pat. No. 4,039,298 (Aug. 2, 1977) address both flux and vacuum brazing, and disclose a composite of complex and highly specified composition as being particularly beneficial in terms of corrosion properties. The disclosed core alloy contains specified amounts of manganese, copper, chromium, silicon and iron as alloying elements with both a solid solution and an alpha-phase, whereas the alloying elements in the cladding are bismuth and silicon. An additional disclosure by the same patentees appears in U.S. Pat. No. 4,093,782 (June 6, 1978) and its continuation-in-part, U.S. Pat. No. 4,167,410 (Sept. 11, 1979), in which the core alloy contains a specified combination of chromium and manganese, with resultant improvements in both corrosion resistance and sag resistance. A similar disclosure appears in Setzer et al., U.S. Pat. No. 3,994,695 (Nov. 30, 1976), where the core alloy contains a chromium-manganese-zirconium combination, the sole claimed benefit however being an improvement in sag resistance. A combination of copper and titanium as primary alloying elements in the core alloy is disclosed in Kaifu et al., U.S. Pat. No. 4,339,510 (July 13, 1982), as providing a benefit in intergranular corrosion resistance.

A different approach is disclosed by Nakamura, U.S. Pat. No. 4,172,548 (Oct. 30, 1979), in which corrosion following fluxless brazing processes in general (including both vacuum brazing and brazing in an inert atmosphere) is controlled by controlling the grain size of the brazing sheet to at least 60 microns in diameter, achieved by a controlled cold work followed by a full anneal.

To summarize, with the exception of Nakamura, prior investigators have approached the problem by introducing specific elements in the alloy composition, rather than introducing processing modifications. Processing modifications, on the other hand, particularly variations in the combinations, degrees and sequence of strain hardening and annealing, are generally used for controlling the ductility and tensile properties of the final product. Setzer et al., referenced above, demonstrates several of these combinations, ranging from those ending with a fully hardened product (maximum cold work) to those ending with a fully strain-free (annealed) product.

In manufacturing operations which require forming of the sheet into various shapes prior to brazing, a certain degree of ductility is required for the sheet to respond in an appropriate manner to the forming equipment. Examples of such forming processes are roll forming, drawing, spinning and shear forming. Conventional methods of achieving the desired degree of ductility involve a partial anneal of the product after it has been cold rolled down to the final gauge. An explanation and description of this is disclosed by Singleton, U.S. Pat. No. 3,963,454 (June 15, 1976) at column 4, lines 34-58. The partial anneal is controlled to leave the desired amount of cold work remaining in the product.

It has now been discovered that intergranular corrosion is particularly acute in composites that have been vacuum brazed, much more so than those brazed by other types of brazing. In addition, it has been discovered that the resistance to integranular corrosion is dependent not on the final hardness of the material before brazing but on the method by which the final hardness is achieved. Thus, a product having markedly superior corrosion properties for a given final hardness is achieved by the use of a procedure whereby a full anneal is followed by a controlled cold work, the latter being the final step before forming and brazing. Finally, and in conjunction with these discoveries, it has also been discovered that a final cold work step within certain ranges of reduction depending on the alloy will provide improved corrosion resistance over degrees of cold work falling outside these ranges.

The process of the present invention is applicable over a wide range of alloys, composites, and processing procedures.

The core alloy may vary widely in composition, although those containing manganese, magnesium or both as the primary alloying elements, i.e., the components present in the highest concentrations other than the aluminum itself, are preferred. The concentrations of these elements in the core alloy may vary widely but typically range from about 0.2% to about 2.0% by weight. Particularly preferred core alloys are those which further contain a small amount of chromium in addition to the primary alloying element(s). Again, the concentration may vary widely, although best results are generally obtained within a range of from about 0.05% to about 0.4% by weight. In further preferred embodiments, the core alloy is a non-heat-treatable alloy, most preferably of the 3XXX series of Aluminum Association designated alloys. Examples are the 3003, 3005 and 3105 alloys.

The cladding alloy (also referred to herein as a "brazing alloy") may be any aluminum alloy capable of flow under typical vacuum brazing conditions to form fillets at the contact points sufficient to produce a sturdy bond. Such alloys will generally contain silicon as the primary alloying element, preferably at a concentration ranging from about 5% to about 15% by weight. Exemplary such alloys are those of the 4XXX series, e.g., 4004, 4043, 4045, 4047, 4104 and 4343.

Cladding alloys which are particularly preferred for their behavior during vacuum brazing operations are those which further contain a gettering agent. The latter is generally an alkali or alkaline earth metal which volatilizes at brazing temperatures and combines in the volatilized state with atmospheric oxygen to prevent such oxygen from combining with the solid surface of the metal article and interfering with the flow of the brazing alloy. The preferred gettering agent is magnesium, generally present in the cladding alloy in an amount ranging from about 0.1 to about 3.0 percent by weight.

The process of the invention is primarily of interest in producing composite aluminum sheet products having a hardness ranging from about one-quarter to about three-quarters hard, hardness being defined as zero hard at the fully annealed state to full hard at the fully strain hardened state. As known to those skilled in the art, full hard is recognized in the industry as being equivalent to the hardness of a product which has been cold worked to a 75% reduction. In preferred embodiments of the invention, the final brazing sheet is approximately one-half hard, i.e., ASTM temper designation H14. In addition, the process of the invention is of particular benefit to brazing sheet products which are intended to be roll-formed into flat tubes such as those used in automobile radiators.

In most applications, the processing steps which precede the final anneal and controlled cold working are a hot working to bond the cladding to the core, a scalping or etching to remove the oxide formed on the surface during the hot working, and a cold working to bring the gauge down to a value somewhat higher than the desired ultimate value, so that the full anneal and final cold working will produce both the hardness and gauge sought. Thus, a first cold working reduction is conducted to leave a critical amount of reduction remaining, annealing is then done to a strain-free state (i.e., full anneal rather than partial), and finally a limited cold working step to achieve the desired gauge.

The strain introduced by the final cold working, provided that it is within the critical range for the particular core alloy, carries the corrosion resistance through the vacuum brazing step. The result is that the ultimate brazed product demonstrates substantially improved corrosion resistance over products of the same hardness made by the conventional process. It has also been discovered that this controlled final cold working increases the degree of flow of the brazing alloy during the brazing step, thus providing stronger fillets.

The cold working and annealing steps may be done according to conventional techniques well known to those skilled in the art. It should be noted, however, that the annealing step is a full anneal, i.e., one which returns the material to a substantially strain-free state. The temperature required to achieve this will vary to some extent depending on the compositions of cladding and core alloys as well as the length of time during which the composite is held at that temperature. In general, however, a temperature of from about 500.degree. F. to about 850.degree. F. (260.degree. C. to 427.degree. C.), for a period of time ranging from about 0.5 hour to about 5 hours, preferably from about 600.degree. F. to about 700.degree. F. (316.degree. C. to 371.degree. C.) for about 1 to 3 hours, will provide the best results.

The critical range of the final cold work step will vary somewhat depending on the amount and type of alloying elements in both the core and the cladding alloys, and will also vary somewhat depending on the processing history of the product prior to the final steps (for instance, the presence, absence or degree of homogenization of the core alloy). In general, however, the critical range extends from about 3.0 percent to about 50 percent reduction after the full anneal. Composites in which the core is comprised of an alloy which contains no more than trace amounts of chromium provide best results over a range of from about 3% to about 20%. Composites in which the core alloy contains in excess of about 0.05% chromium provide best results in the range of about 15% to about 50%.

As mentioned above, the critical steps of the process as well as those which precede and follow them are done according to conventional techniques. In most applications, the cladding layer forms from about 0.1 percent to about 20 percent, preferably from about 1.0 percent to about 15 percent, of the thickness of the entire composite, and is present on one or both sides of the core, depending on the intended use of the brazing sheet. The bonding of the cladding to the core is typically achieved by conventional hot rolling techniques at temperatures ranging from about 450.degree. F. to about 1,000.degree. F. (230.degree. C. to 540.degree. C.). The cladding may be preceded by homogenization of the core typically at a temperature within the approximate range of from 800.degree. F. to 1150.degree. F. (427.degree. C. to 621.degree. C.), although in some alloys and procedures corrosion resistance will be improved by omitting the homogenization step. Cold working is generally done by rolling, stretching or forming at ambient temperature. In general, the initial cold work step (prior to the full anneal) is done to a reduction of at least about 50%, preferably at least about 80%. Finally, products prepared by the process of the present invention may be vacuum brazed according to conventional techniques. Such techniques are done at a pressure of 10.sup.-2 torr or less, preferably between about 10.sup.-4 and about 10.sup.-6 torr, and at a temperature ranging from about 1,000.degree. F. to about 1,200.degree. F. (540.degree.-650.degree. C.), preferably from about 1060.degree. F. to about 1125.degree. F. (571.degree. C. to 607.degree. C.).

The following examples are offered for purposes of illustration only, and are intended neither to define nor limit the invention in any manner.

The following experiment demonstrates the unusual susceptibility of vacuum-brazed composites to intergranular corrosion, and the effect of applying a final cold work to the composites.

A brazing sheet product was formed from the following core and cladding alloys:

The product was 0.020 in. (0.051 cm) thick with the cladding material on both sides, each cladding layer comprising 12% of the total thickness. The product was fully annealed and had been prepared from an ingot which had been homogenized before the cladding layers were added.

Four coupons were cut from the product, each approximately 2.times.5 in. (5.1.times.12.7 cm). The first coupon was heated to and held for two minutes at 1100.degree. F. (593.degree. C.) in an air furnace to simulate air-brazing conditions. Of the three remaining coupons, two were reduced by cold rolling, one to a reduction of 5.6% and the other to 8.6%. All three were then heated in a vacuum-brazing furnace at a pressure of 1.times.10.sup.-5 torr, where they were held at 1100.degree. F. (593.degree. C.) for two minutes.

All four coupons were then tested for corrosion by immersion in a solution consisting of (all figures by weight):

for thirty days, during which time the solution was heated to 194.degree. F. (90.degree. C.) and held at that temperature for one hour every twenty-four hours.

Polished metallographic cross sections were prepared from these coupons. The metallographic sections were etched and photomicrographs were taken and evaluated visually, with the following results: the air-brazed coupon displayed isolated pits in the cladding, the pits not extending into the core; the vacuum-brazed coupon which had not been cold worked displayed a continuous network of corrosion cavities throughout the cladding and extending into the core to a distance of about 25% of the core thickness, clearly defining the grain boundaries; the 5.6% cold reduced coupons exhibited isolated sites of attack that penetrated the core alloys up to about 20% of thickness; and the 8.6% cold reduced coupons exhibited isolated pits that penetrated the core to about 30% of thickness, not defining individual grains.

The comparison demonstrates that vacuum brazing caused a considerably heightened susceptibility of the fully annealed composite to intergranular corrosion in the core alloy than did air brazing, and that this susceptibility was substantially reduced by cold rolling the annealed composite prior to brazing in both of the reductions tested.

This example demonstrates the difference in corrosion susceptibility between two samples of equivalent hardness, one obtained by cold rolling to the desired gauge followed by a partial anneal (to a temper of H24), the other by cold rolling to a thickness greater than the desired gauge, fully annealing, and finally cold rolling to the desired gauge (the resulting temper being H14).

The test was performed on three products. The first was a commercial sheet consisting of core and cladding alloys as follows (on weight basis):

The core alloy had been homogenized after casting by heating to 1120.degree. F. (604.degree. C.) for at least four hours, then clad on one side and hot rolled to a gauge of 0.10-0.15 in. (0.25-0.38 cm), then cold rolled to a final gauge of 0.015 in. (0.038 cm), and finally partially annealed to a temper of H24.

The second product was prepared in the laboratory but was processed in a manner similar to the first, except that the core alloy contained 0.15% chromium and the homogenization step was eliminated.

The third product (also prepared in the laboratory) was prepared and processed in a manner similar to the second except that the cold rolling step was done to a gauge of 0.0215 in. (0.055 cm), and the anneal was done for two hours at 650.degree. F. (343.degree. C.) and followed by a final cold work of 30% to a final gauge of 0.015 in. (0.038 cm), resulting in a temper of H14.

Rectangular pieces measuring 2.5 by 4.0 in. (6.4 by 10.2 cm) were cut from each product, and placed in a three-chamber vacuum brazing furnace, where they were heated in twelve minutes to 1100.degree. F..+-.5 (593.degree. C.) and held for two minutes at that temperature and at a pressure of 1.times.10.sup.-5 torr. The pieces were then further cut into 2.5 in. (6.4 cm) squares, and the nonclad side masked with electroplater's tape. The pieces were then corrosion tested according to ASTM Procedure No. G43 in a 98% humid atmosphere with intermittent exposure to an atomized solution consisting of (on weight basis):

for periods of time ranging from 24 to 200 hours at a temperature of 120.degree. F. (49.degree. C.). The pieces were then rinsed in cold water and cleaned in a solution of chromic and phosphoric acids (ASTM Solution No. G1) at 180.degree. F. (82.degree. C.), and finally baked for at least thirty minutes at 250.degree. F. (121.degree. C.) to remove retained water.

Each square was then pressurized with air at 20 psig (13.8 newtons/cm.sup.2) under water and the number of perforations detected by visual observation of streams of bubbles. The number of perforations per piece is shown below as a function of the number of hours in the corrosion solution:

While an improvement is achieved by the addition of chromium to the core alloy and the elimination of the homogenization step, it is clear that even further improvement is achieved by the modification of the cold work/anneal sequence.

The following demonstrates the effect of the degree of the final cold work step (preceded by a full anneal according to the process of the invention) on the corrosion characteristics of a particular composite.

The composite used was that described in Example 1. After the full anneal, individual strips of the composite were cold rolled to varying degrees ranging from 3.7 to 28.3% reductions, then placed in the vacuum brazing furnace at 1.times.10.sup.-5 torr and 1100.degree. F. (593.degree. C.) for two minutes. The strips were then corrosion tested using the solution and procedure described in Example 1 except that the exposure lasted for sixty days. The strips were then cleaned and polished cross sections prepared for metallographic examination. These were etched in a hydrofluoric-sulfuric acid etchant for sixty seconds and photomicrographs were taken and evaluated visually.

The flowability of the brazing (or "filler") alloy was tested on further pieces cut from the same composites after they had been annealed and cold rolled. The pieces were suspended vertically in the vacuum brazing furnace at the conditions enumerated above. The weight of the filler metal that had drained to the lower third (0.5 inch, 12.7 mm) of each piece was then determined and expressed in terms of the total amount present on the piece.

The results, including both filler metal flow and degree of corrosion, are listed in Table 3.1:

It appears from these observations that optimal results were obtained with final cold reductions of 3.7% to 15.8%.

A test similar to that of Example 3 was conducted on a composite of the following alloys:

The composite had a total thickness of 0.020 in. (0.05 cm), including a 15% cladding on both sides. The range of cold rolling extended from 3.4% to 26.9%, and the photomicrograph and filler metal flow results were as follows:

These observations indicate that optimal results were obtained at cold reductions ranging from 8.5% to 22.6%.

A test similar to that of Examples 3 and 4 was conducted on a composite of the following alloys:

The composite was prepared according to the description in Example 2, including homogenization of the core alloy but not including the chromium addition. Final cold reductions ranging from 6.3% to 16.8% were performed, followed by simulated vacuum brazing as described in Example 2. Samples were then exposed to the corrosion solution of Example 2 for periods of time ranging from 24 to 96 hours, then rinsed, cleaned and baked. Finally, the pressure test described in Example 2 was conducted, yielding the following results:

The tabulated results indicated favorable performance at cold reductions of 6.3% to 13.7%.

A test similar to that described in Example 5 was conducted, with the modification that the core alloy further contained 0.15% chromium and the core alloy was not homogenized. The pressure test yielded the following results:

Optimal results appear in the tests run at 18 to 40% cold reduction, indicating a range of about 15% to 50%.

The foregoing description is offered primarily for illustrative purposes. It will be readily apparent to those skilled in the art that the particular materials and procedures described herein may be further varied or modified in numerous ways without departing from the spirit and scope of the invention as set forth in the following claims.