Research ArticleAPPLIED SCIENCES AND ENGINEERING

Real-time frequency-encoded spatiotemporal focusing through scattering media using a programmable 2D ultrafine optical frequency comb

See allHide authors and affiliations

Science Advances  19 Feb 2020:
Vol. 6, no. 8, eaay1192
DOI: 10.1126/sciadv.aay1192
  • Fig. 1 Principle of the real-time FEST focusing technology.

    (A) A customized ultranarrow linewidth fiber laser, named kHz laser (B, see SM section S1), passes through two orthogonal acousto-optic deflectors (AODx,y) driven by specially designed multitone radiofrequency (RF) signals (section S3). The generated 2D-OFC beam array then feeds a high-speed spatial light modulator (HS-SLM), where the phase of each frequency-encoded sub-beam is controlled by a distinguishing pixel of the HS-SLM (C). The 2D-OFC beam array is subsequently randomized by a scattering medium (SM), resulting in a time-variant multifrequency scattered signal (D). It is then combined with a single-frequency reference beam through a beam splitter (BS) and beats at a high-speed photodetector (PD). The beating signal is sent to a fast signal processing unit, mainly including a fast data acquisition (DAQ) card and high-speed graphics processing unit (GPU), and converted to the frequency domain through fast Fourier transformation (FFT). The phase of each frequency component is retrieved by a four-quadrant inverse tangent operation [Atan2(imag, real)], and shaped into a 2D-conjugated phase map that is immediately addressed to the HS-SLM. A strong spatiotemporal focus (E) is thus generated through spatially phase locking among the optical frequencies. The whole FEST focusing system is controlled by a customized C++ algorithm with a low latency (section S7).

  • Fig. 2 Programmable 2D-OFC beam arrays.

    (A to C) 2D-OFC beam arrays that have 55 × 56, 35 × 36, and 23 × 24 array sizes, respectively. The inset on the right side shows a close-up of the sub-beams inside the red dashed rectangle in (C). (D) Frequency distribution over the 2D-OFC beam array of (A). Similar distributions are also designed for (B) and (C). The total RF bandwidth over the 2D-OFC beam array is about 60 MHz, set by the existing AODs. (E) Typical beating signal after combining the 2D-OFC beam array (B; i.e., the one used in all following demonstrations) with the reference beam at the PD. The total time duration of the beating signal is ~65.5 μs, defined by the pattern lengths of multitone RF signals used in this work. The inset exhibits small features of the beating signal. (F) Frequency spectrum of the beating signal after performing FFT to the beating signal (E). (G) Frequency spacing between the peaks in (F). Except for a few largely spaced frequencies arranged in a “v” shape, ~95% of the frequencies, indicated by the red dashed rectangle in (G), have a consistent frequency spacing of ~26 kHz. (H) Close-up of the frequency spacing at the center of the frequency spectrum, indicated by the yellow rectangles in both (F) and (G). PSD, power spectral density; a.u., arbitrary units.

  • Fig. 3 Spatiotemporal focusing through a scattering medium.

    Two pieces of optical diffusers (Thorlabs DG10-120) stacked together are used as the scattering medium. (A) Conjugated phase map measured by the FEST focusing system. Note that this full-phase map needs to be converted to a binary version when it is addressed to the binary-phase HS-SLM used in this work. (B) Temporal focus generated when the corrected phase map (A) is addressed to the HS-SLM. The temporal focus results from the phase locking among the frequencies presented in the probing location. The red rectangle zooms in the noise floor of the temporal signal. The black rectangle shows a close-up of the generated sharp pulse, compressed to a pulsewidth of ~19 ns. The black curve corresponds to the case that a random phase map is addressed to the HS-SLM. (C) Spatiotemporal focus measured by scanning the fast PD on the xy plane (see also movie S2). (D) Snapshot of the spatiotemporal focus, indicated by the blue dashed rectangle in (C), while (E) shows the case when a random phase map is addressed to the HS-SLM. (F) Temporal shift effect measurement. During the measurement, a phase map, which has been generated for the temporal focus at the center of the time axis (indicated by the red dot), is fixed on the HS-SLM. The temporal focus is shifted by ~34.9 μs when the scattering medium is laterally moved from −40 μm to +40 μm in the x direction. The inset shows the temporal focus at three displacements of scattering medium (indicated by blue, red, and green dots), where they are horizontally aligned and vertically normalized to show the consistency of the temporal focus pulsewidth.

  • Fig. 4 Imaging through scattering media.

    2D imaging is performed by xy-scanning the samples that are placed about 5 mm behind the scattering media, i.e., two optical diffusers and a breast chicken tissue for (A) to (C) and (D) to (F), respectively. (A and B) Bright-field images of a resolution target (group 4, USAF 1951) with corrected and random phase maps addressed to the HS-SLM, respectively. Images of other finer groups are also provided in fig. S10. Note that the intensities of (A) and (B) have been normalized to the maximum of (A). (C) Two line scans of images (A) and (B), respectively, indicated by the dotted lines. (D) Fluorescence signals of the rhodamine 590 dye (inset photo) with corrected (orange curve) and random (black curve) phase maps, detected by an experimental setup presented in fig. S2B. The top left inset shows the sharp fluorescence pulse generated by the spatiotemporal focus. Please note that the orange curve has been vertically offset for better visualization. (E and F) Fluorescence and bright-field images of a mouse kidney tissue (Thermo Fisher Scientific FluoCells Prepared Slide #3), captured at the emission and excitation wavelengths, respectively. [Photo credit for the inset of (D): Xiaoming Wei.]

  • Fig. 5 Dynamically focusing through scattering media.

    (A) Workflow diagram of the continuously streaming FEST focusing system. The bottom left insets show two dynamic scattering samples that are used for dynamic studies, i.e., chicken breast tissue mounted on a motorized translation stage and a living mouse. (B) Correlation curves of the chicken breast tissue moving at two different speeds, i.e., 0.5 and 1.5 mm/s, resulting in decorrelation times of 3.3 and 0.9 ms, respectively. (C) Temporal foci through the moving chicken breast tissue with different decorrelation times. (D) Correlation curve of the living mouse ear. (E) Correlation curve of the living mouse dorsal skin. (F) Peak intensity evolutions of the temporal foci generated through the living mouse ear, when the FEST focusing system continuously updates the phase map (blue curve) and when the phase map updates only once at the very beginning (i.e., t = 0 second, dark teal curve). (G) Sequence of the peak intensities of temporal foci generated through the living mouse dorsal skin. The inset shows a close-up of the temporal focus as indicated by the red circle. [Photo credit for the inset of (A): Xiaoming Wei.]

Supplementary Materials

  • Supplementary material for this article is available at http://advances.sciencemag.org/cgi/content/full/6/8/eaay1192/DC1

    Supplementary Text

    Section S1. Ultranarrow linewidth fiber laser

    Section S2. Full configuration of the FEST focusing system

    Section S3. Design of the driving RF signals for the 2D-OFC generation

    Section S4. Generation of the 2D-OFC beam array

    Section S5. Phase curvature of the 2D-OFC beam array

    Section S6. Mathematical description of the FEST focusing system

    Section S7. C++ control program

    Section S8. Effect of the laser linewidth on the wavefront measurement

    Section S9. Accuracy and reliability of wavefront measurement using FEST technology

    Fig. S1. Ultranarrow linewidth fiber laser and its basic performances.

    Fig. S2. Full configuration of the FEST focusing system.

    Fig. S3. Design of the 2D-OFC spatial map when N = M + 1.

    Fig. S4. Multitone RF signals.

    Fig. S5. Flatness optimization in the x direction.

    Fig. S6. Flatness optimization in the y direction.

    Fig. S7. Typical phase curvature of the 2D-OFC beam array.

    Fig. S8. Effect of the laser linewidth on the FEST performance.

    Fig. S9. Relative phase measurement.

    Fig. S10. Scanned images of the resolution target through the scattering medium using FEST focusing technology.

    Movie S1. Principle of FEST focusing technology.

    Movie S2. Spatiotemporal focusing.

    References (4146)

  • Supplementary Materials

    The PDF file includes:

    • Supplementary Text
    • Section S1. Ultranarrow linewidth fiber laser
    • Section S2. Full configuration of the FEST focusing system
    • Section S3. Design of the driving RF signals for the 2D-OFC generation
    • Section S4. Generation of the 2D-OFC beam array
    • Section S5. Phase curvature of the 2D-OFC beam array
    • Section S6. Mathematical description of the FEST focusing system
    • Section S7. C++ control program
    • Section S8. Effect of the laser linewidth on the wavefront measurement
    • Section S9. Accuracy and reliability of wavefront measurement using FEST technology
    • Fig. S1. Ultranarrow linewidth fiber laser and its basic performances.
    • Fig. S2. Full configuration of the FEST focusing system.
    • Fig. S3. Design of the 2D-OFC spatial map when N = M + 1.
    • Fig. S4. Multitone RF signals.
    • Fig. S5. Flatness optimization in the x direction.
    • Fig. S6. Flatness optimization in the y direction.
    • Fig. S7. Typical phase curvature of the 2D-OFC beam array.
    • Fig. S8. Effect of the laser linewidth on the FEST performance.
    • Fig. S9. Relative phase measurement.
    • Fig. S10. Scanned images of the resolution target through the scattering medium using FEST focusing technology.
    • Legends for movies S1 and S2
    • References (4146)

    Download PDF

    Other Supplementary Material for this manuscript includes the following:

    • Movie S1 (.avi format). Principle of FEST focusing technology.
    • Movie S2 (.avi format). Spatiotemporal focusing.

    Files in this Data Supplement:

Stay Connected to Science Advances

Navigate This Article