Adhesives for bonding cathodically protected rubber to metal devices.

Adhesives for bonding cathodically protected rubber to metal devices

Rubber is often chemically bonded to metal and used in undersea applications. Sonar transducers, oil well platforms, and undersea pipelines are examples where rubber is used to either protect the metal or accommodate the motion of metals relative to each other. When rubber to metal bonded assemblies are used in applications such as these, proper adhesive selection can be critical to the life expectancy of the part. This environment is highly corrosive to metal and exposes the bond to a variety of temperatures. Coupled with the inherent mechanical stresses caused by design or application, one can see the importance of using the proper adhesive system.

Bond failure often occurs for a less than obvious reason. Often the metal is protected against corrosion by the use of a sacrificial anode, impressed current, or a combination of the two. Stevenson (ref. 1) noted that when cathodic potentials are applied to the steel substrate, bond failures would rapidly occur. He later speculated that a build-up of hydroxyl ions, resulting from the reduction of either hydrogen ions or oxygen near the steel surface lays an important role in the loss of adhesion.

Many people (figs. 2-9) have studied the phenomenon attempting to understand the failure mechanisms. Most agree the loss of adhesion can be attributed to chemical reactions that occur within the adhesive film. These chemical reactions can lead to an osmotic pressure build-up as water accumulates in the bondline. Eventually the adhesive and rubber will delaminate from the metal.

Keeping this in mind, we evaluated the resistance of several adhesive systems to cathodic disbonding. We also completed salt spray testing to determine if a relationship existed between the failure mechanisms involved. The adhesive systems chosen for study included several controls which have been studied previously and found to have little resistance to cathodic disbonding. The other systems chosen for study were either expected to provide enhanced barrier properties or employ alternate metal adhering chemistries.

Experimental

The substrate used in this article was SAE 1010 cold rolled steel. Two different substrate preparations were used. The first preparation method involved vapor degreasing the parts in perchloroethylene, blasting them with steel grit, and a final vapor degreasing in perchloroethylene. The second surface preparation method was a calcium modified microcrystalline zinc phosphate treatment.

After surface preparation, the adhesive systems were dip applied. The parts dried thoroughly in a fume hood to remove all the solvents. In instances where multiple coat systems were used, each coat was dried thoroughly before application of the next coat.

Prior to bonding, the parts were masked with tape so that only a one square inch area was bonded to the rubber. The elastomer used in the study was polychloroprene (CR). Table 1 lists the formula for the compound used. Bonding was done via compression molding at 307 [degrees] F. Cure time was 40 minutes.

In some experiments, the effect of prebake on performance was studied. Prebake is defined as the length of time the coated part resides in the mold before introduction of rubber. Prebake times of 0 and 5 minutes at 307 [degrees] F were used.

After bonding, the parts cooled for at least 12 hours at ambient conditions before testing. Three tests were conducted:

* Primary adhesion per ASTM D429-B. Parts were tested at room temperature with a peel angle of 45 degrees. Test speed was 2 inches per minute. Peak peel strength (lbs./inch) and the percent rubber retention on the part were measured.

* 5% salt fog testing at 100 [degrees] F. Parts were stressed by tying the rubber back with stainless steel wire and the bond line was scored to allow penetration of the salt water to the bond interface. Exposure times varied from three to 10 days. After exposure, the rubber was peeled back from the parts and the percent rubber retention was recorded.

* Stressed 28-day cathodic disbonding testing. The test parts were tied back and scored just as they were in the salt fog tests. The tests were performed in a five gallon plastic pail containing tap water with NaCl dissolved in it at a concentration of three weight percent. The parts were connected electrically to a sacrificial zinc anode using stainless steel wire. Care was taken to insure a good electrical connection. This made the bonded parts the cathode of an electrochemical cell. The potential between the zinc anode and a Ag/AgCl reference electrode was adjusted to and maintained at 1 volt. Air was stirred into the pail daily to keep the oxygen supply in the water high. Also, the solutions were replaced with fresh salt water once a week. The parts were monitored daily for signs of delamination. If parts delaminated before the 28 days were up, they were removed from the tank and the time it took for the parts to disbond was recorded. After exposure, rubber was peeled back from the parts that survived and the percent rubber retention was recorded.

Figure 1 shows the cathodic disbonding test apparatus.

Results and discussion

In the figures, the trade name Chemlok is abbreviated CH. Multi-coat systems are always listed with the primer first, followed by any covercoats. For example, a primer coat of Chemlok 205 followed by a covercoat of Chemlok 252 would be shown as CH205/CH252.

A service environment involving cathodic disbonding is very hard on rubber/metal bonded parts. Figure 2 illustrates this. The figure shows rubber retention results from primary adhesion, 3-day salt spray and 28-day cathodic disbonding testing or CR/grit blasted steel bonds using two adhesive systems. The first is an in-house adhesive which we will call Adhesive A. The second is a three-coat system consisting of CH AP134 primer, a second coat of CH 205 primer and a topcoat of CH 220 adhesive.

Adhesive A failed completely in both salt spray and cathodic disbonding tests. The three coat system had no loss of adhesion in the salt spray tests; but adhesion dropped to 58% rubber in the cathodic disbonding test. This shows that the cathodic disbonding tests is tough on rubber/metal bonds.

We used salt spray testing to screen adhesive systems before exposing them to cathodic disbonding tests. Figure 3 shows results from such a screening study. CR was bonded to grit blasted steel with four adhesive systems: In-house adhesives A and C, CH 205 primer coated with CH 250, and the three coat system of CH AP134/CH 205/CH 220. The bonded parts were tested for primary adhesion and exposed to 5% salt spray for three and seven days.

The two in-house adhesives failed after three days in the salt spray. The other two adhesive systems retained a reasonable amount of adhesion after seven days. These two adhesive systems were chosen for further study in cathodic disbonding tests.

Before moving on to cathodic disbonding tests, we ran a study to determine if salt spray resistance would increase if we used calcium-modified microcrystalline zinc phosphated steel instead of grit blasted steel. We bonded CR to both surfaces using four adhesive systems: In-house adhesive A, a two coat system of CH 205 primer coated with CH 233, CH 250 one-coat adhesive, and the CH AP134/CH 205/CH 220 adhesive system. The bonded parts were exposed to 5% salt spray for three days. Figure 4 shows the results.

The zinc phosphated steel gave better performance with all the adhesives except the three coat system. It appears that the choice of substrate for better environmental resistance depends on the adhesive system being used.

Previously, we used salt spray tests to screen adhesive systems before exposing them to cathodic disbonding tests. In our next study, we ran 7-day salt spray and 28-day cathodic disbonding tests on five adhesive systems to determine if salt spray testing really is a good screening test for cathodic disbonding. The adhesive systems were: In-house adhesives A, B and C, the two coat system of CH 205 primer coated with CH 250, and the CH AP134/CH 205/CH 220 three coat system. Figure 5 shows results.

The adhesives which failed in the salt spray test (Adhesive A, B, and C) also failed in the cathodic disbonding tests. The systems which did well in the salt spray test (CH 205/CH 250 and CH AP134/CH 205/CH 220) also did well in the cathodic disbonding tests. The results of this study show that salt spray testing does have some correlation with cathodic disbonding tests.

Our tests thus far found two adhesive systems which are resistant to cathodic disbonding: CH 205/CH 250 and CH AP134/CH 205/CH 220. Our next study looked at the effect of prebake on the cathodic disbonding resistance of these adhesives. Prebake is the period of time the coated parts reside in the mold before rubber is introduced for bonding.

We bonded CR to grit blasted steel with prebakes of 0 and 5 minutes at 307 [degrees] F. The bonded parts were then subjected to 28-day cathodic disbonding tests. The control was in-house Adhesive A. Figure 6 shows results.

Resistance to cathodic disbonding dropped with increasing prebake. The CH AP134/CH 205/CH 220 system retained more adhesion with prebake. Rubber retention for both systems was still over 60%.

Finally, we ran a study to compare our best adhesive systems in cathodic disbonding resistance. In addition to CH 205/CH 250 and CH AP134/CH 205/CH 220, we evaluated CH 205/CH 252. We bonded CR to grit blasted steel and ran prebakes of 0 and 5 minutes at 307 [degrees] F. Figure 7 shows 28-day cathodic disbonding results.

CH AP134/CH 205/CH 220 retained cathodic disbonding resistance better than the other adhesive systems. With prebake, CH 205/CH 250 fell from 90% rubber retention to 60% rubber retention. CH 205/CH 252 stayed constant at 60% rubber retention.

Conclusion

While cathodically protected rubber to metal bonded components present a very unusual and difficult environment for adhesives to resist, the data clearly shows that the proper selection of adhesive will provide a durable, environmentally sound bond. Several commercial rubber to metal adhesives were shown to resist cathodic degradation.

In addition, the data also indicates that some adhesives, while very tolerant of many severe environments, are not a proper choice when used to adhere cathodically protected rubber to metal articles, as early failure can be induced in the laboratory.

The data also points to the use of the salt spray test as a screening test for cathodic resistance, since excellent correlation has been observed.

In summary, laboratory tests show that cathodic protection of rubber to metal assemblies creates a very severe environment that shows up major differences in adhesives not normally observed in other traditional environmental tests. Excellent primary adhesion can be rapidly destroyed in this environment.

Several adhesives passed the testing. These include CH 205/CH 250, CH 205/CH 252, and CH AP134/CH 205/CH 220. Several in-house control adhesives and competitive adhesives failed to resist this environment. The three coat system gives better performance with prebake, but it introduces more expense because of the extra coat of primer required. The two coat systems perform adequately at shorter prebakes and will be less expensive to apply. The particular adhesive system chosen must be based on the customer's own evaluation.

Significant correlation of seven day neutral salt spray test data to the 28-day cathodic resistance test was noted.

There appears to be several acceptable adhesive approaches to bonding rubber to metal assemblies for those applications that require cathodic protection.

PHOTO : Figure 1 - test apparatus schematic

PHOTO : Figure 2 - CR to grit blasted steel ASTMD 429 Method B

PHOTO : Figure 3 - CR to grit blasted steel 5% salt fog exposure testing

PHOTO : Figure 4 -3 day salt spray resistance of CR

PHOTO : Figure 5 - CR to grit blasted steel ASTM D429 method B

PHOTO : Figure 6 - CR to grit blasted steel 28 day cathodic disbonding tests

PHOTO : Figure 7 - CR to grit blasted steel 28 day cathodic disbonding tests

References

[1]Stevenson, A., On the durability of rubber/metal bonds in sea water, The International Adhesion Conference, 1984. [2]Boerio, F. James, Adhesive bonding of neoprene to metals in sonar devices, Report prepared for the Office of Naval Research, August 31, 1989. [3]Alstadt, D.M. and O'Brien, H.C., Methods of minimizing galvanic decay of polymer-metal bonds, 1967. [4]Rushing, R., Evaluation of adhesive systems for bonding 5109S neoprene-to-steel and aluminum substrates, final report submitted to the Office of Naval Research, submittedd by Texas Research Institute. [5]Thornton, J.S., Montgomery, R.E., Cartier, J.F., Cathodic delamination of rubber to metal bonds - cause and control, Texas Research Institute, Austin, Texas 78733. [6]Guy, A., Disbondment of rubbers from cathodically protected steel. [7]Ting, R.Y., Polymer News 10, 197 (1985). [8]Boerio, F.J., Hudak, S.J., Miller, M.A., S.G. Hong, J. Adhesion 23, 99 (1987). [9]Boerio, F.J., and S.G. Hong, J. Adhesion, in press, 1989.