How Converters Can Make RFID Labels

Part One: Low Cost Entry — Wrapping Electronics
RFID is an important enabling technology that appears in an ever-widening variety of applications. In the past, the emergence of barcodes helped create a large label market that lasted for one or two decades before people figured out how to print the barcodes directly on to packaging and products. So, today only 5-15 percent of barcodes appear as labels.

However, there are now so many barcodes required—up to 10 trillion yearly—that even the minor percentage still taking the form of labels constitutes a substantial business. This label business will endure because the labels perform special functions, like serving as a swing tag on retail apparel that would be defaced by having an easily read barcode printed on its outer surface.

Back to the future

RFID is going the same way. One day—perhaps in 2016, perhaps in 2020—RFID functionality will mostly be printed directly on to packaging and products, possibly in the trillions, and will use special electronic inks. In the universities and research laboratories, there are already many potential technologies for printing even the transistor circuit in the RFID chip. However, a large RFID label business will remain because many applications require the ruggedness or clever features RFID labels afford. During the next 15 years, the large and enduring label market may increase production to hundreds of billions of labels yearly.

Converters can participate in many ways

Participation in this business is becoming possible at various degrees of vertical integration. For example, one may take tested RFID inserts, otherwise called inlays, from a company such as UPM Raflatac. These are naked, working RFID tags and may be sourced as reels of polyester film with the RFID microchip and the antenna on top, in a repeated sequence. Most RFID labels use a chip and a connected antenna allowing them to “talk” to an electronic interrogator at a distance. This is a form of radio, thus the term Radio Frequency Identification. Doing the minimum, a converter may simply take these inserts (naked tags) and protect them with paper or plastic on either side, applied by a reel-to-reel process. These unprinted reels of RFID labels may then be sold to a customer.

Typically these labels would also feature graphics, messages, or a barcode. The labels would then be tested for electronic performance before delivery, possibly even cropping them into discrete labels. It takes know-how to wrap the delicate chip and overprint without damaging it. It takes electronic skill and equipment to test the labels at high speed while marking failures. That adds value and financial return, and many converters find this to be a profitable activity today.

RFID on retail apparel

One example of an RFID insert assembly is Paxar using RFID inserts from IER to make swing tags for the retail apparel of Marks and Spencer in the United Kingdom. In other applications, wet (with adhesive) labels are required or even label-like structures such as the flat tags attached to cows’ ears or the smart passport pages made by Charles Oberthur, Gemplus-Setec, and ASK. The latter meet very stringent quality and formatting requirements, and are an illustration of how converters can make more of the label and choose to make more demanding ­designs for more demanding customers.

There is plenty of scope. From the point of view of mechanical and temperature requirements, there are even stitchable RFID labels and units that can be molded into plastic conveyances. Electronic performance of the tag can include wide ranges of electronic interrogation (one millimeter to one kilometer), required minimum percentage of accurate reads, and failures during life, for example. For long range, the label will need a battery, sometimes in the form of an extra laminate.

In the next article we will look further at how the converter can take more added value by replacing part of the inserts in their RFID labels with something they make themselves.

Part Two: Making the Antenna As Well

If RFID label converters wish to add value beyond simply protecting a fully working RFID insert, the key is to use a “strap.” Because integrated circuits are so small, connecting to them is difficult. Converters may wish to add value by printing the antenna of an RFID label onto plastic film or paper feedstock. But chip placement and connection can be troublesome, particularly with something as irregular as paper.

How a strap works

A strap simplifies the connection because it is a chip mounted face down on a small piece of plastic film with metal connections on its surface, created by the supplier, using printing or evaporation. It is then relatively easy to connect the strap to the antenna that the converter may have printed on the paper feedstock, for example. In instances where the popular high-frequency (HF) antenna is needed, there are several printed or etched turns, typically on polyester film substrate, and the strap acts as a crossover, connecting the inner end to the outer end. The alternative would be to make the laminar coil very narrow as it passes under the chip, but this can be expensive and difficult.

The converter makes the antenna

Straps with mounted chips and two connecting conductive surface strips are available from companies such as Avery Dennison and Alien Technology. The antenna can be printed at high speed typically using gravure as chosen by Omron, the largest manufacturer of RFID labels in Japan and by Avery Dennison in the U.S. It is only in the last year that fine, low-temperature, cured conductive inks have been made available for gravure and flexo printing. They provide a cost benefit over screen printing because they use less ink and are faster. In addition, the resulting label can be flexed more before the conductive surface cracks, so there is a performance advantage as well. The antenna surface can then be connected to the strap using suitable conducting adhesives. Electronic testing and overprinting are then required.

All this is an option for both label and packaging converters, particularly at the popular ultra-high frequencies (UHF), where the antennas can be relatively crude in terms of definition and electrical conductance but still work well. UHF RFID tags have been specified for the world’s air baggage by IATA and for incoming cases and pallets by leading retailers and the U.S. Military. The largest bookstore chain in the Netherlands is rolling it out in the form of labels in all its books, and it will not stop there. In all cases, EPCglobal coding and networks managed by GS1 will provide unique identification on each item. All of this creates other opportunities for converters in the form of electronic variants of labels.

Near-Field UHF

First there is the simple antenna possible with so-called near-field UHF. This has short range—typically only 30 centimeters—which is useful for distinguishing one product from another. It is not particularly affected by water or metal. The antenna is a simple printed ring pattern. Tagsys provides this on a strap the size of a postage stamp for converters to make into labels or they can print it themselves and use a conventional tiny strap for assembly. Combined near-field and far-field antennas can be put in a label with the converter printing the larger far-field antenna. It all adds value for the converter. In the Tagsys designs, the far-field antenna does not even have to be bonded electrically to the chip or the near-field antenna.

Challenges

UHF is not entirely a bed of roses even though its use may overtake HF (the most popular frequency) in several years to come. Although conventional far-field UHF gives unusually long range without recourse to a battery in the tag—several meters—the range is unpredictable because of sensitivity to water and metal, even if it is not between the tag and the interrogator. There are workarounds for this and they usually take the form of a wide variety of very different antenna patterns that are screen-, flexo-, or gravure-printed for the particular application. For example, the electronic product code (EPC) chip supplier Impinj offers its “banjo” and “propeller” designs on license to those using its chips and also combines near-field/far-field antenna designs. Others offer asterix, zigzag, paddle, strip, and other weird and wonderful shapes of antenna, all easily printed with conductive ink.

Do not wait for simplicity

A newcomer might say that all this will shake out into very few antenna designs for UHF tags. This is a false hope, even though Avery Dennison has a so-called adaptive antenna that tolerates a variety of metal and water environments. The point here is that there is a wide variety of needs emerging—from the cheapest solution to the one with the highest percentage of good reads, tags that electrically couple with proximate metal to enhance performance, as well as those that need to be small enough to go on the smallest plastic bottle of drugs. This is not academic.

Mandated suppliers of consumer goods to the major Western supermarkets must buy and fit the tags regardless of economics. As with anti-theft tags, they typically get nothing approaching a satisfactory payback so they buy the crudest, cheapest RFID tags they can get away with. Contrast the U.S. Military and many Eastern retailers where they are prepared to pay for quality and functionality and seek mutuality of benefit and enduring supplier-retailer relationships based on economics. The same can be said for that Dutch bookstore tagging its own books, for now at least.

Many antenna patterns

With UHF, in contrast to HF, we therefore have a considerable variety of tag antenna designs, and converters can choose either to make the simplest and most competitive designs or become experts and seek a measure of premium pricing. However, it is not a business for the faint of heart because supplier losses of millions to tens of millions of dollars are being reported in the UHF label, insert, and strap business as supply exceeds demand in the short term and technical problems delay projects.

Massive markets?

The end points of two billion items of air bagage, 35 billion pallets and cases, 50 billion books, and 650 billion postal items to tag yearly remains alluring for those with deep pockets and long-term views. However, a very large number multiplied by a very small price is not necessarily a big business, and prices have been in free fall for UHF tags of late. As the famous economist John Maynard Keynes said, “In the long term, we are all dead.” In the future, it will be increasingly important for UHF RFID label makers to choose the right customers and to find unique items to offer them.

An alternative route for converters seeking to vertically integrate by making the antennas is to look at the traditional HF applications including smart passports, tickets, cards, library books, rented textiles, laundry, and item-level drugs. This choice of HF leads to opportunities to supply up to 30 billion on drugs down to tens of millions yearly on passports. Pricing for quality and conformance are commonplace, as are profitable suppliers of RFID. However, these antennas do not work unless they are very conductive electrically and very precise in definition. Only where degradation of range is tolerable, as with cards, tickets, and passports, do we see printed silver paste antennas, the rest consisting of etched copper or aluminium in the main.

Such problems are opportunities of course, with companies such as Trierenberg Holdings seeking to be first to print HF antennas at high speed. Do not believe the more strident enthusiasts for UHF RFID who argue that it will be used for almost everything. Indeed, in contrast to HF, it will always have serious limitations. There will never be a single signalling protocol, frequency, power level, or bandwidth for UHF RFID across the world. Vested interests in the military, mobile phone ,and other industries will see to that. Near-field UHF for short range and small items has yet to be sold in volume and it has limitations and challenges to offset its undeniably cheap antenna construction. This is detailed in the IDTechEx report “Near Field UHF vs HF for Item level RFID” at www.idtechex.com

Then there are existing specifications for HF in certain applications and the penetration of HF labels into applications such as gas cylinders, beer kegs, and livestock, traditionally served with low-frequency (LF) tags that are plastic or glass moldings, not labels. For the next ten years at least, there will be huge markets for both HF and UHF labels and savvy suppliers are increasingly offering both.

Part Three: Getting Involved With Chips and Batteries

Converters may wish to vertically integrate by making RFID antennas, attaching a “strap” holding the chip, and making them into tested, overprinted, and, if necessary, wet or ruggedized labels. We dealt with this in parts one and two of this three-part introduction to the subject. However, the most clever bit that deserves the greatest added value is the microchip. Can converters even get involved in this?

You will not make chips

The simplest answer to this—whether or not converters can get involved with microchips—is that a silicon chip factory costs $3 billion and only one or two companies can now afford to build them. This problem is aggravated by the rising cost of research needed to maintain the best production technology for chip making. Designing RFID chips for someone else to make is not as expensive, but independent RFID chip designers such as Impinj have raised tens of millions of dollars to compete with established RFID chip designers and makers such as Philips Semiconductors, Texas Instruments, EM Microelectronics, Hitachi, and Infineon. Despite being a start-up company only a few years ago, Impinj succeeded in this, but its skills are a long way from those of a label or package converter.

You can handle chips

For the converter, all is not lost. The converter can get close to leaders in chip supply, such as EM Microelectronic for Marks and Spencer apparel labels or Philips for smart cards and tickets (labels by another name), and chip designers in distant places with very valuable skills. The new G2 Microsystems company in Australia has developed the world’s first very low-power, low-cost, single-chip solution for real-time locating systems (RTLS). RTLS is usually a form of RFID where there is a battery in the tag to give not just identification, but location when it is interrogated electronically.

Disposable RTLS

RTLS tags take the form of plastic moldings on assets, heavy wristbands on people, and so on that can be sensed from a distance away. None are labels as yet, but the creation of G2 Microsystems means that a label version, say like the wristband in a swimming pool or an asset tag, with a printed battery in it, may be possible. The chip is not cheap in the sense of those for pallet and case tagging in billions yearly (soon). It is about $11 in volume. Chips for the pallet/case labels were $0.07 last year and may soon drop to $0.02. However, those using complex $3 RFID chips for labels and pages in passports make good returns because such sophistication can be priced in. Disposable $15 wristbands will find uses in hospitals where sanitizing reusable $100 wristbands is a problem. Some converters will be very happy to do runs of only millions of RFID labels, pages, and wristbands yearly, if they are highly profitable.

Adding value to chips

Although converters will never make RFID chips and are unlikely even to design them, they can get involved in them in other ways. There is a market for RFID tags measuring only 2.5 millimeters across. Hitachi Maxell makes such versions by electroplating a tiny coil antenna on the microchip itself. Range is only one millimeter or so. However, this “label” is fitted to the bottom of millions of tiny test tubes every year. These test tubes are used for blood samples and drug research, for example. The global demand for test tubes is several billion yearly and barcodes are rendered unreadable by the attendant moisture, bending, and abrasion.

Placing and bonding of chips

Another more generally applicable option is to buy a chip-placement machine, from companies like Muhlbauer of Germany, the most successful supplier of these. You can then put chips onto your own antennas or buy antennas on film substrates from companies such as HD Communications in the U.S. or Hanita Coatings in Israel. Symbol Technologies—one of the most successful and technically competent suppliers of UHF RFID labels, interrogators, and systems for books, retailing, and air baggage—also offers mounted RFID antennas as a product.

Shakeout

The RFID value chain has fragmented, with companies variously offering chips, straps, near-field inserts for mounting on far-field labels; conventional inserts; white labels and fully tested overprinted, wet and dry labels that are barcoded; and similar structures. As the market matures, it will not be possible for all to make a profit other than in niches. Certainly, Deloittes has calculated that the promised one cent chip-based labels for everything in the supermarket and even letter post will not be enduringly possible or profitable for all in the RFID value chain. That will call for chipless solutions such as the primitive conductive ink stripes printed at high speed directly onto packaging by ACREO of Sweden and M-real in Finland, if it proves adequate. The Acreo technology recently became available under licence.

Printed thin-film transistor circuits

Alternatively, some of the 55 companies developing printed transistor circuits will provide RFID capability. OrganicID and PolyIC say they will launch the first fully printed RFID labels including transistor circuits next year. Printing technologies used here include ink jet, flexo, gravure, and litho (albeit all in modified form) and with dielectric, semiconducting, and conducting inks. Even light-emitting inks are available for printing electronic displays on the labels where required. However, making chipless RFID labels is not a short-term prospect for converters.

Batteries

Printed, or at least laminar, batteries are needed in RFID labels for more than the dream of disposable RTLS. Tags that work at long range without doing positioning need a battery, as well as tags that can initiate a signal, and units that sense and record parameters such as when tampering occurs or when the label—and therefore the package—was overheated, damaging the vaccines, blood, or food inside. Smart Active Labels SAL and Ubiquitous Sensor Networks USN are hot topics.

Web production

In simplest terms, the laminar and printed battery technologies include environmental, disposable types based on carbon zinc (Thin Battery Technologies or TBT) and Manganese dioxide zinc (Power Paper of Israel and licensees). The TBT product is used by Sealed Air Corporation in smart packaging and the Power Paper product, through licensee Graphic Solutions, has appeared in talking cards and RFID labels. Both technologies can be made in reel-to-reel processes. Alternatively, there are many companies making lithium batteries in laminar form, though not primarily by printing. These can be thinner, hold more power per square centimeter of footprint, and deliver that power faster. However, some have encountered regulations requiring retrieval for disposal in some countries.

About the author
Dr Peter Harrop FIEE is chairman of ­IDTechEx. Previously, he was chief executive of Mars Electronics, a $260 million transaction electronics company, and chairman of 15 high-tech companies over the years. He travels widely carrying out consultancy on RFID, smart packaging, and printed electronics and he has written many reports on these subjects.