Tech Radar | What caught our attention outside the world of oil and gas this month
October 25, 2017
No more tiny flaws
One of the biggest concerns in process control is to ensure that critical components are free of particles or impurities that could affect their operation.
One instance of this is in producing oil pans for vehicle engines. If the process lubricant contains impurities that stick to the areas where the sealant will be applied, the seal will not be tight and the oil pan is likely to leak at this vulnerable point. However, it has not been technically possible to examine every single component for contaminants.
A new inline fluorescence scanner developed the Fraunhofer Institute for Physical Measurement Techniques IPM in Freiburg hopes to alter that.
“This scanner not only enables us to perform inline measurements on every single metallic component – during the production process and without requiring additional time – but also enables us to pinpoint the exact location of the dirt particles,” said Andreas Hofmann, business development manager at Fraunhofer IPM. “The outstanding spatial resolution of this system enables us to identify even the slightest deposits or films of less than ten milligrams per square metre.”
The system uses a point-source UV laser to scan a specific area of the component. If traces of grease, remains of organic cleaning fluids or fibres are detected on the surface, they reflect light in the visible fluorescence spectrum as a response to the laser’s UV light. A detector captures these light frequencies and signals possible contamination.
The scanner is even capable of detecting metallic chips that adhered to the test object in previous machining steps, even though they are not fluorescent.
The scanner is not restricted to applications using metallic components – although the Fraunhofer team says that further studies will be needed to adapt it to other materials.
Playing the palladium
Researchers at the Korea Advanced Institute of Science and Technology (KAIST) have created an ultra-fast hydrogen sensor that can detect the gas at levels of under 1% in under seven seconds.
The sensor also can detect levels of hydrogen gas down to hundreds of parts per million within one minute at room temperature.
The sensor is based on a palladium (Pd) nanowire array coated with a metal-organic framework (MOF). Palladium has traditionally been used in detectors, but the use of an MOF coating – a simple process of dipping the wires in a methanol, zinc nitrate hexahydrate and 2-methylimidazole solution – enables the device to filter hydrogen particles better.
The porous material features micro-pores of 0.34 nm and 1.16 nm, meaning hydrogen gas with a kinetic diameter of 0.289 nm can easily penetrate inside the membrane, while large molecules are effectively screened. This means the detector has a recovery and response speed twenty times faster than pure palladium nanowires at room temperature.
Hydrogen’s lower explosion limit is 4% by volume in air, meaning any sensor must pick up the gas quickly. According to the US Department of Energy (DoE), hydrogen sensors should be able to pick up 1% volume within 60 seconds to ensure an effective response time.
Co-author Professor Il-Doo Kim of the Department of Materials Science and Engineering at KAIST added that other gases could be detected via the use of a variety of MOF layers.
Written by first author and PhD candidate Won-Tae Koo, Professor Kim and Professor R Penner of University of California-Irvine, the study has been published in the online edition of ACS Nano.
Flipping a switch
Wonder material graphene is a promising candidate for electronic devices, thanks to its excellent conductivity. However, there is a catch: electrons move through the material so well that they cannot be stopped. This precludes graphene’s use as a transistor, which must be capable of switching on and off.
Yet a new study from staff at Rutgers University-New Brunswick suggests a method of “taming” these excitable electrons, potentially enabling the material’s use as an ultra-fast transporter of electrons with a low loss of energy. Their study, published online in Nature Nanotechnology, suggests that it may become possible to create a graphene nano-scale transistor. The team managed to control electrons by sending voltage through a high-tech microscope with an extremely sharp tip – the size of one atom. They created what resembles an optical system by putting a voltage across a scanning tunnelling microscope, which offers 3-D views of surfaces at the atomic scale. The microscope’s sharp tip creates a force field that traps electrons in graphene or modifies their trajectories, similar to the effect a lens has on light rays. Electrons can easily be trapped and released, providing an efficient on-off switching mechanism, according to Board of Governors Professor in Rutgers’ Department of Physics and Astronomy in the School of Arts and Sciences Eva Y Andrei, also the study’s senior author.
“You can trap electrons without making holes in the graphene,” she said. “If you change the voltage, you can release the electrons. So you can catch them and let them go at will… In the past, we couldn’t do it. This is the reason people thought that one could not make devices like transistors that require switching with graphene, because their electrons run wild.”
According to the Andrei, the next step would be to scale up by putting extremely thin wires, called nanowires, on top of graphene and controlling the electrons with voltages.
The study’s co-lead authors are Yuhang Jiang and Jinhai Mao, Rutgers postdoctoral fellows, and a graduate student at Universiteit Antwerpen in Belgium. The other Rutgers co-author is Guohong Li, a research associate.
A group of researchers from the UK and the US have used machine learning techniques to predict earthquakes successfully in a lab environment.
Researchers from the University of Cambridge, Los Alamos National Laboratory and Boston University identified a previously hidden sound signal which precedes earthquakes and used it to train a machine-learning algorithm to predict future quakes.
The characteristics of this hidden sound pattern can be used to give a precise estimate (within a few percent) of the force applied to the fault and to estimate the time remaining before failure (with increasing precision as the failure approaches).
The team now thinks that this sound pattern is a direct measure of the elastic energy that is in the system at a given time.
“This is the first time that machine learning has been used to analyse acoustic data to predict when an earthquake will occur, long before it does, so that plenty of warning time can be given – it’s incredible what machine learning can do,” said co-author Professor Sir Colin Humphreys of Cambridge’s Department of Materials Science & Metallurgy, whose main area of research is energy-efficient and cost-effective LEDs.
Their results, which could also be applied to avalanches, landslides and more, are reported in the journal Geophysical Review Letters.