🏆 Foundational Paper

A 512×512 SPAD Image Sensor with Integrated Gating for Widefield FLIM.

Ulku Arin C, Bruschini Claudio, Antolovic Ivan Michel, Charbon Edoardo, Kuo Yung, Ankri Rinat, Weiss Shimon, Michalet Xavier

📰 IEEE journal of selected topics in quantum electronics : a publication of the IEEE Lasers and Electro-optics Society 📅 2019 📊 202 citations
DOI PubMed

Abstract

We report on SwissSPAD2, an image sensor with 512×512 photon-counting pixels, each comprising a single-photon avalanche diode (SPAD), a 1-bit memory, and a gating mechanism capable of turning the SPAD on and off, with a skew of 250ps and 344ps, respectively, for a minimum duration of 5.75ns. The sensor is designed to achieve a frame rate of up to 97,700 binary frames per second and sub-40ps gate shifts. By synchronizing it with a pulsed laser and using multiple successive overlapping gates, one can reconstruct a molecule's fluorescent response with picosecond temporal resolution. Thanks to the sensor's number of pixels (the largest to date) and the fully integrated gated operation, SwissSPAD2 enables widefield FLIM with an all-solid-state solution and at relatively high frame rates. This was demonstrated with preliminary results on organic dyes and semiconductor quantum dots using both decay fitting and phasor analysis. Furthermore, pixels with an exceptionally low dark count rate and high photon detection probability enable uniform and high quality imaging of biologically relevant fluorescent samples stained with multiple dyes. While future versions will feature the addition of microlenses and optimize firmware speed, our results open the way to low-cost alternatives to commercially available scientific time-resolved imagers.

🔬 Techniques

Fluorescence Lifetime Imaging Microscopy Fluorescence Microscopy Widefield Fluorescence Microscopy

🔭 Microscopes

Ti

✨ Fluorophores

DAPI Rhodamine Rhodamine 6G

🧪 Sample Preparation

Fixation

🔬 Cell Lines

HeLa (Henrietta Lacks)

🏭 Microscope Brands

Olympus PicoQuant

🧪 Reagent Suppliers

Sigma-Aldrich Thermo Fisher Scientific

📷 Detectors

Andor EMCCD CCD EMCCD sCMOS SPAD

💻 Software Details

General:
LabVIEW

🏛️ Research Organizations (ROR)

Affiliated research institutions:

🏛 National HIV/AIDS/STI/TB Council 🏛 University of California, Los Angeles 🏛 u00c9cole Polytechnique Fu00e9du00e9rale de Lausanne 🏛 The Classical Association

📊 Figures

Fig. 1.

Fig. 1.

Cross section of the p-i-n SPAD [ 14 ].

Fig. 2.

Fig. 2.

(a) PDP characterization of the p-i-n SPAD [ 15 ], and (b) state-of-the-art peak PDP versus DCR comparison [ 14 , 16 , 17 ].

Fig. 3.

Fig. 3.

Pixel topology inspired by [ 23 , 19 ] but with distinct new features as described in the text.

Fig. 4.

Fig. 4.

Timing diagram of the sensor in rolling shutter mode. The number of frames per gate sequence and the number of gate sequences are user-selectable parameters. In this figure, both parameters are set to...

Fig. 5.

Fig. 5.

Die micrograph of the SwissSPAD2 image sensor, with its basic building blocks superimposed.

Fig. 6.

Fig. 6.

Block diagram of the SwissSPAD2 sensor architecture.

Fig. 7.

Fig. 7.

Block diagram of the SwissSPAD2 system with 8-bit photon counting feature on FPGA. This architecture, the most recent firmware of SwissSPAD2 that was tested so far, still has time intervals where the ...

Fig 8.

Fig 8.

Camera module designed for the characterization of the image sensor.

Fig. 9.

Fig. 9.

Dark count rate (DCR) performance of the sensor, (a) DCR map showing the distribution of hot pixels across the 472u00d7256 array, (b) Population distribution of the array under 5 different excess bias...

Fig. 10.

Fig. 10.

(a) Temporal and spatial SNR of SwissSPAD2 under different illumination levels, (b) Photon count distributions at low and high counts with Poisson fit points. 8-bit images were captured at a frame rat...

Fig. 11.

Fig. 11.

4 and 8-bit grayscale images of a rotating fan at 4.4 kfps and 274 fps.

Fig. 12.

Fig. 12.

Gating characterization of the shortest achieved gate with 5.75 ns length, (a) Gate window shape of each pixel in the array, (b) distribution of gate window rising edge position, and (c) distribution ...

Fig. 13.

Fig. 13.

Gating characterization of the gate with the lowest achieved skew with 22.8 ns length, (a) Gate window shape of each pixel in the array, (b) distribution of gate window rising edge position, and (c) d...

Fig. 14.

Fig. 14.

Intensity profiles of beads with 3 different fluorophore concentrations captured with Hamamatsu ORCA-Flash4.0 V2 sCMOS camera and SwissSPAD2 without microlenses. 625 nm LEDs were used for the excitati...

Fig. 15.

Fig. 15.

Fluorescence intensity images of HeLa cells labeled with DAPI, Alexa 488, and Alexa 555, captured with (a) Andor iXon Life 897 EMCCD camera and (b) SwissSPAD2.

Fig. 16.

Fig. 16.

Conceptual illustration of time gating in SwissSPAD2. The delay between subsequent gate positions is a small fraction of the gate length. Each gate window is sensitive to a different part of the fluor...

Fig. 17.

Fig. 17.

Fluorescence lifetime extraction of (a) Rhodamine 6G (R6G) solution, and (b) quantum dots (QD655) by fitting the decay by a single exponential integrated over a gate profile provided by the IRF sample...

Fig. 18.

Fig. 18.

Phasor plot (20 MHz harmonic) of quantum dots (QD655), Rhodamine 6G (R6G) solution, Cy3B solution for reference, and the IRF. After calibration, the phasor points are transferred to the region around ...

Fig. 19.

Fig. 19.

Fluorescence lifetime maps of the (a) R6G solution, and (b) quantum dots (QD655) captured with SwissSPAD2. Inset: Normalized intensity maps of the samples. In the 3D fluorescence lifetime mapping, 16u...

Figure images are served from the NIH/NLM PubMed Central Open Access Subset or Europe PMC; copyright remains with the publishers and authors.

🏛️ Imaging Facility

🏛️ National HIV/AIDS/STI/TB Council

💬 Discussion

0 comments

No comments yet. Be the first to start a discussion!

Leave a Comment

MicroHub Assistant