NDE 4.0 Podcast Transcript
Episode 18 — How Terahertz technology will influence NDE 4.0
Our Guest: Michael Pepper, Terraview
Editor’s Note: In the interest of time, this transcript is still in rough format and has not been edited for proper grammar and punctuation. If you have a need for a fully edited transcript, please contact us.
[00:00:00] Nasrin Azari: Welcome to Floodlight Software’s podcast, where we interview various experts in industry 4. 0 concepts, issues, and technologies for non destructive testing and inspection. This show is the place to go to learn about the biggest challenges and opportunities around NDE 4. 0 from some of the smartest people in the industry.
[00:00:34] Nasrin Azari: So, sit back and be prepared for a really thought provoking discussion. Hope you enjoy the
[00:00:39] Intro: episode.
[00:00:51] Nasrin Azari: Hello, everyone, and welcome to today’s episode of Floodlight’s NDE 4. 0 podcast, where we pose five questions to NDE 4. 0 [00:01:00] experts in order to understand and advance the world of industrial infrastructure safety and inspections. Today, we are honored to be speaking with Sir Michael Pepper, who is probably the most accomplished individual we have ever interviewed for this podcast.
[00:01:12] Nasrin Azari: He even has a hefty Wikipedia profile. Sir Michael has straddled both academia and industry for over 45 years with an astounding list of impressive achievements. As a physicist in the 1970s, he performed research on localization of in semiconductor structures, then collaborated on the paper that eventually brought a Nobel Prize for the quantum Hall effect to Klaus von Kleitzing in 1980.
[00:01:38] Nasrin Azari: Later, in 1991, Sir Michael founded the first corporate research laboratory of Toshiba outside Japan, now known as the Cambridge Research Laboratory, and was appointed its managing director. His work at the Cambridge Research Laboratory included the development of quantum communications and the pioneering development of terahertz radiation emitted by excitation of [00:02:00] semiconductors where it was shown that terahertz could distinguish between cancerous and healthy tissue for medical applications.
[00:02:06] Nasrin Azari: In order to rapidly develop this technology, it was spun into a corporation called TerraView, where Pepper is currently the scientific director. He has been closely involved with the development of new applications of terahertz in industrial process control, security, and medicine. He became an honorary professor of pharmaceutical science.
[00:02:25] Nasrin Azari: In the University of Otago, New Zealand in 2003 and left his Cambridge chair position to take up the pender chair of nano electronics at University College, London in 2009. He has been associated with many developed. Developments in Semiconductor Physics and Applications of Terahertz Radiation and sits on the Scientific Advisory Committee of Australia’s ARC Center of Excellence in Future Low Energy Electronics Technologies.
[00:02:50] Nasrin Azari: In terms of honors and awards, Sir Michael was elected a Fellow of Trinity College, Cambridge in 1982 and a Fellow of the Royal [00:03:00] Society in 1983. He received the Europhysics Prize of the European Physical Society and the Guthrie Prize of the Institute of Physics in 1985, and then received the Hughes Medal in 1987.
[00:03:12] Nasrin Azari: In 2000, Pepper was awarded the first Mott Prize by the Institute of Physics and was awarded the Royal Medal in 2005 for his work, which has had the highest level of influence in condensed, condensed matter physics and has resulted in the creation of modern field of semiconductor nanostructures. He gave the Royal Society’s Bakerian Prize lecture in 2004 and received a knighthood in the 2006 New Year’s honors list of services to physics.
[00:03:39] Nasrin Azari: He was appointed a fellow of the Royal Academy of Engineering, and in 2010, he won the Swan Medal and Prize. He also has been awarded the 2013 Faraday Medal of the IET, the Institution of Electrical engineering and technology. And in 2019, he was awarded the Institute of Physics, Isaac Newton medal. [00:04:00] That’s so that’s a huge list of accomplishments, Michael.
[00:04:03] Nasrin Azari: It’s a great honor and privilege to welcome you to our pot, to our show today. Thank you for joining us.
[00:04:09] Michael Pepper: Thank you very much. It’s a great pleasure to be here to discuss some of the recent work that my colleagues and I have been performing on terahertz technology and its application to NDE.
[00:04:21] Nasrin Azari: Fantastic. So I’m really looking forward to this conversation today because I know I’m going to learn a lot.
[00:04:27] Nasrin Azari: Although your research and experience is very widespread, today we’re going to focus on the The work you mentioned that’s being done at TeraVue, particularly related to non destructive testing for industrial manufacturing and infrastructure. So let’s get started with a basic question to help set the stage for our audience.
[00:04:44] Nasrin Azari: Can you describe what Terahertz technology is, how it started and, or how it was discovered and how it promotes NDE 4. 0?
[00:04:53] Michael Pepper: Yeah, it’s a pleasure to do so, actually. The Terahertz radiation, actually, in terms of [00:05:00] its Frequency. It sits between the microwave and the infrared. And, um, so that’s basically, you know, if you go down the periodic table for spectrum analysis for light, then you will see obviously about visible light, which is very short wavelength.
[00:05:18] Michael Pepper: Uh, even shorter would be x rays, and then as you increase the wavelength, uh, it’s, you’ll actually come to infrared, and then there’s the terahertz. And even longer than the terahertz is the microwave. Now, terahertz radiation, of course, has always been around. Don’t bother about that. You get a lot of terahertz from a traditional light bulb, uh, basically because that emits the entire spectrum of electromagnetic radiation, of which light is particles.
[00:05:48] Michael Pepper: But it was only in the 1980s and 90s that in fact coherent sources of terahertz were developed. And that means that [00:06:00] essentially the wavelength is more or less uninterrupted. It’s not, it doesn’t randomly change. And because it’s continuous, that means we can investigate phenomena such as interference and diffraction and essentially coherent reflections.
[00:06:15] Michael Pepper: And that’s given rise to the power which terahertz now has and why it’s been exploited in NDE and other applications. And it’s possible to generate it. Because the thing is that the terahertz radiation has energy, which is rather low, much, much less than the energy of room temperature radiation. And that means, for example, that we can’t have something like a television remote control where you use light to actually control the television or produce laser light, because it’s not possible to have the energy less than that.
[00:06:52] Michael Pepper: So we have to basically develop it by exciting currents [00:07:00] in semiconductors, which only persist for a very short period of time. And because they are rapidly terminated by the lattice, we get a current spike, which is extremely short, and then generates the radiation.
[00:07:17] Nasrin Azari: Interesting. Um, so. In terms of, of the industrial world and industrial settings, how is Ertz used today?
[00:07:29] Nasrin Azari: By industrial settings, by which sectors, and what do you see as its greatest future potential in NDE?
[00:07:37] Michael Pepper: Well, the beauty of TZ for NDE is that a lot of materials that most are transparent to it. Uh, so unlike visible light, which will obviously be reflected. and can’t penetrate. The terahertz can. Now infrared of course can penetrate, uh, but it’s very heavily scattered.[00:08:00]
[00:08:00] Michael Pepper: You know, the classic case would be, why, why essentially is the sky blue and the sun red? And the reason for that essentially is that the infrared is really just scattered all over the place by particles in the atmosphere. And so, you know, we get that, um, the visible light will come through, and so we see a red sun.
[00:08:22] Michael Pepper: Uh, now, the microwave is barely absorbed. It tends to be absorbed by water. Of course, 10 hertz is also absorbed by water. But the beauty of it is that the radiation goes in, and every time there’s a change in the optical medium, that is a change in the material, it’s reflected. It goes straight in and straight back.
[00:08:42] Michael Pepper: And then we can analyze the reflections. Because they come at different times. And so we can build up a three dimensional image. And, um, which is very important, because then we can reconstruct the object. We can see if there are any deficiencies, any, for example, cracks, if you’re [00:09:00] looking at something below a coating.
[00:09:02] Michael Pepper: And a general analysis, which is very important in many industrial sectors.
[00:09:08] Nasrin Azari: So do you, what kinds of, um, applications are being used today, uh, through TerraView, um, in. in the industrial sectors.
[00:09:18] Michael Pepper: Well, perhaps I can give a couple of examples. Yeah. One of the first applications that we looked at when we founded TerraView was actually the pharmaceutical sector.
[00:09:30] Michael Pepper: Uh, you’ve seen adverts, uh, which are all over the place. 12 hour relief for this, or 24 hour relief for that, mainly colds, flu, but also for much more serious ailments. You want, uh, controlled release over a long period of time, whereas some, uh, problems are best solved by an instantaneous, uh, release of the active ingredients, such as, so it’s dissolved immediately in the stomach.
[00:09:54] Michael Pepper: If you want to get rid of a headache, for example, take something for that. And then that’s released, but [00:10:00] other illnesses, you want a more gradual release and that’s achieved within a tablet by having a large number of internal coatings. And every time a coating dissolves, so an active ingredient comes out, uh, which helps remove the, the ailment or illness, but you want to do that over a period of time.
[00:10:21] Michael Pepper: So the internal coating takes a bit of time to dissolve. And that’s very well controlled. And we can actually analyze the internal coatings within the tablet. Uh, the terahertz will go straight through. It can be absorbed, uh, after about a centimeter or so, but certainly tablets are much less thick than that.
[00:10:42] Michael Pepper: So we can analyze the internal coating. And that’s a very similar application to the automotive sector, where we’re developing equipment which is being installed by car companies. To control paints. So a typical volume car will have four layers of paint [00:11:00] essentially to prepare the, two of them on, to prepare the final, the colour coating, which is called the base coating, and then on top of that is a clear coat which protects it.
[00:11:12] Michael Pepper: But they want to control that, and they want to be able to see if there’s any problem in the adherence of the paint to a curved surface, for example, where windscreen is, or if there might be any corrosion beneath it. Now, we can actually analyze that, we can analyze the coating thicknesses. Which are typically of the order of 15 to 25 microns in thickness.
[00:11:36] Michael Pepper: Where a micron is one ten thousandth of a centimeter. So they’re very thin. But we can analyze that. And so we can see if there are any problems involved, actually. And they can control the thickness. And also monitor, essentially, the quality of the paint layer on the car. Because one thing they would like to avoid is having any recourse.
[00:11:57] Michael Pepper: So if, for example, paint goes down over. a [00:12:00] corroded region, or if, or if in fact one particular layer of paint is particularly thin, then obviously it could be that after a month or so the car is recalled, uh, because the owner is not satisfied with it and it’s under warranty. So obviously we want to avoid that, so the Terahertz is excellent for that.
[00:12:18] Michael Pepper: Now, the traditional technology which has been used is a form of ultrasound, rather like we have a medical examination of ultrasound, and there the, what you might call the head, the sensor head. is actually placed in intimate contact with the object. So, you don’t want to do that. You’ve got a nice shiny car coming out of the paint shop.
[00:12:38] Michael Pepper: You don’t want to put oil or water on the surface and then clamp a head to it. It’s the last thing you want. So, the terror house is beautiful in that regard because it can actually just sit. It’s moved in a robot and it goes up and down the car investigating particular, uh, points on the car. That, of course, determined unit bounds.
[00:12:58] Michael Pepper: And, and, and [00:13:00] actually it will then produce, uh, a map of the paint layers, so you can see if anything’s going on. So, you can see that essentially those two applications that I mentioned are not entirely dissimilar, that they’re based on the reflection of terahertz from a change in the medium. And, and it will apply to any form of coating.
[00:13:21] Michael Pepper: For example, helicopter blades are laminated, different layers. You can investigate those, and indeed have done so. And catalysis, where you put a layer on for protection. It’s exceptionally good for investigating surfaces and interfaces. Because of the sharpness of the reflection. And x ray itself, although of course has much greater penetrating power, is not sensitive to these small changes.
[00:13:50] Michael Pepper: For example, the different colors on a car. The x ray will just go straight through it, it won’t actually see the changes.
[00:13:58] Nasrin Azari: So it seems like, like [00:14:00] maybe detection of corrosion underneath a painted or a covered surface would be a very common or a very good application of this technology? Absolutely,
[00:14:10] Michael Pepper: definitely.
[00:14:11] Michael Pepper: A lot of companies are interested in it for that purpose. It’s interesting to note, actually, the very first application we did, when we first developed, uh, terahertz technology a long time ago, uh, was actually teeth. We got some old teeth from the dentist and we were able to see the surface layer of the enamel where it actually had disappeared and allowed corrosion to enter into the tooth.
[00:14:36] Michael Pepper: So it’s another example of actually being able to look at very thin layers. And as any dentist will tell you, the x ray will not pick that up actually. The x ray will pick up decay in the tooth, but won’t actually give you the region where the bacteria penetrated because the enamel had disappeared. So our other applications are essentially based around [00:15:00] that.
[00:15:00] Michael Pepper: For example, in the semiconductor industry, um, we can send in a terahertz pulse and we can see how it’s reflected and where it’s reflected in an integrated circuit. So these days, as we know, the emphasis is on miniaturization, getting as much computing power or capacity into a mobile phone, for example, numerous applications.
[00:15:25] Michael Pepper: And to do that, you have a system which is known as essentially a system in a package where you put multiple chips in a package. They’re very thin, and essentially you connect them one to the other. Now, problems can arise in the connection. It can be rather weak or non existent, so that’s an open circuit.
[00:15:45] Michael Pepper: We can send in the terahertz pulse and we can analyze the reflection and we can tell you whether or not you’ve got good contacts, whether they are weak and may give problems in the future or possibly non existent. Uh, so [00:16:00] it’s, it’s, it’s essentially a platform technology with multiple applications.
[00:16:04] Nasrin Azari: Yeah, that’s, that’s, that’s fantastic.
[00:16:07] Nasrin Azari: It’s really interesting. Um, is there a relationship between terahertz technology and AI? Yeah, definitely. Yes,
[00:16:15] Michael Pepper: um, there will be actually at the moment. I mean, essentially, it’s not so much AI as more really massive, fairly simple, basically, applications in which the, which the robot arm, for example, going back to cars, the robot arm can be moved to analyze different regions.
[00:16:35] Michael Pepper: But AI will come in, uh, particularly in where you want to actually follow up, uh, some area of weakness that the terahertz is actually identified. And then it will actually analyze that and say what it should do next. So it will replace the operator. So the operator at the moment might notice some problem, uh, either in, um, different applications in semiconductors [00:17:00] or in paints or, or actually even in, in medicine because, um, the terahertz was the first non ionizing radiation to be able to detect the difference between cancerous tissue and healthy tissue.
[00:17:15] Michael Pepper: We did that, well, it’s one of our first applications as well. But medicine, you know, it basically, one has to have 100 percent certainty. And at the moment, a large number of groups are working on that. So, however, the point is, what I’m saying really, is that you, you, the, the, the Terahertz equipment will notify something, will measure some anomaly.
[00:17:36] Michael Pepper: And from then on, one might have essentially a program, which then takes it over and looks for something else. So it will then be able, on its own, to present a complete description of a particular problem with an object. One, how the object is faring. So from that point of view, yeah, it has a great future in terahertz technology.
[00:17:57] Nasrin Azari: Sure. So, so it’s almost like a [00:18:00] progression of analysis. You analyze one component. Exactly. Notice an anomaly of some sort and then choose a path based on some, um, uh, smart algorithm that might want to look at a further on, um, area. Interesting. That’s right. Yeah.
[00:18:22] Michael Pepper: We go beyond algorithmic. Detection.
[00:18:26] Nasrin Azari: Well, it seems like AI has applications to just about everything.
[00:18:30] Nasrin Azari: Absolutely. It sure does. So, and how about, one of the, a couple of other topics I know we wanted to talk about are how terahertz technology is related to 3D imaging. I can kind of picture that, but I’d love to hear your description of that, um, digital twins and counterfeiting.
[00:18:49] Michael Pepper: Well, it’s actually tailor made for, for three dimensional imaging, because that’s what we do.
[00:18:54] Michael Pepper: Going back to the pharmaceutical tablet, we produce a digital image of the tablet, and it’s [00:19:00] stored, and so it can be reconstructed, uh, from different angles. You can rotate it, the image around, uh, because everything is there. Basically, the Terax technology has dug everything out of it. And so digital printing is essentially what we’ve been doing, except it hasn’t been digital printing.
[00:19:19] Michael Pepper: Uh, so small objects, for example, in the nano region will be tailor made for 3D printing, for example. They’re in conjunction with terahertz. is tailor made again for the digital printing. And indeed, if you were to monitor some particular piece of equipment in industrial process control, as it develops, as time goes on, then essentially that’s a digital threat.
[00:19:48] Michael Pepper: So both digital threading and digital twinning are essential part of NDE4. And so the terror host will be the heart of it. That’s essentially the position that we’re [00:20:00] hoping to move towards. And I think there’s a great realization that this isn’t currently. Yeah. Um, I haven’t mentioned, I just mentioned the medicine.
[00:20:09] Michael Pepper: Again, you know, surfaces and interfaces. are very susceptible to investigation by terahertz. There’s also terahertz spectroscopy, which actually is only being used by research institutions at the moment. It hasn’t really percolated into the industrial arena, but what it does, it helps you to identify particular compounds.
[00:20:32] Michael Pepper: You look at the different frequencies and how they’re absorbed. And from the frequency dependence of the absorption, you can tell what you’ve got. And so, uh, we’ve used that to look at the difference between counterfeit tablets and real tablets, because you can analyze the internal structure. And so, uh, you know, the counterfeit tablets, the ones which are not so good, uh, but are sold under [00:21:00] the incorrect label, uh, they look the same from the outside, but the inside, of course, is completely different, and we can detect that.
[00:21:07] Michael Pepper: So counterfeiting, digital twinning, are all wrapped up in the three dimensionality of the terahertz investigation.
[00:21:15] Nasrin Azari: And I imagine that the technology is pretty, um, fast, meaning that you could Essentially test every single tablet if you wanted to, to ensure sort of 100 percent safety versus, um, just sort of batch checking a few, right?
[00:21:33] Nasrin Azari: Yeah, you could do that. Absolutely. Yeah. Now, when I, when I first saw that word, I was thinking about money, counterfeiting of money, not tablets, but it makes a lot of sense the way you’ve described it, . Yeah,
[00:21:44] Michael Pepper: yeah. That’s right. Yeah. Yeah. Bank notes, for example, have been suggested. We haven’t really done much work on that, uh, that, but I, 1, 1 1, we’ve investigated so many different topics.
[00:21:56] Michael Pepper: One always forgets them, but for example, one of them was crisps. [00:22:00] Uh, um, or potato chips as they’re known in the, in North America, that you can buy a salt and vinegar or plain or barbecue or whatever, whatever. And what happens there is that the chips are made, uh, and then they’re dipped into vats, uh, containing something which gives you a surface layer with the different tastes.
[00:22:23] Michael Pepper: And so, we can tell the difference between plain crisps and salt and vinegar from terahertz spectroscopy. Without
[00:22:30] Nasrin Azari: tasting, without actually tasting them. Yeah,
[00:22:33] Michael Pepper: we did actually, we did actually investigate. Again, I mean, we have actually sold equipment which is used, uh, for looking at foreign objects in rice.
[00:22:43] Michael Pepper: Oh, wow. See, I mean, you know, rice comes along and sometimes you get bits of other things, probably stone or something like that. We can tell that, yes, it has actually been used for that purpose. Again, and [00:23:00] we’ve investigated Stonehenge at the request of a research institution, uh, because the stone can be investigated under the layer of moss.
[00:23:09] Michael Pepper: And of course, you can’t touch anything at Stonehenge or anything like that, absolutely not. So, but the terahertz can, the head can sit a centimeter or a bit more away from the stone, and it will go, the radiation will go through the moss, and you can analyze what’s underneath it. Or paintings for that matter.
[00:23:29] Michael Pepper: You read somewhere someone’s got an old family painting in their attic and turns out underneath, if they’re lucky, they might have a Rembrandt. But we can see what’s underneath it without having to touch it.
[00:23:42] Nasrin Azari: Such a wide range of different applications of the technology. How fun it must be to work in this world.
[00:23:49] Michael Pepper: Yeah, the, the, I mean, difficulty, of course, for a small company is, uh, what do you concentrate on? Yeah. You haven’t got the resources to do everything. Right. So, yeah, so that’s, that’s why, yeah, [00:24:00] essentially, yeah. Yeah, we understand that. As you say, yeah. It’s, there are so many applications and they open you up.
[00:24:07] Michael Pepper: Right.
[00:24:07] Nasrin Azari: Oh, fantastic. So, let’s, let’s finish up with the final question today, which is, what are the limitations of the technology and how might folks overcome those limitations?
[00:24:19] Michael Pepper: Well, we are working on actually increasing the throughput, that is decreasing the time in which it takes, uh, takes to do a measurement, because obviously, again, referring to cars or large objects, you want to measure so many different points, you want to do this very quickly.
[00:24:39] Michael Pepper: Um, and, uh, so we have to sort of speed the process up with continuous improvement, basically. You know, there isn’t a particular point. There are always suggestions coming up that we can actually increase the speed in which the measurements are taken. And you want the flexibility of the, uh, measurements.
[00:24:56] Michael Pepper: And also, of course, the software, as we’ve been talking about AI. [00:25:00] So from that point of view, it’s also extending it to the range of frequencies and combining it in some applications with other modalities. Um, and in order to actually just really enhance the power of the technology in completely characterizing by an object.
[00:25:20] Michael Pepper: So you’ve got an object there. I mean, you’ve got the digital twinning, of course, but other aspects of it you want to actually incorporate into a complete picture. It’s rather like what’s happening in, say, the investigation of people, uh, by modern molecular biology. You know, the. The basic essentially investigations of your molecular structure enable a complete picture to be developed.
[00:25:46] Michael Pepper: Well, we want to do that with basic inanimate objects, uh, and to do that. And that’s what is, is happening. Uh, and, um, it’s going very well actually.
[00:25:57] Nasrin Azari: What about the, the cost of a [00:26:00] system? Like if, if somebody is interested in using the technology for You know, maybe testing their products prior to release. For example, what would it cost?
[00:26:11] Nasrin Azari: Is the cost pretty comparable to other similar types of solutions? Or how does that play into the sort of productization of the technology?
[00:26:22] Michael Pepper: Well, at the moment, the principal driver of cost at the moment is the fact that in order to generate the terahertz, we have to use a laser, and lasers are not cheap, it’s not much a laser diode, but slightly larger laser, and they’re not cheap.
[00:26:39] Michael Pepper: So the cost of the equipment to a purchaser would be in the thousands, a few thousand US dollars, rather than hundreds. And, uh, we are also, of course, attempting to reduce that cost considerably. I think one can say the problem is, uh, in reducing cost is that we need [00:27:00] a pulse of terabytes to come out. So you inject a pulse in and you analyze the reflection.
[00:27:05] Michael Pepper: Whereas, if one uses a continuous distribution, of radiation, then it’s much, much cheaper, actually. There are ways of doing that, but the analysis of it is not so powerful as if you were to use the pulses. So what we’re attempting to develop mathematical theorems, which actually in conjunction with hardware enable us to use the continuous wave more than is the case at the moment.
[00:27:31] Michael Pepper: And that will actually make a lot of But in general, it’s actually, it’s, it’s getting much cheaper all the time. And of course, the thing is that once one starts selling in large numbers, very large numbers, then of course there will be economies of scale and the price will come down and down. But it has actually decreased considerably over the last few years.
[00:27:55] Michael Pepper: Okay,
[00:27:56] Nasrin Azari: that makes a lot of sense. Um, and I’m sure we could go on and [00:28:00] continue talking about other, other interesting aspect, aspects of terahertz technology. Um, but let’s close for today. Michael, I cannot thank you enough for participating in our podcast today. It was such a fantastic and educational session for me to learn about terahertz technology.
[00:28:16] Nasrin Azari: So thank you so much for participating.
[00:28:19] Michael Pepper: It’s a pleasure. Thank
[00:28:20] Nasrin Azari: you very much. And I know that I was probably not the only person who learned a lot today. So if you have been inspired by the topics we covered and would like to reach out to Michael yourself, or we have included some links to learn more about him and Tara View in our show notes.
[00:28:34] Nasrin Azari: If you have feedback or would like to nominate an individual organization to be a guest on a future episode, please send a message to me directly, or you can submit the contact us form on our website. www. floodlightsoft. com. Thanks again for joining us and see you next time.[00:29:00]
[00:29:00] Intro: To learn more
[00:29:00] Nasrin Azari: about NDE 4. 0, Emerging Technologies and Digital Transformation, please visit www. floodlightsoft. com for additional resources, including our blog and several relevant white papers. If you have any questions about today’s episode or suggestions for future episodes, please send an email to info at floodlightsoft.
[00:29:22] Nasrin Azari: com. Thank you so much.
For more expert views on NDE 4.0, subscribe to the Floodlight Software blog at floodlightsoft.com.