Responsivity and NEP Improvement of Terahertz Microbolometer by High-Impedance Antenna

Abstract

The antenna-coupled microbolometer with suspended titanium heater and thermistor was attractive as a terahertz (THz) detector due to its structural simplicity and low noise levels. In this study, we attempted to improve the responsivity and noise-equivalent power (NEP) of the THz detector by using high-resistance heater stacked on the meander thermistor. A wide range of heater resistances were prepared by changing the heater width and thickness. It was revealed that the electrical responsivity and NEP could be improved by increasing the heater's resistance. To make the best use of this improvement, a high-impedance folded dipole antenna was introduced, and the optical performance at 1 THz was found to be better than that of the conventional halfwave dipole antenna combined with a low-resistance heater. Both the electrical and optical measurement results indicated that the increase in heater resistance could reduce the thermal conductance in the detector, thus improved the responsivity and NEP even if the thermistor resistance was kept the same.

Keywords: antenna; electron beam lithography; folded dipole; halfwave dipole; heater; noise equivalent power; responsivity; terahertz; thermal detector; thermistor.

Figure 1

Cross-sectional view of the fabrication processes and top view of the detector: (a) 50-nm-thick Ti thermistor deposition on top of 200-nm-thick thermally grown SiO₂; (b) Ti heater deposition on top of 100-nm-thick ECR SiO₂ interlayer. The heater is aligned on top of the thermistor device for effective thermal coupling; (c) cavity hole fabrication for thermal insulation of integrated heater and thermistor; (d) heater coupled to the antenna gap on top of meander thermistor. Heater width (W_h) is varied to attain wide resistance range; (e) halfwave dipole antenna (top) and folded dipole antenna (bottom) design structure used for THz characterization. Illustrated dimensions are not to scale with the real dimensions.

Figure 2

Simulation results of the designed halfwave (red) and folded (red) dipole antenna: (a) complex impedance characteristics representing real (solid line) and imaginary (dashed line) parts; (b) directivity pattern of halfwave dipole antenna; (c) directivity pattern of folded dipole antenna. The direction of θ = 180° correspond to the substrate side.

Figure 3

OM and FESEM images of the antenna-coupled detector: (a) detector with halfwave dipole antenna; (b) detector with folded dipole antenna; (c) fabricated meander thermistor on SiO₂ interlayer with effective length of 89.5 μm; (d) suspended heater and thermistor devices above the cavity hole for detector’s thermal isolation from the Si substrate.

Figure 4

Equivalent circuits and block for detector electrical and optical measurements: (a) circuit for electrical responsivity; (b) circuit for frequency response; (c) circuit for noise analysis; (d) THz optical responsivity measurement block; (e) electrical connection inside vacuum Dewar.

Figure 5

Material parameters of thermistor and heater: (a) thermistor ($R_t$) and heater ($R_h$) resistance dependence on heater width ($W_h$) with different heater thickness ($t_h$); (b) thermistor resistance and TCR dependence on temperature from 240 K to 300 K.

Figure 6

(a) Measured thermal resistance ($R_{therm}$) of the detectors with different heater resistance ($R_h$); (b) detector’s electrical responsivity ($R_{v-e}$) dependence to heater resistance ($R_h$). Standard deviation between measured and fitting results are 5.43 × 10⁵ K/W and 48.4 V/W for $R_{therm}$ and $R_{v-e}$, respectively.

Figure 7

Thermal resistance model in the microbolometer: a∙R_h represents the contribution of thermal resistance from the heater (red square box in Figure 1c), and R_{therm_base} represents the contribution of thermal resistance from other detector structure, i.e., the thermistor (blue square box in Figure 1c) and SiO₂.

Figure 8

(a) Thermistor output voltage to the change of temperature fluctuation frequency, estimated in CC mode; (b) detector’s cutoff frequencies ($f_c$) dependence to heater resistance.

Figure 9

Noise evaluation of the detector: (a) power spectrum density (PSD) estimated in CC mode; (b) electrical noise equivalent power ($NE{P}_e$) evaluated at 1 kHz, with the standard deviation of 4.5 pW/Hz^0.5 between measured and fitting results.

Figure 10

THz optical responsivity ($R_{v-o}$) and NEP dependence on heater resistance for different coupled antenna: (a,b) halfwave dipole antenna; (c,d) folded dipole antenna (FDA). The standard deviation of the responsivity and NEP is 16.1 V/W and 5.85 pW/Hz^0.5 for the halfwave dipole antenna and 124.5 V/W and 13 pW/Hz^0.5 for the folded dipole antenna, respectively.

Figure 11

Illustration of circuit diagram for heater input power modelling in THz characterization. Note that the complete model for THz responsivity includes the circuit model in Figure 7.