Towards quantum 3D imaging devices

A time-gated 4×128 Single Photon Avalanche Diode (SPAD) line array with a 512-channel Time-to-digital converter (TDC) has been designed and fabricated in a 0.35 μm high voltage CMOS technology for pulsed Raman spectroscopy. In Raman spectroscopy the SPAD array can be designed as a line array with a high fill factor because electronics can be designed on either side of the SPAD array. In this paper cross talk measurements of the array have been carried out to investigate the probability of the cross talk between adjacent SPADs in an array with high density. Measurements showed a cross talk probability of 0.11% with a pitch of adjacent SPADs of 32 μm. In addition, the largest probability of the cross talk was measured to occur with a short delay.

The integrated circuit design of Single Photon Avalanche Diode (SPAD) with quenching circuit in CMOS is highly desirable photon detection at high rates, but low counting rates constrain image acquisition rates and dynamic range. The characterization should primarily define the dead time of quenching SPAD circuit in order to estimate the SPAD performance prior fabrication. This paper reports the development and characterization of the mathematical model SPAD on passively quenched SPAD circuit with ballast resistor. An improved model in defining the dead time (tD) response for SPAD is used to characterize the performance of 8×1 passively quenched SPAD array. The time response for both quenching and recharging time are developed for 180 nm depletion layer which means low voltage technology. Hence that, the 8×1 passively SPAD array circuit is designed by using Silterra 180nm CMOS technology for uncorrelated time measurements with on-chip 4-bit counter to improve the counting rate. The passive quenching circuit design on-chip would enable the capability to perform at higher speed, which is more than 100 kHz. In addition, are presented in this paper.

Single-Photon Avalanche Diodes (SPADs) are an emerging pixel technology able to detect the arrival of single photons. Arrays of SPADs can be used to image a scene in binary frames, indicating whether there has been a photon detection in a pixel during a very short frame exposure (e.g., few microseconds). Like many imaging technologies, SPAD sensors suffer from false photon hits, called Dark Count (DC), which increase the photon count, resulting in abnormally bright pixels, called hot pixels. Correction methods used for common CCD/CMOS sensors, such as dark frame subtraction, are ineffective for SPADs, which have non-linear response in observed counts due to their 1-bit quantization. Thus, we derive an analytical model of the count bias introduced by DC, which is signal-dependent, and provide a pixel-wise, closed-form DC correction algorithm. Remarkably, our correction is optimal in expectation, meaning that on average it removes perfectly the distortion. We validate our method on real SPAD acquisitions and synthetic data, showing that it significantly improves image reconstruction.

Microscopy resolution below the diffraction limit can be achieved by exploiting quantum light properties. NitrogenVacancy (NV) color centers in diamond, dye molecules and quantum dots are examples of single-photon emitters, whose antibunching property allows super-resolution imaging through the measurement of high-order autocorrelation functions. In this work, we present a novel Single Photon Avalanche Diode (SPAD) array architecture optimized for n-fold photon coincidence counting, in each point across the whole sensitive area. It is implemented in a 160 nm Bipolar-CMOS-DMOS (BCD) technology, and it includes 24 × 24 SPAD pixels with 50-μm pixel pitch and 10-μm SPAD diameter. Multi-photon coincidences (within time windows ranging from 2 ns to 500 ns) are identified by post-processing of the in-pixel timing data. Given the expected low photon rate on the detector in quantum imaging applications, on-chip logic discards unwanted information to limit readout throughput and data storage. In fact, reading the whole array would take 3 μs, while skipping rows detecting no photon reduces the readout time to 240 ns in case of no photon detected over the entire array. Moreover, we implemented a multi-gate approach, which avoids halting the array during readout, thus enabling multiple data acquisitions. Thanks to these power-saving expedients and efficient readout, the architecture is scalable towards multiple modules, such as 48 × 48 or 96 × 96-pixel arrays. Finally, it features the possibility of being coupled with a micro-lens array to reach a 78% equivalent fill-factor.

Silicon photomultipliers (SiPMs) are arrays of many single-photon avalanche diodes (SPADs), all connected in parallel. Each SPAD is sensitive to single photons and the SiPM gives an output proportional to the number of detected photons. These sensors are becoming more and more popular in different applications, from high-energy physics to spectroscopy, and they have been significantly improved over last years, decreasing the noise, increasing the cell fill-factor (FF) and thus achieving very high photon-detection efficiency (PDE). In FBK (Trento, Italy), we developed new SiPM technologies with high-density (HD) and, more recently, ultra-high-density (UHD) of cells (i.e. density of SPADs). These technologies employ deep-trenches between cells, for electrical and optical isolation. As an extreme case the smallest-cell, SiPM, i.e. with 5μm cell pitch, has about 40000 SPADs per squared millimeter. Such small SPAD dimensions gives a significantly high dynamic range to the SiPM. These small-cells SiPM have a lower correlated noise (including lower afterpulsing probability) and a faster recharge time (in the order of few nanoseconds), and they also preserve a very good detection efficiency (despite the small SPAD dimension).

Quantum imaging and microscopy profit from entangled photons to surpass the boundaries of classical optics, thus improving image resolution. Thanks to their single-photon sensitivity, readout noise absence, low-voltage operation and high frame-rate, detectors based on Single-Photon Avalanche-Diodes (SPADs) are particularly suited for this application field. We discuss strengths and weaknesses of different SPAD based architectures (classified in SiPMs, SPAD arrays, or SiPM arrays), highlighting those to be exploited as quantum imagers. As just SPAD arrays are capable of spatial resolution at single-SPAD level and, through the possible implementation of quantum-specific on-chip processing, we identified them as the forefront detector type for quantum imaging and microscopy.

We describe the integration of techniques and technologies to develop a Point-of-Care for molecular diagnosis PoC-MD, based on a fluorescence lifetime measurement. Our PoC-MD is a low-cost, simple, fast, and easy-to-use general-purpose platform, aimed at carrying out fast diagnostics test through label detection of a variety of biomarkers. It is based on a 1-D array of 10 ultra-sensitive Single-Photon Avalanche Diode (SPAD) detectors made in a 0.18 μm High-Voltage Complementary Metal Oxide Semiconductor (HV-CMOS) technology. A custom microfluidic polydimethylsiloxane cartridge to insert the sample is straightforwardly positioned on top of the SPAD array without any alignment procedure with the SPAD array. Moreover, the proximity between the sample and the gate-operated SPAD sensor makes unnecessary any lens or optical filters to detect the fluorescence for long lifetime fluorescent dyes, such as quantum dots. Additionally, the use of a low-cost laser diode as pulsed excitation source and a Field-Programmable Gate Array (FPGA) to implement the control and processing electronics, makes the device flexible and easy to adapt to the target label molecule by only changing the laser diode. Using this device, reliable and sensitive real-time proof-of-concept fluorescence lifetime measurement of quantum dot QdotTM 605 streptavidin conjugate is demonstrated.

Time-resolved non-line-of-sight (NLOS) imaging based on single-photon avalanche diode (SPAD) detectors have demonstrated impressive results in recent years. To acquire adequate number of indirect photons from a hidden scene in the presence of overwhelming amount of early-arrival photons, a single-gated SPAD is widely employed to mitigate pile-up. However, additional prior knowledge of the hidden range and relay surface profile are required to preset the gating position and implement the reconstruction. With this work, we propose a gradient-gated technique to alleviate these shortcomings (pile-up and prior knowledge) in NLOS imaging with the development of a 6 × 6 SPAD array fabricated in a standard 55-nm Bipolar-CMOS-DMOS (BCD) technology. The SPAD sensor includes a delay-locked loop (DLL) block, a gating generator and a pixel array, which can be flexibly configured to free-running, single-gating and gradient-gating modes. The rise time of the gates is less than 450 ps (from 10% to 90%). In gradient-gating mode, the interval time between adjacent gates can be configured from 300 ps to 2 ns. In single-gating mode, the minimum gate window is measured below 5 ns. We demonstrated the sensor in a confocal NLOS imaging system that reconstructs hidden scenes with both retroreflective and diffuse objects. The presented scheme enables the development of calibration-free NLOS imaging modality for practical applications.

To increase contrast in positron emission tomography (PET) images, researchers are investigating detectors that reach below the nanosecond timing resolution. This allows a tight coincidence window which reduces random coincidence counts in the acquired data, as well as to include time-of-flight information into the reconstruction algorithms. With this goal in mind, single photon avalanche diode (SPAD) arrays have been under study for their excellent timing performances. However, their spurious dark counts can blur the start of PET signals where timing information is the most precise and create false starts in the acquisition system, introducing dead time. To minimize these problems in digital SPAD systems using a single time to digital converter (TDC) per PET channel, dark count discriminator circuits are required to reduce timing errors and increase the triggering efficiency in presence of dark counts. This paper compares the performance of a probabilistic and a novel delay line based dark count discriminator. Simulations of a SPAD array investigate the impact of dark counts on triggering efficiency and coincidence timing. Results show that the probabilistic discriminator provides excellent event recovery with regard to dark counts at the cost of some coincidence timing resolution. On the other hand, the delay line discriminator maintains the peak coincidence timing resolution but does not provide as much efficiency at high dark count rate levels.

The growing interest for fast, compact and cost-effective 3D ranging imagers for automotive applications has prompted
\nto explore many different techniques for 3D imaging and to develop new system for this propose. CMOS imagers that
\nexploit phase-resolved techniques provide accurate 3D ranging with no complex optics and are rugged and costeffective.
\nPhase-resolved techniques indirectly measure the round-trip return of the light emitted by a laser and backscattered
\nfrom a distant target, computing the phase delay between the modulated light and the detected signal. Singlephoton
\ndetectors, with their high sensitivity, allow to actively illuminate the scene with a low power excitation (less than
\n10W with diffused daylight illumination). We report on a 4x4 array of CMOS SPAD (Single Photon Avalanche Diodes)
\ndesigned in a high-voltage 0.35 μm CMOS technology, for pulsed modulation, in which each pixel computes the phase
\ndifference between the laser and the reflected pulse. Each pixel comprises a high-performance 30 μm diameter SPAD, an
\nanalog quenching circuit, two 9 bit up-down counters and memories to store data during the readout. The first counter
\ncounts the photons detected by the SPAD in a time window synchronous with the laser pulse and integrates the whole
\nechoed signal. The second counter accumulates the number of photon detected in a window shifted with respect to the
\nlaser pulse, and acquires only a portion of the reflected signal. The array is readout with a global shutter architecture,
\nusing a 100 MHz clock; the maximal frame rate is 3 Mframe/s.

Silicon photomultipliers (SiPMs) are arrays of many single-photon avalanche diodes (SPADs), all connected in parallel. Each SPAD is sensitive to a single photon. The SiPM output is proportional to the number of detected photons. These sensors are becoming more and more popular in different applications and they have been significantly improved over last years, decreasing the noise, increasing the cell fill-factor, and thus achieving very high photon-detection efficiency. In FBK (Trento, Italy), we developed several different SiPM technologies and recently, pushing the fabrication process limits, we produced the new ultra-high-density (UHD) technology. This technology features deep trenches between cells, for electrical and optical isolation, and is characterized by a very small cell pitch, from 12.5 down to 5 μm. Such very small cell dimensions pose several issues both in terms of design and fabrication of the SPADs. TCAD simulations have been used to redesign the cell structure, minimizing their border region. As an example, the SiPM with 5 μm cell pitch has about 40 000 SPADs per squared millimeter, thus a very high dynamic range. These very-small-cells SiPM have a lower correlated noise (including lower afterpulsing probability), faster recharge time, and very good detection efficiency (despite the small SPAD dimension).

Scientific experiments often demand the detection of very weak light signals at high-speed or to precisely measure the time of arrival of single photons. Arrays of Single-Photon Avalanche Diodes (SPAD) are ideal candidates when high sensitivity is required together with high frame-rate or precise photon-timing resolution. We designed a linear 32x1 SPAD array using a high-voltage CMOS technology able to provide both good SPAD performance and fast electronics. During frame acquisition all pixels work in parallel, each of them being equipped with anything necessary for photon counting. The array architecture is capable of fully parallel operation of all pixels allowing free running acquisition at high frame-rate. With a low-speed 10 MHz clock frequency, one pixel is read out in 100 ns while the whole array is readout in 320 ns, corresponding to a frame-rate of 312.5 kframe/s. The frame-rate can top to 4 Mframe/s with a clock of 128 MHz. The photon timing modality employs the photon time-of-arrival information provided by each of the 32 outputs. All 32 "timing" outputs feed external Time-Correlated Photon Counting boards. The Full-Width at Half- Maximum using very short laser pulses is 55 ps with few kcps counting rate.

Single-photon detectors are employed to measure faint light signals with high detection efficiency and time resolution. Silicon photomultipliers (SiPMs) are arrays of Single-photon avalanche diodes (SPADs), each one with its quenching resistor; they are connected in parallel giving a signal proportional to the number of detected photons. They offer the typical advantages of solid-state detectors (e.g. ruggedness, low power consumption, small size), high detection efficiency, good time resolution [1] and very good photon-number resolving capabilities. In FBK we produce SiPMs and SPADs with detection efficiency peaked for blue/ultraviolet light (called NUV technology) [2], or green-red light detection (RGB technology), with different geometries. The former are typically employed in a large number of applications exploiting scintillators [3]. Looking for the maximization of the photo-detection efficiency (PDE), e.g. increasing the fill-factor (FF) of the SiPM cell, noise components such as afterpulsing and optical crosstalk between cells (i.e. the correlated noise) become important issues. Recently, we developed new NUV-SiPMs, with high detection efficiency and with different solutions to reduce the noise of the detectors, both at a technological level [4] and a cell-layout level. Here we show the performance of this new NUV-SiPM, produced at FBK, employing a new silicon substrate, with a reduced minority-carriers lifetime, and new SiPM with trenches between cells and high cell density.

Single-photon avalanche diodes (SPADs), integrated in CMOS technology, offer high-gain detection of low-intensity optical signals. SPADs are widely adopted in imaging systems, including both 2D and 3D depth sensing. As SPAD arrays scale to higher resolutions, backside illumination (BSI) is one approach to improving fill factor (FF) and routing, though it increases fabrication complexity. Alternatively, in frontside illumination (FSI), the integration of circuitry within the pixel area results in a significant reduction of the FF. Thus, this brief proposes an external counter architecture with a photon event summing and coincidence detection scheme. Unlike conventional FSI designs, it preserves FF and improves routing efficiency, which facilitates scalability. This architecture is well suited for SPAD-based imaging and 3D sensing, which require high pixel resolution and compact signal integration.

A state of the art Back Illuminated (BI) Single Photon Avalanche Diode (SPAD) array sensor realized via a 90nm CMOS compatible process on a 300 mm silicon platform is reported in the following. The array consists of 10 μm pixels, each using a 7μm thick silicon active layer, which allows to extend the device's optical sensitivity up to the Near-Infrared (NIR) spectrum. Furthermore, buried metal Full Trench Isolation (FTI) was employed to suppress crosstalk (X-talk), a critical feature in a device sensitive enough to be triggered by a single electro-luminescence photon emitted by a neighboring pixel. Finally, in order to maximize the Fill Factor (F.F) and allow a BI structure, a Cu-Cu bonding process was carried out. Through process and device design optimization, not only a Photon Detection Efficiency (PDE) greater than 14% at λ= 940nm and low timing jitter were achieved, but also the best in class Dark Count Rate (DCR) as compared to previously reported SPAD sensors. These combined features pave the way towards efficient, eye-safe and cost-effective time of flight applications.