Photoconductive gain and generation-recombination noise in multiple-quantum-well infrared detectors

Abstract

The photoconductive gain, g, and generation-recombination (GR) noise of multiple-quantum-well infrared detectors are calculated for structures in which the well capture probability, pc, and the fraction of current derived from tunneling are allowed to vary through the structure. For uniform pc and no tunneling current, g=1/(Npc), where N is the number of wells. The GR noise power under the same conditions is 4eĪgB(1−pc/2), where e is electronic charge, Ī is the mean current, and B is the measurement bandwidth. When pc≪1, the noise contributions from carrier generation and decay (recombination) are equal as in a homogeneous photoconductor. However, when pc→1, the recombination noise decreases to zero, and the noise is equal to the shot noise of N series-connected, independent junctions.

Far infrared photoconductivity in a silicon based material: Vanadium supersaturated silicon

“…Therefore, the observed infrared photoresponse of this sample is mainly due to V deep levels introduced in the Si lattice as it has been previously observed for iron deep levels in Si 19 or Zn deep levels in Si. 20 Additionally, these V deep level impurities would act as an important source of localized recombination centers, increasing the electrical noise due to the recombination processes, [21][22][23] which could explain the measurements for this sample in Fig. 1.…”

Langevin approach to the generation–recombination noise of a multi quantum well infrared photodetector

“…As far as the basic physics of bulk semiconductors is used to describe the photoconduction process, the standard model of generation-recombination noise is used to describe the current fluctuations of QWIPs. Several papers dealing with the noise in Quantum Well Infrared Photodetectors have appeared in the literature [16][17][18][19][20][21][22][23][24][25][26][27]. Apart from differences in the relationship between the photoconductive gain g and the capture probability p c , and in the way to perform the sum over the periods, all these works rely on a noise model based on a photoconductivity relaxation dynamics occurring via a simple exponential process.…”

Towards dualband megapixel QWIP focal plane arrays

“…The responsivity of the detector peaks at 4.6 lm and the peak responsivity (R p ) of the detector is 170 mA/W at bias V B = À1 V. The spectral width and the cutoff wavelength are Dk/k = 15% and k c = 5.1 lm respectively. The photoconductive gain, g, was experimentally determined using [13] g ¼ i 2 n =4eI D B þ 1=2N , where B is the measurement bandwidth, N is the number of quantum wells, and i n is the current noise, which was measured using a spectrum analyzer. The photoconductive gain of the detector was 0.23 at V B = À1 V and reached 0.98 at V B = À5 V. Since the gain of a QWIP is inversely proportional to the number of quantum wells N, the better comparison would be the well capture probability p c , which is directly related to the gain [13] by g = 1/Np c .…”

“…The photoconductive gain, g, was experimentally determined using [13] g ¼ i 2 n =4eI D B þ 1=2N , where B is the measurement bandwidth, N is the number of quantum wells, and i n is the current noise, which was measured using a spectrum analyzer. The photoconductive gain of the detector was 0.23 at V B = À1 V and reached 0.98 at V B = À5 V. Since the gain of a QWIP is inversely proportional to the number of quantum wells N, the better comparison would be the well capture probability p c , which is directly related to the gain [13] by g = 1/Np c . The calculated well capture probabilities are 25% at low bias (i.e., V B = À1 V) and 2% at high bias (i.e., V B = À5 V), which together indicate the excellent hot-electron transport in this device structure.…”