Low Cost Open-Path Instrument for Monitoring Atmospheric Carbon Dioxide at Sequestration Sites
DOE Project # 42212

Principle of FM and WM spectroscopy

Phase-insensitive FM/WM spectroscopy over long path

Principle of FM/WM spectroscopy

The priciple of Frequency Modulation and Wavelength Modulation spectroscopy has been well established. Many good websites, books and references are available and here we give links to a couple of the websites.

Phase-Insensitive FM over long paths

To employ the frequency modulation technique over long-path open-air conditions it is necessary to suppress the phase fluctuations. We chose a different approach to detection of the FM signal. We use a narrow band filter to pass the signal at the modulation frequency, and then detect its power. This technique is phase noise immune as power of signal is independent of its phase. Since the power of the received signal is expected to fluctuate due to changes in atmospheric transmission a real time normalization circuit is added to counter such fluctuations.

Figure 1 shows a block diagram of our phase insensitive Two-Tone Frequency Modulation (TTFM) scheme [2]. The high frequency (RF) was chosen to be 2GHz to accommodate the relatively large line width of CO2, ~4.5GHz or 0.15cm -1 , at normal atmospheric pressure. The low frequency LF is chosen to be 5MHz to make sure that the contribution of the 1/f noise is negligible. The field instrument uses a home-build fiber collimator and a commercial reflective 10” Newtonian telescope (Celestron) for launching and collecting the IR laser light, respectively. IR light is launched from the center of the aperture. The IR beam is reflected back at a certain distance by a retro-reflector array, which consists of 22 small aperture corner cube retroreflectors, and has a total aperture of size 14”x15”, figure 3.

Last year, we reported that we achieved a single pass loss sensitivity of 10 -4 in the lab with a sealed 75cm long tube filled with 1atm of CO2 , and over 1,000:1 SNR over a beam path of over 100 meters. Over the past year, we have taken our instrument out for test over 1 , 400m open air path, or 2.8km round trip, figure 2. Figure 6 shows the signal from the local reference channel and the open path as we ramp the laser wavelength (~1cm -1 ) around 1601nm. The signal to noise ratio is more than 500:1 over a 1 minute average and which provides sufficient sensitivity to conduct CO2 monitoring with 1ppm sensitivity during carbon sequestration processes. Figure 5 shows the measured CO2 concentration with our instrument over 24 hours. Figure 6 shows the measured CO2 spikes as we artificially created a CO2 leak on the roof top of our center.

Figure 1. Block diagram of homodyne detection scheme with extraction a reference sine wave from the detected signal.

Figure 2. The optical path (yellow line) between the instrument and retroreflector.

Figure 3. Optical transceiver & electronics on the roof of PEER & retroreflector 1.4km away on top of another roof.

Figure 4. Daily CO2 concentration in Covina , CA , 4 min. average

Figure 5. Sensitivity of the instrument to small manual controlled CO2 leaks.

Figure 6. Screenshot of the interface of custom designed data acquisition software.

In the coming year, we will conduct CO 2 monitoring at sequestration site. We gratefully acknowledge the funding support from DOE NETL under DOE Award No. DE-PS26-04NT42212.