SPIE Membership Get updates from SPIE Newsroom
  • Newsroom Home
  • Astronomy
  • Biomedical Optics & Medical Imaging
  • Defense & Security
  • Electronic Imaging & Signal Processing
  • Illumination & Displays
  • Lasers & Sources
  • Micro/Nano Lithography
  • Nanotechnology
  • Optical Design & Engineering
  • Optoelectronics & Communications
  • Remote Sensing
  • Sensing & Measurement
  • Solar & Alternative Energy
  • Sign up for Newsroom E-Alerts
  • Information for:
SPIE Photonics Europe 2018 | Register Today!

2018 SPIE Optics + Photonics | Register Today




Print PageEmail PageView PDF

Optoelectronics & Communications

Silicon-nanowire detectors exhibit high sensitivity to IR light

Photodetectors employing vertically etched silicon nanowires show broadband absorption down to the near-IR regime because of strong interactions with surface effects.
7 October 2010, SPIE Newsroom. DOI: 10.1117/2.1201009.003161

IR detectors have numerous applications, from imaging to telecommunications. They have traditionally been made from low-bandgap semiconductors such as germanium (Ge) and indium gallium arsenide (InGaAs). Although silicon (Si) has dominated the semiconductor industry and is widely used in photodetectors of all sorts, it is not used for IR detection because of its wider bandgap and detection edge at 1.1μm. Use of Si for photodetection has many significant advantages, including ease of fabrication, large existing infrastructure, and low cost, encouraging researchers to extend the photosensitive wavelength of Si into the IR regime. Some of the proposed structures include SiGe heterostructures on Si,1,2 and deep-level impurity doping.3,4 However, because of both the materials' bulk or thin-film properties and fabrication constraints, these devices exhibit limited sensitivity.

We have demonstrated a new, promising approach by employing the unique properties of nanostructures5 to create a scalable detector from single-crystalline-Si vertical nanowire arrays, which can detect a broad spectrum of light from the UV to the near-IR wavelength range.6 These detectors can be thought of as phototransistors with an optically modulated gate.

The mechanism that allows Si nanowires to become photosensitive to wavelengths below their bandgap originates from their large surface-to-volume ratio, allowing the large number of electronic surface states within the bandgap to play an important role in device behavior (see Figure 1). Although IR light does not have enough energy to excite carriers from the valence to the conduction band in bulk Si, electron/hole pairs can be generated between the valence band and surface states. These electrons annihilate the holes trapped at the surface and modulate the gate field. The generated holes are confined to the center of the nanowire by the gate field. They increase the nanowire channel's conduction for a given lifetime until they are trapped at the surface. The gain of the phototransistor can be significant if this retrapping lifetime is much longer than the hole transit time.

Figure 1.Physical mechanism for IR detection in silicon (Si) nanowires. (a) Because of surface states, mobile charges are captured at the surface, causing depletion and band bending. (b) When IR light illuminates the device, electrons are excited from the valence band to a trap state, where they annihilate trapped holes. (c) Because of band bending, the photogenerated holes are confined to the center of the nanowire, where they increase the conduction of the nanowire channel. (d) After a certain lifetime, the holes are recaptured at the surface. Ef: Fermi level. Ec, Ev, Et: Conduction, valence-band, trap-state energies.

To fabricate truly scalable, large-area, high-sensitivity devices, we employed a top-down approach to form vertical nanowires using nanoimprint lithography and dry etching (see Figure 2). This enables simultaneous printing of large areas and precise control of the wire geometry. Nanoimprint lithography is first used to pattern arrays of nickel dots on the hole-doped substrate. These dots are used as both etch mask and self-aligned contact to the nanowires. The nanowires are then dry etched, which allows formation of very-high-aspect-ratio nanowires. They are then embedded in dielectric and linked to a transparent indium tin oxide top contact.

Figure 2.Scanning-electron-microscopy images of device fabrication after nanowire formation through dry etching. The nanowires are 200nm in diameter, 4μm in length, and 1μm in pitch.

Of notable interest in the IR spectrum is the 1550nm wavelength used in telecommunications. Measurements at this wavelength show a significant improvement compared to that expected in bulk Si. Figure 3(a) shows a detectable increase in photocurrent down to pW/μm illumination levels. As the optical intensity increases, the device shows a saturation effect similar to automatic gain control, which gives it a very large dynamic range. The responsivity of these devices—see Figure 3(b)—exhibits a peak at low intensities of 100A/W, showing a substantial gain.

Figure 3.Optical measurements of 10×10 Si-nanowire arrays using 1550nm light. (a) Photocurrent and (b) responsivity versus light intensity at various temperatures.

While these results prove that it is possible to build a highly sensitive IR detector using Si, significant improvements are needed to make the device properties more controllable. We currently rely on the effects of the native surface states to determine device behavior. The properties of these states can change depending on how the devices are processed. To better control and even engineer the exact properties for optimal device performance, we need to move towards engineering the device surface. One such direction that we are pursuing and that shows promise is to employ an atomic-layer-deposition system that would allow formation of monolayers of precisely controlled material at the nanowire surface. This could be used to create layers from passivation to homo- and heterostructures.

Yuhwa Lo
University of California at San Diego
La Jolla, CA

Yuhwa Lo received his PhD in electrical engineering from the University of California at Berkeley in 1987. He is a professor and the director of Nano-3 (nanoscience, nanoengineering, and nanomedicine). He has published 350 papers and eight book chapters and was awarded 24 patents. He is a fellow of both the Optical Society of America and IEEE.