Check out our science

Discover the stories of researchers and students at the
Faculty of Mechanical, Maritime and Materials Engineering.

The paint reveals the master

In the end, even the most famous painting is just paint on a canvas, says art historian Joris Dik. Studying the paint can expose forgeries, magically conjure up hidden paintings, and allow you to eavesdrop while a Rembrandt or a Picasso practices his craft. ‘Da Vinci did not paint the shadows in the Mona Lisa’s face with a darker hue, but by stacking layers of almost transparent paint,’ explains Joris Dik, professor of materials in art and archaeology at TU Delft. Dik is an art historian, but also a materials scientist. A proponent of the arts in the world of science, but he feels completely at home there. ‘In art, it often revolves around the concept, the image, the idea and the beauty of a painting. In fact, many art enthusiasts actually find technical knowledge disconcerting. That has always surprised me. Because it’s the material that provides us with great insight into the creative process.’ In the past decade, Dik has made a name for himself by introducing the art world to X-ray fluorescence spectroscopy. ‘XRF’ can ‘dissect’ a painting without damaging it. The analysis technique individually reveals different chemical elements in the paint, often pigments. It turns out there is a woman’s face resembling the potato eaters hidden behind Van Gogh’s painting Pasture in Bloom (1887), for example. Van Gogh painted over it because canvasses were expensive. Rembrandts Saul and David (ca. 1660) appears to have been simply cut in half, Dik reveals. ‘Perhaps the paintings earned more money if sold separately?’ Saul and David were reunited, but XRF research has shown that they probably shifted a few millimetres from their original position vis-à-vis each other. ‘Every major museum has its own scanner now,’ Dik says. Nowadays they are important to confirm the authenticity of old masters. ‘Last year another forged Frans Hals was discovered.’ It concerned the painting Portrait of a Gentleman , sold by Sotheby’s for almost ten million euros in 2011. The canvas had surfaced in 2008 and after many art experts examined the painting, it was attributed to Frans Hals based on the typical brush strokes, the composition and the paint that was used. Your own synchrotron Painting research can be conducted with much greater precision with synchrotron radiation, emphasises Dik. ‘It enables you to see exactly where the elements are in a layer of paint. It also provides you with depth, a 3D image. Moreover, you can look at many more elements.’ Dik has already travelled once with a painting to the particle accelerator in Geneva, which generates high-quality synchrotron radiation. ‘But the measuring time is limited, of course. And you don’t want drag around valuable paintings too often.’ In two or three years Dik hopes to have a mobile ‘mini-synchrotron’ at his disposal. Dutch physicists from Eindhoven and Delft are building one, with the support of KNAW in a large research consortium called Smart*Light. The most important applications are in materials science and health care, but Dik is participating as well. ‘Art history is a small field and the world of art a small market. So I have to persuade technologists from completely different scientific disciplines that we’re an interesting niche for the new research technique.’ That’s why he appears regularly on TV. Dik is one of the experts in the popular television series and touring theatre show ‘Het geheim van de meester’ (‘The secret of the master’). And he can be frequently seen on shows such as DWDD, Pauw and RTL Night to talk about the authenticity of paintings. ‘It’s not that I have an urge to be on TV, but as an art historian you’re not building dikes or inventing new medicines, which are things everyone immediately sees the need for. You really have to continue explaining the importance of art history and promoting you work.’ Copyright photo's: Geheim van de Meester When Smart*Light is ready, Dik would like to study Pablo Picasso’s c ubist painting The Old Guitarist . There appear to be several versions beneath it, each possibly more abstract. If you can manage to ‘peel off’ these layers, then you can follow Picasso’s thoughts. ‘Essentially, you’re eavesdropping on one of the greatest innovators of painting,’ Dik says. Palimpsests are on Dik’s wish list as well. These are written parchments from the early Middle Ages, before the age of paper. The valuable, dried animal skins were scraped off multiple times in order to be written on again. ‘The university library in Leiden has a 32-page-long palimpsest that has visible text underneath it,’ Dik says. ‘But it’s too vague to decipher.’ Dik hopes to make the text visible with the Smart*Light by zooming in on a specific element. ‘Who knows, this may lead to us discovering the oldest Dutch text.’ Strange partners Dik is always in search of new alliances and new analyse techniques. For example, he is currently studying the use of optical coherence tomography (OCT) with Jeroen Kalkman. OCT is an infrared technique that’s making deviations in the retina visible in increasing detail. ‘An OCT scan enables you to make a kind of cross-section of the retina. We use it to study layers of varnish. For example to monitor with great precision the removal of a layer of varnish during a restoration. More and more, restoration studios are resembling surgery theatres.’ Dik has also discovered common interests with the Netherlands Forensic Institute (NFI). ‘There are important parallels between art history and forensic research. A forensic investigator is often given objects and asked to explain what happened to them.’ The NFI discovered that XRF is a good way to reveal traces of blood or sperm when the usual UV techniques fail, for example on a deep-black background. And XRF can also reveal traces of gunpowder that have been ‘washed out’ or that are hidden behind a fresh layer of paint. In collaboration with the printing company Océ 3D, Diks’ research group printed a number of old paintings . The Jewish Bride by Rembrandt now graces a wall in the 3mE building at TU Delft. The copy is much more ‘real’ than a poster. The bottom layer was printed in plastic and provides the copy with relief and thick brush strokes and craquelé . ‘The red of the dress is not deep enough,’ the connoisseur Dik points out. ‘We haven’t managed to imitate the colour well, because that also depends on certain pigments in the bottom layer.’ But it’s the lustre that really betrays the fact that it’s a fake, in Dik’s opinion. ‘The lustre has the same height everywhere, whereas it varies in the real painting. A PhD candidate is currently studying the differences in lustre in layers of varnish, and how they affect paintings.’ What needs to be done to make a more realistic 3D copy? ‘More than anything, we want to learn more. How quickly do layers of varnish discolour? How fast does such a layer wear away and how does the lustre change as time passes? And a print enables you to create an image of the effect of a restoration. Make no mistake, discussions about what is really original and whether it should be restored can be extremely heated in the world of art. So we can contribute by giving an idea of the result.’ Who is Joris Dik? His grandfather and uncle restored old paintings. The craft, the studio and the smell of oil paint fascinated Joris Dik (1974). He was well on his way to continuing the family tradition while he studied art history. But during an internship at the J. Paul Getty Museum (Los Angeles), Dik discovered that it was especially the preliminary research preceding restoration that captivated him. Which pigments did the painter use? Have they discoloured? Were any changes ever made to the painting? If so, when and why? He followed up his education with PhD research on Naples yellow. ‘The oldest pigment in the history of art. It can be found on all of the tiles in India from four thousand BC.’ Dik took samples from artworks from different times and places and analysed the composition. This created a ‘historical atlas’ of Naples yellow that helps during restorations and investigations to determine authenticity. In 2003 he started working as a materials researcher at TU Delft. He has been Antoni van Leeuwenhoek professor since 2010.

New grab unloads vessels faster and smarter

There was plenty of reason to celebrate for transport technologist Dingena Schott and her team at TU Delft. Not only did they develop a design method for a new grab, but the grab, built by Nemag, complied with all of the predictions generated by the models, tests and simulations that they validated. ‘There is no precedent for this in the scientific literature. It’s definitely the crowning glory of our work,’ says Schott. This Dutch grab could potentially unload vessels in ports all over the world more efficiently and sustainably. Dingena Schott Every day, huge vessels from Brazil and Australia enter the port of Rotterdam with thousands of tons of iron ore, coal and other dry bulk. Here, crane operators ensure that the cargo is emptied quickly and efficiently, so that it can be stored or distributed to smaller vessels that then transport it via the Rhine to steel factories in Germany and Austria. To unload this kind of cargo, a crane operator has a variety of grabs to choose from, depending on the kind of dry bulk. ‘The question is: do they have the ideal grab for the product they want to unload?’ says Dingena Schott, who is doing research at the Department of Maritime and Transport Technology at the 3mE Faculty on new technologies for port-related transport logistics. ‘The most recent design for grabs is approximately fifty years old. Since that time, no fundamentally new designs have come onto the market.’ So does this mean vessels can be unloaded more quickly so that they can leave the port earlier to pick up the next load? Video created by Stef Lommen Schott thinks so. What’s more, the research that she and her team conducted has already led to an innovative grab that reduces the unloading time of a vessel by at least ten per cent. TU Delft developed the models, the insight and the design method for this, and worked together with grab manufacturer Nemag from Zierikzee, which created the new design. On 13 November, Nemag even received a prestigious award, a prize given by the International Bulk Journal for the most innovative technology for cargo handling. At first sight, the new grab doesn’t seem that much different than existing grabs. All grabs have two shells with a hinge in the middle and a closing mechanism. ‘The aim is to get as much material as possible per grab,’ says Schott. ‘The amount is limited by the crane on the quay, which can only lift up to a certain weight. If you go for a really heavy grab, then it will of course penetrate deeper into the materials, but it can’t take as much per grab. So the ideal grab is as light as possible but can still dig itself in well. It’s kind of a trade-off between mass and force.’ But where do you start when you want to improve a trusted grab in the conservative world of handling and storing bulk? ‘Since 2007, we have been using discrete element software as part of an innovative design method. Since the arrival of this technique, we have been able to understand for the first time exactly what happens when materials are grabbed. It makes it possible for us to model iron ore pellets as well as the grab, monitor their behaviour in a simulation when the pellets are grabbed, and thus determine whether the grab is being filled optimally. Initially we modelled and tested with an existing grab at Tata Steel in Ijmuiden in order to validate the model.’ Thanks to the software, Schott and her team can compute exactly how much force is needed to grab as many pellets of iron ore as possible. ‘In order to be able to move the grab you have to pull the cables at the top. Then the grab tries to close, but because of the resistance of the iron ore, it will become slower and slower. To maintain the same speed, you then have to pull harder on the cables,’ says Schott. To understand the dynamics of the grab together with the iron ore properly, she connected the dynamic grab model to the discrete element model. The difference between the existing and the new grab is in the overall design: both the new mechanism and the new shape of the shell ensure that the ratio between the forces exerted and the filling of the grab is optimised. That’s what ultimately makes it possible to unload a vessel in ten per cent less time. It’s already a considerable improvement in efficiency. In addition to the productivity, which has increased by 10 per cent, the grab also weighs 15 per cent less. As a result, fewer scarce raw materials are needed. Moreover, vessels don’t stay in port as long waiting and unloading, which reduces emissions. But it could be even more effective. Schott’s aim is to speed up the unloading process even more in the future by creating a dynamic grab capable of adapting itself. Moreover, it’s not just about the grab, but also the crane onto which the grab and cables are attached. ‘We only changed the grab design now, because we couldn’t change anything at the top of the crane yet,’ says Schott. ‘But actually we shouldn’t only be perfecting the grab, but also its interaction with the crane.’ Currently a PhD student is developing models for other kinds of bulk materials. The properties of a material, such as the shape of the particles, their texture and stiffness, mean that each material ‘behaves’ differently. This affects the flow inside the grab. Think, for example, of dry sand, which flows more easily than wet sand. ‘By analysing all kinds of substances, we can develop the grab even further. Ultimately, of course, we want a grab that performs optimally for all products and under all conditions.’ Read the press release by Nemag: 'Nemag wins the prestigious IBJ Award with the nemaX®

Thinking and talking like a doctor and a technologist

Thinking and talking like a doctor and a technologist First graduating class of bachelor students in clinical technology New technologies, such as 3D printing and sensor chips are changing medicine. But we can do better when it comes to surgical lights and stethoscopes, for example, as the theses of the first graduating class of bachelors students in clinical technology demonstrate. They want to make the lives of surgeons, doctors and patients easier with new technology. Surgical wrist light The brainwave came when she was sparring with a group about their final assignment, says Tessa van Hartingsveldt, who had just earned her BSc in clinical technology. Despite special surgical lights, surgeons still complain about a lack of light. Their own heads and hands create shadows. ‘We figured out a way of having the light come from underneath the hands,’ says Van Hartingsveldt. This resulted in a prototype of the surgical pulse light: a series of LED bulbs beneath the pulse. Initial tests show that they do indeed provide more light, exactly where it’s needed. For three years now, TU Delft, Leiden University and Erasmus University have been instructing students together in medicine and technology in the clinical technology bachelor’s programme. This has occurred on request of the care sector. Medical technology is playing an increasingly important role in hospitals, rehabilitation clinics and nursing homes. ‘It’s necessary too, because of the ageing population, a lack of staff and the rising cost of care,’ says Arjo Loeve, lecturer and researcher in biomechanical engineering at TU Delft and coordinator of graduations projects. ‘An operation theatre without a clinical technologist is going to become a rarity. He or she will make sure that the technology is used exactly how it should be.’ Bilingual As a clinical technologist you really have to be a jack-of-all-trades, Loeve adds. ‘You play a role that bridges technology and medicine. That means that you have to be excited by “hard technology”, but also that you need thorough knowledge of medicine and the language that doctors speak.’ All of these requirements seem to have come together in the final projects, a kind of master test. For example, one group of students delved into sound technology in order to ‘isolate’ the classic stethoscope. Following a major accident or tragedy, the surrounding noise can be deafening, so imagine trying to pinpoint a weak heartbeat or crackles in the lungs. Another group delved into bot replicas from a 3D printer. When do potential form deviations occur during production, and how extensive are they? The students started to work with scan techniques, 3D printers, measurement techniques and statistics, and they developed a test model to assess the various techniques. ‘The deviations turned out to be minor, a few tenths of a millimetre,’ Ysbrand Willink says,’ and they mainly arise in the printer.’ Paul Roos says that ‘dental surgeons can definitely use the models while preparing for an operation, but they are not suitable for forensic research. The difference between a saw or knife trace on a bone vanishes with a replica. The “copies” are ultimately useful as replacement bone.’ The mortal remains can perhaps be passed on to the survivors while the evidence remains intact. Isolated stethoscope Isolate stethoscope research in anechoic chamber ‘Common’ problems Clinical technologists also have to be able to establish contact with patients, emphasises Lex Linsen, head of Student Education at the Department of General Practice at Erasmus MC. Linsen is the creator and coordinator of the class called ‘Van inleving naar innovatie’ (From empathy to innovation). Students meet a chronically ill patient and come up with an idea of how to help him or her. ‘It’s not about a revolutionary operation technique for parap legia but more about providing help for everyday, bothersome problems,’ says Linsen. ‘Innovation not because it’s possible, but because it’s necessary.’ Linsen has approached patient associations to ask whether there were people with a chronic affliction willing to participate. Students Paul Roos and Amne Mousa, for example, spent a day with a peer suffering from hydrocephalus. They learned, among other things, that these patients have to visit the hospital a few times a year because the pressure on their brains either increases or decreases. When that happens, the drainage speed has to be adjusted. That means a small hole has to be drilled into the skull. Roos and Mousa presented a potential solution to this problem: place a pressure sensor in the valve of the drain so that drilling will no longer be necessary. ‘Brilliant,’ says Linsen. ‘And the neurosurgeons agree. Their immediate reaction was: why didn’t we think of that ourselves?’ The smart-drain isn’t a reality yet, Linsen emphasises. It’s still an idea, an initial design. But hopefully that will lead to a prototype and ultimately to a smart drain that can actually be used. Pioneers On 12 October the more than forty clinical technology students that completed their studies will receive their degree. How does it feel to have been in the very first graduating class? ‘Because you’re the very first,’ Van Hartingsveldt says, ‘there aren’t any practice tests yet. And in the beginning we were often cramped into a lecture room. The three scheduling systems with three universities didn’t work very well yet. But on the other hand there were great opportunities because we were the first ones. For example, we got to witness open-heart surgery. That was really special.’ ‘Everything is new for lecturers too,’ Willink says. ‘That’s why they’re really open to ideas and criticism. That enables you to shape your own education a little bit as well.’ A study with three partners also generates interesting contacts for lecturers, Loeve says. ‘I talk to doctors and researchers that I previously never encountered. That has already led to new partnerships in research.’ That includes institutions outside the Medical Delta, by the way. The final projects also included clients from the AMC and the Jeroen Bosch Hospital. ‘My call for project ideas,’ Loeve says, ‘reached many people by word of mouth, including ambulance services and rehabilitation centres. That’s really great.’ And the reactions to the research work have been full of praise, Loeve says. ‘Doctors are genuinely surprised and impressed by what these students have conceived and implemented in such a short period of time.’ ‘Some doctors had to get used to the idea that a non-doctor was getting involved,’ Van Hartingsveldt says. ‘But I mostly encountered enthusiasm in the hospitals, and people that wanted to talk to us and cooperate with our research.’ Delft-Leiden-Rotterdam The clinical technology bachelor programme is training a new kind of medical professional: an academic with thorough medical and technical knowledge that builds a bridge between the technology and doctor and patient. The programmes began in the academic year of 2013-2014 with a hundred students a year who were admitted via a selection procedure. The Delft University of Technology is coordinating the programme. The partners are Leiden University (LUMC) and Erasmus University Rotterdam (Erasmus MC). Care institutions associated with the Medical Delta are also cooperating. The first forty bachelor graduates received their degrees on 12 October 2017. The future: technical medicine Bachelor graduates in clinical technology Paul Roos and Ysbrand Willink joined the new three-year master programme in technical medicine this September. This bilingual MSc programme provides an opportunity to study medicine and medical technology in more depth. There are two specialisations: Imaging & Intervention, which focuses on imaging techniques, and Sensing & Stimulation, which focuses on tracking and monitoring the health conditions of patients. Anyone with a BSc in clinical technology can also opt for a master in one of two basic disciplines: medicine or biomedical engineering. Those wanting to continue in medicine are required to follow a transition programme. Tessa van Hartingsveldt is considering the master programme in biomedical engineering. ‘The engineering side of it is what appeals to me most. But first I’m going to take a gap year and work and travel.’

Predicting waves at sea

Predicting waves at sea She had already walked onto the deck, the person who had to wave a little flag in eachhand for the helicopter pilot so he would know exactly when to land his aircraft on theship. At the same time, a crane operator was fastening his hook to the place where hewould lift the foundation for an offshore wind turbine. Later he would carefully put itoverboard at the designated place on the bottom of the sea. Elsewhere, another ship wasready to lower a small pilot boat into the water, which would pilot a freighter into theport later that day. All three were looking at the app which has already been commonly used in the maritime industry for some time now: the wave radar. Thanks to this technical gadget, developed by Peter Naaijen, assistant professor of shipand offshore hydromechanics at the Delft University of Technology, everyone at sea cansee whether there are waves in the vicinity in the coming five minutes and how the shipwill respond to them. The neat thing about this new technique is that essentially allships already have radar navigation. Naaijen analyses this radar data, which containsinformation about the location and height of waves. Thanks to his innovation, it hasbecome considerably safer to perform offshore operations. Peter Naaijen image: Jort van der Jagt To a certain extent, wave conditions at sea are already predictable, except these predications are not very specific. ‘They are not that useful, because you want to know exactly where these waves are and when they are coming,’ Naaijen says. Indeed, this knowledge is reducing accidents in the offshore industry: helicopters are ceasing to crash on ships because the deck unexpectedly comes closer and from now on heavy loads can be transferred from one ship to another without them colliding and causing damage. In addition, installers of wind turbines no longer have to wait until the sea is calm again, nor will they have to travel back to the coast because the waves are too high to work in. It is even possible to find a window when it is momentarily calm during bad weather. This saves downtime, on the one hand, the time that ships need to stay ashore and, on the other hand, it ensures that ships do not set out to sea without accomplishing their mission. This saves both time and money, but also fuel. The margins in the wind industry are small,’ Naaijen says. ‘You gain lots of hours by not having to wait and having more hours per day and more days per year to carry out your offshore operations.’ Predicting waves is similar to predicting rain. ‘Compare it to a weather report for the Netherlands that predicts a thirty per cent chance of rain. You cannot be sure exactly where and when this rain is going to fall. A rainfall radar provides more location and time-specific data, which will give you a better chance to avoid getting soaked. That is why the rainfall radar is a good comparison for our wave radar. During my PhD research I showed that it is possible to predict how high waves will be around a ship 2.5 hours in advance,’ Naaijen says. Whether or when a ship leaves port is determined by the ‘sea state’, a kind of coefficient that reflects the intensity of the waves, just as the Beaufort scale does for wind. ‘Imagine that a wave three metres high is the absolute limit for an operation that takes two minutes. A ship will not leave port at sea state 4 because the chances of encountering a three-metre wave are too high,’ Naaijen explains. ‘But in reality 93 per cent of that time consists of intervals of at least two minutes, during which the height of three metres is not reached! So if you know exactly when to expect those high waves, then you can avoid that moment and there will be plenty of time left to safely finish your job. That means you can potentially achieve huge savings because ships can leave port more often during higher sea states and do not have to wait on shore as long.’ Difficulties with working on board because of high waves How exactly does this wave radar work? Most ships already have all of the necessary hardware for predicting waves, and so the technology does not require a major investment. ‘We use the radar navigation on ships. Right now, it indicates where other boats are located and precisely where the coast is, so that ships do not run aground,’ Naaijen says. ‘But actually the radar receives a great deal more information, namely where all of the waves in the water are located.’ This is how it works: a radar transmits electromagnetic waves that collide with the waves in the sea, so that the antennae receives them back again. Right now, that information is being filtered out, because it is not interesting for sea captains. ‘We tap into this raw radar data, so that we can analyse it with all kinds of intelligent algorithms. So we are essentially using a waste product.’ In the meantime, Naaijen has launched a company called Next Ocean with a former fellow student, and he has received a grant from the Netherlands Organisation for Scientific Research to actually develop the wave radar app. ‘Parties from industry, such as Boskalis and Allseas have encouraged me to translate my research into a product, because there is great demand for this in the offshore world,’ he says. The first customer has already been announced: Allseas will be the first company to equip its ships with the Next Ocean Wave Predictor. ‘That is what we decided to call our software, but actually it is more of a ship motion predictor,’ Naaijen says, smiling proudly. He expects the software to be ready in late 2017. In 2018, offshore workers on ships should be able to see how high the waves are on mobile devices, so that they know when the best time is to carry out their work.’