Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group

Monolithic resonant CMOS fully integrated triple-band THz thermal detector

Open Access Open Access

Abstract

Multiband terahertz (THz) detectors have attractive prospects in the areas of THz sensing and imaging. This paper presents a monolithic resonant CMOS fully integrated triple-band THz thermal detector that is composed of a compact loop antenna and an optimized proportional to absolute temperature (PTAT) sensor, leading to an uncooled, compact, low-cost, easy-integration, and mass-production multiband detector. The principles of operation, theoretical calculation, and experimental validation are demonstrated in detail. Calculated responsivities are 34.9 V/W, 51.6 V/W, and 47.6 V/W at the operation frequencies of 0.91 THz, 2.58 THz, and 4.3 THz, respectively, for the natural atmospheric windows. Relatively better experimental results are obtained at 0.91 THz and 2.58 THz due to the scarcity of THz sources, showing maximum responsivities of 29.2 V/W and 46.5 V/W with minimum NEPs of 1.57 µW/Hz0.5 and 1.26 µW/Hz0.5, respectively. The presented triple-band thermal detector has the natural scalability to focal plane arrays, providing great potential for multiband THz sensing and imaging systems.

© 2020 Optical Society of America under the terms of the OSA Open Access Publishing Agreement

1. Introduction

Terahertz (THz) technology has driven a wide range of important applications due to the unique characteristics of THz radiations. THz detectors, which are one of the most key components of THz applications, have drawn much attentions and been in widespread use for imaging, spectroscopy, and sensing fields [13]. By virtue of the assets of enhanced detection probability, improved calibration capability, reduced influence of standing waves and scattering, and achievable more informative images through fusion technology, multiband detectors dramatically improve the overall sensing and imaging ability as they are equipped with corresponding THz sources [46]. Considerable progress has been achieved with regard to multiband THz detectors in the last couple of years [712]. And there has been a strong desire to integrate multiband THz detectors on a CMOS platform because of its excellent advantages of low cost, mass production, and high level of integration for focal plane arrays [13].

Several CMOS integrated multiband THz detectors operating at frequencies below 1 THz have been proposed, which could be roughly divided into active detectors [5] and passive detectors consisting of antennas and MOSFETs [79]. Instead of active devices, passive devices are more suitable for human vision and image processing owing to the diffuse and natural illumination [14,15]. However, operating at higher frequencies is desirable for better spatial resolution. Although multiband CMOS integrated antenna-coupled field-effect transistors for the detection above 1 THz have been presented [1012], these detectors are composed of multiple discrete antenna elements in different sizes or types coupled with various field effect transistors, which inevitably results in lower integration levels, larger chip area occupation, and more expensive cost. Thus, it is desirable to present a multiband detector in a compact unit. However, compact multiband semiconductor detectors are hard to be implemented because the input impedance of the semiconductor transistors differs at different frequencies [8]. On the other hand, performances of MOSFET-based detectors degrade dramatically at frequency above 1 THz owing to the effects of frequency-dependent parasitic elements, showing decreased responsivities of 550 V/W, 30 V/W, and 4.6 V/W at 0.763 THz, 2.9 THz, and 4.1 THz respectively [7,10]. In contrast to multiband semiconductor detectors, CMOS integrated multiband thermal detectors with several assets of operating at room temperature, allowing radiation to be detected on a wider spectrum, and avoiding difficult multiband impedance matching constitute promising candidates [16].

This paper proposes an innovative compact triple-band THz thermal detector using a standard CMOS process for the first time to our knowledge. This type of CMOS fully integrated triple-band THz thermal detector is made up of a strong antenna and a sensitive PTAT sensor, which dramatically reduces the fabrication complexity and production cost compared to the detectors based on the present dominant technology such as micro-machining process or post-processing procedures. There exists a trade-off between better penetration of radiations below 1 THz and better spatial resolution at frequency above 1 THz. Therefore, the proposed triple-band thermal detector is chosen to operate at 0.91 THz, 2.58 THz, and 4.3 THz for natural atmospheric windows. In addition, the proposed triple-band THz thermal detector adopts an optimized PTAT sensor with considerably higher temperature sensitivity compared to the basic PTAT sensor applied in the single-frequency thermal detectors. The designed detector achieves relatively better characteristics with maximum responsivities of 29.2 V/W and 46.5 V/W and minimum NEPs of 1.57 µW/Hz0.5 and 1.26 µW/Hz0.5 at the operation frequencies of 0.91 THz and 2.58 THz respectively. The new approach to detect multiband THz waves and the natural scalability to focal plane arrays due to the availability of the biasing, read-out and addressing of arrays in the CMOS process present a significant advance towards an uncooled, compact, cost-effective, easy-integration, and mass-production multiband detection system [17].

2. Theoretical analysis

The schematic diagram of the proposed triple-band THz thermal detector is shown in Fig. 1. The incident single-frequency THz wave is first selectively absorbed by a compact on-chip triple-band loop antenna loaded with a polysilicon resistor. In this way, an instantaneous frequency-dependent current is excited in the antenna and flows through the polysilicon resistor at the termination of the antenna as the THz wave interacts with the passive antenna. By this means, the selectively absorbed electromagnetic (EM) energy is immediately dissipated into Joule heat by means of ohmic loss, dielectric loss, and conductive loss, which lead to a localized temperature increment before dissipating totally by the substrate [18]. The THz wave is indirectly detected by the bottom PTAT sensor which transforms the rising temperature into an increased output voltage. The triple-band detection is accomplished as THz waves of different frequencies are incident on the triple-band thermal detector successively. It is obvious that the detection of the triple-band THz thermal detector is depending on three energy transfer processes: selective absorption of EM energy, thermal energy conversion, and electrical response.

 figure: Fig. 1.

Fig. 1. Schematic diagram of the proposed triple-band THz thermal detector.

Download Full Size | PDF

The proposed triple-band THz thermal detector was fabricated in the 55 nm CMOS process with nine metal layers. The antenna was implemented in upper metal layers and dielectric layers, while the PTAT sensor was located in lower layers and in close proximity to the polysilicon resistor.

Responsivity and noise equivalent power (NEP) are two major performance indicators to evaluate the CMOS integrated THz thermal detectors. Responsivity which is defined as the ratio between the output voltage variation and the input power that the antenna receives, can be expressed as follows [19]:

$${R_v} = \frac{{\varDelta {V_{out}}}}{{{P_{in}}}} = \frac{{{P_{absorb}}}}{{{P_{in}}}} \cdot \frac{{\varDelta T}}{{{P_{absorb}}}} \cdot \frac{{\varDelta {V_{out}}}}{{\varDelta T}}$$
where Rv is the responsivity, ΔVout is output voltage variation, and Pin is the input power that the antenna receives. Pabsorb is the absorbed EM power by the frequency selective antenna and ΔT is the temperature increment caused by the absorbed EM power.

The coupling efficiency ηEM of the antenna can be defined as follows:

$${\eta _{EM}} = \frac{{{P_{absorb}}}}{{{P_{in}}}}$$

The thermal conversion efficiency βheat, indicating the ability of the device to convert the absorbed EM energy into the temperature increment, can be defined as follows:

$${\beta _{heat}} = \frac{{\varDelta T}}{{{P_{absorb}}}}$$

The electrical conversion efficiency θelec, which is applied to characterize the temperature sensitivity of PTAT sensors, is defined as the ratio of the output voltage variation to the temperature increment. It can be defined as follows:

$${\theta _{elec}} = \frac{{\varDelta {V_{out}}}}{{\varDelta T}}$$

Therefore, the responsivity of the detector can be expressed by:

$${R_v} = {\eta _{EM}} \cdot {\beta _{heat}} \cdot {\theta _{elec}}$$

According to above equations, the responsivity of detectors is dependent on the conversion efficiency of these three physical processes including selective absorption of EM energy, thermal energy conversion, and electrical response. Improved responsivity of detectors is especially desired for better image quality, further stand-off distance, and higher resolution of imaging systems [6]. It is readily accessible by optimizing collaborative design process of detectors such as improving signal coupling efficiency of absorbing structures, enhancing temperature sensitivity of temperature sensors, and reducing thermal losses of devices.

NEP is defined as the ratio of the noise spectral density (NSD) to the responsivity, as expressed by:

$$NEP = \frac{{{V_{rms}}}}{{{R_v}}}$$
where Vrms is the NSD and Rv is the responsivity of the proposed thermal detector. Thus minimum NEPs could be obtained by means of reducing the NSD and increasing the responsivity of detectors effectively.

3. Designs and simulations

3.1 Selective absorption of EM energy

Frequency selective antennas for coupling the radiation present advantages of readily shaping radiation patterns and achieving impedance matches within a wider bandwidth compared to FSS-like absorbers [20]. Antennas of high coupling efficiency are of course desirable. A compact triple-band loop antenna loaded with a polysilicon resistor was designed to absorb the incident THz radiations of any polarization at three different frequencies, as presented in Fig. 2(a). The proposed antenna was designed and optimized using High Frequency Structure Simulator (HFSS) tools where three-dimensional simulations were performed. The triple-band antenna was made up of a set of rings in different sizes, and smaller ring was embedded in bigger ring. The length of the loops is equal to the wavelength for the desired frequencies, thus the triple-band absorption of 0.91 THz, 2.58 THz, and 4.3 THz is contributed to the absorption of the outer loop, middle loop, and inner loop respectively [9]. The designed antenna was composed of antenna layers implemented in metal 9 and metal 8 layers, metallic ground plane, polysilicon resistor, and inter-metal dielectric regions. The outer loop was constructed in metal 9 layer, while the middle loop and the inner loop, which were connected to the outer loop by the connection structure in metal 8 and corresponding vias, were constructed in metal 8 layer. The metallic ground plane was realized in metal 3 layer to prevent the THz wave from penetrating into the substrate, while metal 1 and metal 2 layers were used for electronics routing. The polysilicon resistor at the termination of the antenna was fabricated in the polysilicon layer and vias from the loop antenna down to the polysilicon resistor formed a transmission line. Additionally, a grounded wall composed of metal layers and vias layers surrounding the antenna was adopted in order to trap the EM energy and prevent external interference. The inter-metal dielectric regions were modelled using SiO2, and metal layers were modelled as aluminum. The thicknesses and attribute of the metal layers and inter-metal dielectric regions were process dependent. The optimized geometric parameters of the triple-band loop antenna are as follows: Metal 9 outer loop length= 55 µm and width = 9 µm. Metal 8 middle loop length = 16.5 µm and width = 5.5 µm. Metal 8 inner loop length = 6.8 µm and width = 2.5 µm. The length and width of the connection structure between the outer loop and the inner loop are 24.5 µm and 5 µm respectively. The length and width of the metallic ground plane are 137 µm. The length, width, and height of the grounded wall is 137 µm, 7 µm, and 10.335 µm respectively. The thickness of metal 9, metal 8, and metal 3 layers is 1.325 µm, 3.35 µm, and 0.215 µm. The thickness of dielectric layers from the metal 9 layer to the metal 3 layer is 8.046 µm.

 figure: Fig. 2.

Fig. 2. (a) 3D model of the designed triple-band loop antenna in HFSS; (b) Simulated return loss of two kinds of antennas.

Download Full Size | PDF

The polysilicon resistor was optimized to obtain a maximum coupling efficiency and the increased Joule heat. It was set to be 100 Ω for impedance matching with the proposed loop antenna (sample A). In order to verify the frequency selective absorption of the incident THz waves at three different frequencies, sample B with only a metallic ground plane was also simulated for comparison. Figure 2(b) shows the simulated return loss (S11) curves of the two kinds of antennas. The triple-band loop antenna is resonant near 0.91 THz, 2.58 THz, and 4.3 THz, and there is a second harmonics at 1.8 THz of the outer loop which operates at 0.91 THz. Besides, there is no frequency selective characteristics of the sample B. The proposed triple-band loop antenna has simulated gains of 4.02 dBi, 4.92 dBi, and 4.07 dBi in the Z-axis direction, and simulated radiation efficiencies are 60.5%, 84.99% and 87.14% at the operation frequencies of 0.91 THz, 2.58 THz, and 4.3 THz, respectively.

3.2 Thermal energy conversion

The frequency selective absorbed EM energy by the antenna was dissipated into Joule heat which resulted in a localized temperature increment to the PTAT sensor. For the purpose of quantizing the temperature increment caused by the incident THz waves of three different frequencies, thermal energy conversion process was performed in COMSOL multi-physics simulation tools where a three-dimensional model of the triple-band THz thermal detector was built. Multi-physics modules including the frequency domain module of EM fields and heat transfer module were applied. Lumped ports and far-field domain were set in the frequency domain research of EM fields, enabling selective absorptions at three operation frequencies. EM energy of same power value at different resonant frequency was set in the excitation port in turn. The metal layers, dielectric regions and polysilicon resistor of the antenna were set as heat sources in the research of solid heat transfer. The increased temperature was obtained by collaborative simulations of frequency domain simulation and steady-state research.

Heat distributions of the CMOS integrated triple-band THz thermal detector operating at 0.91 THz, 2.58 THz, and 4.3 THz are shown in Fig. 3. Simulated temperature increments with an assumed incident power of 0.1 mW are about 0.56 °C, 0.59 °C, and 0.53 °C at 0.91 THz, 2.58 THz, and 4.3 THz, respectively. Therefore the thermal conversion efficiency is calculated to be 5.6 °C/mW at 0.91 THz, 5.9 °C/mW at 2.58 THz, and 5.3 °C/mW at 4.3 THz. Obviously, there exists a trade-off between coupling efficiency of absorbing structures and physical dimension of antenna elements. It can be seen that the polysilicon resistor which generates the maximum Joule heat constitutes the major heat source. Besides, the conductive loss of the metallic antenna which operates at different resonant frequencies and the dielectric loss in the first few hundred nanometers of inter-metal dielectric beneath the antenna also generate some Joule heat [6]. The generated Joule heat is transferred into the lower CMOS layers to the PTAT sensor which is located below the triple-band antenna and in close proximity to the polysilicon resistor.

 figure: Fig. 3.

Fig. 3. Heat distributions of the triple-band THz thermal detector: (a) Operating at 0.91 THz; (b) Operating at 2.58 THz; (c) Operating at 4.3 THz.

Download Full Size | PDF

3.3 Electrical response

PTAT sensors whose fundamental principle is based on the mechanism that the voltage difference between two bipolar junction transistors is proportional to the absolute temperature, are a type of common temperature sensors [21]. They could transform the absolute temperature increment into a desired output voltage variation proportionally. Although PTAT sensors occupies larger size and introduce higher noise because of its multiple transistors, it provides higher temperature sensitivity, better linearity, and greater measurement accuracy than other types of CMOS fully integrated temperature sensors such as pn diodes and MOSFET. The basic PTAT sensor, which is composed of a current mirror circuit, two bipolar junction transistors, and an output circuit, has a transducer gain of 3.06 mV/°C at 25 °C deriving from the slope of the plot in Fig. 5(a) [18]. Post-simulated NSD of the basic PTAT sensor is shown in Fig. 5(b).

For the purpose of further improving the responsivity of the CMOS integrated triple-band THz thermal detector, an optimized PTAT sensor with an improved transducer gain was designed using Cadence tools. In contrast to the basic PTAT sensor, the optimized PTAT sensor was mainly added in an operational amplifier circuit, which was applied to improve the temperature sensitivity of the PTAT sensor, as shown in Fig. 4. Post-simulated results of the optimized PTAT sensor are shown in Fig. 5. There is a highly linear plot from 10 °C to 100 °C and considerably higher transducer gain is obtained of 10.3 mV/°C at 25 °C compared to the basic PTAT sensor as shown in Fig. 5(a). Figure 5(b) shows the NSD of the PTAT sensor at different modulation frequency. The noise spectrum reveals that the noise is composed of flicker noise mainly existing in lower modulation frequency, and thermal noise dominating in higher modulation frequency. Post-simulated NSD values of 156.748 µV/Hz0.5 and 47.37 µV/Hz0.5 are obtained at modulation frequencies 0.2 Hz and 2 Hz, respectively.

 figure: Fig. 4.

Fig. 4. Schematic diagram of the optimized PTAT sensor.

Download Full Size | PDF

 figure: Fig. 5.

Fig. 5. Post-simulated results of the two kinds of PTAT sensors: (a) Output voltage against ambient temperature; (b) Noise spectral density versus modulation frequency.

Download Full Size | PDF

3.4 Calculated performances

According to the simulations of these three physical processes, the calculated responsivities of the CMOS integrated triple-band THz thermal detector using Eq. (5) are 34.9 V/W, 51.6 V/W, and 47.6 V/W at 0.91 THz, 2.58 THz, and 4.3 THz respectively. The NEP of the detector can be calculated using the simulated NSD derived from the Cadence tools and the calculated responsivity, showing calculated values of 1.35 µW/Hz0.5 at 0.91 THz with a chopping frequency of 2 Hz, as well as 3.03 µW/Hz0.5 and 3.29 µW/Hz0.5 at 2.58 THz and 4.3 THz with a chopping frequency of 0.2 Hz.

4. Measurements and results

The layout of the designed triple-band THz thermal detector and a device with only a PTAT sensor which is used to verify the resonant absorption of the incident THz wave at three different frequencies, is shown in Fig. 6(a). The proposed detector occupies the chip area of 150 µm × 210 µm, while the area of loop antenna is 146 µm ×146 µm, and the area of the PTAT sensor which is located below the compact triple-band loop antenna is140 µm × 190 µm. Figure 6(b) shows the die micrograph of thermal detectors with interface pins to facilitate the measurement. The die micrograph includes six detectors, while this work involves two of them.

 figure: Fig. 6.

Fig. 6. (a) Layout of the triple-band THz thermal detector and a device with only a PTAT sensor; (b) Die micrograph of thermal detectors.

Download Full Size | PDF

4.1 Temperature sensitivity

The temperature sensitivity of thermal detectors was characterized by the transducer gain of temperature sensors. The experimental transducer gain of the proposed triple-band thermal detector was determined by the measured output voltages at different ambient temperatures. The output voltage of the PTAT sensor was measured by placing the designed thermal detector in an environmental chamber and sweeping the temperature from 0 °C to 80 °C with a temperature interval of 5 °C [22]. The triple-band thermal detector has experimental transducer gain of 10.1 mV/°C at 25 °C as the PTAT sensor is biased at 2.5 V.

4.2 Responsivity

The responsivity of detectors was determined by the output voltage difference and the incident power that the antenna receives. The output voltage difference was obtained by monitoring the output voltages of the proposed detector as the THz wave was on and off [6]. The incident power on the detector was established by the effective area of the on-chip antenna and incident power density of the THz beam, which is derived from the total power of the THz beam measured by a power meter and the spot size at the focus point. In addition, the effective area of the on-chip antenna is calculated by the gain and the EM wavelength of the antenna at three different operation frequencies. The responsivity can be expressed as follows [13]:

$${R_v} = \frac{{\Delta {V_{out}}}}{{{P_{in}}}} = \frac{{\Delta {V_{out}}}}{{{J_{in}} \cdot {A_{eff}}}}$$
$${A_{_{eff}}} = \frac{{G \cdot {\lambda ^2}}}{{4\pi }}$$
where ΔVout is the output voltage difference, Pin is the THz beam power, and Jin is power density that the antenna receives respectively. Aeff represents the effective area of the antenna.

The block diagram of the output voltage measurement setup of the triple-band thermal detector as it operates at 0.91 THz is shown in Fig. 7(a). A backward-wave oscillator (BWO) with an average power of ∼125 µW at the operation frequency of 0.91 THz was applied as the incident THz source. Two parabolic optical mirrors collimated and focused the THz beam down to a diameter of about 1.15 mm. The triple-band thermal detector bonded on a PCB was mounted on a three-dimensional stage and positioned at the focus point of the THz beam. A mechanical chopper with a minimum chopping frequency of 2 Hz was placed between the THz source and the optical mirror to modulate the THz beam. A chopper controller was adopted to modulate the chopper and the SR830 lock-in amplifier synchronously. The detector was biased at 2.5 V and the output voltage signal was measured from a SR830 lock-in amplifier. The responsivities of the proposed triple-band thermal detector and the device with only a PTAT sensor at different operation frequencies are shown in Fig. 7(b). The proposed THz thermal detector resonates at 0.91 THz with the highest responsivity of 29.2 V/W at a chopping frequency of 2 Hz, which is close to the calculated value of 34.9 V/W. Furthermore, the responsivity of the device with only a PTAT sensor is close to zero since there is no significant voltage variation as the THz radiation is on or off.

 figure: Fig. 7.

Fig. 7. (a) Block diagram of the output voltage measurement setup as the proposed detector operates at 0.91 THz. (b) Responsivities of the two kinds of detectors.

Download Full Size | PDF

Figure 8(a) shows the output voltage measurement setup of the triple-band thermal detector as it operates at 2.58 THz. The incident THz source was a quantum cascade laser (QCL) operating at 2.58 THz with a maximum power of 60 mW and a duty cycle of 4%. The focused beam with a diameter of ∼500 µm was collimated and focused by two parabolic optical mirrors. The triple-band THz thermal detector was mounted on an x-y-z stage and positioned at the focus point of the THz beam. The responsivity as a function of modulation frequency ranging from 0.2 Hz to 5 Hz was acquired by modulating the THz source and the SR830 lock-in amplifier simultaneously using a signal generator. The output voltage was measured from the lock-in amplifier. Figure 8(b) shows the measured responsivities of the proposed triple-band thermal detector and the device with only a PTAT sensor with respect to the modulation frequency from 0.2 Hz to 5 Hz at room temperature. The proposed THz thermal detector resonating at 2.58 THz has the highest responsivity of 46.5 V/W at 0.2 Hz compared to the calculated responsivity of 51.6 V/W. Besides, the device with only a PTAT sensor shows responsivities of close to zero at different modulation frequency.

 figure: Fig. 8.

Fig. 8. (a) Block diagram of the output voltage measurement setup as the proposed detector operates at 2.58 THz; (b) Responsivities of the triple-band detector and the device with only a PTAT sensor.

Download Full Size | PDF

4.3 Thermal time constant

The thermal time constant which represents the duration for the temperature change and output signals rising from 0 to 63.2% of the stead-state values, is derived from the responsivity versus modulation frequency [6]. The thermal time constant value of the proposed triple-band THz thermal detector operating at 2.58 THz is extracted as 330 ms at a bias voltage of 2.5 V.

4.4 NEP

The experimental NEP was established by the measured NSD and responsivity of the proposed triple-band thermal detector. The measurement setup of the NSD was composed of a R&S FSV40 spectrum analyzer, a R&S HMP 4040 voltage source, the proposed triple-band detector, and a DC block located between the spectrum analyzer and the detector. Experimental NSD was obtained by measuring the noise voltage from 0.1 Hz to 10 kHz at room temperature from the spectrum analyzer as the detector was at a bias voltage of 2.5 V [22].

Figure 9(a) shows the NEP varies with different operation frequencies as the proposed triple-band detector operates at a fixed modulation frequency of 2 Hz, showing a minimum NEP of 1.57 µW/Hz0.5 at 2 Hz. The NEP as a function of modulation frequencies from 0.2 Hz to 5 Hz is shown in Fig. 9(b) as the proposed detector operates at 2.58 THz, showing a minimum NEP of 1.26 µW/Hz0.5 at 3 Hz. Thus, minimum NEPs of 1.57 µW/Hz0.5 and 1.26 µW/Hz0.5 occurs at 2 Hz and 3 Hz of the triple-band thermal detector as it operates at 0.91 THz and 2.58 THz respectively.

 figure: Fig. 9.

Fig. 9. (a) NEP varies with operation frequency at a modulation frequency of 2 Hz. (b) NEP with respect to modulation frequency as the detector operates at 2.58 THz.

Download Full Size | PDF

4.5 Performance summary

The designed triple-band THz thermal detector are summarized and compared with several passive CMOS fully integrated multiband detectors in Table 1. Dual-band and triple-band detectors consisting of antennas and MOS transistors operating at frequency below 1 THz have been demonstrated [79]. These three semiconductor detectors achieve better characteristics results, while operating at higher frequencies is desirable for better spatial resolution. Multi-band detectors in Ref. [12] and Ref. [13] achieve wideband detection of frequencies above 1 THz, but they are composed of multiple discrete antennas and FETs, resulting in lower integration levels, larger chip area, and more expensive cost. Although the proposed triple-band THz thermal detector has higher NEP and slower response speed than multiband semiconductor detectors, it constitutes promising options for the assets of realizing a compact multiband structure, allowing a wide spectrum detection, achieving better responsivity at higher THz bands, and avoiding difficult multiband impedance matching. In contrast to thermal detectors with micro-machining process or post-processing procedures, the proposed triple-band detector enables considerably reduced fabrication complexity and lower production cost.

Tables Icon

Table 1. Performance Summary and Comparison.

5. Conclusions

In conclusion, an uncooled monolithic resonant antenna-coupled triple-band THz thermal detector is proposed. This kind of detector which is composed of a compact triple-band antenna and an optimized PTAT sensor, is fully implemented in a standard CMOS process, dramatically reducing the fabrication complexity and production cost. The principles of operation, theoretical calculation, and experimental validation are demonstrated in detail. The triple-band THz thermal detector resonating at 0.91 THz, 2.58 THz, and 4.3 THz achieves relatively better measurement results. Moreover, the compact antenna structure is readily to achieve multiple resonant absorption at a wide spectrum for uncooled multiband detections. The proposed triple-band detector has the natural scalability to focal plane arrays which adopt necessary thermal insulation processes to trap the heat caused by THz radiations for bigger temperature increment and enhanced responsivity, showing significant advances to develop uncooled, compact, cost-effective, and mass-production multiband THz detection systems. Considerable efforts will be made on this type of CMOS fully integrated multiband THz thermal detectors for higher responsivity, reduced noise, and better response speed. The author will also try his best to get access to the 4.3 THz source and report the measured characterization results in future works.

Funding

National Key Research and Development Program of China (2016YFA0202200).

Disclosures

The authors declare no conflicts of interest.

References

1. W. Guo, Z. Dong, Y. Xu, C. Liu, D. Wei, L. Zhang, X. Shi, C. Guo, H. Xu, G. Chen, L. Wang, K. Zhang, X. Chen, and W. Lu, “Sensitive Terahertz Detection and Imaging Driven by the Photothermoelectric Effect in Ultrashort-Channel Black Phosphorus Devices,” Adv. Sci. 7(5), 1902699 (2020). [CrossRef]  

2. B. Globisch, R. J. B. Dietz, T. Göbel, M. Schell, W. Bohmeyer, R. Müller, and A. Steiger, “Absolute terahertz power measurement of a time-domain spectroscopy system,” Opt. Lett. 40(15), 3544–3547 (2015). [CrossRef]  

3. J. F. Federici, B. Schulkin, F. Huang, D. Gary, R. Barat, F. Oliveira, and D. Zimdars, “THz imaging and sensing for security applications—explosives, weapons and drugs,” Semicond. Sci. Technol. 20(7), S266–S280 (2005). [CrossRef]  

4. T. Takekoshi, T. Minamidani, S. Nakatsubo, T. Oshima, M. Kawamura, H. Matsuo, T. Sato, N. W. Halverson, A. T. Lee, W. L. Holzapfel, Y. Tamura, A. Hirota, K. Suzuki, T. Izumi, K. Sorai, K. Kohno, and R. Kawabe, “Optics Design and Optimizations of the Multi-Color TES Bolometer Camera for the ASTE Telescope,” IEEE Trans. Terahertz Sci. Technol. 2(6), 584–592 (2012). [CrossRef]  

5. K. Statnikov, J. Grzyb, B. Heinemann, and U. R. Pfeiffer, “160-GHz to 1-THz Multi-Color Active Imaging With a Lens-Coupled SiGe HBT Chip-Set,” IEEE Trans. Microwave Theory Tech. 63(2), 520–532 (2015). [CrossRef]  

6. I. E. Carranza, J. Grant, J. Gough, and D. R. S. Cumming, “Metamaterial-Based Terahertz Imaging,” IEEE Trans. Terahertz Sci. Technol. 5(6), 892–901 (2015). [CrossRef]  

7. M. Khatib and M. Perenzoni, “Response Optimization of Antenna-Coupled FET Detectors for 0.85-to-1-THz Imaging,” IEEE Microw. Wireless Compon. Lett. 28(10), 903–905 (2018). [CrossRef]  

8. R. Huang, X. Ji, Y. Liao, J. Peng, K. Wang, Y. Xu, and F. Yan, “Dual-frequency CMOS terahertz detector with silicon-based plasmonic antenna,” Opt. Express 27(16), 23250–23261 (2019). [CrossRef]  

9. L. J. Xu, J. N. Guan, X. Bai, Q. Li, and H. P. Mao, “A novel cmos multi-band thz detector with embedded ring antenna,” J. Infrared, Millimeter, Terahertz Waves 38(10), 1189–1205 (2017). [CrossRef]  

10. S. Boppel, A. Lisauskas, M. Mundt, D. Seliuta, L. Minkevicius, I. Kasalynas, G. Valusis, M. Mittendorff, S. Winnerl, V. Krozer, and H. G. Roskos, “CMOS integrated antenna-coupled field-effect transistors for the detection of radiation from 0.2 to 4.3 THz,” IEEE Trans. Microwave Theory Tech. 60(12), 3834–3843 (2012). [CrossRef]  

11. M. Bauer, R. Venckevicius, I. Kasalynas, S. Boppel, M. Mundt, L. Minkevicius, A. Lisauskas, G. Valusis, V. Krozer, and H. G. Roskos, “Antenna-coupled field-effect transistors for multi-spectral terahertz imaging up to 4.25 THz,” Opt. Express 22(16), 19235–19241 (2014). [CrossRef]  

12. K. Ikamas, D. Čibiraitė, A. Lisauskas, M. Bauer, V. Krozer, and H. G. Roskos, “Broadband Terahertz Power Detectors Based on 90-nm Silicon CMOS Transistors With Flat Responsivity Up to 2.2 THz,” IEEE Electron Device Lett. 39(9), 1413–1416 (2018). [CrossRef]  

13. Z. Liu, L. Liu, J. Yang, and N. Wu, “A CMOS Fully Integrated 860-GHz Terahertz Sensor,” IEEE Trans. Terahertz Sci. Technol. 7(4), 455–465 (2017). [CrossRef]  

14. E. Grossman, C. Dietlein, J. Ala-Laurinaho, M. Leivo, L. Gronberg, M. Gronholm, P. Lappalainen, A. Rautiainen, A. Tamminen, and A. Luukanen, “Passive terahertz camera for standoff security screening,” Appl. Opt. 49(19), E106 (2010). [CrossRef]  

15. J. Sun, Y. Zhu, W. Feng, Q. Ding, H. Qin, Y. Sun, Z. Zhang, X. Li, J. Zhang, X. Li, Y. Shangguan, and L. Jin, “Passive terahertz imaging detectors based on antenna-coupled high-electron-mobility transistors,” Opt. Express 28(4), 4911–4920 (2020). [CrossRef]  

16. D. Nguyen, F. Simoens, J. Ouvrier-Buffet, J. Meilhan, and J. Coutaz, “Broadband THz Uncooled Antenna-Coupled Microbolometer Array—Electromagnetic Design, Simulations and Measurements,” IEEE Trans. Terahertz Sci. Technol. 2(3), 299–305 (2012). [CrossRef]  

17. I. E. Carranza, J. Grant, J. Gough, and D. R. S. Cumming, “Terahertz Metamaterial Absorbers Implemented in CMOS Technology for Imaging Applications: Scaling to Large Format Focal Plane Arrays,” IEEE J. Sel. Top. Quantum Electron. 23(4), 1–8 (2017). [CrossRef]  

18. F. Chen, J. Yang, and Z. Li, “Modeling of an Uncooled CMOS THz Thermal Detector With Frequency-Selective Dipole Antenna and PTAT Temperature Sensor,” IEEE Sens. J. 18(4), 1483–1492 (2018). [CrossRef]  

19. F. Chen, J. Yang, and Z. Li, “Modeling of an uncooled CMOS THz thermal detector with frequency-selective cross-dipole antenna and NMOS temperature sensor,” Microw. Opt. Technol. Lett. 59(12), 3160–3166 (2017). [CrossRef]  

20. D. Corcos, N. Kaminski, E. Shumaker, O. Markish, D. Elad, T. Morf, U. Drechsler, W. T. S. Saha, L. Kull, K. Wood, U. R. Pfeiffer, and J. Grzyb, “Antenna-Coupled MOSFET Bolometers for Uncooled THz Sensing,” IEEE Trans. Terahertz Sci. Technol. 5(6), 902–913 (2015). [CrossRef]  

21. A. Syal, V. Lee, I. Andre, and J. Altet, “CMOS differential and absolute thermal sensors,” J. Electron. Test 18(3), 295–304 (2002). [CrossRef]  

22. I. Escorcia, J. Grant, J. Gough, and D. R. Cumming, “Uncooled CMOS terahertz imager using a metamaterial absorber and pn diode,” Opt. Lett. 41(14), 3261–3264 (2016). [CrossRef]  

Cited By

Optica participates in Crossref's Cited-By Linking service. Citing articles from Optica Publishing Group journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (9)

Fig. 1.
Fig. 1. Schematic diagram of the proposed triple-band THz thermal detector.
Fig. 2.
Fig. 2. (a) 3D model of the designed triple-band loop antenna in HFSS; (b) Simulated return loss of two kinds of antennas.
Fig. 3.
Fig. 3. Heat distributions of the triple-band THz thermal detector: (a) Operating at 0.91 THz; (b) Operating at 2.58 THz; (c) Operating at 4.3 THz.
Fig. 4.
Fig. 4. Schematic diagram of the optimized PTAT sensor.
Fig. 5.
Fig. 5. Post-simulated results of the two kinds of PTAT sensors: (a) Output voltage against ambient temperature; (b) Noise spectral density versus modulation frequency.
Fig. 6.
Fig. 6. (a) Layout of the triple-band THz thermal detector and a device with only a PTAT sensor; (b) Die micrograph of thermal detectors.
Fig. 7.
Fig. 7. (a) Block diagram of the output voltage measurement setup as the proposed detector operates at 0.91 THz. (b) Responsivities of the two kinds of detectors.
Fig. 8.
Fig. 8. (a) Block diagram of the output voltage measurement setup as the proposed detector operates at 2.58 THz; (b) Responsivities of the triple-band detector and the device with only a PTAT sensor.
Fig. 9.
Fig. 9. (a) NEP varies with operation frequency at a modulation frequency of 2 Hz. (b) NEP with respect to modulation frequency as the detector operates at 2.58 THz.

Tables (1)

Tables Icon

Table 1. Performance Summary and Comparison.

Equations (8)

Equations on this page are rendered with MathJax. Learn more.

R v = Δ V o u t P i n = P a b s o r b P i n Δ T P a b s o r b Δ V o u t Δ T
η E M = P a b s o r b P i n
β h e a t = Δ T P a b s o r b
θ e l e c = Δ V o u t Δ T
R v = η E M β h e a t θ e l e c
N E P = V r m s R v
R v = Δ V o u t P i n = Δ V o u t J i n A e f f
A e f f = G λ 2 4 π
Select as filters


Select Topics Cancel
© Copyright 2024 | Optica Publishing Group. All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies.