Imaging CO2: An interview with laser imager Vic Miller

No one has yet captured an image of the CO2 spewing from car tailpipes.  Such video images could revolutionize how we think about the copious carbon pollution (20 lbs. of CO2/gallon) we emit when we drive.

To better understand how and when we might expect to get images of CO2 tailpipe emissions, I interviewed Vic Miller, an engineering post-doc at Stanford University, and an expert in laser imaging of gases. Vic is working on a technique to image CO2 from a tailpipe (among other projects).  His team’s laser-enhanced video of a match strike was recently featured in the New York Times.

Q. Why has CO2 imaging been so elusive?  What are the principal technical obstacles?

A.   CO2 is an infrared active molecule, meaning it absorbs (and emits) light in the infrared (IR).  And so I’m pretty sure qualitative measurements or images of CO2 emissions can be made with currently available off-the-shelf infrared cameras that are sensitive in the mid-infrared.  The tricky part is being quantitative – or getting an actual measurement of CO2 concentration with well-characterized uncertainty bounds.  This is what the NASA OCO-2 satellite does – it’s a quantitate instrument for measuring CO2 concentration.  The challenges lie in CO2 spectroscopy (i.e., understanding how CO2 absorbs light in the infrared, and what other species also absorb in the infrared and how they could interfere with CO2 spectroscopy) – and there are engineering challenges with making good measurements (is the signal on your detector actually what you think it is?).

Q.  What are the promising new techniques/technology for imaging?

A.  New infrared light sources (e.g., lasers) and infrared detectors are constantly being improved – these new pieces will provide access to better wavelengths (i.e., wavelengths that provide a more sensitive measurement with less interference) for CO2 detection.

Q.  Is it possible to measure it with non-visual means to model how CO2 would move out of a tailpipe?

A.  CO2 measurements can be made relatively cheaply with a capnograph.  CO2 measurements are readily made in medicine to monitor patients’ respiratory health (and other conditions) – the same thing could be done near a tailpipe. Computational fluid dynamics (CFD) could be used to model and observe tailpipe emissions.

Q. When do you think the new imaging techniques might be ready for providing car tailpipe images?

A.  We might be able to have a short little paper out in maybe a year – would like to do it in 6 months, but we probably won’t finish it that quickly.