Steven Cherry: Hi, this is Steven Cherry for IEEE Spectrum’s “Techwise Conversations.”
Have you heard about the tracking device in U.S. banknotes? There’s a plastic strip embedded in $20 bills that lets the government track them and count how much money is in your wallet by sending a signal to a satellite.
That’s an urban legend that dates back to at least 2001, but something a bit like it is coming true in Saudi Arabia.
Researchers at the Functional Nanomaterials and Devices Laboratory at the King Abdullah University of Science and Technology are working on putting a memory device into banknotes.
As in the U.S. $20 bills, the purpose is to cut down on counterfeiting, but unlike them, the Saudi system has some electronics on board. It can send an RFID radio signal to a scanner, and it contains some nonvolatile memory that can store a record of it every time it’s scanned.
The research is described in a new article in the Wiley journal Advanced Materials. The article is titled “High-Performance Non-Volatile Organic Ferroelectric Memory on Banknotes,” and its authors are M.A. Khan, Unnat Bhansali, and Husam Alshareef, who is my guest today. He’s an associate professor at King Abdullah University of Science and Technology. His area of expertise is materials science, the field in which he earned a Ph.D. from North Carolina State University in 1995.
Husam, welcome to the podcast.
Husam Alshareef: Thank you very much, Steven
Steven Cherry: Husam, this is essentially a chip that’s small enough to be embedded in a banknote, and it can survive all the wear and tear that paper currency is subjected to. Your group is working on one part of it—the chip. But I gather the trickiest thing was getting the materials right.
Husam Alshareef: Correct. So, this project actually is quite ambitious because, as you mentioned, the banknote gets a lot of bending, a lot of folding, and it is necessary to get the right materials to actually be able to handle all kinds of bending and flexibility in these banknotes, as you know. So, usually the RFID tag has three components: It has, actually, the antenna; and it has—if you will, just to simplify things—transistors; and it has a memory part of it, a logic and memory part to it. So in the first part of this work, we actually worked on the logic and the memory. The antenna part has not been implemented yet, but we are currently working on that. So these materials, to be maximally flexible, actually, were all organic. So, what we set out to do is an all-organic...including the contacts, including the transistors, the capacitors, all these circuit components had to be organic to withstand the significant amount of flexibility that is needed for handling a banknote.
Steven Cherry: I guess heat would have been an issue with silicon as well.
Husam Alshareef: Yes. So to be able to do it in silicon, you would have to heat, as you know, to very significant temperatures—you know, hundreds of degrees Celsius. But in the case of these materials, they can actually be processed at very low temperatures—like 140 C or below, the banknote, we have shown, can handle.
Steven Cherry: And in the U.S., banknotes are even designed so they can survive being laundered—you know, put in a washing machine or something.
Husam Alshareef: And ironed as well, actually. The students, before they actually started building the devices, what they did, we went through a rigorous exercise of dipping the banknote in different solvents, ironed it directly with an iron, actually, heating it to different degrees, and actually they are more tenacious than I thought.
Steven Cherry: Very good. Now, the chip draws power from the scanner when it’s scanned. I guess that’s not uncommon, but it means the power requirement has to be very slight.
Husam Alshareef: Exactly. And this is basically why we set out to implement the ferroelectric memory, because the ferroelectric memory is nonvolatile. That means it does not need to continually be refreshed, as in DRAMs, for example. DRAMs, they need to be continually refreshed with an onboard source of power, otherwise you lose your information. The ferroelectric memory is unique, in that sense; it does not need any refresh, so you don’t need power. So whenever the power comes from a signal, that can be used to read or write the memory.
Steven Cherry: The Saudi riyal is a very pretty, colorful currency. Does this change the appearance at all?
Husam Alshareef: It can. This is one thing that one has to optimize. If you look at in our publication, actually, because in this case we did use the top layer as gold contact, and the thickness of the layers may have not been optimized. Intrinsically, the layers we used other than gold are transparent, so in principle, when all the thicknesses are optimized and the roughness of the layers, you should not be able to see this by eye.
Steven Cherry: It takes money to make money, and you just mentioned gold. I gather you would like the final system to not use gold at all?
Husam Alshareef: Yeah. Actually, we are already at a stage where we do not need to use gold. But if you look in the paper, we have devices that use both to try to make the comparison. Gold will ultimately not be used in those. And also, we are thinking these things could probably be for the higher-value currencies, you know, instead of, like, small denominations.
Steven Cherry: Now, exactly how will this system cut down on counterfeiting?
Husam Alshareef: So, there have been many ways to try and combat counterfeiting. A lot of them, like the holograms, like the raised features on the banknote and maybe embedded threads of metal, you know, all these things—we’re all familiar with those. We thought with electronic counterfeiting, actually, it may be more difficult to reproduce. For example, you could make raised text, but if you have a unique number that is actually encrypted for each banknote, then it would be very difficult to reproduce that, we thought.
Steven Cherry: That could be done without the nonvolatile memory, right?
Husam Alshareef: You could. Right now, the RFID tags are using what’s called EEPROMs. It’s a different kind of memory, but its disadvantage is that it has limited endurance, which means the number of times you can scan will be limited, but it also consists of complex circuits. You will need multiple transistors to implement a bit of memory. The beauty of these ferroelectric memories is that you can have a single transistor actually for the memory, which is a much simpler circuit. Also the EEPROMs that are currently used are slower—not that you need extremely high speed for this application, but still, the ferroelectric memory would give you an advantage in terms of speed.
Steven Cherry: Once there are enough bills and scanners out there, and bills routinely have this embedded in it and are scanned at every transaction—which is in principle possible—there’s the potential for quite a database for how money changes hands. Would this let a government go after drug traffickers, for example?
Husam Alshareef: Absolutely, especially in large denominations. I think the circuit can become as complex as you want it to be, right? I mean, in its simplest form, the circuit is not very expensive, and it can be quite simple to implement, actually, but if you want inventory and tracking and positioning like this, you could basically add functionality to the circuit. And this could also again be done with organic materials and go beyond knowing [whether] this is just a bona fide, government-issued banknote or not. You could go into tracking and locating banknotes and so forth; it’s just you design the circuit you want to appear on the banknote.
Steven Cherry: And so it could also help a government go after the so-called “gray economy,” like waiters and cab drivers and everybody else who works mainly in cash and might be underreporting their income.
Husam Alshareef: It could, but that was not our intention. [laughs]
Steven Cherry: I mean, it could be used by a government for any reason, right? I mean, if a government wants to track who’s purchasing alcohol or something.
Husam Alshareef: Yeah, I think actually you’re bringing up some ideas we didn’t entirely think about, all of those. You know, counterfeiting is definitely one of the leading, but these are all interesting applications. And in principle, I would say RFID tags have many reasons and many areas of uses for RFID tags, and this could basically whatever one wants to do with banknotes with this robust circuit that requires low power and flexes very nicely and can be produced cost-effectively. You can address many of these things that you’re talking about because the cost will be reasonable.
Steven Cherry: Yeah, it occurs to me that the same advantages that it presents here would also make it the RFID chip of choice to embed into clothing and all sorts of other places.
Husam Alshareef: Yes, that is definitely ongoing, many people are interested in that, including government. Actually, I got invited to give a talk in Dubai about this. There’s a conference that happens there about IDs and security and printing on banknotes and things like that, so there’s interest in that.
Steven Cherry: Well, thank you very much. It’s a really interesting piece of research, and I wish you luck with it.
Husam Alshareef: Thank you very much, Steven. Have a great day.
Steven Cherry: We’ve been speaking with materials scientist Husam Alshareef about work being done in Saudi Arabia to create an embedded RFID chip for paper currency.
For IEEE Spectrum’s “Techwise Conversations,” I’m Steven Cherry.
Announcer: “Techwise Conversations” is sponsored by National Instruments.
NOTE: Transcripts are created for the convenience of our readers and listeners and may not perfectly match their associated interviews and narratives. The authoritative record of IEEE Spectrum’s audio programming is the audio version.