Mar., 02/ Inadequate wear resistance and low seizure loads prevent the direct use of aluminum alloys in automotive parts subject to intensive friction combined with high thermal and mechanical loading, such as brake discs, pistons, and cylinder liners. To enable the use of aluminum alloys in the production of automotive brake discs and other wear-resistant products, the insertion of a monolithic friction cladding rather than surface coating has been considered in this work. Three experimental approaches, two based on the pressureless infiltration of porous ceramic preforms and one based on the subsequent hot rolling of aluminum and metal-- matrix composite strips, are currently under investigation.

INTRODUCTION

Over the last decade, automotive industry representatives identified automobile weight reduction as an effective way of improving their product competitiveness as well as their ability to make profits (apparently through improved fuel efficiency, exhaust emission reduction, enhanced safety, and vehicle driving performance). Perhaps, as a result, aluminum use in passenger cars is increasing steadily.1

The chassis, which accounts for about 27% of the weight of the entire average car, is under serious consideration for additional weight reduction. Particular attention is to be paid to reducing the weight of brake components,2 which are currently fairly heavy. However, the improved fuel efficiency and consequent lower exhaust emissions are not the only benefits of weight reduction. Lightweight aluminum-based rotors also provide increased acceleration, reduced braking distance, reduced noise (groan and squeal), higher wear resistance, and uniform friction.

Attempts to replace cast iron in disc brake rotors, drums, and calipers with aluminum-based materials have been ongoing since the mid-1980s.

BRAKE ROTORS

As with most soft metals, aluminum suffers from poor resistance to adhesive wear or galling when in relative motion with another metallic or non-metallic surface. Several solutions to this problem have been proposed, including high silicon3 and beryllium-aluminum4 alloys and various composites.5-8

Two main design concepts have been applied: a homogeneous brake rotor and a brake rotor consisting of a core lined with friction cladding.

Although brake rotors made from high silicon and beryllium- aluminum alloys may reach minimal friction and wear performance standards, they still cannot achieve the performance level of the traditional cast gray iron brake rotor. The main problems are poor wear resistance and a low coefficient of friction.

To improve the wear resistance and to increase the coefficient of friction, aluminum metal-matrix composites (MMCs) reinforced with 20- 35 vol.% of ceramic particulate were proposed.9 Because of their superior friction and wear properties, as well as the considerable alteration in mechanical behavior (tensile, yield, fatigue, creep, and impact strengths) and physical properties (thermal expansion and thermal diffusity) that the addition of ceramic particulate provides to the base alloy, these materials are capable of functionally replacing cast iron and cast gray iron in automotive brake componens. However, due to the high cost of composite materials relative to non-- reinforced aluminum alloys, as weel as costly brake rotor forming and machining procedures, current aluminum MMC brake rotors cannot be cost-competitive with cast-iron equivalents.

Monolithic ceramic materials are also an interesting alternative in replacing cast and gray cast iron in brake rotors. Ceramic brake rotors are already available for racing and luxury cars.10 The discs are produced in two halves from a blend of carbon fibers and liquid polymers thermally compressed in a mold. Once hardened, the rough discs are carbonized by pyrolysis (in a furnace with a nitrogen atmosphere) at temperatures approaching 1,000 deg C. In this way, all polymers not made of pure carbon are converted into carbon. After that, the discs are "siliconized" at 1,800 deg C in a 20 kPa vacuum. In spite of a 25% improvement in friction coefficient over gray cast-iron discs, a 50% weight reduction, and a lifetime equal to the lifetime of the car, the monolithic ceramic brake disc is too expensive to be used in passenger vehicles.

As it became more apparent that homogeneous composite and ceramic materials cannot provide a cost-effective lightening of the automobile brake disc, nor impart all the necessary properties, a brake disc consisting of a core and friction cladding was introduced.2 To provide maximum weight reduction and optimal economy, the core material was selected to be a traditional non-- reinforced aluminum alloy. On the other hand, advanced multifunctional materials that impart unique wear, friction, and thermal properties and effective bonding with the core were proposed for the cladding.

Two manufacturing directions for brake disc cladding were investigated: insertion casting2,11 and surface-coating technologies.12 In both cases, the basic idea is the same-to process the cladding separately, using the most efficient and cost- effective technique. This approach allows engineering of the required properties locally-into particular functional zones of the cladding.

Various thermal and cold-spraying techniques are particularly effective in the production of such multi-zone cladding. Due to their inherent simplicity, these methods are cost-effective in scale- up manufacturing and also provide near-net shape processing capabilities. A coated brake disc technology that optimizes the aluminum brake disc and also makes the disc compatible with current brake pad technology is already in use.13

Apart from the many benefits of the spray technique, the cladding obtained in this way has only a limited thickness of the friction zone (maximum 1-2 mm), which significantly reduces the lifetime of the component. In addition, the deleterious effects of high- temperature oxidation, evaporation, melting, crystallization, residual stresses, de-bonding, gas release, and other common problems of spray methods reduce the lifetime of such cladding.

In order to prolong the lifetime of the frictional zone of the cladding, other producers are practicing insertion casting. There are several modalities and commercial names2,11 for this method reported in the literature, depending on the porosity level in the applied ceramic preform and the reactivity between the preform and molten aluminum alloy.

Basically, a porous or dense ceramic preform is fed into the cavity of the mold. The molten metal for core preparation is then poured into the mold, adhering to the preform in the mold. In the case of a dense ceramic insert, adherence is achieved by reactive penetration. In the case of a porous insert, bonding is caused by infiltration.

Until now, insertion casting has been mainly practiced for the local reinforcement of aluminum-alloy substrates with ceramic particulate.14 Although surface reinforcement of an aluminum-alloy brake rotor with ceramic particulate increases the friction coefficient and improves wear resistance, other problems remain, in particular, thermal degradation of the discontinuously reinforced composite, as will be discussed later. In some cases, the lack of a solid lubricant such as graphite represents another serious problem. The lack of graphite in the system results in low braking efficiency, adhesive wear, and galling. In the traditional cast- iron rotor, graphite is always present in the iron. As the brake wears, the graphite is freed from the iron matrix to be utilized as a solid lubricant on the wear surface. To overcome this problem, nickel-coated graphite particles have been incorporated into aluminum-- silicon-carbide composite cladding by a patented insertion casting procedure named the CastCon process.2 Prototype brake rotors were developed and reported as having better wear resistance than a cast-iron rotor. However, the temperature that the rotor reached during tests with the same braking input as a cast- iron rotor exceeded the maximum operating temperature for the composite material. This demonstrates that a single-layered cladding could not meet all the necessary multi-functional requirements, which may be completely addressed by multi-- layered cladding structures.

To achieve the optimal balance of properties at the friction surface and inside the multi-layered cladding, the composition and thickness of each particular layer should be carefully determined by modeling its thermomechanical behavior as well as the behavior of the integrated multi-layered structure.

RESIDUAL STRESS MANAGEMENT

In most multi-layered structures, stresses caused by differential thermal expansion or contraction of individual layers as the layers heat or cool together should be carefully considered.18

These stresses can be calculated by applying simple elastic beam theory based on knowing the material's properties (Young's modulus and coefficient of thermal expansion [CTE]), the dimensions, the boundary conditions, and the temperature gradient, provided the system remains elastic.

Consider two bonded beams by (Delta)T heated from a stress-free state so that a misfit strain resulting in equal and opposite forces, of magnitude F^sub (CTE)^, is set up in both the layers. In the first approximation, the compressive stress generated in this way can be expressed as proportional to (Delta)E \* (Delta)alpha (Delta)T where (Delta)E, (Delta)alpha, and (Delta)T represent the difference in Young's modulus, the difference in CTE values, and the temperature difference in the two layers, respectively.

In order to reduce the mechanical residual stress in a multi- layered structure, a proper tailoring of the physical properties inside particular layers is necessary. Metal-matrix composites are particularly suitable for gradient adjustment of their physical properties based on the selection of the type and volume fraction of the reinforcing phase, as well as the matrix composition. By optimizing these parameters in two neighboring layers, one can achieve a situation where (Delta)E0 and (Delta)(alpha)0. In a typical arrangement used for brake cladding, an intermediate MMC layer, consisting of the same ceramic and metallic phases as the top ceramic friction layer and an aluminum-- alloy substrate, should be applied to minimize the CTE mismatch and the residual stresses.

ALTERNATIVE ROUTES

Three different manufacturing routes to obtaining multi-layered macro-- composites consisting of a high-- temperature wear- resistant friction cladding and an aluminum-alloy-based core are currently ongoing in the author's laboratory.

The first processing route involves bonding of silicon-carbide preforms of different porosity. In a typical experimental arrangement, silicon-- carbide preforms with 30 vol.%, 45 vol.%, and 60 vol.% of porosity are combined into a three-layered structure and pressurelessly infiltrated by an Al-- Si-Mg aluminum alloy. The spontaneous infiltration is achieved by the oxidation of the silicon- carbide skeleton prior to its infiltration and by an appropriate magnesium content. All preforms are cold uniaxially pressed. The three-- layered cladding obtained is then bonded with the core by a conventional insertion-casting procedure.

A three-layered aluminum MMC is also manufactured by simultaneous hot rolling of aluminum-MMC strips appropriately combined with an aluminum-- alloy strip. The aluminum-MMC strips are obtained by hot rolling of commercially and laboratory-prepared slabs. The resulting 50 mm thick strip, consisting of three-layered aluminum-- MMC cladding and an aluminum-alloy core, is then hot forged into a prototype core-cladding brake disc.

In the second processing route, a dense mullite is partially (to one-third of its height) reactively penetrated by aluminum alloy. As an alternative, a dense mullite/SiC/Al^sub 1^O^sub 3^/ZrO^sub 2^ preform, obtained by reaction-bonding aluminum-oxide technology, may also be used. The partial reactive penetration is achieved by insertion casting, applying the dense ceramic preform as an insert. The three-layered structure obtained consists of the friction ceramic layer on the top of the cladding, the intermediate an aluminum-MMC composite layer reinforced with 35 vol.% of Al^sub 2^O^sub 3^, and an aluminum-alloy core.

The third processing route combines the traditional friction material-gray iron-in cladding, with a light core made from conventional aluminum alloy. Also in this case the bonding between core and cladding is achieved by an insertion-casting method. First, an insert consisting of a gray iron top layer and a porous TiB^sub 2^ substrate is prepared by pressureless infiltration of molten cast gray iron to approximately one-half of the height of the porous TiB^sub 2^ preform. After that, the non-infiltrated part of the TiB^sub 2^ preform (i.e., the porous part of the two-layered insert) is infiltrated by molten aluminum alloy. In a separate set of experiments, an insert consisting of a gray cast-iron TiB^sub 2^ composite layer on the top (iron matrix reinforced with 30 vol.% of TiB^sub 2^ particles) and porous TiB^sub 2^ as a substrate layer is used. During insertion casting, the porous part of the insert is fully infiltrated with molten aluminum alloy for the purpose of adequate bonding with the core. At the Fe-Al interface inside the TiB^sub 2^ preform infiltrated by cast gray iron from one edge and by molten aluminum alloy from other edge, Fe^sub 3^Al and FeAl intermetallic alloys (and also traces of FeAl^sub 3^, Fe^sub 2^Al^sub 5^, FeAl^sub 2^) are localized. In some cases, a cermet based on Fe-40Al has also been detected. Less expensive ceramic preforms will also be evaluated.

ACKNOWLEDGEMENTS

The author gratefully acknowledges the financial support of the Slovene Ministry of Science and Technology.

BRAKE DISC TEMPERATURE MANAGEMENT

During braking, through friction in the brakes, the kinetic and potential energy of the moving vehicle are converted into thermal energy distributed between the rotor and pad. The percentage of the total braking energy gamma absorbed by the rotor, which can be expressed in terms of material properties, is about 70% for a medium- size passenger car.

Consider a midsize passenger car with a mass of about 2,000 kg, decelerating at 0.8 n/s2 from a speed of 128 km/h without brake lock- up. For this case, one can calculate that the average braking power absorbed by one half or one side of one front brake would be about 20 kW.15 Hence, the heat-flux into the friction area of the brake rotor with a friction area of 300 cm^sup 2^ would exceed 600 kW/ m^sup 2^.15

In the standard SAE D-212 friction test for vehicles sold in North America16 consisting of 15 stops from 96 km/h with a deceleration rate of 0.5 m/s^sup 2^ and 35 s intervals between braking, the temperature measured on the friction surface of gray cast-iron brake rotor increased to 484 deg C.17 Under the same test conditions, the temperature measured on the friction surface of an aluminum MMC rotor (alloy A356 reinforced with 20 vol.% of SiC particles) was even higher (542 deg C).17 The surface temperature of a ceramic (SiSiC) brake disc exceeded 800 deg C.7

Even if the rotor and pad are made from high-temperature wear- resistant materials, the high temperatures on the friction surfaces adversely affect their wear rate and lifetime. To protect both rotor and pad, an operating temperature of <370 deg C is recommended.17 To remain below 370 deg C, brake rotor designers tried to intensify the convective heat transfer from friction surface to air by introducing air-cooling channels and cross-drilled holes. However, apart from some limited success in lowering the friction surface temperature, air cooling tends to be ineffective because the disc fits inside the wheel well. It is particularly ineffective in the case of the ceramic brake disc, where six-pot front calipers and four-pot rear calipers are standard units, apart from a ceramic heat shield on each piston to prevent heat flow from the high disc temperatures boiling the brake fluid.

The most effective way of managing the friction surface temperature is by tailoring the thermal conductivity of the friction layer. One approach is based on applying a high-temperature wear- resistant friction layer with a low thermal conductivity and a high- temperature wear-resistant pad material with a high thermal conductivity. The low thermal conductivity of the friction layer minimizes heat flow from the cladding to the core region and favors the accumulation of heat into the friction zone of the rotor. Hence, the temperature at the friction surface increases rapidly and a strong temperature gradient through the friction layer is generated. However, at the same time, the high temperature at the friction surface significantly promotes convective heat transfer from the rotor to air. In this respect, an equilibrium temperature, T^sub eq^, at the friction surface between 370 deg C and 600 deg C would be managed to promote competitive lifetimes of both pad and rotor.

The opposite approach is based on applying a high-temperature wear-resistant friction layer with a high thermal conductivity. Due to its high thermal conductivity, most of the heat absorbed by the friction layer will be conducted toward the core or the air-cooled channels. As a result, the temperature at the friction surface will be kept below 370 deg C.

References

1. M.N. Becker, JOM, 51 (11) (1999), pp. 26-38.

2. SX. Huang and K. Paxton, JOM, 50 (8) (1998), pp. 26-28.

3. A.S. Reddy et al., Wear, 171 (1994), pp. 115-127.

4. G. Valentini, Alluminio Magazine, 15 (2) (1999), pp. 32-33.

5. D.M. Stefanescu and S. Sen, editors, Cast Metal Matrix Composites (Des Plaines, IL: AFS, 1993).

6. F. Pinna, Al Alluminio E Leghe, 105 (1998), pp. 75-81.

7. European Automotive Design, 5 (1) (2001), p. 17.

8. A. Blomberg et al., Wear, 171 (1994), pp. 77-89.

9. Adv. Mat. & Proc., 143 (6) (1993), pp. 223-234.

10. P. Ponticel, Automotive Engineering International, 108 (10) (2000), pp. 86-88.

11. P. Mayer et al., Cast Metal Matrix Composites (Des Plaines, IL: AFS, 1993).

12. N.B. Dahotre et al., JOM, 53 (9) (2001), pp. 4446.

13. Adv. Mater & Proc., 156 (3) (1999), p. 17.

14. Adv Mater & Proc., 158 (1) (2000), p. 50.

15. R. Limpert, Brake Design and Safety (Warrendale, PA: SAE, 1999), p. 111.

16. M.G. Jacko et al., U.S. patent 5,339,931 (1994).

17. J.A.E. Bell et al., Processing, Properties, and Applications of Cast Metal Matrix Composites (Warrendale, PA: TMS, 1996).

18. R.YC. Tsui, Comprehensive Composite Materials (New York, NY: Elsevier, 2000).

V. Kavorkijan is principal scientist at Independent Researching p.l.c in Maribor, Slovenia.

For more information, contact V. Kevorkijan, Independent Researching p.l.c., Betnavska c. 6,2000 Maribor, Slovenia; e-mail kevorkijan.varuzan@amis.net.