Advertisement
Articles
Advertisement

Beating the Heat

Thu, 02/21/2008 - 6:26am
Paul Livingstone

Ceramics show promise in R&D for new jet turbines, where heat is the main cause of failure.

Hot section components in aircraft engines can degrade over time, causing catastrophic failure of the turbine section of aircraft engines. This is a problem that has plagued jet turbine development since the earliest days of jet-powered flight.

Even today, jet engine failure is still one of the most common factors in aviation accidents. Failures of the turbine section in airplane engines continue to be largest cause of lost aircraft to the U.S. Air Force outside of controlled flight into terrain. While such accidents are relatively rare, the visual and psychological effects of such mishaps can be dramatic.

“In-flight turbine failures in civilian airliners have made spectacular news headlines and can make for a bad day in commercial aviation,” says John Wrbanek, curator of Physical Sensor Instrumentation Research at NASA’s Glenn Research Center (GRC), Cleveland, Ohio.

Wrbanek is heading an effort to help minimize this wear and tear. His team at the sensors and electronic branch is investigating new forms of thin-film sensors that are able to withstand tremendous levels of heat (1100°C), tremendous strain (turbine engines run past 42,500 rpm), and vicious thermal cycles (some sensors have been tested at 1000°C for more than a million cycles).

“The new aircraft engines that are being developed for the Joint Strike Fighter and new civilian airliners are pushing the limits of what can be modeled inside the hot section of turbines,” says Wrbanek. “Any good model needs validation points, and good measurements in these severe environments are required for confidence in the design and operation.”

As engine technology progresses, noble metal thin films fall further and further behind. The temperatures that routinely exist in these engines are beyond the upper use limits of noble metal thin films, requiring higher temperature materials such as ceramic thin films.

The effort is part of the Aircraft Aging & Durability Project in NASA’s Aviation Safety Program. The scientists are examining new fabrication methods for thin films of basic nitride, oxide, and boride ceramics as well as various nano-structured materials.

Ceramics are the answer
According to Wrbanek, the primary advantage of thin film sensors over wire sensors is that the sensors are made part of the surface of the component that is being characterized without modifying the component.

Wire and foil sensors require machining into components or significant build-up of material that will change how the component reacts to the high temperatures and flows of the engine environment. With thicknesses of 10 µm or less, the thin films do not have as much mass as a wire or foil, so their transient response is much quicker as well, particularly to temperature.

The development of sensors to measure high temperatures and levels of static strain are a high priority for GRC’s instrumentation goals. Temperature sensors are, at a glance, simple—the most basic type, says Wrbanek. Temperature data is important because understanding the temperature change in operating engine conditions will reveal other characteristics.

The study of static stresses, however, is more complex and requires a different sort of sensor that would be used in many of the same areas as a temperature gauge.

“Understanding the static stresses that an engine component is undergoing from centrifugal, thermal, and vibrational stresses is critical, as well as stresses resulting from different materials in contact with each other in these conditions,” says Wrbanek.

These stresses are essentially steady-state phenomena, requiring sensors with output that is very insensitive to thermal gradients. Conversely, the output of sensors measuring high temperatures needs to be fairly sensitive to thermal gradients.

In transitioning to a new materials like ceramics, researchers at NASA were forced to adapt to the differing abilities and requirements for instruments made from such materials. Ceramics are highly sensitive to process changes, which means that slight modifications in the construction of sensors, whether intentional or unintentional, can have a huge impact on the film properties and appropriateness as a high-temperature sensor.

The trick, says Wrbanek, is to find the right combination of ceramic, process, and application to allow reliable measurements.

“The big breakthrough for us in developing ceramic sensors was demonstrating a fine-lined thin-film tantalum nitride rosette strain gauge using patterning technology originally developed for platinum thin films. This gauge demonstrated that anywhere we can put a metal film sensor on a component we can put a ceramic sensor and have feature sizes of 50 µm or less,” says Wrbanek.

This multi-functional rosette sensor was fabricated from tantalum nitride at NASA’s Glenn Research Center. Images: Glenn Research Center
This thin-film multi-functional sensor was made with an on-axis strain rosette and a Type R thermocouple.
A YSZ-alumina multi-layer insulates a resistance temperature detector on a nickel-alloy fan blade.


After extensive research by the GRC sensor team, several ceramic-metallic combinations have been identified as being the most promising for near-future sensor applications. Conductive ceramics based on high temperature metals, such as titanium and zirconium, have the most promise as thin-film sensors. The most popular for researchers are nitride and oxide ceramics, because they are easy to fabricate and modify in processing and can be used to fairly high temperatures.

Silicide and boride ceramics can be used to high temperatures as well but are more sensitive to variations in film purity and require poisonous gases to form from base metals.

“High-temperature carbides are also attractive, but for the oxidizing environment we are dealing with, probably are not the best choice,” say Wrbanek.

Fabrication process for thin-film ceramics
Tolerances are extremely tight in high-velocity turbine engines. For sensors to be useful, they must be very thin, and GRC is now assembling them to be less than 10 µm thick. This presents a significant but important issue for NASA, because these are physical sensors.

Wrbanek and his team have investigated a number of ways to assemble the films, but for high-temperature turbine component characterizations, physical vapor deposition (PVD) is the mature technology and is in common use for thin-film sensor applications.

PVD allows engineers to rely on the action of vaporizing the ceramic to form the film, or, alternatively, forming the ceramic film by vaporizing the base material in a reacting atmosphere. Sputtering PVD is appropriate for the sensor element fabrication because it limits deposition to a line-of-sight region on the component. In addition, the option of sensor patterning by lift-off or shadowing is defined by the chosen application. Lift-off patterning can allow fine features nearing the micron level but requires several processing steps to achieve.

Lift-off patterning requires photolithography processes, typically with a manufactured photomask, to define the sensor. Shadowing requires the pattern cut-out of a stencil mask laid over the component. The features of the shadow masks are limited to the mask thickness and ability to hold the pattern and are generally used for features on the order of a millimeter or larger.

“We have used both successfully in different applications,” says Wrbanek.

Industrial sectors for ceramic sensors
Turbine engine manufacturers, both large and small, will be the primary customers for this type of sensor. In turn, their primary client for high-performance engines is the military, not just the Air Force, but all branches. Advanced turbine design is important to military forces throughout the world, and technologies that improve engine durability and efficiency will be in demand everywhere.

“Besides the Joint Strike Fighter development and improvements to the Black Hawk helicopter,” says Wrbanek, “new propulsion systems are being designed for naval ships that include high-performance turbine engines.”

With tighter design tolerances on new civilian aircraft engines, the commercial aircraft industry benefits from this technology as well.

Another industry that is investing thin film sensor development is the power generation industry. Electric power is mostly derived from turbines, and thin film sensors are applied to turbine generators to better understand the conditions of their operation, which can be extremely hot and corrosive environments.

To a lesser extent, from Wrbanek’s observations, the automotive industry has shown interest in applying the sensors to components to validate and optimize design engine models.

The Univ. of Rhode Island’s chemical engineering dept., a partner in GRC’s development work, is developing nano-structured oxides for use as thermocouples and static strain gauges under a NASA Research Announcement grant.

In addition, NASA’s Fundamental Aeronautics Program is supporting thin film silicide sensor development for studying silicon carbide-based carbon matrix composites (CMCs) in high-temperature applications as part of their Supersonics Project. Fundamental Aeronautics is one of four programs funded under NASA’s Aeronautics Research Mission Directorate.

These projects indicate that demand for high-temperature thin-film sensors is high. According GRC’s research, the capability for thin-film sensors to operate in 1500°C environments for 25 hours or more is considered critical for future ceramic turbine engine developments. And as the industry moves toward space transportation vehicles, propulsion system components of at least 1650°C to 3000°C are expected. Ceramic materials have shown they can stand up to these extremes.

Resources

• NASA Glenn Research Center, Cleveland, Ohio, 216-433-4000,
www.grc.nasa.gov

• GRC’s Physical Sensor Instrumentation Research,
www.grc.nasa.gov/WWW/sensors/PhySen/index.htm

• Univ. of Rhode Island, Kingston, R.I., 401-874-2655,
www.uri.edu

Published in R & D magazine: Vol. 50, No. 1, February, 2008, p.32-33.

Advertisement

Share This Story

X
You may login with either your assigned username or your e-mail address.
The password field is case sensitive.
Loading