Q&A: Trends in MEMS Fluid Handling Systems

Experts Discuss How Micro Technology Plays In Real-World Applications

Paul Galambos, Ph.D., is a principal member of the technical staff for the MEMS Group at Sandia National Laboratories. Dr. Galambos has 15 patents in the area of microfluidics and has taught the ASME short course on microfluidics packaging.

Stephan Braun is an R&D project manager and resident MEMS technology expert for Sensirion AG in Switzerland. In recent years, Mr. Braun has made major contributions to the development of Sensirion’s MEMS liquid flow sensor technology and the associated production processes.

Q: MicroElectroMechanical Systems (MEMS) have been getting increased attention in the fluid handling segment in recent years. In your opinion, what potential do MEMS technologies hold for fluid measurement and control applications?

Galambos: Any application that requires measurement and control of very small fluid volumes is ideal for MEMS. By small, I mean less than 100 milliliters/min for gas flows and less than 100 microliters/minute for liquid flows. Examples of applications that might require such small flowrates might include chemical and bio-chemical sensors (for instance, gas or liquid chromatography), vacuum applications, and drop generation (for instance, ink-jet printing and droplet-based fabrication). Another question is: Can MEMS be applied to larger-scale applications of importance to a larger segment of industry? I think they can in the sensing area, by potentially providing cheap, less intrusive, more easily distributed and integrated sensor networks that can measure and monitor a wide variety of fluid properties (such as density, viscosity, voltage, current, chemical reactions), as well as flowrate and pressure.

Braun: Sensirion has been working with MEMS technology more than 10 years. MEMS flow measurement and control solutions have made significant progress during this time. For the liquid flow sector, the most important evolution of technology in the MEMS area going forward will be new combinations of actuators and liquid flow sensors in biomedical systems. Sensors will be used for process monitoring only, but they can even be employed as part of fast-acting feedback control loops.

Meanwhile, the success of electrokinetic pumping is still difficult to forecast, even though this technology seems to have an interesting potential when combined with appropriate sensors.

For gaseous media, the low energy consumption and high speed of MEMS solutions will allow for more applications in the HVAC markets.

Q: How have MEMS technologies evolved over the past five years? How are the MEMS of today better than the MEMS of a few years ago?

Galambos: The past five years have seen the continued movement of MEMS from the lab and the university into the world of industry. There are now a significant number of MEMS companies* that allow easier access to MEMS technologies. In addition, packaging and systems integration technologies have been developed that allow integration of MEMS components to form functional microsystems.

* www.google.com/Top/Science/Technology/Mechanical_Engineering/MEMS/Companies/; edablog.com/2007/04/17/mems-companies/

Braun: The most important progress in real-world application environments has been made in the capabilities for volume production on a high level of quality. At the same time, a higher level of integration has helped to reduce total cost of ownership. Along this line, an important step has been the introduction of highly integrated MEMS sensors that provide analog and digital signal processing on the same chip. For the customer, this, in combination with simplified packaging technology, has improved the usability and effectiveness of MEMS sensor solutions.

Q: What are some common applications of MEMS technology for fluid measurement and control applications today? How are MEMS currently being employed in live fluid handling systems?

Galambos: The most common application for MEMS fluidics (microfluidics) is ink-jet printing, with companies such as Hewlett-Packard and Canon leading the market. Other applications include vacuum process sensing and control (e.g., Redwood Microsystems mass flow controllers and valves), fluid handling for chemical analysis (e.g., Upchurch Scientific) and biochemical analysis (e.g., DNA chips by Cephied, analysis cartridges by Micronics, point-of-care handheld blood analysis by Istat).

Braun: Apart from the well-known ink-jet systems, there are few applications with liquids that make use of active MEMS elements. Most sensor applications are in high-end devices for the biomedical sector. For example, some nano-liquid chromatography systems work with MEMS liquid flow sensors. In process automation, MEMS sensing and control solutions offer a much better price/performance ratio than conventional devices like, for example, conventional Coriolis flowmeters.

MEMS-based fluid measurement of gaseous media is already more common than liquid flow sensing because the compressibility of gas requires more sensing capabilities for precise process control. In addition, liquids getting in touch with MEMS sensor elements make the design requirements much more demanding regarding process compatibility and long-term stability.

An example of a real-world fluid handling application based on MEMS flow sensors is medical ventilation in systems used during surgeries. Such high-end ventilation systems benefit from the high speed of MEMS sensors.

Q: What are some of the common limitations of MEMS technology for fluid measurement and control? For what application scenarios are MEMS not a particularly good fit? Why?

Galambos: The primary limitation of MEMS is the flip side of the primary advantage of MEMS – i.e., small size. It does not make sense to use MEMS to move large volumes of fluid, gas, or liquid. Even ganging up a large number of MEMS actuators to move a larger volume often is less efficient or cost-effective than using a single larger actuator that is more properly sized for the fluid volume. This leads to the development of meso-scaled actuators and meso-systems that are between microsystems (100s of microns) and macro-systems (centimeters or larger) in size for many small-scale fluid handling applications.

Braun: MEMS is very good for relatively small mass flows. (Liquids: About 10 to 100 ml/min down to nl/min; Gas: All below some liters per minute). In applications with low flowrates and high accuracy and speed requirements, a MEMS flowmeter is nearly unbeatable. When considering large industrial systems handling very large volumes, MEMS systems are generally not a good fit. For most applications in this area, macroscopic solutions are usually more robust and flexible.

Q: How do you see MEMS technology evolving over the next 5-10 years, particularly in regard to fluid measurement and control applications? How will MEMS technology improve in the near term?

Galambos: I think MEMS can evolve into the sensing area of fluid measurement and control within the next few years. MEMS can be used to sense flowrate and pressure, as well as many other fluid properties, such as chemical concentration, density, viscosity, and electrical properties. MEMS sensors can be made cheaply, particularly in batch fabrications, ruggedly packaged, and are by definition small and unobtrusive.

In addition, MEMS can be integrated into distributed sensor networks with wireless communication between system elements. All these attributes can lead to competitive advantages for fluid measurement and control applications if the limitation associated with bringing disruptive technologies into a new market can be overcome.

Braun: Industry will still need some more time to get used to this new technology. However, in the long term this will help to drive MEMS costs down and bring more and more MEMS solutions into low-cost applications.
Process control for liquids in the range from nl to ml has not been common up to now. A market for this kind of measurement needs to develop while the demand for process monitoring and improved safety profiles is growing, for example, in the healthcare industry.

Q: Longer-term, what are some of the key obstacles you see that need to be overcome for MEMS to reach their ultimate potential? What needs to happen for those obstacles to be overcome?

Galambos: In order for a MEMS product to enter a new market, several factors need to be met. First, high-volume production is needed to take advantage of batch fabrication cost advantages for all but specialty (e.g., military or medical) products. This means there must be a demand for many MEMS sensors in a given market. This high volume of fabrication forces high yield and high reliability for the entire sensing system, not just the MEMS component. As a result, a somewhat long and expensive development process will be required. The specific costs for this product development depend on the specific application and product (for instance, medical products require a lengthy testing and approval cycle), but it is not unreasonable to expect that a development cycle might take multiple years and over $10 million.

All of these factors add up to the so-called “valley of death” – a several-year period of development in which there is little profit, significant development costs, and difficulty in obtaining funding. These challenges aside, the main reason MEMS companies fail is a mismatch between the technology and the market*. The key to a successful MEMS product introduction in the field of flow measurement and control is to synchronize the rollout of a new product that has significant competitive advantages to a market primed to be ready for that product.

* See comments by Kurt Peterson, Ph.D., in MEMS Investor Journal www.memsinvestorjournal.com/2010/07/mems-industry-overview-the-past-the-present-and-the-future.html

Braun: Reaching large-scale production (economy of scale) of MEMS flow sensors is a real challenge and this is not just depending on technology, but also on markets. For new applications in the liquid sector, crossing this chasm is a main obstacle.

The gas flow sensing sector has already made more progress today because the features of such MEMS products do fit in a better way with the requirements from markets that already exist today.

mems.sandia.gov
www.sensirion.com

Search for more Issue Archives

FOCUS AREA: MONTH: YEAR:

Current Issue

Flow Control Sponsors

Q&A: Trends in MEMS Fluid Handling Systems

Search for more Issue Archives

Request More Information

Most Viewed Articles

Most Viewed Products

Featured Jobs