 |
Will printed circuits replace barcodes
on tomorrow's soup cans?
By David Pescovitz
The future of the ubiquitous bar code is looking grim. In development
at Berkeley are circuit-laden smart tags printed directly on product
packaging that could revolutionize commerce beginning with your
weekly trip to the supermarket.
 |
| Wafers such as the
one Subramanian is holding are used to fabricate printed circuits.
Peg Skorpinski photo |
Imagine filling your shopping cart and walking right out of the
store past a sensor that automatically identifies what you're
buying and instantly charges your credit card. Of course, the
store itself would always be fully stocked because the electronically-enabled
shelves would take their own inventory and automatically reorder
supplies as necessary. Your refrigerator might even generate its
own shopping list, sensing when your milk is sour or your egg
carton empty.
"We're focused on disposable electronics," says
Professor Vivek Subramanian of the Department of Electrical Engineering
and Computer Sciences. "The question is -- can we print a
circuit on a package so that when you ping it with a radio signal
it'll reply ‘hey, I'm a can of soup.' Just
as importantly, can we do it very inexpensively?"
For these printable radio frequency identification (RFID) tags
to catch on, they need to be dirt cheap -- adding less than one-half
a cent to the price of existing product packages, Subramanian
says. To meet that price point, Subramanian and his research group
have embarked on a multi-disciplinary project spanning chemical,
electrical, and mechanical engineering. The result is an extraordinary
inkjet printer and a family of electronic inks that enable circuits
to be patterned onto paper, plastic, or cloth without damaging
the material.
 |
| Liquid gold synthesized
in Subramanian's lab is printed in computer-generated patterns
onto the wafer by the inkjet printer to form transistor contacts,
wires, inductors, and other components used in RFID circuits.
Peg Skorpinski photo |
An RFID tag consists of passive components -- the inductors,
capacitors and wires that handle the communication, interconnection,
and power coupling; and active components -- the transistors and
diodes that handle signal modulation and switching.
"In the long term, you'd like to have a bit of programmability,"
Subramanian says. "For example, every can of soup could have
the same identification number, but each batch could be programmed
with a different expiration date." To introduce this capability,
the group is also working on adding memory to the tags.
At the April meeting of the Materials Research Society, Subramanian's
group presented their success in developing a printed conductor
system that could be used to fabricate the RFID tags' power
scavenging and communication circuitry. The key is "liquid
gold." Synthesized in Subramanian's laboratory, liquid
gold consists of gold nanocrystals that are only 10 atoms across
and melt at 100 degrees Celsius, 10 times lower than conventional
gold films. The size of the gold nanocrystals is engineered to
reduce the gold's stability at elevated temperatures, to
reduce the melting point.
The gold nanocrystals are encapsulated in an organic shell of
an alkanethiol (an organic molecule containing carbon, hydrogen,
and sulfur) and dissolved in ink. Then, an inkjet printer -- either
the group's cannibalized commercial model or one they have
built from scratch -- deposits the material on the plastic, paper,
or fabric in the desired circuit pattern. The liquid gold is also
suitable for screenprinting, commonly employed to print product
packaging. As the circuit is printed, the organic encapsulant
is burned off, leaving the gold as a high-quality conductor.
"Gold is already used in semiconductors, and given the amount
you need in our system, the raw material cost is not very high,"
Subramanian says.
The next stage of the research is to develop high-quality printable
transistors, probably a year or two away, Subramanian says. One
challenge, he explains, is protecting the printed transistors
from corrosive oxygen and moisture. In collaboration with the
College of Chemistry, the researchers are exploring the use of
the same polysiobutylene rubber-type material used in automobile
tires as a screen printable packaging for the printed transistors.
In the meantime, the researchers are working with their existing
organic transistors, as well as with models of what they postulate
their future transistors will look like.
"We want to know just how good the transistors need to be
for the system to work," he says. "After all, this project
is truly at the intersection of economics and engineering."
|
