Just the ticket

ON JUNE 26th 1974 history was made at a supermarket in Troy, Ohio, with a ten-pack of Wrigley’s Juicy Fruit chewing gum. It was the first time a commercial item bearing a Universal Product Code (UPC) was scanned by a cashier at the checkout. Forty years on, what became known as a barcode has transformed the world of commerce by providing reliable product identification, tracking and pricing. Nearly everything now comes with a barcode.

As revolutionary as it was, the barcode has limited abilities. It can impart only the information it was printed with, represented by a series of horizontal stripes or a matrix pattern that can be read by an optical device, like a laser. The next generation of labelling will be more adept, containing tiny printable electronics able to generate, store and share information. These smart labels are about to become a big part of “the internet of things”.

Some products already have small electronic devices, such as a radio-frequency identification (RFID) tag, attached to them. Unlike a barcode, an RFID tag does not have to be in sight of the device that reads it. They are used as security tags, to track products in a factory and they are implanted into pets for identification. Most are passive, emitting only an identifying signal, such as their UPC, in response to a radio transmission. An RFID tag, though, remains relatively expensive compared with a barcode, which can be printed as part of the design on packaging.

The idea behind smart labels is to produce a flexible film of electronics that can be printed like a barcode. One company commercialising this is Thinfilm, based in Oslo. It has pioneered a way to print memory circuits which can be used by smart labels to store information.

The Thinfilm memory is printed as a film of ferroelectric polymer sandwiched between two electrodes. When a voltage is applied to the top or bottom electrode, the polymer’s magnetic polarisation can be switched. When the voltage is removed, it remains in the same state. This means it can be read as a one or a zero.

The amount of printable memory that can be produced is not huge. But for some applications a little can go a long way. As Thinfilm points out, a tiny 20-bit memory label can store over 1m combinations, which is enough for companies using them in things like toys or loyalty cards.

The next thing is to add logic, which requires printing transistors. For this, Thinfilm has teamed up with PARC, a Californian subsidiary of Xerox now run as an independent research laboratory. PARC has been working on printing flexible electronics for more than a decade.

PARC develops inks with the properties of the components it wants to print, such as a semiconducting material for transistors. Silicon is used to make transistors, but being difficult to dissolve does not make a good ink. So PARC uses organic materials instead. These can be printed with inkjet machines or, as Thinfilm does, with commercial reel-to-reel presses.

In January Thinfilm bought Kovio Technology, a Silicon Valley firm with expertise in Near Field Communication (NFC). This is a system which allows a user simply to tap an NFC tag with a portable device, like a smartphone, to send or receive data. NFC is a more sophisticated version of RFID and is already used by some contactless payment systems. By incorporating NFC, smart labels will be able to communicate wirelessly.

Besides conveying product codes, the applications which the company already has in mind include recording storage times and temperatures for perishable goods like food and pharmaceuticals. These data can be displayed on a flexible screen along with other consumer information. Smart labels might even be programmed to automatically discount their prices in response to marketing campaigns. Confirming the identity of products is another potential use. Davor Sutija, Thinfilm’s chief executive, thinks smart labels can printed for less than the cost of holograms, which are currently used to authenticate some goods.

To gain widespread use, smart labels will need to be cheap. Basic printed-memory labels could be produced for around two cents, reckons Dr Sutija. Printed sensor-labels might cost 50 cents or less, compared with $10 or more for a system using conventional microelectronics. Prices would be bound to come down if smart labels become anywhere near as popular as barcodes. That would open up all sorts of possibilities. Imagine a label that would inform a smartphone app that you have put on the wrong trousers.