Communications - October 2010

er than any other means of production that humanity currently has.”

new Design frontiers

In addition to enabling people to manu-
facture objects they never could before,
3D printers could lead to radically new
designs that are not possible with tra-
ditional fabrication techniques. “Your
first instinct when you have one of these
machines is that instead of making
something in the machine shop, you are
just going to print it,” says Lipson. “But
at some point you realize you can make
new things with complicated geometry
that you cannot make any other way.
You don’t have to stick to straight edges
and flat surfaces that can be easily ma-
chined or thin walls that can be injec-
tion molded. You can make absolutely
any shape that you want.”
For instance, Lipson’s team has ex-
perimented with printing objects with
both hard and soft materials. When the
materials are printed at a random 50%-
50% ratio, the results are ordinary. How-
ever, when the dots of hard and soft ma-
terial are printed in special patterns, the
material, when stretched like an elastic,
actually gets thicker.

Indeed, one of Lipson’s favorite 3D printer materials is Play-Doh. He recently used it to create miniature copies of the U.S. space shuttle during a school visit as part of the Fab@School project, led by himself and Glen Bull, a professor of instructional technology at the Uni-versityofVirginia.Th e Fab@School’s goal is to use 3D printers to show K– 12

students the combined power of math, science, and engineering. The MacArthur Foundation and Motorola have awarded $435,000 to the Fab@School group to develop curriculum, build more 3D printers, and expand the project.

a Larger ink Palette

Although Play-Doh and other squishy substances can be used in 3D printers, melted plastic remains the primary material. Other desirable materials, including various metals and ceramics, are more challenging to use. Progress has been made in printing with metal, but more experimentation is needed to make the process easier and overcome fundamental properties in the materials like melting point and viscosity.

For Lipson’s Fab@Home project, the ultimate goal is to design a robot that can walk out of the printer. Before that can happen, “inks” for batteries, actuators, wires, transistors, and numerous other pieces must be developed. However, Lipson’s lab has already developed an actuator that operates with low voltage and a printable battery.

Adrian Bowyer at the University of Bath has had success making a printable conductor that melts at a lower temperature than the plastic does. Due to the temperature difference, the 3D printer can manufacture plastic channels that do not melt when filled with the hot conductor for wires or other electrical circuitry.

“At the moment the way we manufacture goods is from economies of

news

scale,” says Bowyer. “It is more efficient to make lots of one thing in one place and that’s how conventional industry works all over the world. But there are many things we used to do that way that we don’t do anymore. For instance, I’m old enough to remember my parents getting personalized letterhead printed at a local printer, whereas now we have computer printers. Imagine the idea of a whole industry disappearing, and everybody making what they want in their own home. That would be a pretty profound economic change.”

Foldit Research Paper’s 57,000+ Co-authors

Since May 2008, tens of thousands of Foldit video game players have competed online against each other, and a computer program, in figuring out how 10 different proteins fold into their three-dimensional configurations. in the end, the players managed to outperform the computer program—and are cited as co-authors on the resulting paper, which was published in Nature.

while scientists understand the general process of how the primary structure of a protein is transformed into a three-

dimensional structure, the method of using statistical and related software algorithms to predict protein structures is computationally demanding.

“if you were blindfolded and all you’re doing is picking pieces at random, that’s more or less what the computer is doing,” says Zoran Popovíc, an associate professor of computer science at the University of washington. “the computational methods are eating up huge amounts of resources.”

Foldit’s top scores are
posted online, allowing
the players, who compete
individually or in groups, to
compare their scores. in the
10 separate protein-folding
puzzles, the players matched
the results of the computer-
generated solutions in three
of the puzzles, outscored them
in five puzzles, and created
significantly better solutions in
two puzzles, according to the
scientists.

author list in appreciation of
the more than 57,000 players’
contributions “through their
feedback and gameplay.”
Such extensive author lists
will soon become commonplace
given the increasing online
collaboration between citizen
volunteers and scientists, says
Popovíc, who plans to establish
a center for Game Science at the
University of washington this
fall, and will work on problems
that can be solved with the
symbiosis of human volunteers
and computers.