Starry future for printed RFID

7 February 2006



Report points to printed tags in bid to drive down prices, reviews Pauline Covell


"It is clear that the largest RFID volumes will occur when costs go well below five cents (per tag), enabling deployment of item level RFID. Given the gap between this and the cost of silicon RFID, there is an opportunity for alternative RFID technology. A strong candidate is all-printed RFID." So writes a leading researcher in polysilicon technology and organic electronics, Vivek Subramanian in his 66-page report 'Developments in Printed RFID', published by Pira International at £295.

His cogent and informative survey of work today and explanation of tags currently used is a good start for anyone trying to get to grips with RFID technology. But it is the potential of printed antenna and, even more, the use of nanotechnology in materials and inks, complete printed circuits and transistors that made the reading of this report actually quite exciting. As the author expands along a clear path from basics to high technology, it is possible even for the reader without an electronics degree to gain a relatively clear understanding of a complex subject.

Tag manufacture

Comprising five chapters, the report covers RFID classification, Printing silicon based RFID devices, Printing circuitry, Value added technology and Major players.

Chapters two and three are particularly interesting. The first comprises a review of the RFID tag manufacture, focusing on conventional RFID tags that use silicon chips as their main electronic components and the challenge of printed antenna, whilst the latter looks at the potential for all-printed RFID tags.

"Cost reduction is a critical driver in RFID R&D. To achieve the promised volumes and market penetration, particularly for item-level tracking, costs must be reduced - and one proposal is to use printing techniques for some of the fabrication steps.

"In all existing silicon based RFID tags, the antenna is moved off-chip. In other words," he explains, "everything but the antenna is fabricated on the silicon chip, which is then attached to a separate antenna stage. This partitioning strategy obviously helps to reduce the size of the silicon chip, but it adds extra components and extra costs to the RFID tag."

Silicon chip costs will fall, he predicts, but main challenges are to reduce costs of the antenna and of its attachment. "Those challenges create the main opportunities for printing in RFID applications." Using fast technologies such as screen, flexo, gravure or offset printing it is possible to produce antennas much more cheaply than using etched copper technology, he says. "As printing typically involves only one or two steps, such as printing followed by drying, it is possible to use a fast, simple process to produce antenna straps rapidly. And since printing is additive – there is no etching process – the waste disposal costs are much lower than in etched copper technology."

However, in practice, complications arise from material choices and printing techniques, he warns. In general, the resistivity of printed materials is higher than the resistivity of etched copper. Consequently, printing is used for UHF antennas, but not for HF antennas with their tougher resistivity requirements.

The report contains a well researched resumé of printing processes considered for antenna. Main limitations of screen printing arise from its viscosity requirements and resolution limits. He writes: "A typical screen printing ink consists of conductive particles or flakes interspersed with a binder material such as polymer within a solvent base. The binder is to give the required high viscosity. After screen printing and subsequent drying, the resulting pattern typically consists of metallic flakes and particles with interspersed binder. Since the binder is relatively non conductive, the overall resistivity of screen printed material tends to be many times larger than an equivalent film of pure metal, and to lower their resistance, features have to be made thicker or wider."

Flexo and gravure have lower viscosity requirements than those of screen printing; they usually have a lower binder content, hence a lower resistivity. Both processes, he reports, have made some inroads into UHF antenna fabrication, but are still rare for HF antenna due to their more stringent resistivity requirements. "Given that gravure appears to deliver higher resolution, including linespace, it should be possible to print HF antenna as long as the ink formulations offer low resistivity at a small enough linespace."

The inkjet process typically allows lower viscosity inks than any other printing technology, so it is possible to produce inkjet compatible inks with no binders and create patterns having higher conductivity. "Unfortunately, there is generally concern over its stability, although this is being addressed by advanced heads specifically for electronics applications. Drop-by-drop delivery typically requires several passes to produce relatively thick films. Most inkjet printed lines are less than 1mm thick and this reduces process throughput for antenna production. Since inkjet tends to be a relatively slow technique, high throughput needs large arrays of heads." There has been a resurgence of interest in inkjet printed antenna of late, he reports.

On the materials front he introduces us to ultra small particles having diameters less than 100nm, often called nanoparticles. "These small particles are often extremely stable in colloidal suspensions and can be used in inks with high mass loadings of particles. It is possible to produce inks having a high concentration of particles in a solvent plus polymer binders to adjust viscosity. It is also frequently possible to achieve fairly high conductivity compared to conventional flake based pastes." And since very high mass loading using such small particles can be achieved, "the binder content for a given viscosity is often lower, allowing higher conductivity inks for a given viscosity requirement.

"When the diameter of metallic nanoparticles is reduced well below 100nm, they exhibit an intriguing physical phenomenon. As the diameter of a particle is reduced, the ratio of the particle's surface area to its volume increases. When particle diameters go below 10nm, many metallic nanoparticles show a dramatic reduction in melting point. For example, gold nanoparticles with diameters of approximately 2nm melt at about 100degC whereas bulk gold melts at about 1,000degC . Films formed using such small particles may be annealed at very low temperatures, causing the particles to melt and fuse together, at least locally. This produces much better contact between particles than if they were merely touching each other, raising the possibility of films with conductivity much closer to bulk conductivity than obtainable using larger particles or flakes. Conductivity as high as 30–70 per cent of bulk conductivity has been reported."

Since the sintering process used may be carefully tuned to ensure very low residual organic content, it is possible to produce films with conductivities much closer to bulk metal conductivity than obtainable using paste and particle inks.

One cent

"Printed antennas are one of the more near term applications of RFID printing. They have a definite cost advantage over etched copper. The main trade-off is between antenna performance and cost. Although etched copper antennas typically perform much better than printed antennas, particularly in HF applications, they do cost more. Similar trade-offs exist between the different printing technologies. Paste inks and high throughput printing technologies such as screen and flexo typically cost less than novel inks and their printing technologies. There are tremendous opportunities to design new inks that mean fewer trade-offs."

Perhaps the biggest boost for printed antennas will be the development of all-printed circuits, the subject of his third chapter. "Depending on the assumptions about material costs, equipment capital expenditure and deprecation and substrate costs, the cost of an all-printed RFID tag may be as low as one cent." If an all-printed tag is achieved, the economic costs will be compelling, provided the cost of printing the rest of the tag is less than the cost of silicon plus attachment, he enthuses. In other words, given the current scaling trends of silicon and attachment costs, all-printed RFID is essentially competing with attachment.

"Playing devil's advocate, it may be argued that solving the attachment problem is an easier problem than developing an entirely new printed circuit technology. This is possibly true; however, the potential elegance and efficiency of an all-printed, reel-to-reel RFID process is extremely compelling and has driven substantial activity in printed RFID," he writes.

The chapter focuses on the printed transistors that are crucial to all-printed RFID. Most of the opportunities relate to materials and printing technology. "There is substantial interplay between material requirements and printing capabilities. For example, current high speed printing technologies typically have minimum line widths greater than 40 micron and layer-to-layer registration worse than 15 micron. Clock frequencies above 100kHz will probably be needed for long range RFID applications and this is likely to require mobilities much higher than 0.1cm2/Vs. But if transistors with gate lengths of 10 micron and overlap of 5 micron are realised through advances in printing technology, then the existing performance of many printed organic semiconductors will probably suffice."

"Current high speed printing technology has poor layer to layer registration, requiring devices with large overlaps, which degrades performance. If registration accuracy were improved to better than 5 micron, it would reduce demands on semiconductors, and bring forward all-printed RFID. Most device and circuit architectures for producing all-printed RFID are relatively conventional and this leaves big opportunities to innovate," he predicts. "Over the next three to five years, all-printed RFID will increase in maturity, and based on current trends in materials and printing technology, some demonstration prototypes will be produced, followed by product development."

Also included in the report are some useful insights into related added value products such as printed batteries and e-paper. Finally there is an excellent snapshot of the major players involved in aspects of printed RFID.

Contact

Pira International Tel: +44 (0) 1372 802080




External weblinks
Converting Today is not responsible for the content of external internet sites.

Pira International

Printed HF RFID tag: schematic block diagram Printed HF RFID tag: schematic block diagram
Archetypal printed transistor: schematic cross-section Archetypal printed transistor: schematic cross-section


Privacy Policy
We have updated our privacy policy. In the latest update it explains what cookies are and how we use them on our site. To learn more about cookies and their benefits, please view our privacy policy. Please be aware that parts of this site will not function correctly if you disable cookies. By continuing to use this site, you consent to our use of cookies in accordance with our privacy policy unless you have disabled them.