Broadband infrared LED for spectroscopy applications

The OSRAM Podcast: Episode #10 with Dr. Carola Diez


Welcome to the Photonstudio, the OSRAM Podcast. My name is Dieter Schierer and I am an OSRAM employee in the Digital Communication department. I am very pleased to present the new episode on infrared spectroscopy. Imagine you go into a supermarket, buy your food and there is this apple and you are not sure how fresh it is and how many nitrates and vitamins it has. So you reach for your smartphone, put the device briefly on the piece of fruit and a few moments later you get the information you are looking for via an app. Or you are somewhere on a trip around the world and have a headache. You get a painkiller, but you have doubts whether it is just a piece of chalk instead of the desired ibuprofen pill. After a scan, you can be sure that the active ingredient is indeed ibuprofen. What sounds like science fiction is in reality the description of the spectroscopic procedure.

The method is not so new and is used in countless laboratories every day. What might be new, however, is the fact that your smartphone could be that laboratory. But how does it work and what does that have to do with light and OSRAM? To answer this question, I invited my colleague Dr. Carola Diez from Regensburg to the Photonstudio today. Carola Diez is the product manager responsible for infrared emitters and near-infrared LEDs at the Opto Semiconductors business unit. I am very pleased to discuss this exciting and futuristic topic with her.

Dieter: Carola, thank you for taking time for us today and welcome to the Photonstudio!

Carola: Hello Dieter, thank you very much for the invitation to the Photonstudio!

Dieter: You are working with us in the business unit Opto Semiconductors in Regensburg. What is your exact job there?

Carola: I am a product manager in the product line Sensing. In this function, I am responsible for a very broad portfolio of infrared LEDs that can be used in a wide variety of applications. One example is LEDs that are used in surveillance cameras for lighting, for example to enable night vision. In this way we can help to make our environment safer. But I am also responsible for a second, quite exciting portfolio. This includes the so-called near infrared broadband LEDs. This is a somewhat unwieldy word, so we always call it NIR-LED. This LED has the potential to open up completely new fields of application and use cases.

Dieter: Before we come to the main topic, I would like to clarify an essential question. What is the difference between a near infrared broadband LED and a normal LED?

Carola: The near-infrared broadband LED emits light over a wider spectral range. Here the LED emits light from 650 to 1050 nanometers. That means we have a spectral width of 400 nanometers. A classic infrared LED, which typically emits at 850 to 940 nanometers, has only a bandwidth of 40 nanometers. This means that our NIR LED emits ten times as wide as a classic IR LED.

Dieter: I would like to briefly read out a quote from a press release from 2018: "Measuring how fresh the vegetables are and how sweet the strawberries are in the supermarket, calculating the calories of lunch in the canteen or checking whether the supposed headache tablet really is one - in future, consumers will be able to test this themselves quite easily with their smartphone. This is made possible by the development of broadband infrared LEDs (IRED), which emit in a broad wavelength range. It all sounds a bit like science fiction to me now. But I have the feeling that we're not that far off now from having these gadgets in our smartphones soon. Can you explain to us how it's all going to work?

Carola: Let's imagine there's a red tomato on the table in front of us. When lit by daylight, it absorbs all colors except red. The red color is therefore reflected and we perceive the tomato as red. In other words, the tomato's property is to reflect red light when it is illuminated with visible light. If we illuminate the tomato with invisible light, i.e. infrared light, the tomato also absorbs parts of the light and reflects parts of the radiation. Depending on how the tomato is constructed, we obtain a unique molecular fingerprint of the tomato. A tomato consists of a lot of water and water absorbs light in a radiation of 970 nanometers. We cannot see this radiation with our eyes, but we can use tools to see it. In this example, this would be a so-called spectrometer. This takes up the measured radiation and looks at wavelengths selectively. Downstream there is a software that processes the information from the sensor and compares the recorded spectrum with a database. The database contains spectra of different tomatoes. These are then compared, so that a statement can be made, for example, how high the water content of the tomato is. And the water content is an indication of how fresh the tomato is. The higher the water content, the fresher the tomato.

Dieter: Do I understand correctly that we do not need a fabric sample for this procedure? So I don't have to cut any piece out of the tomato and have it analyzed by a device, but I just put the device on the surface or how does it work?

Carola: Exactly, the whole thing works purely optically. You simply place the device on the tomato or measure from a small distance. You send the light there, the light interacts with the object and the reflected light is then recorded. This means that nothing is destroyed and you can measure it really non-destructively.

Dieter: But if we put the smartphone or gadget on the tomato, in my opinion only the surface is scanned. Don't we have to look into the depth of the tomato or does that happen automatically?

Carola: It actually happens automatically. The NIR radiation penetrates a little bit into the surface and interacts there. That is then a few millimeters or even centimeters.

Dieter: There are many spectroscopic methods, such as microwave spectroscopy, X-ray spectroscopy and laser spectroscopy. Why is near-infrared spectroscopy of all things used here and not laser spectroscopy, for example?

Carola: The spectroscopy methods you mentioned are all methods that are mainly used in special laboratories. The measuring devices usually cost several 10,000 Euros and some require special training to be able to work with them. The nice thing about near-infrared spectroscopy is that the technology has now reached the point where there are spectrometers that are getting smaller and smaller and cheaper. In the last few years we have seen that well-known sensor manufacturers, but also smaller startup companies have developed miniature spectrometers, which can be used in the broad masses at a reasonable price. But the sensor alone is not enough, as discussed before, the light source is also crucial. Here it is very important to have a very good and well defined light source. For the laboratory devices, which have been available for decades, a halogen light source is used. But it is quite big and unwieldy. If you want to make everything smaller, you also need a smaller light source for your spectrometer. And OSRAM has closed this gap by developing the first broadband LED. This means that the market now has a technology to develop new use cases. In addition, the radiation is absolutely harmless in near-infrared spectroscopy. Really everybody can handle it, which is probably not the case with X-rays.

Dieter: Some time ago your colleague Christoph Göltner was a guest of mine in the Photonstudio. We talked about emitters, green LEDs and photodiodes in Smartwatches and Fitness Trackers. Among other things, we talked about the fact that the new generations of fitness watches not only contain one module, but several modules. And he explained to me that the signal-to-noise ratio is very important here. In other words, if I have several modules that measure simultaneously, the best signal is selected and compared with the database. How important is this topic for near infrared spectroscopy? Are there also these problems or is it harmless for this area?

Carola: In principle one can say that wherever radiation is emitted and measured again, one needs a very good signal-to-noise ratio to be sure that the measured signal is correct and no interferences influence the result. Therefore it is also valid for NIR spectroscopy. In the example of the tomato it is very important to be sure that the signal is coming from the tomato and that you do not measure ambient light because you have placed the sensor in the wrong position. Therefore it is very important that we have a light source with a very high radiant intensity to filter out the ambient light later, but also that we have a sensor with a high sensitivity and a good spectral resolution.

Dieter: When the scanning process is finished, the result will be compared with your database. Who actually makes this database? Where does this information come from?

Carola: This is actually a very good question. We are of course the component manufacturer. We focus on having the best LED on the market to drive the issue forward. Regarding the software and the database we have a large partner network that we rely on and with whom we all try to push these applications and use together. So we do not do it ourselves, but we have partner companies that work together and support us.

Dieter: About two weeks ago we unveiled our latest broadband emitter for near-infrared spectroscopy, this OSLON® P1616. This is the smallest LED on the market suitable for near infrared spectroscopy. What else is special about this OSLON® LED?

Carola: It is actually super small, i.e. 1.6 x 1.6 millimeters, and can therefore be used for a very wide range of applications. But the much more important feature is the very high radiant power. With this new development, we have managed to significantly increase the radiant power. Compared to our predecessor products, the radiant power has been increased three times. We have also managed to design the emitted spectrum in such a way that the performance is good and constant over the entire spectral range.

Dieter: This reminds me a bit of the green gap phenomenon with green LEDs, i.e. a drop in power above a certain wavelength in the green spectrum. But at 650 nanometers, we are already much further along, aren't we?

Carola: Exactly, that is much further. With green LEDs, the light comes directly out of the semiconductor material and is emitted directly. In near-infrared spectroscopy, where you want a wide range of wavelengths, we use a classic blue LED. And on top of this blue LED we put a phosphor material. This phosphor material absorbs blue light and then converts it to another spectral range. This means we start with blue light at 450 nanometers and convert it to the near-infrared spectral range of 650 to 1050 nanometers. We have a super wide spectral range here and the special thing is that even at high wavelengths, above 950 nanometers, we still get a lot of light. We have found a very good converter material that converts very efficiently and emits a lot of light above 950 nanometers.

Dieter: I can see from your stories that a lot of research has been done on this. Especially with the phosphor layer, the quality must be 100% right. Even the smallest deviations would otherwise not be able to achieve the performance in the LED, would they?

Carola: The whole phosphor development takes place internally and we have a very competent team that is really on the pulse of time. Three years ago we were the first on the market with this near infrared LED and have already won an innovation award for it.

Dieter: I can imagine that there would be many interested people who would like to know if their food is still fresh. That is part of the consumer sector. What other examples are conceivable?

Carola: Near-infrared spectroscopy is a very powerful technology and can be used in many areas, including industry. In agriculture, for example, it can help farmers determine the ideal time to harvest. Or in the food industry, it can also help to support quality control. In fact, it can be used wherever you need to know more about materials and their composition.

Dieter: Don't farmers already determine the perfect time for harvesting with soil or grain samples? And what benefit will farmers have when they use OSRAM technology in the future?

Carola: Professional farmers will of course analyze their soil or grain. But at the moment they may take a sample every few weeks and send it to an external laboratory. There the sample is then analyzed. That takes a few days and also costs a lot of money. A week later they will have the result and will only know the quality of this one spot. If a farmer now had a gadget like this that he always has with him, he could really take a measurement at any time and any place and would receive the result directly.

Dieter: And what about drugs? Can they be checked with near infrared spectroscopy?

Carola: You can do that as well. Of course you have to make some concessions. In terms of resolution, it's not as good as an expensive laboratory device that costs a few 10,000 euros. That's clear, but you can still draw initial conclusions from it.

Dieter: I can imagine that one simply wants to know at the pharmacy whether a tablet is, which helps against headache or only a piece of chalk.

Carola: I would already trust my local pharmacy. But when I think of online shopping and buy an expensive leather handbag, I naturally want to know whether it is really the brand I ordered or whether it is a cheap fake product. And if there is a corresponding app on the smartphone to check this out, that is of course wonderful.

Dieter: Super interesting topic! Where is the development in this area heading and what are the current challenges you are facing?

Carola: On the hardware side, the goal of development is to improve the light source and increase the radiant intensity. Also on the detector side the sensitivity has to be improved to enable applications and use cases. But also on the software side there is a lot to do. For each material analysis an appropriate database is also necessary. Here hardware and software companies have to work together to advance the topic.

Dieter: Thank you very much for your time and for these insights. And of course, good luck to you and your colleagues. I am looking forward to the next products!

Carola: Also from my side many thanks for the invitation to the Photonstudio. It was a lot of fun.

In today's episode of Photonstudio, Carola explained to me how near-infrared spectroscopy works, what role light plays in it and how diverse the possible applications for this method are. As always, you can listen to this episode and all other podcast episodes on iTunes, Spotify, Soundcloud, Google Podcast and many other platforms. Have fun with it and see you next episode in the Photonstudio!

Broadband infrared LED for spectroscopy applications