Noise-Induced Limits of Detection in Frequency Locked Optical Microcavities

Shuang Hao, Judith Su

Research output: Contribution to journalArticlepeer-review

8 Scopus citations


Ultra-high quality (Q) whispering gallery mode (WGM) optical microcavities have been shown to be sensitive biomolecular sensors due to their long photon confinement times. We have previously experimentally demonstrated that a system known as FLOWER (frequency locked optical whispering evanescent resonator) can detect single macromolecules. FLOWER uses frequency locking in combination with balanced detection and data processing to greatly improve the sensitivity, stabilization, signal-to-noise ratio (SNR), and the detection limit of ultra-high-Q microcavities. Here we present the analytical basis for FLOWER and explore its limits of detection via numerical simulation. We examine the effects of key parameters such as Q-factor and frequency modulation depth on the SNR of FLOWER. We demonstrate that the frequency locked optical microcavity system is limited by the shot noise from the receiver, as well as the laser intensity noise. Using median filtering in combination with step-fitting algorithms, frequency locked ultra-high-Q microcavities can detect resonance shifts as small as 0.05 attometers at one millisecond time intervals. Our results can guide the choice of experimental parameters to achieve better sensing performance in a variety of target applications, including fundamental studies of protein-protein interactions and medical diagnostics and prognostics.

Original languageEnglish (US)
Article number9145600
Pages (from-to)6393-6401
Number of pages9
JournalJournal of Lightwave Technology
Issue number22
StatePublished - Nov 15 2020


  • Biosensing
  • frequency locked loops
  • microcavity resonators
  • noise cancellation

ASJC Scopus subject areas

  • Atomic and Molecular Physics, and Optics


Dive into the research topics of 'Noise-Induced Limits of Detection in Frequency Locked Optical Microcavities'. Together they form a unique fingerprint.

Cite this