Autore: ExtremeTech

Self-assembling nanowires, via aerotaxy

Kepler, Descartes, Hooke — a few of the greatest minds in history linked through the simple question of the child: How does nature create the intricate structure of a snowflake, seemingly out of thin air? Advances made by a group at the University of Lund in Sweden have now elevated the synthesis of semiconductor structures off the chip and into the air, speeding things up 1000-fold in the process.

Building things on the 2D surface of a piece of silicon takes time and patience. As molecules are deposited from a gas phase, new material can be supplied only as fast as it can diffuse to the region of synthesis. In the air however, the growing 3D structure itself diffuses through the medium, and so enjoys maximal supply at all times. Not only that but the growing ends see opportunity to bind new material from virtually all directions, as opposed to just from above. Since the growing structure is free from any substrate, it can add material to any side, expanding the range of things that can be produced.

While their arms grow at similar rates, each snowflake is exposed to a slightly different microenvironment, and no two are the same. The challenge faced by the Lund group was to control conditions so that their process was repeatable. In 1998, they demonstrated that nanocrystals could be grown in the gas phase [paywalled]. They named the process “aerotaxy,” in a nod to the older methods know as epitaxy, wherein metals are deposited onto a chip surface.

Self-assembled nanowires, process diagram

Now the group has been able to synthesize GaAs (Gallium Arsenide) nanowires, seeded from small crystals of gold. The diameter, length, and shape of the nanowires can be controlled by changing the Au particle size, the growth temperature, and the growth time, respectively. The magic happens in a long heated tube, called a tube furnace. Typically gases are introduced from one side in a tightly controlled fashion using mass flow controllers — specialized valves that essentially count how much gas is entering, and then regulate that level as required. A current passing through a sensing lead in the controller feels a resistance proportional to its temperature. When a lot of gas is flowing through past the sensor, it cools faster and voila — instant gas atom counting.

For nanowire synthesis, TMGA (trimethylgallium) and AsH3 (arsine) were the gases used to provide the metal feed material. These gases react and deposit their metals on the growing ends of the wire in its preferred direction of growth as dictated by the crystal structure. The ability to functionalize an ever-increasing palette of materials in this fashion is opening up new ways to build structures. Not just by using new materials but also by using new methods like for 3D printing of conductive materials

nanowire growthIncreasing speed of synthesis 1000-fold will no doubt attract attention from semiconductor manufacturers. Increasing the volume yield per furnace shouldn’t hurt either. Synthesis on 2D surfaces doesn’t stack well and nonuniformity would be introduced both at the edges and throughout different levels in the furnace. In 3D — as in real 3D, not just Intel’s 3D FinFETs — the full volume is available with unhindered access. There will still be issues perhaps with clumping or deposition on the walls of the reaction chamber but new ways to confine and manipulate the process may be found.

The Lund group was able to collect their wires for inspection by manipulation with electric fields. If the wires can be also capped at a certain length, and binned according to size and end group, the molecular tinker toy set may follow. Aerotaxy will perhaps be given its full due in the zero gravity of space. Removing gravity’s influence eliminates one more bias detracting from the uniformity of the build. Perhaps Kepler and his 17th century friends imagined for themselves the era of this achievement. No doubt they would be proud.

Now read: Atom-by-atom sub-22nm chip fabrication

Research paper: doi:10.1038/nature11652 – “Continuous gas-phase synthesis of nanowires with tunable properties”