INVITED PAPER Quantum-Dot Infrared Photodetectors By optimizing dot growth processes, materials have been produced that demonstrate potential, with further development, for high-performance infrared sensing. By Joe C. Campbell, Fellow IEEE , and Anupam Madhukar ABSTRACT | We present a study of a series of n-i-n InAs quantum-dot infrared photodetectors (QDIPs) with uninten- tionally doped active regions. Different quantum-dot capping layer materials (GaAs, InGaAs, and AlGaAs) are utilized to tune the operating wavelength and modify the QDIP performance. Normal-incidence operation with high detectivity in the mid (3–5 "m) and long (8–12 "m) wavelength regimes and the potential for multicolor operation is demonstrated. KEYWORDS | AlGaAs; GaAs; infrared; InGaAs; photodetector; quantum dot; quantum-dot infrared photodetector (QDIP) I. INTRODUCTION Numerous civilian and military applications, including night vision, missile tracking, and environmental moni- toring, require high-sensitivity and low-noise infrared sen- sors, primarily for the midwavelength (MWIR, 3–5 "m) and long wavelength (LWIR, 8–12 "m) infrared transmis- sion windows [1], [2]. To date, the most widely used material for these applications is HgCdTe [3]–[5]. The excellent performance of HgCdTe is attributable to two material characteristics. By changing the ratio of Hg to Cd in the alloy, the bandgap can be tuned across the spectrum from 1 to 9 25 "m. The large absorption coefficients enable high quantum efficiencies. In addition, the recombination mechanisms yield high-temperature oper- ation and low-defect density material can be grown on transparent substrates. On the other hand, there are also several reasons that an alternative to HgCdTe has long been sought. These include difficulties in epitaxial crystal growth and processing which lead to spatial nonunifor- mities, low yield, and high cost. In contrast, mature materials growth technologies for III-V compound semi- conductors, namely, molecular beam epitaxy and metal– organic chemical vapor deposition, can provide large wafers (9 2 in) with very accurate control of compositions and layer thicknesses. The most advanced III-V MWIR and LWIR detectors, to date, are quantum well infrared detectors (QWIPs), which utilize intersubband or subband to continuum transitions in quantum wells [6]–[9]. QWIPs have demonstrated excellent imagery performance [7]; however, in order to achieved performance compa- rable to that of HgCdTe detectors, QWIPs require lower operating temperature, owing to their higher thermionic emission rates. Also a serious drawback is the fact that the n-type QWIPs cannot detect normal incidence radiation due to polarization selection rules [8]. Consequently, QWIPs require the addition of light couplers such as surface gatings, which add to cost and complexity. Re- cently, quantum-dot infrared detectors (QDIPs) have emerged as a potential alternative to HgCdTe and QWIPs. The motivation for interest in QDIPs is rooted in two characteristics of quantum dots. The first is that QDIPs are sensitive to normal-incident infrared radiation, a con- sequence of the 3-D confinement of electrons in the quantum dots [10]–[14]. The other attribute is the weak thermionic coupling between the ground state and excited states. This should result in lower thermal excitation and, thus, lower dark current and higher operating tempera- ture. The concomitant increase in the lifetimes of excited carriers should enable higher responsivities as carriers have more time to escape and contribute to the photo- current before relaxing to the ground state [15], [16]. In the past few years, QDIPs with peak photoresponse between 3  18 "m have been reported with detectivities ranging from 10 8 to 10 11 cmHz 1=2 =W at 77 K [17]– [25]. In this paper we review our approach to the design Manuscript received November 29, 2006; revised March 23, 2007. This work was supported by the U.S. Department of Defense Multi-disciplinary University Research Initiative (MURI) program in Nanoscience, administrated by the Air Force Office of Scientific Research under Grant AFOSR F49620-98-1-0474. J. C. Campbell is with the Department of Electrical and Computer Engineering, University of Virginia, Charlottesville, VA 22904-4743 USA (e-mail: jcc7s@virginia.edu). A. Madhukar is with the Nanostructure Materials and Devices Laboratory, Departments of Materials Science and Physics, University of Southern California, Los Angeles, CA 90089-0241 USA (e-mail: madhukar@almaak.usc.edu). Digital Object Identifier: 10.1109/JPROC.2007.900967 Vol. 95, No. 9, September 2007 | Proceedings of the IEEE 1815 0018-9219/$25.00 Ó2007 IEEE