Advanced sensors for power generation applications

The deployment of sensors that are capable of operating in the high temperatures and harsh environments of conventional fossil fuel-based power generators may significantly improve the efficiency of existing utility-scale power plants and reduce their greenhouse gas emissions. In addition, advanced sensors and controls are required to enable the widespread installation of highly efficient power generation technologies and processes, such as gasification, solid oxide fuel cells, gas turbines, advanced boilers, and oxy-fuel combustion (see Figure 1).^1,2 Temperatures involved in these processes range from 800–1600°C and the environments can be highly oxidizing, reducing, erosive, or corrosive. Such extreme conditions make the development of stable sensors challenging, particularly for chemical sensing applications.

Figure 1. New functional sensor materials that are stable within the hot zone of fossil fuel-based power generation technologies are currently being developed. They will be suitable for use in (a) fuel cell stacks, (b) gas turbines, and (c) advanced boilers.

Sensing technologies that are currently compatible with the harsh environments include chemiresistive,³ electro-chemical,⁴ and surface acoustic wave sensors.^{5, 6} All of these require electrical wiring, contacts, or connections at the sensing location, which are common sources of device degradation and present a safety concern when they are used with potentially explosive gas mixtures. A new optical-based sensing approach, however, offers several advantages.^7–9 With this system there is the potential for broadband wavelength interrogation, as well as compatibility with remote and distributed sensing technologies.²

Our work focuses on the development of advanced, stable, and functional sensor materials that respond to the critical parameters of interest. We synthesize and characterize material systems that undergo optical property changes in response to changes in the chemical composition of gas streams at relevant temperatures.^10–16 In parallel, we are working towards the fabrication and theoretical modeling of sensors. We are also developing packaging and integration methods that will ultimately allow our prototype sensors to be demonstrated successfully at appropriate conditions.

We investigated the common material systems that are used in chemiresistive sensing applications, e.g., tin dioxide (SnO₂) and titanium dioxide (TiO₂).^{10–12, 16} Our results showed that SnO₂-based sensing layers are not stable under the high-temperature and reducing conditions of interest. We also demonstrated that the incorporation of gold nanoparticles into these oxides can greatly enhance their optical signal response (see Figure 2). This is due to modifications to the characteristic gold localized surface plasmon resonance (LSPR) absorption peak.^15–18 We observed that gold nanoparticles preferentially occupy defects, grain boundaries, and twin boundaries in the crystals. This causes enhanced oxide stability in high-temperature and reducing conditions.^{11, 16}

Figure 2. (Top) Scanning electron microscope image of titanium oxide (TiO₂) with incorporated gold (Au) nanoparticles. Such materials demonstrate enhanced optical gas sensing responses under high temperature conditions relevant to power generation processes. (Bottom) The film transmittance of Au-incorporated TiO₂ at 700°C, for two different wavelengths (λ). The optical response varies as the volume of hydrogen (H₂) in an oxygen-nitrogen background gas stream changes.

We explored the optical sensing response of oxides, which are typically considered inert for gas-sensing purposes (e.g., silicon dioxide—SiO₂—and aluminum oxide), with incorporated gold nanoparticles. We successfully demonstrated high-temperature gas sensing responses up to 900°C.¹³ We also fabricated optical sensors based on SiO₂ materials with incorporated gold nanoparticles. These sensors displayed significant responses to common oxidizing and reducing gases at temperatures of 850°C, including a shift in the wavelength of the gold LSPR absorption peak. The characteristic temperature dependence of the gold LSPR peak wavelength also enabled optical temperature sensing using the same probe. These wavelength characteristics mean that it may ultimately be possible to achieve simultaneous monitoring of gas stream temperature and chemical compositions using a single sensor element through broadband interrogation. However, due to instabilities of SiO₂-based optical fibers at high temperatures, further work is required to improve the stability of the optical sensor devices.

As we searched for a broader range of materials that are suited to high-temperature stable optical gas sensing, we also studied transparent conducting oxides (e.g., aluminum-doped zinc oxide—AZO—and tin-doped indium oxide). Thin films of AZO have high levels of free carrier absorption and can therefore potentially be used for near-IR wavelength sensing applications.¹⁴ Although we conducted proof-of-concept studies for AZO films, these films are not stable in highly reducing conditions at temperatures greater than about 500°C. We have thus begun to focus on demonstrating similar effects in systems with better temperature stability.

We researched and developed new sensor materials that are stable in the high temperature, highly reducing, and highly oxidizing conditions typical of power plants. We fabricated the first prototype sensors based on these material systems. We plan to test these prototypes in realistic embedded sensing environments shortly. We also continuously seek to improve the technology readiness level of systems and to eventually create commercialized versions of our technologies.