Long wave infrared detection using probabilistic spintronic bolometer arrays

TL;DR

This work targets high-speed, high-pixel-density LWIR detection using digital-mode probabilistic spintronic bolometers (SUN) integrated with plasmonic nanoantennas. They demonstrate a scalable 2x2 row-column multiplexed SUN bolometer array with a plasmonic transduction layer, achieving broadband readout from 9 kHz to 3 GHz and high count rates on the order of tens of thousands to millions counts per second. The device relies on stochastic Magnetic Tunnel Junctions whose magnetization flips follow Poisson statistics; infrared heating increases the transition rate, yielding a digital count signal. The paper discusses advantages of row-column multiplexing for scaling to sub-micron pixels, the potential CMOS-compatibility, and the first array demonstration of row-column multiplexed stochastic MTJs, with measurements including NEDT $NEDT = rac{C_n}{(dC/dT)}$ and broadband operation, suggesting promise for near-field IR sensing and microscopy.

Abstract

The use of probabilistic spintronic devices for infrared radiation detection has introduced a shift in approach to thermal imaging. The integration of probabilistic magnetic tunnel junctions with infrared plasmonic nano-antennas achieves high-sensitivity digital-mode infrared sensors at room temperature. Here, we present a scalable approach towards multipixel plasmonic-spintronic bolometer array fabrication and readout. We fabricate proof-of-concept 2x2 row-column multiplexed probabilistic plasmonic sprintronic arrays and show their response to long-wave infrared radiation (8-14um) with high readout speeds (10K-1M counts per second). These spintronic, ultrafast, nanoscale (SUN) bolometers can result in novel high-pixel density CMOS compatible infrared detection platforms. Our work provides a broadband (9kHz to 3GHz) readout platform for future digital probabilistic detector applications. Furthermore, our approach addresses a key challenge associated with scaling infrared pixel sizes that can drive progress towards high pixel density detector arrays for infrared sensing and microscopy applications.

Figures (4)

• Figure 1: (a) The schematic of the row-column multiplexed spintronic bolometer array (b) SUN Bolometer device schematic demonstrating the operating principle of the device. The incident radiation is absorbed by the transduction layer. The transduction layer consists of plasmonic nanoantennas which enhances the absorbed radiation towards the spintronic nanopillar. This heat causes an effective temperature increase in the sensing layer. (c) Shows the SEM image of the single pixel SUN bolometer. (d) Measured absorptance of the SUN bolometer plasmonic nanoantenna array (orange) compared to 300K black body spectral radiance (purple). (e) The magnetization flips of the sensing layer are readout as transitions between two resistance states of the device. These transitions increase as the temperature in the sensing layer increases. (f) Histogram of the count interarrival times (i.e. the time between two nearest count events). The curve exhibits an exponential dependence indicating that the count statistics follow a Poisson process. Inset shows the measurement of interarrival times for a readout signal.
• Figure 2: (a) Schematic of row-column multiplexed array. (b) Optical image of a 2x2 SUN bolometer array and SEM image image of a single SUN bolometer with plasmonic integration (c) Readout schematic of the 2x2 multiplexed array- To select a device, each row and column are addressed using RF switches controlled by a digital control circuit. All the other devices are connected to AC ground on both terminals reducing leakage or splitting of their signal in the array. Through the RF switches, one row is biased with an input voltage and one of the column lines is connected to the readout path. The row line is AC grounded and the column line is DC grounded using a bias-tee. This ensures that the AC signal from only one device is readout by the column readout path. The other devices are either unbiased or have AC ground connected to both terminals. (d) Readout waveform of two devices on the same array. (e) Shows the count extraction algorithm where suitable thresholds and time windows are set to capture these ultrafast transition events.
• Figure 3: (a) NEDT measurement setup with f/0.5 LWIR optics. The extended array blackbody source has a stability of 1 mK. To measure NEDT the blackbody source is set to a fixed temperature and the device response is measured using 5 sequential measurements. Then the blackbody temperature is changed and device is measured again. (see supplementary s1(c)) (b) The setup is characterized using FLIR A325sc thermal camera. (c) Response of 4 SUN bolometer array devices to thermal radiation in NEDT measurements. The linear response and standard deviation are used to find NEDT. SUN bolometer error bars represent standard deviation of 5 measurements with 20ms integration time (50 Hz).
• Figure 4: (a) Advantage of digital detection mechanism for capturing the signal in high background scenarios (supplementary s1). (b) Comparison of readout architectures for infrared arrays - (i) Schematic representation of pixel level ROIC architecture. (ii) Schematic of proposed row-column multiplexed architecture. (c) Comparison of the NEDT trade-offs with the two architectures (suplementary sII). The dashed arrows mark the smallest detector devices currently fabricated in comparison with spintronic nanobolometers.