Beyond Gas Sensing Panel Discussion | FAQs

FAQs

Welcome to our FAQ section! This is where you'll find answers to some of the most common questions. If you don't see your question addressed here, feel free to reach out to our support team and we'll be happy to assist you.

What gases can be measured using optical gas detection?

Toxic gases have strong, unique absorption bands in the mid-IR region known as the fingerprint region. These stretching vibrations can be measured using a variety of optical gas detection techniques such as NDIR (Non-Dispersive Infra-Red), TDLAS (Tunable Diode Laser Absorption Spectroscopy and DOAS (Differential Optical Absorption Spectroscopy). Examples of these absorption wavelengths of commonly measured gases are 3.3 microns (Methane), 4.3 microns (Carbon Dioxide), 4.7 microns (Carbon Monoxide), 5.3 microns (Nitric Oxide), 7.7 microns (Hydrogen Sulphide), and 8.9 microns (Ammonia).

What is an InAsSb detector?

This is an exciting, new opto-semiconductor mid-IR detector that is ideally suited for gas detection. It is a direct replacement for lead salt (PbS and PbSe) and Mercury Cadmium Telluride (MCT) detectors that contain toxic lead and Mercury and are not RoHS compliant. The InAsSb detectors are RoHS compliant and have higher speed responses, better linearities, and stabilities compared to lead salt and MCT detectors. Additionally, InAsSb wavelength sensitivity is much wider than PbS and PbSe.

Down to what detection levels can gases be measured using these new InAsSb detectors and Quantum Cascade Lasers (QCL)?

Using a QCL as a source and an InAsSb detector, levels down to parts per billion (ppb) can be measured. This compares to detection levels of only down to % levels when a filament lamp source and pyro-electric detector are used.

What are the common light sources used in optical gas detection?

Either light sources that cover a broadband or a narrow region of the near-IR and mid- IR spectrum can be used. Examples of broadband sources are the graphene MEMS light source (emitting between 1-7.5 microns) and Xenon flash lamps (185 nm- 2500 nm). Examples of narrow sources include mid-IR LEDs (emitting at 3.3 microns, 3.9 microns, and 4.3 microns). Quantum Cascade Lasers are available at key wavelengths (emitting at 4.6 microns, 5.3 microns, and 8.9 microns) and customizable wavelengths to fit application needs.

In the oil and gas industry, there are significant new requirements for gas sensing and detection, particularly for CO₂, CH₄, and H₂, specifically for net-zero emission monitoring purposes. We're always looking for better H₂S monitors for major hazard toxic gas detection. Given that every photon is precious from the detector’s point of view, how weak are the signals you're dealing with, and what is needed to detect them?

H₂S is particularly difficult to detect because it is optically very inert. However, you stand a chance of detecting it with a quantum cascade laser device (QCL) because you start with a significant number of photons. To detect H₂S, very long optical paths are required, so the beam is reflected back and forth between mirrors and the cell itself. That's why you need upfront power since you lose some reflection each time you go through the cell. By the time the beam comes out, you don't have very much signal left.

 

From a beam quality perspective, if you want to launch a beam through such a system, you need a nicely dressed beam. Otherwise, after multiple passes, the beam quality may degrade significantly.  Those are all factors that come into play. Although it is feasible to detect H₂S using this method, it needs to be well-engineered and cost-effective for the industry.

 

The industry has long sought an optical method to replace electrochemical sensors for H₂S detection. While it is possible, it needs to be well-engineered.

 

Hydrogen is interesting as it hardly absorbs anything in the mid-infrared. MIR spectroscopy of hydrogen has been performed with kilometer path lengths, but many researchers are focusing on Raman spectroscopy. Raman has a one over lambda to the fourth power relationship with signal strength, so most work is conducted at shorter wavelengths.

 

In the oil and gas industry, the complexity of the gas samples often contaminated with various other substances makes choosing the optimal spectral region critical. QCLs are advantageous due to their narrow linewidth, so you can target specific spectral lines. Part of the modelling work we do is to tease out the optimum spectral region. If you have a complex gas sample, you can hunt through the spectroscopic database, simulate your gas mix, and see if you can pick out optimum lines. However, it’s crucial to have enough knowledge of the interfering species. Often, people want to measure a specific gas in an application without knowing what else is present.

When can we expect to see low-cost QCLs on the market?

Of course, we want everything to be cost-effective, but the direction of technological development is also crucial. This topic came up amongst the panelists and is something we have considered for some time.

 

QCLs’ active medium has a relatively broad spectrum. Consequently, a significant challenge with QCLs is that they tend to generate several wavelengths during manufacturing.

 

The actual manufacturing tolerances are very tight to hit a specific spectroscopic absorption line. This generally leads to low yield, increasing the price of QCLs.

 

That's the problem that needs to be solved. If you can solve that at the point of manufacturing and get your yield up, the price will come down. We have performed a simple analysis looking at how to recover the wafer cost, which involves growing, processing, and manufacturing QCLs.

 

The initial cost is high, and you need to break this wafer up into little chips and sell the chips. At a minimum, you've got to recover your costs, and ideally make a profit. With a low yield, only a few functional lasers cover the cost of the entire cost of the wafer, resulting in high prices. It's something like a one-over-X type characteristic.

Other questions? Contact us.

Contact us