Interfacial adhesion of elastomers to HNBR.

It became evident in the mid-1970s that the workhorse oil-resistant elastomer, nitrile rubber, was approaching its heat resistant limits in many automotive and industrial applications. It was during this period that the use of high performance elastomers, such as CSM, AEM and ECO, increased to meet these more demanding conditions (ref. 1). As we enter the 1990s and demands continue to increase, many of the high performance elastomers are themselves being replaced by more expensive high performance elastomers. As raw material costs skyrocket, part manufacturers are looking to new production methods to minimize the impact of these higher raw materials costs. The use of thin veneers of fluoroelastomers is one such method commonly used.

Hydrogenated nitrile rubber (HNBR) is a family of high performance elastomers. These rubbers meet many of the most stringent application requirements. Currently, many parts manufacturers are reluctant to use HNBR because of its high cost relative to many of the traditional high performance rubbers. However, HNBR is considerably more cost effective than fluoroelastomers in many applications. In order to minimize the cost premium while maintaining the performance advantages of HNBR, veneers of HNBR with other elastomers might be used.

In the construction of rubber articles, such as belts and hoses, it is important to form good interfacial adhesive bonds between the layers of rubber. If HNBR is to be considered for use in articles manufactured in this manner, it is important to show the ability of HNBR to develop good interfacial adhesion when covulcanized to the other elastomers. A number of test methods have been developed to determine the interfacial adhesion of rubber-rubber interfaces (refs. 2-4). Unfortunately, these test methods were not suitable for these studies with HNBR and a new modified test method was developed.

This new modified test method will be outlined in this article. In addition, the level of interfacial adhesion observed between HNBR covulcanized to other elastomers will be presented. Finally, examples of potential applications for this technology will be discussed.

Experimental

Criteria for development of adhesion test

Some of the test intended to measure interfacial adhesion of elastomers are based on a 180$degrees] peel test (refs. 3 and 4). Because of the type of adhesion expected and the fundamental characteristics of the HNBR, it was felt that a modified peel test was required. In order to effectively measure the interfacial adhesion of an elastomer to another elastomer using a peel test, the following points must be considered:

* When tested, the samples must separate at the interface between the elastomers. If the samples separates by tearing within one of the vulcanizates, the only conclusion that can be drawn about the adhesion is that it is greater than the force measured. Separation is commonly directed to the interface by using an insert to separate portions of the vulcanizates.

* The insert used between the covulcanized elastomers must be effectively designed to restrict the peel to the interface while minimizing the edge effects of the insert.

* The insert material used must not adhere to both of the elastomers being tested. However, the insert should have sufficient tack to the polymers to restrict sliding of the insert between the rubber sheets during pressure molding of the test sample.

* The adhesive surface between the elastomers to be covulcanized must be kept clean.

* The rubber sheets should be backed with a material which gives sufficient adhesion to the elastomers and adequate mechanical strength to the samples. This provides the test with the capability of differentiating between samples with high interfacial adhesion.

* The samples should be cured in a constant pressure mold to eliminate any test variability caused by pressure promoted adhesion. Use of low pressure will also better simulate the actual curing conditions found in the construction of hand-built hose or the fabrication of co-extruded hose. Furthermore, adhesion is more difficult to obtain under these conditions compared to compression or drum curing of hoses or belts.

Test method

The following steps summarize the modified 180$degrees] peel test developed to measure the interfacial adhesion of covulcanized elastomers.

Preparation of test specimens

* Prepare the insert as shown in figure 1 by dieing out four 1 cm wide by 10 cm long slots in the 15 cm square rubberized nylon carcass fabric.

* Prepare two sheets of backing material (15 cm square) using heavy woven cotton.

* Prepare two sample top tabs from heavy gauge plastic of dimensions 15 cm long by 4 cm wide.

* The two elastomer stocks required for the adhesion test must be freshly remilled to a thickness of 0.25+/-0.1 cm.

* Allow the milled rubber stock to rest on a flat surface until cool and all shrinkage has ceased. Vulcanization of the test samples must start within four hours of remilling.

* Construct the test sample as shown in figure 1.

Vulcanization of test specimens

* Preheat the press and constant pressure mold to the desired cure temperature.

* Place the test sample in the cavity of the constant pressure mold (15 cm square).

* Cure the test specimen for the required time. Maintain constant pressure throughout the curing.

* Once the proper cure time has been achieved, remove the sample from the mold and allow to air cool.

* Cut the cured test sample in order to produce four identical test pieces.

Measuring adhesion of test specimens

* Using a tensile test machine, pull two pieces separating the component elastomers at an angle of 180$degrees] using a head speed of 50 mm/minute at ambient temperature.

* Determine the interfacial adhesion as given by the median force peak of the two force traces. If there is only one force peak in each trace, its value will be considered the median. Report results as kiloNewtons force per meter of adhesive surface width. See figure 2 for examples of measuring the force readings from the strip chart printout of the test machine.

* Observe the nature of the adhesion failure and report as:

- Interfacial peel - rubber separates from the other rubber at the interface. This gives a true measure of the interfacial adhesion between the two elastomers.

- Stock tear - rubber tears or chunks from one of the rubber samples. This gives a measure of the tear strength of the weaker elastomer being tested.

- Backing failure - rubber separates from the backing or the backing tears. This gives a measure of the weaker polymer adhesion to the backing or the tear strength of the backing itself.

Reproducibility

Because of the various methods of sample failure noted above during testing and the large variety of elastomers tested, it would have been very difficult to calculate a precise and meaningful test reproducibility. In lieu of a detailed evaluation, the test precision was estimated using the differences in duplicate testing values measured. From these differences and assuming a normal sample distribution, the 95% confidence level has been estimated to be approximately 14% of the average measured force.

Results

A number of elastomers commonly used in hose and belting applications were covulcanized to HNBR to assess the interfacial adhesion. In addition to assessing a wide variety of elastomers, these elastomers were also tested using a variety of cure systems. Table 1 summarizes the pairs of elastomers on which adhesion tests were conducted. (TABULAR DATA OMITTED)

As shown in table 1, two families of cure systems were used with the HNBR (one peroxide and three sulfur donor cure systems) and the HNBR cure system selection was based on the compatibility with the cure system of the other elastomer. The HNBR compound formulations used, their physical properties and curing characteristics are shown in tables 2 and 3 respectively. The ODR rheograms for the HNBR compounds are shown in figure 3.

The compound formulations for the other elastomers are available. The physical properties of these materials and their interfacial adhesion to HNBR will be discussed in the subsections headed "Elastomer adhesion."

Pressure affects the interfacial adhesion of HNBR to the other elastomers. A study was done to assess the extent of the effect of pressure on adhesion. This work is summarized in the section headed effect of pressure on interfacial adhesion. During the testing phase of this study, it was determined that a particular chemical added to the HNBR compound enhanced interfacial adhesion. Further work was conducted to confirm the extent of this effect. These results are discussed in the section headed "Effect of adhesion promoter XX."

Effect of pressure on interfacial adhesion

The effect of pressure on the interfacial adhesion of covulcanized elastomers was assessed for the pressure range of 0.45 MPa to 9.0 MPa (50 psig to 1,300 psig). Test samples construced from HNBR (S1) covulcanized to CSM (S2) were used for this study. The adhesive force observed is shown versus mold pressure in figure 4. It is apparent that pressures between 0.45 and 9.0 MPa only have a small effect on the interfacial adhesion of HNBR to CSM. Because the effect was small, all further work was done using molding pressures of 0.45 and 1.48 MPa (50 psig and 200 psig). No significant differences were observed between the adhesive strength of the test samples when molded at these two pressures. Possibly, the effect of pressure would be more pronounced if the test samples had been held under varying pressures for a time prior to curing. This intimate contact under pressure would allow the polymer chain ends to more effectively diffuse across the rubber-rubber interface, thereby enhancing interfacial adhesion after curing.

Elastomer adhesion

CSM adhesion

Three sulfur cured CSM compounds were covulcanized to HNBR (S1) and HNBR (S2). The CSM formulations were selected for the various formulations outlined in the references (ref. 5). The physical properties and the interfacial adhesion to the HNBR compounds are show in table 4. The ODR rheograms of the CSM compound are shown in figure 5. Excellent adhesion was obtained between all of the HNBR and the CSM compounds. All of the measured force values exceeded 25 kN/m except for the force required to separate CSM (S2) from HNBR (S1), which was about 16 kN/m. Comparing the cure curves of the HNBR and the CSM in figures 3 and 5 respectively, it is apparent that any differences in cure rate could not have caused the differences noted in the interfacial adhesion of CSM to HNBR. The decrease in adhesive strength could be related to incompatibility of these specific sulfur cure systems.

          Table 2 - HNBR compound formulations used
                       for adhesion study
Compound name        HNBR         HNBR      HNBR      HNBR
                      (P)          (S1)      (S2)      (S3)
Tornac A3835         100.0          -         -         -
Tornac C3830           -          100.0     100.0     100.0
N550 black            40.0         50.0      50.0      50.0
Magnesium oxide        5.0          3.0       3.0       3.0
Zinc oxide             5.0          2.0       3.0       3.0
Stearic acid           1.0          1.0       1.0       1.0
Vulkanox OCD           1.1          -         -         -
Vulkanox ZMB-2         0.4          -         -         -
TP-95                  -           10.0      10.0      10.0
Ricon 153-D            6.5          -         -         -
Varox DBPH-50         10.0          -         -         -
Spider sulfur          -            0.4       0.5       0.4
Vulkacit DM            -            4.0       1.5       -
Vulkacit thiuram       -            -         1.5       -
TETD                   -            -         -         1.3
Vulkaciat CZ           -            -         -         1.3
Vulkacit D             -            1.3       -         -
Rhenocure S/G          -            3.8       -         -

Note: All compounds mill mixed using the following procedure: 1) band rubber on cool mill (40[degree] C); 2) at 2 minutes, add all ingredients except curatives; 3) makes 3 - 4 cuts; 4) at 9 minutes add curatives; 5) at 11 minutes, remove from mill; 6) refine. Table 3 - physical properties of HNBR compounds

              used for adhesion study
Compound name               HNBR  HNBR      HNBR      HNBR
                             (P)   (S1)      (S2)      (S3)

Cure characteristics Monsanto ODR (model R100) 1[degree] arc, tested at curing temp.

ts2 (min)                    1.2   12.0       4.0       3.8
t'90 (min)                   7.2   24.8      10.4       6.8
ML (dN.m)                    7.2    3.0       3.2       3.6
MH (dN.m)                   60.8   43.4      42.0      38.4

Physical properties

Modulus at 100% elong. (MPa) 6.7    3.4       3.4       2.4
Modulus at 200% elong. (MPa)19.1    7.8       8.4       5.4
Modulus at 300% elong. (MPa)  -    12.1      13.4       8.4
Tensile strength (MPa)      25.8   21.4      22.4      18.8
Ultimate elongatin (%)       250    530       510       690
Hardness (Shore A)            77     73        73        70

Note: All test sheets were cured for 30 minutes at: - 160 [degrees] C for sulfur cured HNBR; - 180[degrees] C for peroxide cured HNBR

AEM adhesion

Two diamine cured AEM compounds were covulcanized to HNBR (S1) and HNBR (S2). These AEM formulations were obtained directly from the elastomer supplier as typical for a hose cover and a hose tube.

Poor adhesion was obtained between all of the HNBR compounds and the AEM compounds. All samples peeled at the interface. Since the AEM cure rates are similar to HNBR (S2), cure rate can be eliminated as the cause of the poor adhesion. Consequently, the poor adhesion must be caused either by the incompatibility of the cure systems or the incompatibility of the polymers.

CO, ECO and ECO terpolymer adhesion

One diamine cured ECO compound, three ETU cured compounds (one each of CO, ECO and ECO terpolymer), one sulfur cured ECO terpolymer compound and one peroxide cured ECO compound were covulcanized to all the various HNBR compounds. The ECO formulations were selected to represent the various cure formulations possible, as outlined in reference 6.

The ECO terpolymer shows excellent adhesion to HNBR when cured with sulfur (ECO(S)) or ethylene thiourea (ECO(E3)). CO and ECO show good adhesion to all three sulfur cured HNBR compounds when cured with ethylene thiourea (ECO(E2) and ECO(E1) respectively). However, ECO when diamine cured (ECO(D)), shows moderate-to-poor adhesion to HNBR (S1) and HNBR(S2) but good adhesion to HNBR (S3). The scorch time of HNBR(S1) is much longer than that of ECO(D). Consequently, the ECO would have been mostly cured before HNBR(S1) started to cure. These differences in curing characteristics could have caused the poor adhesion noted in this case. Surprisingly, HNBR (S2) and HNBR (S3) have similar curing ingredients and curing characteristics with the exception of CBS being substituted for MBTS in HNBR (S3). Since the curing intermediates for both CBS and MBTS are similar, the large differences noted in interfacial adhesion were not expected.

ECO terpolymer, when peroxide cured with (ECO(P)) shows good interfacial adhesion to HNBR(P).

NBR adhesion

One sulfur cured and one peroxide cured compound of 34% bound acrylonitrile NBR were covulcanized to HNBR (S2) and HNBR(P) respectively. The NBR formulations used similar curatives as the HNBR formulations and the levels of these chemicals used were adjusted to give similar cure rates.

The 34% bound acrylonitrile NBR shows relatively poor interfacial adhesion to HNBR when sulfur cured (as NBR (S)). Peroxide cured (NBR(P)) shows relatively good interfacial adhesion to HNBR. One would expect that NBR should adhere well to HNBR independent of cure system because of good solubility of one polymer in the other. Surprisingly, the adhesion was not as good as expected. However, chemical adhesion promoters are available to enhance the interfacial adhesion of NBR to HNBR as discussed in a later section of this article.

Halogenated butyl adhesion

One peroxide cured brominated butyl compound and two sulfur cured halogenated butyl compounds (one each of brominated and chlorinated butyl) were covulcanized to all of the respective HNBR compounds. These formulations were developed to provide premier heat resistant sulfur and peroxide cured halogenated butyl compounds.

Halogenated butyl rubber, when sulfur cured (BIIR(S)) and (CIIR(S)), show moderate-to-poor interfacial adhesion to sulfur cured HNBR. Brominated butyl, when peroxide cured (BIIR(P)), shows moderate interfacial adhesion to peroxide cured HNBR. The peroxide cured brominated butyl was cured for 30 minutes at 180 [degrees] [C] and was overcured during preparation of the adhesion sample piece. Unfortunately, this was necessary to allow full curing of HNBR(P). Better interfacial adhesion may be possible if the compounds are optimized to better match the cure times of the brominated butyl and the HNBR compounds.

EPDM adhesion

One sulfur cured and one peroxide cured EPDM were covulcanized to the sulfur cured HNBR compounds and HNBR(P) respectively. The sulfur cured formulation was selected as a representative formulation rom reference 7. The peroxide cured formulation is considered by Dunn et al. (ref. 8) as a premier heat resistant EPDM formulation.

EPDM shows poor adhesion to HNBR independent of the cure systems used in the EPDM or in the HNBR. The poor adhesion with the sulfur cured EPDM compound (EPDM(S)) may be attributable to the high levels of plasticizer in the formulation. The peroxide cured EPDM formulation used (EPDM(PP)) was very difficult to process and had a very rough surface which may have contributed to the poor adhesion to the HNBR.

Fluoroelastomer adhesion

One bisphenol-cured FKM compound and one peroxide-cured FKM compound were covulcanized to HNBR(P). These compounds were felt to be typical for bisphenol and peroxide cured fluoroelastomers.

The bisphenol-cured FKM (FKM(B)) shows relatively good interfacial adhesion to HNBR(P). However, since the tensile strength of this material is very low, the fluoroelastomer tore when a force was applied. The peroxide-cured FKM compound (FKM(P)) shows poor interfacial adhesion to the peroxide-cured HNBR. However, chemical adhesion promoters are available to enhance the adhesion of fluoroelastomers to HNBR as discussed in a later section of this article.

Effect of adhesion promoter XX

Various adhesion promoters were tested to evaluate which one could enhance interfacial adhesion to HNBR. One such chemical was found and its effect on the interfacial adhesion of various elastomers covulcanized to HNBR is shown in table 5. The use of this chemical to promote adhesion of other elastomers to HNBR is currently being patented.

The effect of Chemical XX on the physical properties of HNBR is also shown in table 5 and the ODR rheograms for HNBR, both with and without Chemical XX, are shown in figure 6. The adhesion promoter slightly decreased the state of cure, as indicated by the decrease in modulus and decrease in elongation. The adhesion promoter also acted as a cure retarder, as shown by the rheograms in figure 6.

From table 5, it is clear that Chemical XX significantly enhances the interfacial adhesion of HNBR when covulcanized: 1) to itself (peroxide cured); 2) to NBR (peroxide and sulfur cured) and, 3) to fluoroelastomers (bisphenol and peroxide cured.) In fact, Chemical XX can, in some cases, double the force required to overcome the interfacial adhesion of the polymers. Conversely, Chemical XX shows no adhesion enhancement in peroxide cured systems with EPDM rubber and brominated butyl rubber. No conclusions can be drawn on the adhesion of epichlorohydrin to HNBR because the failure mode was either stock tear or backing failure.

Potential applications for thin-layered HNBR covulcanized to other elastomers

HNBR has an excellent balance of heat and oil-resistance, as indicated by its position in the SAE J200/ASTM D 2000 classification. There are many applications requiring HNBR because the rubber article is being exposed to aggressive environments. In some of these applications, only certain areas of the rubber article are exposed to these stringent conditions and part costs could be minimized by using a multilayered construction of HBNR with other elastomers. By using this type of construction, HNBR can be used to protect only the critical areas of the rubber article which are exposed to more aggressive environments, minimizing the cost premium required with HNBR. Potential applications are:

* Automotive or industrial hoses used to transport hot fluids or oils in hot and/or aggressive environments (e.g. power steering hose, transmission cooler hose, hydraulic hose). These hoses could be constructed with an HNBR veneer inside a co-extruded tube with a cover made from an elastomer with lower fuel and oil-resistance, such as CSM or XIIR. Commercially, some automotive power-steering hoses use a construction similar to this utilizing an HNBR tube and a CSM cover.

* Automotive or industrial hoses used to transport fluids or oils in an aggressive environment, where the heat resistance requirements are more stringent either inside, or exterior to, the hose. This type of hose could be constructed with HNBR veneer, with a co-extruded laminate of a lower-heat-resistant elastomer, such as ECO or NBR.

* Automotive hose used for the transport of fuels, possibly flexible fuels (fuel blends with alcohol). This hose must have good fuel-resistance and low fuel permeability. This type of hose could be constructed from a fluoroelastomer veneer in the tube with an HNBR tube and cover. The fluoroelastomer will give the excellent fuel impermeability while the HNBR will reduce costs and also provide improved mechanical strength and reasonable fuel-resistance. Work has been done (refs. 9-11) to develop a laminated hose using fluoroelastomer and NBR. These patents outline how to enhance the interfacial adhesion by addition of compounding ingredients into the fluoroelastomer compound. These principles should also be applicable to hoses constructed of fluoroelastomer and HNBR.

* Conveyor belting requiring good oil-resistance and/or abrasion-resistance and/or good heat-resistance. These belts could be constructed with an NBR center core with an HNBR laminated cover.

* Industrial drive belts requiring a rib compound with good abrasion-resistance and/or heat and oil-resistance could be constructed with an HNBR rib compound with a cover made from a less resistant, lower cost elastomer compound such as CSM.

Summary

A modified 180 [degrees] peel test was developed to effectively measure the differences in interfacial adhesion when HNBR is covulcanized to another elastomer. The test shows good reproducibility and is capable of measuring interfacial adhesion forces up to 25 kN/m.

HNBR shows good-to-excellent adhesion when covulcanized to: CSM (all types of curing systems); ECO (sulfur, ethylene thiourea and peroxide curing systems); NBR (peroxide curing system); and fluoroelastomer (bisphenol curing system)

HNBR shows moderate adhesion when covulcanized to: ECO (diamine curing system); NBR (sulfur curing system); XIIR (sulfur and peroxide curing systems); and fluoroelastomer (peroxide curing system).

HNBR shows poor adhesion when covulcanized to: curing systems). However, the poor adhesion to the EPDM can be attributed to poor choices for EPDM compounds. If the EPDM compound was optimized, it is possible that good adhesion to HNBR could be obtained.

A compounding ingredient was found which could effectively improve the interfacial adhesion of HNBR when covulcanized to itself, to NBR and to fluoroelastomers. The use of this ingredient is currently being patented.

There are a number of potential applications which could utilize this technology. Of primary interest would be applications in hose and belting. Use of this technology with HNBR would allow the part manufacturer to maintain the superior performance provided by HNBR while minimizing the material cost premium required.

References

1. R.J. Weir, G.C. Blackshaw, "The emergency of hnbr as an important high performance elastomer," International Rubber Conference, Sydney, Australia, 1988.

2. G.S. Fielding-Russel, R.L. Rongone, "Fatiguing of rubber-rubber interfaces," Rubber Chem. Tech. 56, 4, (1983).

3. P. Loha, A.K. Bhowmick, "Modification of the peel test for testing of rubber-rubber joints," Polymer Testing 7 (1987).

4. U.S. patent 4,645,793, issued to Polysar Ltd., Feb. 1987.

5. Manufacturer's literature for Hypalon chlorinated polyethylene rubber.

6. Manufacturer's literature for Hydrin epichlorohydrin elastomers.

7. Polysar Rubber Corp. literature for EPDM.

8. J.R. Dunn, D. Keller, J. Patterson, "EPDM compounding for high temperature ageing requirements in automotive hose," presented at the 132nd Rubber Div. Meeting, 1987.

9. U.S. patent 4,457,799 issued to Polysar Ltd., July 1984.

10. U.S. patent 4,522,852 issued to Polysar June 1985.

11. U.S. patent 4,806,351, issued to Toyoda Gosei Co. Ltd, September 1987.

Acknowledgements

"Interfacial adhesion of elastomers to HNBR" is based on paper given at the October 1991 meeting of the ACS Rubber Division.

"Isobutylene-based polymers in tires" is based on paper given at the October 1990 meeting of the ACS Rubber Division.

Advances in EPDM flame retardancy, testing" is based on paper given at the October 1991 meeting of the ACS Rubber Division.

"Styrenic block copolymers as bitumen modifiers for improved roofing sheet" is based on paper given at the October 1991 meeting of the ACS Rubber Division.