COMSOL recently released a MEMS module for its FEMLAB simulation software. We recently spoke with Dr. John Dunec, the manager of the company's Palo Alto branch, about COMSOL current offerings in the MEMS area and its plans for the future. We also asked Dr. Dunec about emerging MEMS applications that he thought had promising prospects for growth over the next few years.
MEMS Investor Journal: There are a number of companies on the market today with MEMS simulation products. You've also recently launched your own MEMS module for your FEMLAB product. How does your software compare with competition such as Coventor, Intellisense and others and what makes it different?
Dr. John Dunec: We normally do not comment on our competitors work. However, our product has strengths that we believe set us apart. We focus on multiphysics analysis in which any combination of physical relations may be combined and solved simultaneously or sequentially. I think we are unique in being able to freely link arbitrarily coupled phenomena into one simulation, at the level of the describing equations. This basically gives unlimited multiphysics capabilities without having to write a single line of code (no need for creating and linking user subroutines). In addition, all variables known to the simulation are available to be used in custom definitions of materials, boundary conditions, or in defining custom feedback loops. These qualities are especially important in MEMS applications where non-standard materials and multiphysics couplings are common.
MEMS Investor Journal: Which are the most popular simulations that people do with your MEMS module?
Dr. John Dunec: Based on the Proceedings from our conference and our top 10 downloaded example models, we can see that our users model mostly sensors, actuators, and MEMS-microfluidic devices (e.g. Fluid-solid interaction models, lab-on-a-chip). Our ability to solve fully-coupled moving boundary problems, (electrostatic pull-down, squeezed-film damping, fluid-solid interaction…), piezo-electric materials, electro-osmotic and electrokinetic flow & diffusion problems have been very popular. The adaptability of our software and its ability to arbitrarily combine any set of physics has been well-utilized by the research community to simulate ferroelectric pumps, actuated shunt switches for RF MEMS, or atomic-force microscopy MEMS cantilevers. Here, it is our adaptability that is the popular feature. It is very exciting to see what users have done!
MEMS Investor Journal: Are you planning any additions to the MEMS module in the near future? What kinds of new features are you developing?
Dr. John Dunec: Our product already has many ready-made multiphysics couplings for the most common combinations of physics important to MEMS simulation. And while we have made it quite easy to define your own combinations, this requires that you know the physics, and corresponding mathematics pretty well. It is natural that we will add more ready-made multiphysics combinations for "turn-key" applications that will include material properties, proper boundary conditions etc. These are features we will continue to develop. In addition we will continue to build on our already strong foundation of solvers and application modes.
MEMS Investor Journal: You recently held a conference for your current users. Were there particular simulations that you were especially impressed with?
Dr. John Dunec: There were so many excellent papers that it is hard to single out particular simulations. However, if forced to choose, there are two that I myself was particularly excited about: Dr. Amy Duwel’s work at Draper Labs with High Q MEMS Resonators and Prof. Carl Meinhart’s Simulation of Electrokinetic Instabilities at UC Santa Barbara. Dr. Duwel leads a new project at Draper on using radioisotope sources to power a MEMS-scale battery. She initiated and leads the RF MEMS effort at Draper, which is focused on high frequency resonator development. She used COMSOL Multiphysics to analyze and optimize these resonators. Prof Meinhart is investigating highly nonlinear electrokinetic driven instabilities to promote mixing in microfluidic channels. Animations of the transient nonlinear mixing simulations he performed with COMSOL Multiphysics were both very impressive and beautiful to watch. I was particularly impressed when he correlated his COMSOL simulation results with experimental visualizations of the same phenomena at Stanford. This is a very nonlinear problem and I was very impressed with Prof Meinhart’s ability to simulate it and the apparent accuracy of the COMSOL result in its ability to capture the overall structure, periodicity and character of the experimental result.
MEMS Investor Journal: Based on your observations, which MEMS applications do think will see the most growth over next year or two?
Dr. John Dunec: From my perspective we will see continued penetration of MEMS sensors into commercial products, continued growth of microfluidics and rapid growth of BioMEMS and nano-technology. We can expect much commercialization of MEMS devices in life sciences both in the actuator and sensor arenas and in microfluidics and drop-delivery systems for drug discovery. In the commercial sector, RF tagging is poised to explode as are MEMS sensors and controls in consumer products – such as active-control anti-vibration systems in camera lenses. Finally I see continued growth of MEMS and MEMS sensors in the automotive industry – a strong traditional market for MEMS. All of these will require sophisticated modeling to design and develop into commercial products. We expect to see COMSOL Multiphysics emerge as the tool of choice for simulating cutting-edge designs and processes such as these.
Comments