Some of you will recognize that this a reprint of a blog topic from September 19th, 2009. It's a wonderful explanation of time-domain terahertz, and is worth having as a separate page.
Time domain, THz systems, (such as the API/Picometrix T-4000) are PULSED, which means they send out a series of broadband THz pulses (about 100 million of these per second). Pulses allow you to derive information because you can determine exactly when you send the pulse and exactly when it returns (time-of-flight information). You can imagine, this allows for 3-D imaging and with a high degree of precision, even when the objects move slowly or are static. This 3-D imaging approach takes advantage of the fact that it takes longer to reflect off of a surface that is farther away than one which is closer. This "time-of-flight information makes it possible to construct 3-D images, as well as to determine exact distances to target. In addition, this THz is BROADBAND. The term broadband means that it contains every frequency from about 50GHz to 4 THz (or 4,000 GHz). Different materials absorb different THz frequencies (providing a unique fingerprint so to speak). For example an explosive material may uniquely absorb at 1.15 THz and at 1.85 THz. It may be the only material that absorbs THz at these 2 frequencies. As a result, if the broadband THz pulse that is sent out comes back minus these 2 frequencies (meaning only these 2 frequencies were absorbed), the T-4000 can demonstrate it is the explosive material. This is what provides the spectroscopic aspect of THz, the unique absorption of certain THz frequencies. Since this sends out a broadband pulse, it can be determined what virtually any material is made of that comes into contact with this THz pulse (as long as it is absorbed uniquely by one or more of the THz frequencies). Thus, having PULSED THz provides API with 3-D and distance information, and having BROADBAND also provides spectroscopy (the ability to determine the chemical composition of the material due to its absorption and delay at unique THz frequencies). This is why this type of broadband THz technology is employed. It is my lay understanding that API, generates these THz frequencies using photoconductive techniques (techniques that use laser energy to activate it's patented optoelectronic semiconductors that in turn generate and detect THz). Photoconductive THz technology generates pulsed plus broadband THz, creates the lowest noise THz signal and has the highest THz output efficiency (relative to the input power).
In a nutshell, this technology is the best performing for subsurface imaging, uniquely identifying subsurface material and identifying subsurface feature defects. Contrast pulsed broadband THz (such as used by API) to a continuous wave (CW), single frequency THz technology (think of a flashlight which is always on). CW THz is very difficult (if not impossible) to construct 3-D images, since you can never really tell when the pulse that was sent out returns. Think of a flash light that you are shinning on your hand which is projected on the wall. You can see the image of a rabbit ear made by your hand on the wall in 2 dimensions but not 3 dimensions, and can't really tell how far away it is from the wall. In fact, you could not tell anything else about the "rabbit ear", even that it was made by a hand. In some cases this may be enough information, but not usually. In addition, CW THz is narrow band ( meaning a single frequency) , let's 1.15 THz for example. This is a problem in many real world applications. In the previous example, you would need not only 1.5 THz but also 1.85 THz in order to identify the explosive material. Thus a CW THz system running at 1.15 THz would not be able to identify the explosive material, it would need more frequencies for that. As you can see, THz systems based upon CW technology are much more limited but could offer a lower cost solution in some limited applications. CW systems might not have a lower cost advantage in the future as the cost of pulsed THz systems comes down driven by higher volume economies of scale and cost reductions from reduced laser costs. CW systems have historically been sub THz (20 GHz up to 100GHz) and usually deploy all electronic methods to generate the signal. That is the allure, but electronics is at the edge and the edge is nowhere near 1,000 GHz at this stage. Trying to move it there is a struggle against the laws of physics. One advantage of electronic GHz systems is that they can generate more power in that narrow frequency and thus could potentially send a 100 GHz signal a greater distance than pulsed broadband THz systems, which would be potentially handy for Satellite communications
(THIS POST is purely the opinion of the unnamed author, and it is not posted with the knowledge of, or consent of Advanced Photonix or it's management. I will correct any inaccuracy or remove any portion upon request).