Check out our science

Discover the stories of researchers at the
Faculty of Mechanical Engineering.

A thrilling nail clipper

A thrilling nail clipper Engineering for forensics A good nail clipper can mean the difference between a murderer behind bars and one on the loose. ‘You can have the best high-tech machinery in your forensic laboratory, but if you’re not able to enter the right evidence into it, you’ll never find the perpetrator.’ Cotton swabs, tape and tweezers are important tools for forensic investigators. Gathering evidence at the scene of the crime is done manually. ‘Low tech. And there’s nothing wrong with that,’ emphasises researcher Arjo Loeve, project leader of Engineering for Forensics at TU Delft’s ME. Loeve loves technology, but also simplicity. ‘The simpler the instrument, the easier it is to use and clean. And that also reduces the chances of contamination.’ But low tech certainly does not mean that the work is easy. Forensic investigators have to make quick decisions under time pressure. Where are traces of evidence likely to be found? How can we secure it? After all, the scene of the crime can be a busy motorway, a cramped cellar space or a beach at low tide. Experience and expertise play a key role. ‘That’s precisely why I believe that engineering can make a major contribution,’ Loeve says. ‘By developing more reliable or faster instruments, but also through modelling. What, for example, is the most efficient sequence of action for gathering evidence in a given setting?’ Textile polyester fiber with trace material Stamp with adhesive tape for collecting traces Loeve, who trained as a biomedical engineer, ended up in forensic research more or less by coincidence. ‘I had an exciting conversation at a birthday party with someone from the Netherlands Forensic Institute.’ The talk led to a joint research project. They wanted to figure out how serious the risk of DNA cross-contamination is during physical examination following a sexual offense. Cross-contamination is the unintentional and undetected ‘transfer’ of DNA from one place to another during the investigation. Loeve experimented with true-to-life models in the laboratory and was shocked by the findings. The probability of cross-contamination using the classic method – speculum and cotton swab – was more than sixty per cent. ‘And to think, finding the perpetrator’s DNA on or in the victim can ultimately mean the difference between acquittal and conviction.’ So it’s high time for a new method. ‘We developed a ring with a kind of plastic bag on it that shields the cotton swab. There wasn’t a single incidence of cross-contamination in ninety tests that we performed in the laboratory. We hope that it will be field tested soon.’ More research projects soon followed. Together with the Academic Medical Center and NFI, Loeve is trying to more accurately predict the victim’s time of death. ‘Right now, there’s often still a margin of error of hours. We think we can reduce that to less than half an hour with new sensors (TU Delft) and calculation models (AMC) that take into consideration body weight and a variety of environmental factors.’ Loeve also developed practical tools for forensic investigators who have to wear an airtight suit at dangerous crime scenes. Some of the research projects clearly have common ground with Loeve’s biomedical research. One example is a study on abusive head trauma, the damage that babies suffer as a result of violent shaking. ‘In order to acquit or convict people, we need more knowledge about what kind of damage is caused by shaking as opposed to falling, for example,’ Loeve says. ‘And of course that would also make it possible to give parents better advice.’ Loeve is using knowledge from other ongoing research on how and when bones break for this study. ‘Increasingly I’ve noticed that our medical and forensic research projects have common ground and even overlap.’ Series such as CSI and NCIS were a huge success. How exciting is Loeve’s ‘crime lab’? ‘We don’t do investigations ourselves at crime scenes or in real cases. I do talk a lot with forensic investigators and read case reports to get a good idea of what happens in the field.’ Loeve stresses that these never involve ongoing cases. ‘And believe me, I’d rather not have been exposed to some of the details. Anyone who thinks that the writers of these series have a morbid imagination are wrong.’ The knowledge in the files may be shocking, sad and intense, but that’s also precisely what drives Loeve. ‘It does make it clear that you’re contributing to something important. Unfortunately there’s a need for what we do.’ What’s needed, for example, is the new nail clipper that Loeve designed with his students and NFI. The prototype was extensively tested in the laboratory and is ready to be field tested by the police. ‘A nail clipper doesn’t sound very sexy or exciting,’ Loeve says, laughing. But you can find traces of the perpetrator or victim’s DNA under nails. Or material that reveals something about recent activities or the environment that he or she was in. The new Delft nail clipper makes it possible to clip the nails of a stiffened fist without having to bend the fingers. The nail cannot jump away while being cut. And you can easily replace the front with a clean part. It makes it easy to clip the left and right hand individually. ‘Knowing that something was done with the left or the right hand can turn out to be crucial evidence,’ Loeve says. What was the most difficult part, technically speaking? ‘There wasn’t really a eureka moment. We adapted the design step by step until all of the requirements were met. We aimed for the simplest possible tool that could be cleaned easily and quickly, which you could use with two hands in all situations.’ It’s especially important, when developing a new tool, to be able to let go of ideas and listen closely to what’s really needed, according to Loeve. ‘The simplest solution often turns out to be the best.’ Arjo Loeve Co van Ledden Hulsebosch Center TU Delft’s ME Faculty has been an official partner of CLHC, the Co van Ledden Hulsebosch Center, since last year. This research consortium in the forensic sciences was established in 2013 by scientists from the University of Amsterdam, the Amsterdam Medical Center (AMC) and the Netherlands Forensic Institute (NFI). It is named after the first Dutch forensic scientist Co van Ledden Hulsebosch, who co-founded the International Academy of Criminology in 1929. Loeve is the coordinator at Delft. ‘CLHC provides additional contacts and expertise. This kind of consortium is satisfying: short lines of communication with all kinds of scientists and meetings where you speak to new people and share problems.’ Lowlands Crime Lab A ‘murder’ was committed by science 175 times during Lowlands 2016. It all happened in the Crime Lab set up by the Amsterdam University of Applied Sciences, TU Delft, the Police Academy and NFI. More than six hundred festivalgoers participated. Almost two hundred of them smothered, with painted hands, a mannequin head with a pillow, which yielded 175 pillows with ‘choke traces’. Before the murderers had a go at the mannequin head, they put a pillow case neatly around the pillow, also with painted hands. Arjo Loeve was there: ‘That allowed us to examine whether there is a recognisable difference between regular depression marks and marks caused by violence.’ Online forensic engineering course Forensic investigation is used in many more places than you would think. For example, forensic engineers are called in to analyse all kinds of situations in which technical systems have or seem to have failed. Examples include gas extraction in Groningen, aircraft crashes or patients in hospitals who are infected by bacteria that are transmitted through surgical instruments. Together with Karel Terwel (civil engineering) and Michiel Schuurman (aerospace engineering), Arjo Loeve set up a massive open online course. This free course, entitled ‘Forensic Engineering – Learning from Failures’ can be found at www.edX.org and will start in mid-October.

New watch mechanism out of a single piece

New watch mechanism out of a single piece Innovation from the lab into the real world Nima Tolou, a researcher and entrepreneur at Delft, is very proud of a new development in the core of mechanical watches. In a close cooperation with the LVMH Watch Division (TAG Heuer/Zenith) and Flexous, a start-up from TU Delft that he co-founded with physicist and entrepreneur Oleg Guziy, he used his expertise in compliant mechanisms and MEMS technology to co-develop a watch with a totally new time mechanism. Along with this, he’s working on another innovative device that converts vibrations into free energy. Nima Tolou ‘For me, an innovation is not complete until the finding actually works in the real world. In other words, outside the lab,’ he emphasises. But the real world is often stubborn. Unexpected problems emerge that ruin even the most fabulous innovations. A great discovery can end up in the archives of science, carefully described in a paper or patent. That’s very important but for Tolou not enough. ‘As a child I made my own toys. At school in Iran I regularly won awards for my creations. I’m a thinker, a designer, but definitely also a maker.’ Tolou is convinced that the developers themselves know and understand their technology better than anyone. ‘In case of the watch, we started all over many times. You don’t give up easily when you’re convinced that there’s a solution. The faith that it can work is a key aspect of the development.’ Flexible The size of the control mechanisms in watches has been considerably miniaturised since Dutch scientist Christiaan Huygens invented the pendulum in 1675, but the principle hasn’t changed. The pendulum in the clock, or the spring drive in a watch, ensures that the hands always indicate the right time through a complex system consisting of many parts. But Tolou, who has a background in physics and mathematics, together with the LVMH Watch Division and Flexous, discovered an alternative. Most mechanical systems consist of rigid structures , usually metals. Think of the wheels, pinions and bearings. Tolou and his research team specialise in micro mechanisms that contain flexible, elastic components, referred to as compliant micro mechanisms. In other words, matter that gives, bends, quivers or vibrates. The team discovered that a single circular part with complexly shaped cut-outs can combine multiple essential functions that take more than 30 parts in a conventional watch. This new approach, along with smart use of the nonlinear nature of compliant mechanisms, lead to an innovative mechanical system that is more accurate, more precise and more energy efficient at the same time. The other advantage of this idea is that with high precision production techniques, this kind of watch mechanism can be etched from a single flat piece of silicon. And it will be simpler, small, maintenance-free and robust. Tolou was convinced that you can apply these compliant micro mechanisms in a watch. Who dares to realize such a wild idea? Mechanical watch makers tend to fancy classical craftsmanship. After 3 years, the TU Delft team, Flexous and the LVMH Watch Division, managed to realise this. The product was introduced by LVMH to the market on 14 September 2017 under the name Zenith Defy Lab as “the world’s most accurate mechanical watch”. Zenith Defy Lab Watch With 15Hz Movement Is 'World's Most Accurate' | aBlogtoWatch A series of publications provided general knowledge about the accurate design of oscillators in compliant mechanisms and lead to the granting of two Best Paper awards at international conferences in France (MARSS) and the United States (ASME IDETC). Free power A second finding by the Delft researcher is about to be made public: an energy harvester. It’s a device that harvests energy when it vibrates after passing of a car, for example, or because it’s standing on a machine. The crux of the discovery, however, lies in the use of extremely small, flexible components that makes the device sensitive to low frequencies and irregular vibrations. ‘It’s about mass on specially shaped elastic material,’ Tolou says. ‘This mass starts to move as a result of vibrations, and we subsequently convert this movement into power.’ The prototype looks like a thick classic battery (size D). ‘But one that charges itself,’ Tolou says. The energy harvester can provide sensors with power and make it possible to send data. And these types of sensors are becoming increasingly prevalent. There are expected to be twenty billion of them in the world in just a few years’ time. ‘Nowadays we connect all devices to each other: the Internet of Things. Our environment is becoming increasingly interactive,’ Tolou says. Sensors that derive their energy from vibrations don’t have a plug. Their battery lasts ‘forever’. That makes them considerably less expensive, particularly in hostile environments and places that are difficult to access where battery replacement is inconvenient or expensive. Tolou’s energy harvesters are therefore useful for monitoring moving objects: sea containers, sheep in mountainous environments or a stolen bike. But they’re also useful for detecting forest fires or the sound of illegal chainsaws in the rainforest. You simply hang the sensor on a drooping branch. Most of all, Tolou hopes to make the world a little bit safer, more efficient and more productive . For example, sensors on railway tracks or train wheels that automatically warn you when a train is approaching. And many more sensors on industrial estates. ‘If sensors are inexpensive and can function independently, then you can attach them to every storage tank, pipe, pump and valve.’ Overheating will be signalled in good time and maintenance will be carried out based on measurements instead of annual cycles. ‘What’s more, you’re doing all this with energy that would otherwise be wasted,’ Tolou explains. Free and environmentally friendly electricity in other words. The amount of power generated by his energy harvester is limited to a maximum of 10-20 milliwatts, however. ‘Not enough to charge your mobile phone, in any case. For that you would need to fill your backpack with harvesters. Which isn’t very practical, of course.’ Thanks to the unstoppable rise of sensors, harvesting energy from our environment has become a hot field of research with a great deal of competition. So what is it that makes Tolou’s discovery so promising? ‘Our energy harvester can convert fast and slow vibrations into power. In many other systems, the vibrations have to be within a specific range. But it makes no difference with our system whether you are cycling quickly or slowly over cobblestones.’ Also in this case the breakthrough was generated because of the combination of thorough knowledge (the Veni-award of Tolou) and his drive to turn innovations into products. Gap Tolou is working on new products with the spin-off company Flexous that he co-founded in early 2014 with physicist and entrepreneur Oleg Guziy. Tolou is technical advisor, Guziy CEO. The division that focuses on marketing the motion energy harvester has been named Kinergizer. “Flexous bridges the gap between academy and production. If you introduce your idea to the world by means of a patent license, it will take years before you actually have a product. If you work on the development as an inventor, as a fundamental researcher, then it will happen more quickly, in two to three years’ time. But most of all, I want to be involved in it all, contribute by thinking and acting myself . The process is at least as exciting as the end result.’ Tolou certainly proved that with the recent products. What will his next project be? ‘I’m focusing mainly on research on autonomous systems: self-sufficient electronics. Devices that observe autonomously and subsequently respond to their environment. And they can do that thanks to an energy harvester. Nima Tolou (1982) studied applied mechanical design in Iran. He has been working at TU Delft since 2008, first as a PhD student, and now as a university lecturer. ‘I wanted to continue doing research, but the opportunities were limited in Iran.’ Tolou searched for a reputable university in the area of applied research. He actively seeks partners in the business sector to convert findings and research results into products with his own team.

Using chemistry to close the CO2 cycle

Create fuels out of it, with the aid of green electricity If we want to make the world more sustainable, then we need to find a solution for CO2. Professor Wiebren de Jong (TU Delft) from the Department of Process & Energy (Large-Scale Energy Storage section, LSE) is working hard on this problem. He wants to capture emissions from fuels and other bulk chemicals. A project will be launched in April to convert that into formic acid. De Jong sees opportunities for the long term for transport fuels and the storage of sustainable electricity. It’s the problem of our time: carbon dioxide (CO2). Carbon dioxide is released when we burn gasoline, diesel, gas or coal, and it accumulates in our atmosphere. There it acts as a warm blanket that raises the planet’s temperature and disrupts the climate. ‘If we want to address the climate problem, then we need to handle CO2 in a smarter way,’ De Jong says. ‘The key to a sustainable future lies in closing the CO2 cycle. By making fuels out of CO2 again, for example, or raw materials for industry.’ That makes sense, but it’s going to take serious technological development. The problem is that in chemistry this essentially harmless, colourless gas is a kind of terminal destination. The CO2 molecule is highly stable, and anyone intending to do something useful with it will have to put a great deal of energy into their effort. Figuratvely speaking, but also literally. Indeed, chemists are pretty jealous of nature. Plants use sunlight to produce sugars from CO2 and water, and from there they subsequently produce yet other substances. So it’s possible to use CO2, the only problem is that biological photosynthesis is not easy to replicate chemically. That’s why Wiebren de Jong has opted for a different approach. ‘There is an increasing amount of sustainable energy available in the form of electricity, derived from solar panels and wind turbines. In the future, we’re going to use that green electricity to drive chemical processes that convert CO2. That’s how we can produce fuels from CO2 again.’ Professor Wiebren de Jong was appointed full professor of large-scale energy storage at the ME Faculty’s Department of Process & Energy. Several months prior to that he was also named part-time professor of integrated thermochemical biorefineries at the University of Groningen. Wiebren de Jong is an expert in the areas of thermal & chemical conversion and biorefinery. De Jong has been affiliated with TU Delft since he started conducting research for his PhD in 1996. Storing electricity If De Jong pulls that off, he will kill two birds with one stone. First, it will make transport by lorry, ship and aeroplane much more sustainable. ‘Electric power is too heavy for that, and the range is too small,’ De Jong says. ‘With “CO2 fuels” in the tank, the transport sector can create a closed CO2 cycle relatively easily, without having to make huge investments or changes.’ The second, equally important application, is derived from the fact that the processes De Jong is working on essentially store electricity in the CO¬¬2 fuels. This makes it possible to create a balance between the supply and demand of green energy. After all, the production of green electricity is rarely equal to demand. A solar cell delivers electricity when the sun is shining, but people only turn on their lamps once it has gotten dark. If you generate electricity with a wind turbine, it would be great if you could use it when there’s no wind. On top of that, production and consumption always have to be in equilibrium in the electricity grid. There’s little point propping a country full of solar cells if no one is consuming all that electricity on sunny days. ‘It’s crucial for a stable sustainable supply of electricity that you can store green electricity,’ De Jong says. ‘Of course that can be partly achieved with batteries. But chemical storage in CO2 fuels makes it possible to store electric energy on a huge scale.’ A plant running on CO2 fuels can then deliver supplemental electricity during those moments when solar cells and wind turbines are not producing enough. ‘This enables you to buffer supply and demand,’ De Jong says, ‘and reduce your dependence on conventional energy sources, such as coal and nuclear energy.’ Moreover, he adds, the CO2 fuels can also play a valuable role in ‘linking up the seasons’. For example, it would be possible to ‘save’ sustainable energy from the summer (sun) and autumn (wind) until the winter, when houses need to be heated up. Once CO2 has entered the atmosphere it’s not easy, chemically speaking, to use it anymore. ‘The concentration is 400 ppm,’ De Jong says, ‘a few hundredths of a per cent. That’s too high for the climate, nor is it optimal in terms of process technology.’ Indeed, he sees possibilities for using CO2 that is released in high concentrations, for example in power plant chimneys, waste incinerators and steel plants. Another possibility is to allow trees and plants to grow (because they absorb CO2) and subsequently use the biomass. Indeed, Wiebren de Jong is devoting a great deal of attention to that, in particular when it concerns the use of organic by-products. From lab to factoryDe Jong’s research is characterised by the ambition to ‘make something that works,’ as he puts it. His room borders the process hall with the ‘skids’: the technology set-up where new processes are developed and researched. ‘This is where ideas are transformed into reality,’ De Jong says. ‘We take what was conceived in a lab and carried out on a small scale in a fume cupboard and elaborate on it here in such a way that industry can really do something with it on a large scale. That means making optimal use of the thermal effects from the chemical conversion, for example. That’s how we create an integrated process that is essentially ready for large-scale use.’ The industry, or research institutes such as TNO and ECN, with which De Jong frequently collaborates, can then take the developments at the pilot plants to the next level.We have not reached this stage yet in the area of CO2 conversion. You can convert CO2 into CO with electricity, he explains. Carbon monoxide is much less stable and can therefore serve as a starting point for the synthesis of fuels and chemical raw materials. As is often the case, this is all easier said than done. Efficient electrodes are needed to introduce electric energy into the reactor; the CO2 has to be dissolved in the reaction solvent; good catalysts are needed; and it’s still a challenge to direct the complex interaction of chemical reactions is such a way that you end up with the desired product. ‘Still plenty of work to do,’ De Jong summarises. The ME Faculty’s Process & Energy Lab, which has just entered its second year, is the hub for all of TU Delft’s large-scale research in the area of process and energy technology. It is the only one of its kind in the Netherlands. The research being conducted at Delft in this area takes place in six sections: intensified reaction and separation technology; energy technology; large-scale energy storage; fluid dynamics; multi-phase systems; and engineering thermodynamics. Formic acid An interesting example of the use of CO2 is its conversion to formic acid. That is a ‘fuel’ for battery-free electric cars, but also a sustainable raw material for the chemical industry. The latter is keen to become ‘greener’ and is therefore prepared to invest. ‘Of course, the knowledge that we are developing,’ De Jong says, ‘is important for the real large-scale processes that will take place in the long term, such as the production of transport fuels and “sustainable natural gases”.’ A special project was launched in April 2017 in which De Jong further developed the path from CO2 to formic acid together with TNO (project coordinator), start-up company COVAL Energy, Manure Processing Fryslân BV , and CE Delft. The FAST student team from Eindhoven, which developed a car based on formic acid, was involved as well. So far, the research has generated a prototype the size of a paving stone. ‘Now it’s important to push the process in a direction where it becomes relevant for industry. That means, among other things, bigger electrodes and higher pressure, for example. Together with TNO and in close collaboration with the Voltachem platform for electrochemical conversion, we are going to find out what that means for process technology.’ Formic acid An interesting example of the use of CO2 is its conversion to formic acid. That is a ‘fuel’ for battery-free electric cars, but also a sustainable raw material for the chemical industry. The latter is keen to become ‘greener’ and is therefore prepared to invest. ‘Of course, the knowledge that we are developing,’ De Jong says, ‘is important for the real large-scale processes that will take place in the long term, such as the production of transport fuels and “sustainable natural gases”.’ A special project was launched in April 2017 in which De Jong further developed the path from CO2 to formic acid together with TNO (project coordinator), start-up company COVAL Energy, Manure Processing Fryslân BV , and CE Delft. The FAST student team from Eindhoven, which developed a car based on formic acid, was involved as well. So far, the research has generated a prototype the size of a paving stone. ‘Now it’s important to push the process in a direction where it becomes relevant for industry. That means, among other things, bigger electrodes and higher pressure, for example. Together with TNO and in close collaboration with the Voltachem platform for electrochemical conversion, we are going to find out what that means for process technology.’

Mechatronics 2.0: sustainable form of all-in-one

The Netherlands is good at mechatronics, the multidisciplinary field centred on integrated mechanical systems that carry out their work by means of a clever combination of sensors, actuators and control engineering. Just Herder, professor of interactive mechanisms and mechatronics and new chairman of TU Delft’s Department of Precision and Microsystems Engineering, likes to look into the future, at what is unofficially called ‘mechatronics 2.0’. Whereas the elements of mechatronics have traditionally been separate entities, Herder is trying to integrate these elements with each other to the fullest extent possible. ‘We can do it on a smaller scale and with more precision, using fewer materials and less energy.’ Crowd surfen Why does a mobile phone have separate parts to process zeros and ones, to probe the outside world and to emit a vibration?’ Herder asks. His research group is attempting to integrate different functions with each other in a smart way. An important advantage of this approach is that it is highly suitable for miniaturisation. And that is how Herder found himself in the magical world of nanotechnology. ‘If we want to continue to make increasingly small computer chips, then the machines that do it are going to have to become more and more precise as well, without their energy consumption going off the charts.’ Herder imagines a point on the horizon in his field of research: ‘Now that chips are being produced with the precision of a millionth of a millimetre (the scale of the nanometre), we also have to start thinking about transformations at the nanoscale in the moving container that transports the chip from A to B during the manufacturing process. An actuator layer between the container and the chip can compensate for transformation at the nanometre scale in the container, thus protecting the chip. But would it also be possible to completely replace the container with thousands of miniscule fingers, which allow the chip to crowd surf with the highest possible accuracy? Perhaps these fingers could inspect the chip in the meantime too.’ Manufacturing and controlling these kinds of smart fingers is one of the single biggest challenges for the new mechatronics. Pliable Elastic mechanisms with extremely low stiffness are an important tool. Replacing traditional mechanical systems, for example based on layers, with pliable alternatives will create a material surface onto which active layers can be placed. Electro-active polymer layers, for example, can act as sensors and piezoelectric layers as actuators to generate movement. Thus movement, actuation and sensing would be combined in one continuous structure. This structure wastes less energy and is also easy to scale down. ‘This is another way of looking at mechanical engineering equipment,’ Herder says. ‘At the moment equipment is being made that is increasingly large in order to achieve greater precision, and for that reason they use up more space, material and energy.’ Energy harvesting A special property of mechanisms with extremely low stiffness that is being studied in Herder’s group is the fact that they can be used to ‘harvest’ energy in slow movements. ‘Think, for example, of a sea container,’ he explains. ‘In order to figure out where it is located, you would basically want to build a GPS tracker in it that emits an occasional signal. But you want to avoid having to constantly replace the batteries. A simple mechanical system that gets energy out of the movements that the container makes while it crosses the ocean does not need a battery.’ The applications are countless, and Herder is working in close cooperation with industrial partners. The spin-off company that emerged from Herder’s research on flexible mechanical parts, Flexous BV, has already launched a subsidiary company called Kinergizer BV that focuses specifically on energy harvesting applications. Autonomous microrobots ‘In five years,’ Herder says, ‘I hope we will have succeeded in fully integrating these low-stiffness mechanisms into microelectromechanical systems (MEMS), manufactured by means of prevailing chip manufacture techniques. ‘If we can apply these commonly used techniques to also build integrated MEMS/mechatronic systems, it will open up a whole world of opportunities.’ During the same period, he expects advances to take place in the area of bio-inspired robots, which can manoeuvre autonomously through unstructured environments, such as certain organs in the body, in order to inspect them around the clock for long periods of time. ‘Nature does not use propellers to generate movement through a fluid. Instead, this task is performed by cilia or flagella. The combination of mechatronics and nanotechnology provides opportunities to build these kinds of systems on the smallest possible length scale. We have already achieved initial results on a larger format. In the coming years I hope that prototypes will be able to move autonomously, similar to the way robot vacuum cleaners do.’ Nano-Engineering Research Initiative Herder is enthusiastic about the potential of the combination of mechatronics and nanotechnology in his department. ‘I think there is a great deal more we can achieve. Innumerable interesting phenomena are there for the taking now that we can inject nanotechnological expertise into the area of mechatronics. It will enable us to produce materials, instruments and equipment that work thanks to nanotechnology (“nano-enabled”), and, conversely, develop machines that produce products on a large scale based on nanotechnology (“enabling nano”).’ Indeed, he has great expectations of the Nano-Engineering Research Initiative (NERI) initiative. ‘The great thing about NERI is that it combines fields of expertise into one department. I have limited understanding of nanotechnology but all the more of automation. For others it is precisely the opposite. The lines are short. We are thinking up the most exotic plans and keep discovering that these plans are actually feasible as long as we do it together.’ In addition to the projects being carried out in partnership with industry, Herder views NERI as a new way of pursuing productive partnerships and safeguarding continuity in the long term. ‘The idea is to provide group support for certain lines of research. Companies from different branches of trade interested in the same research topic join forces. Because the participants have different applications in mind, there is a great deal of leeway for collaboration instead of competition.’ At the moment Herder and his colleagues are very busy with the first NERI contracts. ‘This is the tipping point. NERI will enable us to showcase what we have to offer and what we are worth.’ Prof. Just Herder

Quality of transport improves through better understanding of steel

The Titanic was the largest ship in the world. In 1912, during its maiden voyage, it struck an iceberg and sank within three hours. More than 1,500 passengers died. ‘It is the most famous maritime disaster in history, because everyone thought it was unsinkable. But in those days people barely knew anything about steel,’ says Jilt Sietsma, professor of microstructure control in metals at the Department of Materials Science and Engineering. ‘All knowledge about steel at the time was based on practical knowledge. For example, a smith would throw a new sword into water to cool it off and would then hold it over a fire and hit it with a hammer to change its shape. But the smith had no idea what was going on in the material.’ In the meantime, not only do we know why steel is more deformable at high temperatures, but also that it becomes extremely brittle at low temperatures, at which point it can easily fracture. Ships that navigate through floes therefore have to be made of materials that can handle such conditions. ‘We know that steel must not become brittle anymore at temperatures around zero, but we still do not know exactly how we are going to achieve that,’ says Marcel Sluiter, associate professor at the same department as Sietsma. ‘That is one of our greatest research challenges. In short, we are still in the middle of fundamental research into steel.’ Materials scientists at the ME Faculty are constantly trying to gain a better understanding of the structure of steel. By structure they mean that every atom likes to surround itself with certain other atoms. Just like the Atomium in Brussels, where one atom is surrounded by eight other atoms: this kind of structure is also called a crystal. The Atomium represents the structure of an iron crystal, but a grain of cooking salt is just as good as a crystal. Steel also contain crystals, but they are much smaller than the ones in cooking salt and they cling firmly to each other. As a result, we perceive this material as a single entity. (Photo: Dochter van de Smid. Photo made by Richard Alma) When you heat steel, the atoms have a tendency to take on another structure. This phenomenon is called phase transition. Many people think of this as the transition from a gaseous to a liquid to a solid state of matter or vice versa. But this type of phase transition occurs within solid states of matter as well. It is important that we understand them, because ultimately it is the structure of a material that determines everything: how strong it is, how deformable it is and how good it is at withstanding corrosion. Marcel Sluiter studies this structure at the atomic scale, while Jilt Sietsma does the same at the micron scale, which is a slightly larger scale. ‘Suppose a material has to have a number of properties, including a strength of 384 megapascals. Then you first have to determine what the ideal structure is and subsequently how you can manage to achieve that structure in practice.’ The greatest paradox in the use of steel is that people have been using it for thousands of years, for example in construction, in cars and in cans in the packaging industry, but we have only started to increase our understanding of it in the last decade. ‘X-radiation was only discovered in 1905,’ Jilt says, ‘which is what we use to analyse how the structure of the material works. Before that time, we had no way of seeing from which crystals steel was composed. So while it may be easy for us to use steel, understanding it is much more difficult.’ Yet we have made a great deal more progress in improving the properties of steel than we have with most other metals. One example is the car: modern types of steel are much better at absorbing the force of a collision than in the past. If you collided with another car in 1970, then your car offered barely any protection. Nowadays a large part of the force in a collision ends up in the crumple zone, as opposed to your body, because today’s steel can store much more strain energy. Luckily the Delft scientists are increasingly improving their understanding of what happens in steel. In the meantime they are dreaming about the things they can achieve in the future. ‘For example, we want to understand how fractures materialise in railroad switches. If we were to understand that, then we could make sure that switches never break anymore. Perhaps that would prompt us to encourage the use of public transport,’ says Sietsma. That may sound easy, but many different kinds of steel were used in the switches that were installed in the Netherlands in the past forty years. Another dream for the future is even better crumple zones in cars, for example for sideway collisions. The most ambitious application of steel is in nuclear fusion, or generating energy. ‘It involves enormously high temperatures and levels of radiation,’ says Marcel. ‘Materials are really the bottleneck there, because the temperature to which you can expose them is limited, and we still do not know how they handle the radiation damage.’ The fundamental research on steel is largely being conducted in collaboration with Tata Steel. ‘Once we have a better understanding of the fundamental basis of steel and can start modelling well, then we can develop new types of steel much more efficiently and stop relying on trial and error,’ Jilt says. ‘Tata Steel uses our fundamental work in their research department and in applications in their factory, such as making steel for cars and packaging steels.’ ME is also doing projects with ProRail: in light of the fact that all tracks and switches are made of steel, the research is focusing on where damage occurs, why, and how that is related to the structure of steel. Ideally, these Delft scientists would want all engineers to have a better understanding of materials science. Preferably from the moment they become students. ‘Everything is made of materials,’ Marcel says, ‘and every engineer should know more about that. In practice, they are often not capable of choosing the ideal material, for example for a switch, a biomedical application or a bridge. You have different needs for all of these areas: the material cannot be sensitive to corrosion if it is being used outside or in the human body, whereas steel in a car has to be high-strength and high-deformability. And it is not just about choosing the best material but also choosing one that stands the test of time. Engineers have to understand that there are more aspects to a material than the table that states exactly how strong it is.’ Jilt Sietsma Marcel Sluiter

Making precise structures at the smallest scale

Attractive tools for biomedical and pharmaceutical research Imagine a tiny sensor setting off an alarm when a healing wound gets infected, or the effects of an experimental drug being tested not on a human or animal, but using a tiny chip packed with biological cells and sensors. The biomedical field is more and more turning to technology at the smallest scales. Dr. Luigi Sasso, at the Department of Precision and Microsystems Engineering (PME) works to enable the large-scale fabrication of such small-scale technologies. The latest addition to the set of tools at his disposal is an advanced 3D printer (called the Nanoscribe) that routinely produces structures as small as 20 millionth of a millimeter. “We’re very happy that we can now cover the full resolution range,” Sasso says. Organ-on-chip An exciting avenue being pursued in the area of personalized medicine is to test treatment options for each individual patient using only a small amount of cells or tissue. How do you channel nutrients or drugs towards these cells and achieve a real-time read out of how they respond to these drugs? Sasso uses soft plastics, called polymers, to build tiny structures incorporating, on the one hand, channels for the transport of liquids and, on the other hand, sensors. “Polymers are biocompatible: cells like them better than other materials. They can also be made to conduct electricity, which allows them to be crafted into electrochemical sensors.” These sensors are sensitive to biomedical parameters such as the local acidity (pH), temperature or glucose levels. As Sasso explains, “The blood’s glucose level, for example, provides information about a cell’s metabolism, while changes in the pH of a wound that’s in the process of healing can indicate an infection taking place.” When such sensors are embedded in structures designed to hold cells and channel liquids, miniature laboratories are obtained (an approach called ‘organ-on-chip’) that are very attractive tools for biomedical and pharmaceutical research. Replicating With the basic concepts proven in laboratory conditions, for Sasso and his team, the challenge lies in ways to precisely and reliably replicate the systems, paving the way for large-scale production and use. Their main technique to replicate polymer-based structures is that of soft embossing. In this approach, a mold is created and used to ‘stamp’ the desired structures into soft polymer material. The length scales of these structures vary from the micro (1 micrometer = 1 thousandth of a millimeter) to the nanoscale (1 nanometer = 1 millionth of a millimeter). “The smaller and more precisely the structure can be made and, most importantly, replicated, the better,” explains Sasso. Drug development Take, for example, research on drugs that help prevent kidney stones. Kidney stones are due to substances found in urine that initially form crystals and eventually macroscopic stones. . In order to speed up drug testing, biomedical researchers are looking for tiny test substrates that can host the basic material making up these crystals, and that allow the drug under study to be introduced and its effect on the crystal growth monitored. It has been found that the growth of the kidney stone crystals on polymer substrates can be controlled by carefully designing the surface structure: the more the surface mimics the crystal structure of the kidney stone material, the faster the material grows and the quicker it can be used to evaluate new drugs. Sasso’s approach to mass-produce finely structured polymer substrates can therefore help speed up drug development. 3D printing on the nanoscale Attaining the highest resolution is key to Sasso’s work, which is why his latest tool, nicknamed the ‘Nanoscribe’ after the company that supplied it, is such a big deal. “We now have access to all relevant length scales, from 20 nanometer to the millimeter scale.” To print 3D structures on these small scales, the Nanoscribe uses a technique inspired by conventional chip fabrication. During the chip fabrication step called lithography, a thin polymer layer is partly illuminated with UV light. The illumination affects the local chemical structure of the layer, so that, in a next step, the non-illuminated areas can be dissolved and removed, leaving only the parts that were originally illuminated. Instead of UV light, the Nanoscribe uses a 2-photon absorption process to illuminate the polymer layer, one 3D pixel of 20 nm in size at a time. “We are very excited to welcome the Nanoscribe. It opens new avenues for our group. Well before it arrived, my students were already planning one project after another.” The smallest Rietveld chair in the world The Nanoscribe will undergo a series of tests to verify its ability to produce nanoscale features as small as 20 nanometer. As a first exercise in operating the machine, Sasso programmed it to 3D-print a Rietveld chair of 100 micrometer (one tenth of a millimeter) in size. “Although huge in terms of what the Nanoscribe can do, we thought it was fun to start simple. The result is beautiful.” The new machine could reproduce the entire Madurodam miniature park in just a few square centimeters. Diversity In the development chain stretching from idea to product, Sasso focuses on enabling technologies, developing the methods and techniques that will be vital to fabricating future products on a larger scale, most notably tools for biomedical research. In other words, he brings exploratory laboratory concepts to users worldwide. Sasso comments: “Delft is a very fruitful place for this type of research. There is a very good collaborative spirit in the Department of Precision and Microsystems Engineering. Our researchers cover all relevant disciplines, from theoretical modeling to experimental approaches. This diversity allows us to look at challenges from different angles. Our engineering mentality makes us an attractive partner to complement biomedical experts.” Next frontier In his quest to find solutions to reproducibly structure materials and to increase the functionality of these structures, Sasso’s next frontier is that of composite materials, where non-polymeric materials, such as nanoparticles, are added to the mix, introducing new functionalities to the system but also new challenges to the production and replication process. “These are challenges we’re ready to take on,” concludes Sasso. NERI The department PME initiated the NanoEngineering research Initiative (NERI). NERI is a platform for long-term collaboration of industry and academia to jointly turn nanoscience into industrially relevant applications: Moving nano from lab to app. Moving nano from lab to app requires a new knowledge and technology foundation that allows the development and engineering of repeatable and reliable relevant functions and applications at an industrially relevant scale. This challenge requires cooperative efforts from both industry and academia. Luigi Sasso

Photonics for faster Internet

A major challenge in photonics is the accurate alignment of photonic chips, lenses and mirrors that are used in miniscule devices. Researcher Marcel Tichem works on a method to let this part of the assembly take place inside the device itself, known as on-chip assembly. That would make production more accurate and much faster than current manual production processes. A promising application for this novel assembly concept are devices for the transfer of data over fibre optic cables. Marcel Tichem is an Associate Professor at TU Delft’s department of Precision and Microsystems Engineering (PME). His work is set in the world of high-tech systems, where precision, speed and reliability are of the essence. “Companies like ASM International, NXP, ASML, and FEI create complex, high-performance machines and instruments. We develop new concepts for the control and actuation of precision systems and microsystems, and for automated design”, says Tichem. “Such concepts make it possible to meet the high demands of optical systems: the precision alignment of lenses and mirrors, for example.” Fibre optic data transfer One application of Tichem’s work is fibre optic data transfer. The bandwidth of traditional copper-based cable and the amount of data that can be sent are limited, but the demand for data traffic keeps growing. In fact, the explosive growth of online video and mobile data threaten to grind the internet to a halt. Then there is the sustainability factor. “Copper is not very energy-efficient”, he says. “That is why you find these colossal datacentres next to even more colossal cooling systems, often in a port area so enough water for cooling is available.” Using fibre optics could solve both problems. “You can transfer so much more data over fibre optic cable, and it doesn’t produce excess heat.” Chip design: TJ Peters, Photo: Hans de Lijser “Photonic chips work with light instead of electrical signals. They can detect or generate light waves, change their colour and so on”, explains Tichem. “Such micro-sized structures are very difficult to produce. Imagine, for example, that you have to make a tiny laser source that emits a stable beam and can continue to do so over a lifespan of a few decades.” After years of research, however, The Netherlands are now at the forefront of photonic chip development, and Photonic Integrated Circuits (PICs, the optic version of traditional microchips) can be produced at a reasonable price. But photonics are an order of magnitude more complex than electronics. Photonic devices, called packages, contain micro‐optics and electrical components, as well as PICs. On-chip MEMS for photonic alignment “Such a package is really a complex photonic system. Particularly the integration and alignment of all the parts make the assembly extremely complex, and industry does not have the technology yet to automate this process”, Tichem says. He is working on that though: his current research focusses on microscopic mechanical devices, or MEMS for on-chip photonic alignment. “We are integrating tiny mechanicals systems on the chips, so they can help align their photonic elements.” This research is part of an STW project as well as an EU-funded project PHASTFlex (Photonic Hybrid Assembly Through Flexible Waveguides) that is looking into automated assembly systems in response to the needs of the data communications sector. Within this project, cooperation exists with major European players working on applications and photonic chips (LioniX, Oclaro) and precision assembly (ficonTEC, Aifotec). Photo: TJ Peters” Micromachines 2016 , 7(11), 200; doi: 10.3390/mi7110200 A first achievement on the road towards full automation would be the automatic alignment of two PICs in one package. “In electronics assembly, machines can pick and place thousands of chips per hour. In optical assembly, this is still operator based, using microscopes and precision stages”, says Tichem. Project PHASTFlex is working on a two-step process. First, a machine picks up the PICs and places them on one substrate, with an accuracy of a few micrometres. The machines can do this, but this is still far too inaccurate for perfect alignment. That is the task of the on-chip MEMS. “On one of the chips we integrate an assembly machine consisting of tiny motors, or actuators. With the help of these actuators, you can slightly adjust the position of flexible waveguide beams on one chip in relation to the other.” As the name suggests, the assembly functionality is added solely for the purpose of assembly. “You only use it once. That takes space on your systems, and it costs money, but you have to offset that against current production methods that are labour-intensive.” PME The department of Precision and Microsystems Engineering (PME) is one of TU Delft’s departments working on technologies that can be used in optics. PME’s specific mission is to use the phenomena and technologies of the small-scale to create breatrhough innovations of smart materials, systems and devices. “We are mostly involved with mechatronics, meaning systems that have to move very accurately”, says Tichem. “Optics are only one field of application. Smart materials give new possibilities for improved performance of systems. Tichem explains: “The alignment of mirrors and lenses is critical in optics. However, alignment can change under the influence of temperature fluctuations, for example. You can try and minimise the effects of that by designing a structure, parts of which will not expand under the influence of heat.” That is called topology optimisation, a field that is also important for modern additive manufacturing technology. Optic technologies – or photonics, the term we use in relation to optical microsystems – are on the rise. Hence, a future chair of the department will focus solely on micro-optics and micromechatronics for photonics. Already the department is developing important enabling technologies. “A lot of what we do here can help others improve optic systems, whether these are medical instruments, inspection cameras or measuring instruments.” The department PME initiated the NanoEngineering research Initiative (NERI). NERI is a platform for long-term collaboration of industry and academia to jointly turn nanoscience into industrially relevant applications: Moving nano from lab to app . Moving nano from lab to app requires a new knowledge and technology foundation that allows the development and engineering of repeatable and reliable relevant functions and applications at an industrially relevant scale. This challenge requires cooperative efforts from both industry and academia. PME also takes part in the recently initiated Dutch Optics Center (DOC). Marcel Tichem