Broadband Terahertz Computed Tomography Using a 5k-pixel Real-time THz Camera

Georgios C. Trichopoulos, Kubilay Sertel

Research output: Contribution to journalArticlepeer-review

16 Scopus citations


We present a novel THz computed tomography system that enables fast 3-dimensional imaging and spectroscopy in the 0.6-1.2 THz band. The system is based on a new real-time broadband THz camera that enables rapid acquisition of multiple cross-sectional images required in computed tomography. Tomographic reconstruction is achieved using digital images from the densely-packed large-format (80×64) focal plane array sensor located behind a hyper-hemispherical silicon lens. Each pixel of the sensor array consists of an 85 μm × 92 μm lithographically fabricated wideband dual-slot antenna, monolithically integrated with an ultra-fast diode tuned to operate in the 0.6-1.2 THz regime. Concurrently, optimum impedance matching was implemented for maximum pixel sensitivity, enabling 5 frames-per-second image acquisition speed. As such, the THz computed tomography system generates diffraction-limited resolution cross-section images as well as the three-dimensional models of various opaque and partially transparent objects. As an example, an over-the-counter vitamin supplement pill is imaged and its material composition is reconstructed. The new THz camera enables, for the first time, a practical application of THz computed tomography for non-destructive evaluation and biomedical imaging.

Original languageEnglish (US)
Pages (from-to)675-686
Number of pages12
JournalJournal of Infrared, Millimeter, and Terahertz Waves
Issue number7
StatePublished - Jul 5 2015
Externally publishedYes


  • Focal plane arrays
  • THz camera
  • THz computed tomography
  • THz imaging

ASJC Scopus subject areas

  • Radiation
  • Instrumentation
  • Condensed Matter Physics
  • Electrical and Electronic Engineering


Dive into the research topics of 'Broadband Terahertz Computed Tomography Using a 5k-pixel Real-time THz Camera'. Together they form a unique fingerprint.

Cite this