What goes around, comes around

GLASSMAKING began 4,500 years ago, in Mesopotamia. The industry’s first products were trinkets, such as beads and pendants, cast from moulds and carved by hand. But craftsmen quickly worked out how to make more practical stuff, such as jugs, bottles and drinking vessels, by coiling strands of molten glass around a sand or clay core of appropriate shape, which could then be shaken or scraped out after the glass had cooled.

Since those early days, many other ways of forming glass have been invented. These range from blowing forcefully through a tube to inflate a hot gob of the stuff, creating a hollow vessel, to floating it as a liquid on a bed of molten tin to produce perfectly flat window panes. But ancient wisdom often still has value, and now a group of researchers at the Massachusetts Institute of Technology have had another look at the coiling method, pronounced it good, and modernised it. Their principal updating is to dispense with the core. Instead, they have turned to the field of 3D printing—or additive manufacturing, to give its formal name. Objects of rare beauty, and possibly of great utility, result.

Neri Oxman and Peter Houk, the team’s leaders, started with the form of 3D printing most familiar to hobbyists. Fused-deposition modelling, as it is known, works by extruding a filament of semi-molten material (in desktop applications this is usually a thermoplastic) through a mobile nozzle in a pattern controlled by a computer. This builds up whatever object the software is programmed to create. Using glass instead of plastic requires higher temperatures, but the principle is the same.

The researchers’ first efforts failed because the filament—pulled by gravity from a hole in the bottom of a crucible heated to 1,000°C—had an irregular diameter. This, and too-rapid (and therefore uneven) cooling to room temperature, led to poor adhesion between successive coils as an object was built up, and also to stress within the filament as it solidified. That meant the finished product was often fragile.

To overcome this problem, the team added a heated ceramic nozzle which they could use to shape and control the flow of glass through the hole, and also an annealing chamber, heated to 500°C, in which the object being manufactured is built up. This chamber acts as a staging post on the journey from the temperature of the crucible to the temperature of a room and thus reduces the risk of uneven cooling. At present, the flow of glass is turned on and off by cooling and heating the nozzle by hand, using compressed air and burning propane respectively. In future versions, this process might be automated. A plunger could also be employed to propel the filament, thus controlling the rate at which it is extruded.

Dr Oxman and Mr Houk have already used their device to print a range of objects, including optical prisms and decorative vessels. It could also, they think, turn out things like specialist lighting fixtures and special glassware for biological experiments. One useful feature for this sort of work is that, unlike a blown-glass vessel, which necessarily has a smooth internal surface, a printed vessel can have complex surface features on the inside as well as the outside. That can help control the circulation of liquid inside such a vessel. If the process can be industrialised, then, 3D printing looks like a clever way of making bespoke glass devices that can, according to need, be either extravagantly decorative, or impressively utilitarian.