018-019-020_EETE-VF

EETE JUNE 2013

Fig. 3: IIIV/SiGe/Ge heterogeneous FinFETs for CMOS scaling. age level memory cell transistors. The new challenges for such memory architectures are about finding the materials that can withstand the temperatures of multilayer semiconductor processing, while enabling high aspect ratio etching to obtain the vertical channels. Wait and see who pays the bill for EUV The only speaker to really raise the issue of an economic barrier that could block future nodes was KLA Tencor’s CEO and president Rick Wallace. With billions of transistors on the most advanced semiconductor chips and hundreds of processing steps with 40 to 50 mask layers, things only get harder as design rules shrink. Wallace noted that while the costs of building fabs increased roughly 5 folds moving from the 90nm node to the 22nm node, the cost per transistor didn’t follow the same trend. In fact it may rise, and the only way to make it economically viable is to improve yield. You would expect that sales pitch from a semiconductor inspection tools vendor. The argument here is that you can’t control what you can’t measure, and you certainly can’t fix what you can’t see. Talking about EUV lithography which has yet to happen, Wallace reminded us that no equipment vendor can afford to develop the corresponding metrology tools purely based on speculation that this technology will be adopted, especially by only a few end-users who may decide in the end, that EUV is not what they want. “It is not a truism that everything is going to scale, economics could block that path” Wallace said, calling for more industry collaboration. “We have some ideas for the tools needed but if there’s no funding, we’ll wait and adoption will slow down”, he added. Healthcare goes mobile Another key topic covered during the imec technology forum was how the miniaturisation of health monitoring and diagnostic instruments could drastically cut public spending by encouraging consumers to wear their own health monitoring devices. Health spending drivers include hospital stays, physician or clinical care and nursing home care, often only to keep a tab on heart-beat, temperature, eventually get a blood sample for glucose monitoring or for further lab analysis. The ultimate goal would be to create disposable micro-fabricated test modules that would connect wirelessly to a dedicated mobile phone application or send their analysis results to web services for a better diagnostic – see figure 4. Such modules could host on-chip microscopy for cell identification and counting, on-chip DNA amplification and detection, on-chip protein and metabolite measurements for specific disease biomarkers, microfluidics for sampling blood, saliva or urine, and a low power wireless communication unit. The smartphone and its cloud-based apps would replace the expensive dedicated data processing instruments currently in use in clinical labs, while the disposable unit would only cost from 0.5 to 10 dollars. Imec has already demonstrated several building blocks for such a device. The on-chip microscopy element is achieved through the implementation of an in-line hologram, effectively the interference pattern of a reference beam with the light scattered by the tissues, an image can then be reconstructed from the interference pattern. In cooperation with Panasonic, imec has also developed a microreactor for DNA amplification. To replace costly flow cytometry equipment which can typically count 1000 cells/s, the lab has prototyped a cell sorting microfluidics platform that it believes is scalable to count over a million cells per second. Similarly, miniaturized bioreactors with integrated imaging and biochemical monitoring could be used to perform highly reproducible cell production for regenerative therapy, to develop organ-on-a-chip models for absorption, distribution, metabolism, and excretion (ADME) testing and toxicology. This could also be used as a tool to predict drug efficacy and unwanted toxicity/side effects, or even for personalized therapy. In his presentation, Peter Peumans, director for bionanoelectronics at imec, also discussed biocompatible miniature electronics to interface directly with the body’s control system through neurons. Featuring 456 electrodes able to connect to neuron terminations, the miniature prototype built in his lab integrates on-chip amplification, filters and analog-to-digital conversion for 52 channels. Aimed at neurosciences, this miniature instrument exhibits a very-low-noise (4μVrms) and is able to record and stimulate electrical signals. Another prototype development project is to use the verylarge scale integration of photonics and electronics for genome sequencing, effectively replacing multiple $500k machines with a single disposable chip. Peumans expects such a solution would increase genome sequencing throughput by two or three orders of magnitude. Program manager for body area networks at imec/Holst Centre, Julien Penders believes that turning healthcare into a consumer product with dedicated smartphone apps is a good way to prevent diseases while encouraging a healthier living. Junk food, the lack of physical activity and tobacco use are the biggest risk factors associated to chronic diseases. By putting medical data into consumers’ hands together with software applications to help them change their behaviour, Penders believes that wearable health monitoring solutions could save lives at a minimal cost. Fig. 4: Disposable micro-fabricated test modules could connect wirelessly to a dedicated mobile phone application. www.electronics-eetimes.com Electronic Engineering Times Europe June 2013 17


EETE JUNE 2013
To see the actual publication please follow the link above