A brief history of metal foams

The first process
The first record of metal foams dates back to the late 1940's, when Benjamin Sosnick filed a patent^[1] on a "Process for making Foamlike Mass of Metal". His method exploited the fact that in alloys containing different phases with widely different melting and boiling points, the phases can melt and boil independently. In the process, a multiphase alloy is heated, with the composition of the alloy chosen so that one of the components boils while the other has only melted. The alloy was held in a pressure vessel during heating, so that the gaseous metal could not escape the liquid. Releasing the pressure led to sudden boiling of the mixture - which could then be cooled to form a solid full of closed pores. Suggested uses exploited the improved impact resistance of the foam, and its heat and sound absorbing properties. Obviously, only a few alloys were really suitable for this process, as something in the alloy had to have an unusually low boiling point, and this is reflected in the suggested compositions (all of which include relatively volatile elements like zinc, cadmium and mercury). The process was fairly expensive, and only produced smallish amounts of a fairly irregular foam (containing some fairly nasty metals!). It's maybe just as well that this process was never widely used.

Open-cell foams
Some time later, in the late 1950s, the first open-cell metal foams were produced^[2]. This involved pouring molten aluminium into a preform of compacted rock salt grains, and then dissolving out the salt to leave open pores. This produced a more reliable foam structure than Sosnick's method, but at the time it was still seen as a curiosity rather than a serious engineering material. Suggested applications included shock-proof material to house the fragile electric circuitry in missiles and rockets (which became unnecessary as microchips became more robust), high temperature metal filters, and large-area metal cathodes. A variety of castings were produced, including (for reasons never fully explained) a door handle. The inventors published an article on the subject, and then moved on to other things and this method was forgotten.

Enter the foaming agent
In 1959, a completely different approach was discovered. Research staff at United Aircraft Corporation filed a patent^[3] for a method of foam making where a powdered metal was mixed with a powdered gas-forming material. The gas-forming material had to be carefully chosen to decompose and release a large amount of gas at about melting temperature of the metal. This mixture of powders was cold-compacted and extruded, to give solid metal material containing a dispersion of powdered foaming agent. When this solid was heated to the metal's melting temperature, the 'foaming agent' decomposed to release gas into the molten metal, creating a metal foam. Metal hydrides (titanium hydride in particular), which decompose to metals and hydrogen gas at convenient temperatures, and are used in many of the current commercial foam-making methods, made their first appearance as foaming agents in this patent.

Cooling the foam was a problem. Induction heating meant that the heat source could be turned off fairly quickly, but the metal would still be hot, and was prone to collapsing back to molten metal before it got cold enough to solidify. The patent suggested that using water-cooling or heating the foam only locally, would generally avoid this problem, but this problem was later to become a significant challenge to reliable foam production.

Foaming directly from the melt
Four years later, Hardy and Peisker filed a fairly broad patent^[4], covering a method of adding foaming agents directly to a semi-molten metal. This had the advantage of producing a significantly cheaper foam than the powder-based route. A new development was the addition of siliceous materials to the molten aluminium (vermiculite was used in practice), to increase its viscosity and trap the evolved gas in the structure. Another new idea was that, as an alternative to using metal hydrides as foaming agents, hydrated clays could be used - which would release the water molecules trapped in their structure as water vapour when added to the melt.

A few institutions (mainly in the UK and central Europe) continued to work on powder-based foams, though the cost of making metal powders tended to keep the foams produced as academic curiosities rather than commercially useful materials. Melt-route foams were the main focus of commercialminterest. A number of universities and companies started work on alternative methods to produce foams, and successfully developed several new processes based on infiltration, deposition, and other systems, with a wide range of cost and foam quality. As a rule they tended to produce extremely high quality material at high cost, or affordable foam that was too irregular mto be commercially viable.

Commercialisation of foams
Melt-based foams started to dominate the (still embryonic) commercial production of foams. Several companies started to produce metal foams. The Shinko Wire company in Japan developed Alporas foam (produced by direct injection of TiH₂ into a melt whose viscosity is enhanced by additionn of calcium). Hydro in Norway and Cymat in Toronto developed melt-based foams using direct injection of gas to molten metal, with Cymat later producing the first high-speed production line (Hydro was later acquired by Cymat). Inco and ERG aerospace developed high-quality open-cell foams (see the production section for details). Alulight in Austria produces powder-based foams. Several of these ventures are going well, with the first large-scale applications of foams emerging.

However the market for metal foams is still limited, due to the relatively high cost of production of good quality foams. Developments such as the FORMGRIP and FOAMCARP foaming processes at Cambridge aim to raise the quality of melt-route foams to the point where they are commercially useful, while avoiding the high cost inherent in powder-based or infiltration-based production processes. This is mainly in developing techniques which are more stable and produce a more uniform cell structure that the current methods. There has been good progress so far, though the true test of metal foams is whether they can move on from the niche markets they currently occupy into the field of conventional engineering materials.

Foaming other metals
To date, almost all the research has been concerned with aluminium foams (because of its low density, corrosion resistance, and a relatively low melting point which makes it easy to handle). However, iron, nickel and lead foams have also been produced. The nickel foams are used as chemical filters; the others are of dubious quality and are really more of an academic curiosity at the moment.

References
[1] Process for Making Foamlike Mass of Metal - United States Patent No. 2,434,775, Benjamin Sosnick, 20th Jan 1948.
[2] Cellular Materials: New Concepts Provide Unique Possibilities - Feature article, The Iron Age, February 8, 1962, pp 119-121.
[3] Method of making Foamed Metal - United States Patent No. 3,087,807, Benjamin C Alen et al, filed 4th Dec 1959, granted 30th April 1963.
[4] Method of Producing a Lightweight Foamed Metal - United States Patent No. 3,300,296, Paul W Hardy and Glenn W Peisker, filed 31st July 1963, 24th Jan 1967.
[5] Method for Manufacturing Porous Articles - United States Patent No 5,181,549; V.I. Shapovalov, 1993.