Detectors

The system strategy in the SMM/THz has often been to compensate for the relatively low powers of sources by taking advantage of sensitive detectors, which are made possible by the very low noise backgrounds in the SMM, or put another way the high brightness of cw electronic sources. Here we will discuss detectors in more detail.

Square Law Detectors

For the large majority of SMM experiments, square law detectors have become the detector of choice. While room temperature diode detectors are appropriate for many applications, because of their convenience and reliability, cooled bolometers have been widely adopted [1].and these detectors can approach the background limit. However, for systems in which 1 mW of probe power is available, Townes noise will have an impact and a detector that is limited by the very low thermal background limit is not useful.

The relative simplicity of square law detectors, especially at room temperature, makes them attractive candidates for use in focal plane arrays in imaging systems. In general the sensitivity of room temperature arrays has required the use of active illuminators to achieve adequate sensitivity [6]. but innovative cryogen free cooled systems have been successfully demonstrated [7].

Heterodyne Detectors

Heterodyne detectors offer a sensitive, room-temperature option, but at the cost of added complexity due to the requirement for a local oscillator. However, with the rapid development of mass-market microwave electronics, the penalty associated with this complexity is rapidly diminishing.

A heterodyne system with a receiver noise temperature of TN = 3000 K, 1 MHz of IF bandwidth, and 1 Hz of post detection bandwidth (see Eq. 1) has a noise power of \begin{equation}P_N = kT_N (\Delta v B)^{1/2} = 4 x 10^{-17} W\end{equation} Since heterodyne detectors use the same square-law circuit elements (typically diodes or microbolometers) as square law detectors, it is interesting to consider the relation of a diode used as a classic direct (square law) vs. its use as a mixer in a heterodyne receiver. To provide a simple comparison, let us consider an absorption spectrometer in which an FM modulated source, for which the discriminator action of the spectral line produces an AM sideband that contains the spectral information. If the carrier signal is comparable to that associated with a local oscillator (for room temperature diodes, ~1 mW), the diode mixes the sideband carrying the spectral information with the carrier to produce the spectroscopic signal at the modulation frequency. This conversion has exactly the same efficiency as the diode with an external local oscillator operating as a mixer in a heterodyne system!

Why then might we choose to have the extra complexity associated with the heterodyne system and why is the fundamental noise (equation above) of the heterodyne system so much lower than that of the square law detectors of the previous section? There are two reasons:

  1. In real diodes, especially those with considerable carrier or local oscillator power on them, there can be considerable $\frac{1}{f}$ noise. In an FM modulated system with a direct detector, the modulation frequency is limited to the spectral linewidth if one does not want to introduce modulation sidebands on the spectra. This limit is typically under 1 MHz. Often much lower signal recovery frequencies are used. In heterodyne systems the IF can be chosen arbitrarily by the offset between the probe and local oscillator frequencies.
  2. In many spectrometer systems, molecular saturation limits the probe power to several orders of magnitude below 1 mW. This significantly reduces the conversion efficiency of the diode. The detector noise then becomes several orders of magnitude smaller than the noise associated with the diode and its following preamplifier. This is particularly true for room temperature diodes as opposed to helium bolometers. Thus, heterodyne systems provide a path to sensitive room temperature systems, especially in the limit for which the carrier power is limited by molecular saturation.

Additionally, many applications are not absorption spectroscopic systems in which a small amount of power is absorbed from a more powerful probe. Common examples include imaging and communications systems in which the received signal will be relatively weak. In these cases, the ability of heterodyne systems to look at much smaller regions of the background noise spectrum and the efficient conversion mechanism of the high level mixer (to overcome the electronics noise) make a heterodyne system the system of choice. Additionally, these applications often do not require the complicated tracking of transmit frequency and detector frequency that is required in broad band spectrometer applications that use the frequency selective heterodyne receivers.

References

  1. Helminger, P., De Lucia, F. C. & Gordy, W. Extension of Microwave Absorption Spectroscopy to 0.37-mm Wavelength Phys. Rev. Lett. 25, 1397-1399 (1970). Google Scholar
  2. Helminger, P., Messer, J. K. & De Lucia, F. C. Continuously Tunable Coherent Spectroscopy for the 0.1- to 1.0-THz Region Appl. Phys. Lett. 42, 309-310 (1983). Google Scholar
  3. Winnewisser, G. Spectroscopy in the terahertz region Vib. Spectrosc. 8, 241-253 (1995). Google Scholar
  4. Pickett, H. M., Cohen, E. A., Brinza, D. E. & Schaefer, M. M. The Submillimeter Wavelength Spectrum of Methanol J. Mol. Spectrosc. 89, 542-547 (1981). Google Scholar
  5. De Lucia, F. C. Noise, detectors, and submillimeter-terahertz system performance in nonambient environments J. Opt. Soc. Am. B 21, 1273-1297 (2004). Google Scholar
  6. Luukanen, A., Miller, A. J. & Grossman, E. N. Active millimeter-wave video rate imaging with a staring 120-element microbolometer array Radar Sensor Technology VIII and Passive Millimeter-Wave Imaging Technology VII 5410, (2004). Google Scholar