POWDER METALLURGY AND ITS IMPLICATIONS FOR ENGINEERS - Part Four

By D. GORDON JONES (Visitor) 

CONCLUSIONS

In the space of this paper, it has not been possible to cover the whole field of powder metallurgy as it interests the engineer. One notable omission is the hard metal field and the production of tungsten carbide. This is an industry which is well established in the Republic and the products are of the greatest importance for machine tools and bits in both the general and the mining industries. The author felt that the process of their manufacture is well known. The processes of pressing, lubrication and sintering practised at these plants are common to all powder metallurgical operations and those who have visited the hard metal factories will be impressed by the high standard of cleanliness and the good working conditions.

It is hoped that the paper has achieved its object in bringing to the notice of engineers the wide range of applications of powder metallurgy products in the engineering field.

In the future, powder metallurgy will play an important part when the high strength and the specific modulus of whiskers and fibres will be utilized. In his fascinating lecture Materials and the engineer gave to The Institution of Mechanical Engineers, London(21), Sir Robert Cockburn dealt with materials of the future which include reinforcement of metals and non-metals by high strength fibres and whiskers. Fig. 17 is taken from his lecture and indicates the possibilities that lie ahead. It may be fitting to quote from Sir Robert's conclusion.

'It is primarily the job of the engineer to appraise trends in demand and anticipate future requirements. It is not sufficient merely to identify the direction of advance; the pace is equally important. The exponential expansion of scientific discovery has its counterpart in engineering practice. Technology is advancing rapidly in every industrial country and new engineering projects must withstand competition from the rest of the world. It is of the greatest importance, therefore, to reduce the delay between discovery and innovation.'

Acknowledgements

In the preparation of the paper, the author would like to acknowledge his indebtedness to:

Murex Ltd., and member firms of the British Metal Sintering Association for literature and samples; Messrs G. J. Wevell (Pty.) Ltd., who provided the Hoganas Iron Powder Handbook which contains a tremendous fund of information on every aspect of iron powder metallurgy; Messrs Simasco, Cape Town, and Messrs Alrite Engine ring Johannesburg, for photographs and samples. The author thanks International Nickel Ltd., London, for permission to give this paper and his colleagues in International Nickel Ltd. for their generous assistance in the provision of technical data and production items.

Reference

1. Iron smelting in pre-industrial communities. J. Iron & Steel Inst. Vol. 203 Pt. 4. April 1965, pp 340-348.

2. Hoganas Iron Powder Handbook, Vol. 1. Basic information.

3. Carbonyl iron powders. International Nickel Ltd., Pub. No. 2531.

4. Self-diffusion of metals-ALEXANDER. B. H. & BALLUFI-Report NYO - 663: 1950, New York Atomic Energy Commission. Grain growth during sintering-HAUSNER, H. H. Iron & Steel Inst. Special Report No. 58.

5. The fundamental principles of powder metallurgy. JONES, W. D., Arnold, London, 1960.

6. Oil retaining bearings. Brochure. Bound Brook Ltd.

7. Fluon-impregnated self-lubricated bearings-BLAINEY, A. Iron & Steel Inst. Special Report No. 58, 1954. pp 222-236.

8. Bronze filters. Brochure-Bound Brook Ltd.

9. NAESER, G. & ZIRM, F., Stahl u. Eisen, 1950. Vol. 70. pp 995-1003.

10. WORN, D. K & PERKS, R. P.-Powder metallurgy, 1959. No.3. pp 45-71.

11. WORN, D. K-Powder Metallurgy, 1958, No. 1/2. pp 85-93.

12. MIDDLETON, A. B., PFEIL, L. B. & RHODES, E. C. J. Inst. Metals, 1949, 75. pp 595-608.

13. KNIGHT, J. R. & TAYLOR, B. Powder Metallurgy, 1962 (10) 108.

14. BETTERIDGE, W. Powder Metallurgy, 1964, Vol. 7, No. 14. pp 142-151.

15. IRMANN, R. Iron & Steel Inst. Special Report No. 58. pp 236-241.

16. TRACEY. V. A. Schweiger Archiv., 1964. 30 July. pp 213-223.

17. REDDEN, T. K & BARKER, J. F. Metal Progress-Vol. 87, 1 Jan. pp 107-113.

18. DECKER, R. F. & DE WITT, R. W. Jour. Metals, 1965, 17 Feb. pp 139-145.

19. ROBERTS, D. H. & RATCLIFF. N. A. Metallurgia, Vol. 70, No. 421, Nov. 1964. pp 223-227.

20. RICHARDS, C. E. Electrical Manufacturing, Vol. 60, No. 60. Dec. 1957.

21. COCKBURN, Sir Robert, Materials and the Engineer, lnst, Mech. Engs., 51st Thomas Hawksley Lecture. Proc. 1964/5, Vol. 179. Pt. 1.

DISCUSSION

Dr H. O. Reisener (Visitor): This paper has come at an apt time, for powder metallurgy, in all its different phases, is making its impact felt in every field of engineering where new and better materials are required and are being developed. As the author has explained, the principle is old and is extensively used in many industrial processes. It is quite interesting to note that even in the Nibelungen saga, Siegfried's sword Baldung was made of iron filings which had been fed to geese. Their droppings were placed in a fire and forged to form a blade which afterwards performed wonders. This is another ancient example of powder metallurgy and sintering.

The author has given us a concise and valuable contribution to this interesting and contemporary subject. He is fully qualified to talk about powder metallurgy, having carried out considerable research on the rolling of sintered nickel powders over 30 years ago, when powder metallurgy was still in its infancy.

Large quantities of metal powder in the form of flakes are used extensively in the paint industry. The most commonly used ones are those of aluminium. The flakes are mixed with a suitable vehicle and are applied to protect surfaces or to improve appearance. This type of paint is familiar to everyone. When the vehicle is one of the modern resins, these paints show remarkable resistance to corrosion and the influence of the weather.

Reference was made to tungsten carbide cutting material. This is a universal application of powder metallurgy and has reached a high stage of perfection. It is produced in considerable quantities.

It should be mentioned, however, that a vast amount of research went into its development. The writer well remembers, as a young student overseas, the many setbacks which were experienced by research workers when trying to improve the grading and milling of the powders, their compaction and their subsequent sintering in self-made apparatus and furnaces. For sintering, special hastily built hydrogen furnaces were used which had the nasty habit of exploding at a crucial moment.

As an engineer, you will probably have made use of sintered bearing shells and bushes, based either on a copper alloy or a ferrous alloy. These are used quite extensively in the machine tools. Caro-bronze and oilite, as examples, are well known to many. Their performance under arduous conditions is remarkable. In many cases, they can be regarded as 'life-time' bearings.

Powder metallurgy, in its more refined form, is now being used to manufacture micro-mesh sieves and filters of extreme fineness, ranging down to a few micro-inches. The materials used are corrosion resistant and possess high strength, thus permitting high pressures to be employed. The author has given this field a good coverage. These filters are essential items in the chemical industry.

The production of the powders is quite a complex operation, requiring a considerable amount of experience and know-how. The old atomization method was to pour a thin stream of molten metal through a rapidly revolving shaped disc into a water bath. The fineness left much to be desired so that other methods had to be found. This led to those described. The metal shot is manufactured similarly. In some plant’s atomization is carried out in a vacuum or an inert gas to avoid oxidation and to obtain a powder of high purity.

Particle size distribution is the most important. In some cases, the grading is so selected that the densest packing is obtained. This is achieved by using varying proportions of fine to coarser particles so that the void is filled. On the other hand, if porosity is required, then the grading is more selective and uniform.

New very refractory materials are now being produced by sintering the oxides of various metals, using metal as an interfacial binder. The compacts are first fritted at a temperature above the melting point of the metal which forms a film around the refractory particles. It is then sintered at a high temperature, resulting in the metal diffusing into the oxides and so forming a tough material, very resistant to thermal shock. Cermets are another development for which more and more industrial uses are being found. These provide the engineer with many new and exciting materials, unheard of a few decades ago.

Mr Gordon Jones has given us a resume of the various processes and a good insight into the development of powder metallurgy. It has been most interesting to hear about the various rolling techniques and that powder metallurgy method are now applied to produce strip from which coins are struck.

It is quite clear that many more industrial uses will be found for powder metallurgy. It will provide the engineer with new materials, enabling him to use them in new designs or processes. It is also essential that the modern and progressive engineer should know something about these materials and their possible uses. We are indebted to Mr Gordon Jones for providing us with such an informative paper.

Dr A. Blainey (Visitor): I am glad to contribute to Mr D. Gordon Jones' most interesting and informative review of powder metallurgy. Our work at Harwell on p.t.f.e. bearings began in July 1949, when there was an urgent requirement for a thrust bearing to operate in corrosive gases (volatile fluorides) with a rubbing speed of 15 000 ft/min and at a bearing pressure of ? psi. Conventional thrust bearings proved useless under these conditions; it was thought that p.t.f.e. (polytetrafluorethylene, i.e. polymerised C2F4) could be used in some way, since it is completely inert in corrosive media and has a very low coefficient of friction, thought to be due to mutual repulsion of adjacent molecules by electrostatic forces. The p.t.f.e. molecule consists of a long carbon chain (probably some thousands of atoms long) with attached fluorine thus: -

Due to charge transfer and the small size of the fluorine ion, the C- F bond is very strong, conferring chemical inertness and resistance to thermal decomposition and giving a preponderance of electrostatic charge at the periphery of the molecule, that was responsible for the low coefficient of friction. The coefficient of friction of p.t.f.e. against bronze coated with p.t.f.e. was measured at various normal pressures and temperatures, with surprising results as indicated below: -

Unfortunately, p.t.f.e. has three serious shortcomings as a bearing material, namely: a low shear strength, a low thermal conductivity, and a high thermal expansion (about 4 x 10-4) per degree C compared with metals. This combination would render solid p.t.f.e. useless except for light loads and low speeds. These disadvantages were overcome, whilst still retaining chemical inertness and low friction, by incorporating the p.t.f.e. in a porous metal, which supports it and conducts away from the frictional heat. During use, the surface of the metal becomes coated with the p.t.f.e. which adheres probably by a quasi-chemical bond via the fluorine ions and forms a continuous film.

Bearings of this type, using porous bronze or stainless steel (prepared by sintering powders) as the supporting structure were made by either of two methods, according to requirements. Flat (e.g. thrust) bearings were made by pressing 0.01 in thick p.t.f.e. sheet into the surface of the porous metal at 350° -450°C (higher temperatures for deeper penetration). Bushings and more complex shapes were impregnated through a 40 per cent by volume dispersion of p.t.f.e. in water (p.t.f.e. latex) which, by repeated vacuum impregnation and drying would largely fill the pores to a depth of about 0.05 in.

These bearings performed satisfactorily under the corrosive conditions already mentioned and gave interesting test results in other applications. To cite two examples:

(a) P.t.f.e.-stainless steel bearings were run in 50 per cent nitric acid with a shaft surface speed of 200 ft/min and a pressure of 42t psi. At best the wear rate of the journal was nil, and of the bushing 0.007 in/hr. (The stainless steel was not completely corrosion resistant.)

(b) P.t.f.e. bronze bearings were run in wide-cut gasoline with a surface speed of 5 000 ft/min and pressure 50 psi. Bushing wear was 0.001 in/100 hours and journal wear less than this.

These tests indicate possible applications in a chemical plant and in internal combustion engines where the bearings could be cooled and kept clean by the fuel, thus eliminating the oil lubrication system. The high-temperature frictional properties of p.t.f.e. indicate possible uses in furnaces and similar equipment.

Methods for surface impregnation of metals, strip, etc., were developed, the metal surface being first prepared to provide furrows or craters about 0.005 in deep and wide for the retention of p.t.f.e. The (clean) surface was then coated with p.t.f.e. either by hot pressing or rolling on a 0.005 in thick sheet or by dipping in a 10-20 per cent by volume aqueous dispersion and sintering at 380°-400°C. Aluminium bearing shells prepared in this way were run in gasoline at a bearing pressure of 200 psi and a journal speed of 5000 ft/min, the wear being only about .001 in per 100 hours. A clock gear train was treated in this way and when assembled was found to require only about half the normal torque to drive it and being oil-free did not tend to collect dust and grit during operation. This type of material has obvious applications in sewing, knitting, weaving and spinning machinery, calculating machines, typewriters, food processing equipment, etc. It is cheaper to manufacture and for many applications more effective than the impregnated porous metals.

Mr M. van der Spuy (Member): Being a user of specialized sintered articles. and not a manufacturer, I will confine my observations solely to the electrical field.

During the last two decades, powder metallurgy has made a great inroad in the electrical field. Today, a large part of heavy-duty electrical contact materials for switches, switchgear, and circuit breakers is produced exclusively by powder metallurgy techniques. Copper powders have been employed successfully for such diverse applications as current collectors, brushes, commutator segments, and rotors for squirrel cage motors. A somewhat specialized application which has been developed is porous electrodes of aluminium, iron, or nickel for use in condensers and batteries. The manifold applications of tungsten, tantalum and molybdenum, in incandescent lamps, X-ray tubes, and in the electronics field in general, have already been discussed.

Electric Contact Materials

A considerable proportion of electric contact materials is produced exclusively by powder metallurgic techniques. Power metallurgy permits not only the production of refractory metal contacts but also the manufacture of composite or agglomerated products from components which are not alloyed able. The composite contact materials combine the low contact resistance, high current-carrying capacity, and high thermal conductivity of a metal such as silver or copper, with the high mechanical and temperature, resistance-particularly arcing resistance-of a refractory metal such as tungsten or molybdenum, or a semi-refractory material such as graphite, nickel, cobalt, or cadmium oxide.

The development of electrical machinery promoted the use of sliding contacts to enable the transfer of electrical current between the moving and stationary elements of the equipment. Current c 11 tor brushes were first produced from carbon. Carbon (e.g. lampblack, coke, graphite) powders were pressed with a pitch binder into bars that were sintered into bulk graphite. The required brush shapes were formed by simple machining. These carbon brushes, however, did not fill the needs of modern electrical engineering, except in high-voltage, low-current applications where brushes made only from carbon usually p dorm satisfactorily.

Metal-graphite brushes and similar products were developed from pure carbon brush materials and represent one of the earliest practical applications of powder metallurgy. The most frequent applications are as brushes for motors, generators, rotary converters, and similar machines. Also, the material is used for moving parts of rheostats, switches and current-carrying washers, bushings and rollers.

Satisfactory brush performance requires that sparking, electrical losses and mechanical losses be kept to a minimum. For high-voltage, low-current applications, these requirements are usually satisfactorily met by all-carbon brushes, while high-current, low voltage applications require a higher current-carrying capacity than is obtainable with all-carbon materials.

The high electrical and thermal conductivity values of copper and bronze meet the requirements of high current applications, but the high frictional characteristics of these metals prevent their use as brush material without proper lubrication. Since the insulating properties of lubricants such as oils and greases impair the electrical performance, the use of graphite as the lubricant is indicated. Graphite films between the moving parts reduce friction and wear, and the good conductivity of graphite prevents excessive contact drop. The formation of stable graphite films requires a uniform structure of the metal-graphite product, which cannot be produced by fusion. Early attempts to prepare uniform products by emulsifying the graphite in the molten metal were also unsuccessful.

Powder metallurgy techniques, on the other hand, permit the production of uniform, fine-grained products which combine the desired characteristics of the two components. Copper-graphite and bronze graphite brushes are produced with graphite proportions varying between 5 and 70 per cent. In high graphite compositions, the metal content is not sufficient to secure a satisfactory coherence of the sintered product, so that these products require a binder such as tar or pitch. High-metal products require no binder. In the case of bronze-graphite materials, the powder contains up to 10 per cent of tin and 0 to 10 per cent each of zinc and lead.

The compacted powders are sintered at temperatures between 750· and 900·C (1 280· -1 650·F). In the case of bronze products, the sintering takes place in the presence of a liquid phase.

The current-carrying capacity varies with the carbon content from 35-50 amperes per square inch for all-carbon products, 70 amperes per square inch and higher for high-graphite, to 125-250 amperes per square inch for high-metal products. High-metal products permit momentary loads up to 450-500 amperes per square inch.

Standard types of metal-graphite brushes were found unsatisfactory for equipment in planes flying above 20 000 feet. At these altitudes, the rise in current density and lower temperature have but little effect on the wear of the brushes, but the low pressure and moisture content of the atmosphere tend to pulverize the structure. Thus, the development of improved material became necessary. The high-altitude copper-graphite type of brush which was perfected contains a complex organic lubricant that resists oxidation and provides a stable film which gives satisfactory brush life and contact drop and low coefficient of friction even under extremely adverse service conditions.

Composite materials of the systems tungsten-silver, molybdenum-silver and tungsten-copper, the latter generally containing small additions of nickel, are representative of heavy-duty contact materials used for switches, circuit breakers, and other switchgear, and are also used as welding electrodes. The lifetime of contact material is limited mainly by:

1. The heat developed by the passage of currents.

2. Arcing caused by the switching operation; and

3. Wear from mechanical motion.

The refractory metals are superior as contact materials, not only as far as temperature resistance and resistance to mechanical wear are concerned, but also regarding the detrimental effect of arcing.

Results of arcing are evaporation and material transfer (pitting) from one contact part to the other. In addition to arcing, glow discharge and cold electron discharge also produce a material transfer. However, since the material transfer is proportional to the current, and the current density involved in these two processes is small compared with arcing, for practical purposes only material transfer by arcing need be considered.

The minimum current density required for arc formation at a given voltage is a characteristic material property. Critical arcing currents for several contact materials are summarized in Table.1

The figures indicate the superiority of the refractory metals, tungsten and molybdenum, as far as resistance to arc formation is concerned.

Material transfer by arcing depends on the arc characteristics, and thus on the operating conditions, as well as on the characteristics of the contact material. In a short arc, the fast-primary electrons reaching the anode cause evaporation of the anode material; while in a long are, the ionization of the dielectric medium causes evaporation of the cathode material.

Another advantage of the refractory metal is the absence, under contact pressure, of metal-to-metal sticking. The factor mainly responsible for this resistance to sticking is the presence of oxide films preventing metal-to-metal contact. These oxide films are, however, also mainly responsible for the limited applicability of pure refractory metal contacts. The high resistivity of the oxides prevents satisfactory current transport unless the applied voltages are high enough (at least about 110 volts) for a fritting of the oxide films and the formation of metallic bridges.

To utilize the high contact qualities of the refractory metals for applications at lower voltages, it is necessary to combine these properties with the good conductivities of metals such as copper and silver. As these combinations are non-alloy able, the application of powder metallurgy is indicated. This non-alloy ability is an advantage since alloying would change the characteristic properties of the components, while the sintered products combine the desired properties of the components. Sintered composite materials of the systems tungsten-silver, tungsten copper and molybdenum-silver satisfactorily meet the requirements for heavy-duty applications. These products consist of a skeleton of the refractory metal, the pores of which are filled with copper or silver.

Several methods are available for the production of these materials. The first method consists of pressing and sintering the refractory metal to a porous skeleton and then, using capillary action, impregnating the pores of the skeleton with molten silver or copper. The skeleton can be produced in the desired final shape, but it is also possible to work the impregnated product by forging, rolling or extruding.

Another method applied to the manufacture of composite contact materials consists in pressing the mixed powders at 40-80 tons/in" and sintering below the melting points of copper or silver, respectively.

The selection of the contact material depends upon the application. Molybdenum-silver materials are recommended for low or medium voltage and high currents, and for cases where conductivity is more important than high wear resistance. Tungsten-copper materials are suitable for high voltages were pitting and burning must be overcome and long life is required, particularly for oil-immersed switchgear. Tungsten-silver compositions are preferred for all voltages and medium currents, especially where wear is more important than conductivity. Among other factors to be taken into account in the selection of contact, materials are frequency of operation, the inductance of load, D.C. or AC. application. Special compositions have been developed for each particular application.

The lifetime of composite contact materials is considerably higher than that of either copper or silver contacts. It has been demonstrated that 60:40 tungsten-copper material used as a contact in an oil circuit breaker exceeds the lifetime of ordinary copper contacts by a factor of six.

Fig. 1 shows the life of copper contacts compared to the life of tungsten-copper contacts. Carrying a load of 10 kilowatts the copper contacts gave approximately 25 000 operations, while the composite materials gave 150 000 operations.

Permanent magnets

The author has already pointed out the advantages gained by the application of powder metallurgy to the production of permanent magnets. A very important group of magnetic materials not mentioned is the ceramic magnet.

The term ceramic is applied to an extensive class of magnetic materials. This choice of term is based on the similarity between the manufacturing processes used to produce this class of material and those used to produce pottery. Since the ingredients of ceramic magnets are oxides of certain metals incorporated initially either as oxides or carbonates, the description oxide magnets are often applied.

The ceramic materials are usually characterized by high electrical resistivity, modest remanence and high coercivity; their temperature coefficients in respect of magnetic properties are generally higher than those of alloy magnets.

On account of the possibility of producing good permanent magnets from raw materials that are moderate in price and free from strategic restrictions, considerable research effort has been devoted to the investigation of ceramic permanent magnets based on various oxide systems. The most important type in quantity production is barium ferrite (BaO. 6Fe2O3) and there are modifications in which some or all of the barium is replaced by elements such as lead and strontium.

Isotropic barium ferrite (i.e. random grain material)

This compound is not new; it was described as long ago as 1926. It is formed by heating a well-mixed compact of ferrous oxide and barium oxide or carbonate.

This mixture is compacted under a press and then fired at a temperature of 1 200-1 300°C (2 190-2 370°F). Such material is magnetically isotropic with remanence 0.21 weber/metre (2 100 gausses) and maximum energy product of about 8 000 joules/metre (1.0 MGO).

Anisotropic barium ferrite (i.e. aligned grain material)

The procedure, given above, is extended by preferring the mixture to produce a ceramic powder with crystal grain size larger than single-domain size. This ceramic powder is then ball-milled until a particle size averaging about 1 micron is produced. This powder is again compacted and fired but this time the particles are first oriented with their preferred axis parallel either by pre-magnetizing the dry powder, or by forming a slurry, and applying a strong magnetic field immediately before and during pressing.

The essential operation is the pressing in a magnetic field. If the material is magnetized to saturation, by applying a strong field for a short time, before being introduced into the die, the aligning field need be no more than about 400 000 ampere-turns/metre (5 000 oersteds); if not, then a saturating field of about 800 000 ampere-turns/metre (10 000 oersteds) must be applied at some time after loading.

The remanence of the finished product at 0.38-0.40 weber/metre (3 800 - 4 000 gauss) is nearly double that of the isotropic ferrite. The maximum energy product ranges from 24 000-29 600 joules/metre (3.0-3.7 MGO) and some higher figures are quoted.

Fig. 2 illustrates the superior demagnetization properties of the aligned grain barium-ferrite material, curve (b) as compared with the random grain material, curve (a).

References

1. Dr P. Schwarzkopf, Powder Metallurgy-Its Physics and Production (1947).

2. H. H. Hausner, Powder Metallurgy Bulletin 2, 6 (1947).

3. D. Hadfield, Permanent Magnets and Magnetism (1962).

Dr B. Young (Visitor): Another powder metallurgical technique which is of growing usage is that of infiltration. By this term, we mean the drawing of liquid metal by capillary action into a loose or lightly sintered powder mass. Thus, strong compacts can be readily produced from materials which do not sinter readily or only do so at excessively high temperatures.

This technique is widely used in the manufacture of diamond drill crowns where the properties of high strength, good holding power for the diamonds and high abrasion resistance must be met. A restriction which must be observed is that high temperatures be avoided to avoid reversion of the diamond to graphite. These conditions are met by bonding the diamond in a compacted mass of fine tungsten carbide powder and infiltrating with a brass binder. Such crowns withstand satisfactorily high stresses and abrasion at the bottom of holes 10 000 to 15 000 feet deep.

Applications of infiltration are of increasing importance in the manufacture of compacts for rockets, space vehicles, etc., where the strength, density and phase distribution can be controlled by the manufacturing technique. An example of this is the fabrication of rocket nozzles for solid-fuel rockets. The requirements of these components are: -

(i) Temperature resistance, up to the melting point of tungsten (3410°C).

(ii) Maximum heat dissipation.

(iii) Minimum chemical reaction with fuel products.

(iv) Minimum material loss by erosion from the propellant gases.

(v) Resistance to thermal stress cracking.

The most satisfactory combination of these requirements is obtained by infiltrating with silver or copper the lightly sintered pre-shaped nozzle of tungsten.

Using such techniques, other components such as compressor blades, ablators, etc., are also being manufactured.

Mr T. R. M. Aberdein (Member): I read in Compressed Air, April 1965, another use for powder metallurgy. Teflon, a fluorocarbon resin, was discovered by a large chemical factory in the U.S.A. in 1938. Because nothing would dissolve it, there was difficulty in fabricating Teflon into useful products. t could not be bonded or laminated to other materials because of its slipperiness; it could not be moulded or extruded in conventional equipment because its melting point was so high.

Techniques were devised, however. With the aid of powder metallurgy, resins were compressed and sintered into blocks that were machined into the desired shapes. Then, dispersions of resins in water were used to coat glass cloth and make enamels. To coat wire and make tubing, a powder was developed that could be blended with hydrocarbon and cold compressed to the desired shape.

Almost nothing will stick to fluorocarbon polymers because of their extremely low coefficient of friction. It also resists attack by highly corrosive compounds. It is now being used to make non-stick frying pans, gaskets, valves, seals, and bearings. Non-lubricated air compressors are fitted with Teflon compression and wearing rings.

Mr G. E. Mavrocordatos (Visitor): When in ordinary life we want to express weakness and lack of solidity we commonly use the phrase 'it crumbles like powder.' We have been introduced to a comparatively new field in the metallurgical and engineering world, a field which is based on the very negation of all our metallurgical taboos and prejudices.

The comparatively useless 'crumbled powder' of the conventional engineering world is deliberately produced and then it is carefully reconstituted and compressed and heated until a new object not only of beauty but of extreme utility and durability emerges.

In the conventional world, we are all familiar with the continuous arguments between the user and the foundryman about existing pinhole porosity and blowholes. In this modern technology not only is porosity accepted, not only is it deliberately left in the product to be filled with oil for subsequent lubrication of the moving parts, but it is stated to confer extra strength in wires and sheet, when the voids act as a 'dispersoid,' to use the term coined to define a disposed of phase which is neutral to the matrix.

And this brings me to the third unconventional phenomenon put to good effect in the modern field of materials engineering. In the conventional world of yesterday impurities such as oxides and other substances, when left in the metal, made the metallurgist shudder and the inspector reject such offending products. DurviIIe specially designed a casting method so that they inevitably formed alumina film on the surface of molten aluminium was left undisturbed as the metal was cast into the ingot. Today alumina is deliberately produced in powder aluminium components to confer high strength at temperatures nearing the melting point of the article.

Thoria is added in tungsten and nickel wire for good creep properties.

Mr Gordon Jones stated without explaining the reason that if the applied pressure produces local deformation of the particles then bonding is better. I would like to enlarge a little on this subject by suggesting that the reason for better bonding is the enhanced and accelerated rate of diffusion. The deformation causes a large disturbance to the regularity of the lattice by producing a large number of dislocations and vacant lattice sites which allow a greater rate of diffusion across the particles.

In describing the S.A.P. alloys the author stated that the alumina film remains in the final alloy around the grains and it thus hinders grain growth. I am sceptical as to the accuracy of this statement since if the alumina film was not disrupted during processing no bonding would have been possible. The most probable mechanism is that the fragments of the alumina film, perhaps spheroids or tend generally to coalesce, new aluminium grains grow at the points of bonding but grain growth is hindered in the usual manner due to the existence of a third phase.

If the grains were coated with alumina the product must be weak. Witness the method of production of powders, by forming deliberately a film of an impurity around the grain, such as sulphide of oxide or chromium carbide in the case of stainless steel.

Mr R. Rees (Visitor): One of the most stimulating discoveries for the engineering industry was the development of cemented tungsten carbide in 1929. It was a well-known fact before this date that carbides of Group VI metals exhibited extreme hardness but were brittle. Turning tools of fused tungsten carbide had been tried but life and performance were unpredictable.

Research carried out at a German steelworks enabled a material made by cementing fused carbide powder with metals of the iron group to be developed and from that time spectacular strides have been made in perfecting complex carbide mixtures for all sorts of engineering applications. Of course, it is no longer necessary to fuse tungsten and carbon, crush the product and then mix it with a binder. The process is now carried out by a powder metallurgy technique and while temperatures over 2 600°C were necessary to fuse the carbide, the mono carbide is now produced at temperatures below 1 500°C, which simplified furnace design and made furnace linings in common use suitable.

Before the advent of tungsten carbide, wire production was a laborious process consisting of drawing rods through a steel die plates or hardened carbon steel dies and through diamond dies for the smaller sizes of wires. Today tubes up to 12 in diameter are drawn through tungsten carbide dies. Many complicated sections such as hexagon, strip, curtain rail and trolley wire are produced by drawing the material through shaped tungsten carbide dies made by powder metallurgy technique. Diamond dies are seldom used these days for wire sizes above .008 in diameter.

As far as the Republic is concerned the introduction of cemented carbide to the mining industry was an extremely important step. Before 1947 blast holes were drilled with forged chisel steels and it was frequently necessary to use a new drill for every 3 ft 6 in the depth of the hole. One can imagine the amount of transport and blacksmith services utilized on a moderately sized mine. To some extent, this was alleviated by the introduction of the throw-away detachable bit which although it did not drill more than its 3 ft 6 in depth before being blunted was much easier to handle.

At the end of the Second World War, the tungsten carbide industry had assumed considerable momentum due to the demand of the engineering industry for lathe tools for munitions. Every effort was made to find new uses for carbide to keep these factories working. Small scale tests of carbide tipped rods were carried out on the Witwatersrand early in 1947 and these showed great promise. In 1948 two or three concerns started manufacture in South Africa and the mines gradually changed over to carbide tipped drill steel. While in 1948 a lift of 100 ft per rod was considered good, a similar rod today would average 400 ft. Much development in the grade of carbide and the method of assembling into the steel has been responsible for this substantial improvement in performance.

A more recent development in rock drilling has been the introduction of down-hole bits up to 12 in in diameter incorporating tungsten carbide cutting inserts.

Mechanical coal cutting machines which have become more and more popular, use tungsten carbide cutters and this innovation has increased coal production considerably, often with less labour utilization.

AUTHOR'S REPLY TO DISCUSSION

Mr D. Gordon Jones thanked the many contributors to the discussion for the presentation of useful and valuable additional data which added materially to the value of the paper. Only one point called for the reply and that was raised by Mr Mavrocordatos concerning S.A.P. aluminium. He rightly points out that the oxide film has to be ruptured before bonding

is possible. The oxide film is less than 0.01 micron thick while the thickness of the aluminium particle considerably more than this. The bonding of the aluminium particles during extrusion is facilitated by the plasticity of the aluminium groundmass. The non-plastic oxide envelope is broken by allowing diffusion and bonding to take place.

Mr Gordon Jones, M.Sc., studied metallurgy at Swansea where he graduated with first-class honours, being awarded a University of Wales research studentship. He joined the Mond Nickel Company Limited in 1929 as a research metallurgist at their Birmingham laboratories. In 1944 he accepted an appointment with Engelhard Industries at Messrs. Baker Platinum, London, and in 1949 came to South Africa to manage Precious Metals Development Limited.

In 1954, International Nickel Limited opened a Technical Office in Johannesburg and he was appointed manager. He is now a Director of International Nickel (S.A.) (Pty.) Limited and Manager of Primary Nickel Division. He has taken an active part in various technical societies. He is Chairman of the Overseas Panel in Southern Africa of the Institution of Metallurgists, a Vice-President of The South African Institute of Mining and Metallurgy, Past Chairman of the Base Metals Division, a Past President of the South African Institute of British Foundrymen and a Vice-President of the Institute of Welding.