Rubber to metal bonding.

Brass plating is still widely used in the tire industry for bonding NR skim stocks to steel cords. The widely studied brass-rubber adhesion mechanism has served as a model for the development of two new classes of steel tire cord coatings with improved properties. One class is that of a highly corrosion-resistant metallic coating (refs. 1-3). Its composition is largely Zn1.5% Co, but it has a nickel-rich thinner topcoating (Ni2O%Zn). Thus the overall composition is approximately 75% Zn, 22% Ni and the balance Co. The study of this alloy has initiated a revision of the model that has been widely accepted for rubber-brass bonding (refs. 4-5). A new technique that has proven useful for the study of rubber-metal interfaces is Time-of-Flight Secondary Ion Mass Spectrometry (TOF-SIMS). An updated adhesion mechanism will be presented.

The second type of material proposed for bonding steel to rubber is a thin film of polyacetylene deposited from a plasma of argon and acetylene (refs. 6 and 7). Since such PP-[C.sub.2][H.sub.2] films have a structure which resembles that of NR, they are chemically crosslinked to rubber during vulcanization (refs. 4 and 5). Some results, and a design for coating wires in a continuous process are presented and discussed.

Experimental

Materials

The Zn/Ni/Co coatings were deposited on 1.4 mm wires which were then drawn to 0.25 nun filaments and were corded to 5 x 0.25 mm steel cords in the pilot line of Pirelli (ref. 8). Laboratory experiments were carried out by depositing the alloys on 25 x 4 [cm.sup.2] cold-rolled steel (CRS) strips of 0.5 mm thickness (ref. 1). The PP-[C.sub.2][H.sub.2] films were deposited on type 1010 CRS coupons of 5 x 5 [cm.sup.2] and 0.5 mm thickness (refs. 6 and 7). The rubber compounds used for measuring the performance of the ZnNiCo coatings were sulfenamide-cured NR compounds with variable levels of sulfur and cobalt-containing adhesion promoter. For the testing of the PP-[C.sub.2][H.sub.2] coatings, a commercial test compound was used.

Procedures and test methods

Adhesion of the tire cords was determined using the standard ASTM D 2229-73 cord pull-out test with 12.5 mm embedment length. Aged adhesion was determined by exposing the cured blocks to an environment of 90% relative humidity at 65 [degrees] C for eight days prior to adhesion testing.

The adhesion of the PP-[C.sub.2][H.sub.2] films was determined using two 8 x 8 [mm.sup.2] test samples with a 4 x 20 mm handle in a lap-shear arrangement. A thin layer of rubber was cured between two of such samples which were pulled apart in a tensile tester (ref. 9).

Corrosion of the cords was determined by immersing them in an aerated 4 wt.-% NaCl solution and monitoring the time for the formation of white rust (zinc corrosion products) and red rust (iron corrosion products). Also, potentiodynamic scans of the cords were recorded in a range of [+ or -] 100 mV around the open circuit potential.

From these Tafel plots the corrosion potential and the corrosion current density were calculated.

Electroplating of the alloys was done from acid baths which were specifically developed for ZnCo and NiZn deposition (ref. 2). They contained the metal sulfates and an excess of [Na.sub.2][SO.sub.4] only. The temperature of both baths was 60 [degrees] C and the pH was 2-2.5. Drawing of the plated 1.4 mm wires was done using drawing conditions and lubricants used for brass-coated cords.

Deposition of PP-[C.sub.2][H.sub.2] films was done in a laboratory plasma reactor of 15 cm diameter inductively coupled to an RF power supply (13.56 Mhz) (refs. 6 and 7). The argon was introduced in the RF coil region, the monomer [C.sub.2][H.sub.2] was introduced downstream from the argon introduction (remote plasma). The CRS substrates were first cleaned in an argon plasma of 20 W for 10 min. Deposition was typically done at 125 Pa pressure, 50 W power and flow rates of 20/25 sccm for argon/[C.sub.2][H.sub.2]. Around 100 nm PP-[C.sub.2][H.sub.2] film was obtained in 30 min.

Sulfidation of the alloys and PP-[C.sub.2][H.sub.2] films was studied using the squalene method (ref. 10). The samples were immersed in squalene mixtures containing rubber chemicals in the same ratios as in the actual rubber compounds. The mixture, through which nitrogen gas was bubbled, was held at 150 [degrees] C and samples were withdrawn at regular intervals. The surface and bulk composition was investigated with a range of analytical techniques, e.g. RAIR, XPS, AES, ellipsometry and TOFSIMS.

Instrumentation

The following instruments were used:

* Electroplating - a high speed rotating circular cathode was used in the laboratory depositions of the ZnNiCo alloy and a one-wire pilot line for depositions on wires;

* Plasma reactor - custom built tubular RF reactor (13.56 MHz);

* TOFSIMS - Kratos prism instrument with a 25 [kV.sup.69][Ga.sup.+] primary source; used for surface analysis of samples treated in the squalene mixtures;

* XRD - Siemens diffractometer no. 11 with CuK[alpha] for phase analysis of the coatings on wires;

* AES - Perkin Elmer 590A scanning auger microprobe using 5 [Angstrom]/min sputtering rate; used for depth profiling of alloy and PP-[C.sub.2][H.sub.2] coatings on steel;

* SEM - Jeol JX840 with a Tracor Northern EDX system used for surface morphology studies of the alloy coatings;

* AAS - Perkin Elmer 460 for analysis of the alloy coatings after dissolution in acid;

* RAIR (Reflection-Absorption IR) - Perkin Elmer model 1800 FTIR instrument; 100 scans at 4 [cm.sup.-1] resolution; used for structural characterization of thin PP-[C.sub.2][H.sub.2] films on polished steel;

* XPS - Perkin Elmer model 5300 instrument with MgK[alpha] X-ray source at 300 W; take-off angle was 45 [degrees]; used for surface analysis of PP-[C.sub.2][H.sub.2] films;

* Ellipsometry - Rudolph Research model 436 ellipsometer for thickness measurements Of PP-[C.sub.2][H.sub.2] films;

* Corrosion - an EG&G Parc 342-1 corrosion measurement system with calomel and graphite as the reference and counter electrodes.

Results and discussion

Adhesion mechanism

In figure 1 a modified model is shown of the interface between rubber and metals such as brass, zinc and the new ZnNiCo alloy. This model is based on TOFSIMS, XPS and RAIR analyses of the surface of samples treated in the squalene mixtures and also on analyses of the rubber-metal interface. New is the role that the accelerator plays in the mechanism. TOFSIMS showed that in the first stages of the vulcanization, the accelerator is adsorbed on the metal surface. A bond of the type Me-S-X is formed at the metal surface, similar to the thiomercaptide that ZnO forms in the rubber from soluble zinc and X radicals where X is the [C.sub.6][H.sub.4]-NS-CS* sulfenamide accelerator fragment. Stearic acid plays an important role, as it dissolves metal oxides, so the metal can react more readily with the X fragments. The role of cobalt in this mechanism is to depolarize metal oxides. For instance, in the case of the Zn/Ni/Co alloy, no adhesion is obtained if the rubber does not contain small amounts of cobalt. This cobalt activates the passive NiO present at the metal surface. This passivity is removed by Co and Ni then forms the same type of Me-S-X bonds as copper and Zn do.

In the intermediate stages of the cure, [S.sub.8] insertion into the Me-S bond occurs, resulting in complexes of the type Me-[S.sub.y]-X, similar to the zinc and cobalt perthiomercaptide in the rubber itself In the latter stages of the cure, these complexes decompose into metal sulfides. In this model, the initial reaction is thus the formation of metal-organic complexes, the final result is the formation of an inorganic sulfide. In this mechanism rubber molecules become locked up in the growing sulfide film. Whether metals will bond to rubber will in this mechanism thus depend on the stability of the Me-[S.sub.y]-X complex and its rate of decomposition into sulfides. The actual adhesion mechanism is still an entanglement of rubber molecules and sulfide films.

Performance of Zn/Ni/Co alloy system

In tables 1 and 2 performance data in ten-ns of adhesion and corrosion are given for this new alloy system. The results demonstrate that the coating system performs as well as a standard brass coating to most compounds, except for NR compounds without cobalt, where the Zn/Ni/Co system does not bond for reasons described above. The aged adhesion in humidity is clearly improved as compared to brass.

  Table 1 - adhesion of Zn/Ni/Co cord to NR
            (refs.1-3)(*) 5 x 0.25 cord

Cord           Initial   4 d aged   8 d aged(**)

NiZn/ZnCo      517/100     475/100       383/100
CuZn           495/100      330/50        340/25

(*) Standard test compound with cobalt decanoate
(**) Aging in 90% r.h., 65 [degrees] C

Table 2 - corrosion of Zn/Ni/Co-coated cords
  (ref.1) immersed in aerated 4 wt-% NaCl

Wires           White rust         Red rust
                     (min)            (min)

Zn                      20            1,815
ZnCo                    45            4,320
NiZn                 1,380            1,815
NiZn/ZnCo               60            3,630
CuZn                     -               20
Steel wire               -               90

Cords
Zn                      10              240
ZnCo                    20            1,152
NiZn                     -                5
NiZn/ZnCo            1,140            1,720
CuZn                     -               20

The corrosion performance is the area where the most spectacular advantages of this system are seen (table 2). These results demonstrate why a single layer coating of either ZnCo or NiZn does not work as well as the double layer. The double layer combines the advantages of the passivating properties of Ni, which forms passive NiO, with the cathodic protection properties of the ZnCo base layer. These properties are better than those of pure zinc. Thus, the Zn/Ni/Co coating affords better corrosion protection to the steel than Zn, ZnNi or ZnCo alone and considerably better than brass, which actually accelerates steel corrosion.

Performance of thin PP-[C.sub.2][H.sub.2] films

A summary of results with this system is given in figure 2 for small steel coupons cured in contact with a high-sulfur, high-Co tire cord test compound. The control results were obtained with polished 65/35 solid brass samples. It is seen that the PP-[C.sub.2][H.sub.2] films perform as well as pure brass in initial and steam aged adhesion, but that they lag behind in salt aging resistance (ref. 12).

Important aspects of optimizing the plasma process for deposition of this type of films on wires in a reproducible way are the methods of cleaning of the wires, the stability of the films, the homogeneity that can be achieved, the rate of deposition, etc.

A reactor for deposition of plasma films is currently being built in our laboratories. The deposition will be carried out completely under computer control. The type of discharge in this reactor will be, by DC rather than RF, as DC has several important advantages over RF (ref. 11).

Conclusions

The application of TOFSIMS to the study of metal-rubber interfaces has provided new information on the mechanism of bonding NR to metals.

An updated mechanism has been presented; metal-accelerator complexes are formed, which are converted to metal sulfides at a larger stage of the cure; the result is an interlocked mixture of metal sulfides and rubber.

A highly corrosive-resistant new alloy system Zn/Ni/Co has been developed which has both passivation and cathodic protection properties.

Thin films of plasma-polymerized acetylene (PP-[C.sub.2][H.sub.2]) deposited onto steel can provide outstanding adhesion to NR compounds, comparable to that of pure brass.

References

[1.] J Giridhar and W.J. van Ooij, Surf. Coat. Technol, 52, 17 (1992); 53, 35 (1992); 53, 234 (1992). [2.] J. Giridhar, W.J. van Ooij and M. Cipparrone, Rubbercon '93, New Delhi, February 8-10, 1993. [3.] J. Giridhar and W.J. van Ooij, Wire J Int., 26, 30 (1993). [4.] W.J. van Ooij, Rubber Chem. Technol, 52, 605 (1979); 57, 421 (1984). [5.] W.J. van Ooij, J. Giridhar and J.H. Ahn, Kautsch. Gummi, Kunstst., 44, 348 (1991). [6.] Y.M. Tsai, U.R. Aggarwal, F.J. Boerio, D.B. Zeik, S.J. Clarson, W.J. van Ooij and A. Sabata, J. Appl. Pol. Sci. Appl. Pol. Symp., 54, 3 (1994). [7.] Y.M. Tsai,' F.J. Boerio, W.J. van Ooij, D.K. Kim and T. Rau, Surf. Interface Anal., 23, 261 (1995). [8.] J. Giridhar, Ph.D. Thesis Colorado School of Mines, 1991. [9.] W.J. van Ooij and A. Kleinhesselink, Appl. Surf. Sci., 4, 324 (1980). [10.] W.J. van Ooij, W.E. Weening and P.F. Murray, Rubber Chem. Technol., 54, 227 (1981). [11.] W.J. van Ooij, S. Eufinger, S. Guo, P. Barto, N. Tang and T.H. Ridgway, submitted to Plasma Polym. [12.] Y.M. Tsai, M.S. Thesis 1992, Univ. of Cincinnati, Cincinnati, Ohio.