The working principle of quantum well infrared detectors is based on subband transitions. After absorbing infrared radiation, the detector excites electrons in the quantum well, causing them to transition from the ground state to the continuum state, thereby achieving infrared detection.
According to a report by MMS Consultancy, a recent research team from Shanghai University of Technology and the Shanghai Institute of Technical Physics, Chinese Academy of Sciences, published an article in the journal "Infrared and Millimeter Waves" with the theme "Material Characterization and Device Performance Study of Non-Uniform GaAs/AlGaAs Quantum Well Infrared Detectors." The first author of the article is Su Jiaping, and the corresponding authors are Chen Pingping and Chen Zezhong.
This work focuses on non-uniform GaAs/AlGaAs quantum well infrared detectors for the focal plane array (FPA), providing a foundation for related long-wave focal plane quantum well infrared detectors in the range of 10 μm - 11 μm. The main characteristic of non-uniform quantum wells is the introduction of non-uniform barrier widths and doping concentrations in the quantum well, which alters the band structure and internal electric field distribution. This also provides new insights for the design of novel optoelectronic and semiconductor devices.
Experimental Process: The samples in this paper were grown using the Compact-21 molecular beam epitaxy (MBE) system from Riber, France, on a 3-inch (1,0,0) semi-insulating GaAs substrate. The MBE system is equipped with a single-zone Al source furnace, a dual-zone Ga source furnace, and an As cracking furnace as the As source furnace, all of which use solid-state sources.
As the growth progressed, the barrier width linearly decreased from 75 nm to 15 nm, while the doping concentration in the well increased from 1.0×10¹⁷ cm⁻³ to 1.0×10¹⁸ cm⁻³. The distribution of doping concentration and barrier width in the well is shown in Figure 1. Most of the doping concentration changes in the quantum well occurred in the last five wells, increasing from 3.0×10¹⁷ cm⁻³ to 1.0×10¹⁸ cm⁻³.
Figure 1: Distribution of Doping Concentration and Barrier Width in the
Non-Uniform Quantum Well
After the sample growth, the growth of the non-uniform quantum well period was characterized using a Talos F200X high-resolution transmission electron microscope (HRTEM) to ensure that the sample growth matched the design parameters. The doping concentration of the non-uniform quantum well period potential well was tested using a Cameca 7f secondary ion mass spectrometer (SIMS).
In this study, standard procedures were followed to prepare the test devices, as outlined below: First, a 200×200 μm² mesa was obtained through photolithography and wet etching. Then, top and bottom electrode layers were grown using electron beam evaporation with AuGe/Ni/Au metal layers of 100/20/400 nm, and Ohmic contacts were formed under appropriate annealing conditions. Finally, the sample was ground into a 45° inclined facet for coupling the incident light and secured onto an oxygen-free copper heat sink with low-temperature adhesive. To evaluate the photoelectric performance of the devices, they were mounted in a low-temperature Dewar cooling device, and parameters such as blackbody response, dark current, and photocurrent spectra were measured.
Results and Discussion Microstructural Characterization of Non-Uniform Quantum Well Epitaxial Material In this study, the non-uniform quantum well epitaxial material was characterized using high-resolution transmission electron microscopy (HRTEM) to investigate its epitaxial layer and interface properties. The crystal quality of non-uniform quantum well epitaxial material and deviations in material uniformity from the growth parameters are important factors influencing detector performance. HRTEM, as an essential characterization tool, provides detailed information on nano-scale structures in molecular beam epitaxy materials, such as interface morphology, crystal defects, and structural defects. Observation of the material's interface morphology and lattice defects can further guide the optimization of quantum well infrared detector materials and improve their performance. Therefore, sample A was selected for HRTEM characterization, and Figure 2 shows the HRTEM images of sample A.
Figure 2: High-Resolution Transmission Electron Microscopy Image of Sample A
Furthermore, in the characterization of microstructures using HRTEM, element composition and content analysis of the central region of the non-uniform quantum well structure were performed using an energy dispersive spectrometer (EDS). As shown in Figure 3, the leftmost image is a high-angle annular dark-field (HAADF) image. From the image, it can be observed that both sample A and sample B have steep interfaces, and the quantum well's GaAs barrier layers (white layers) and AlGaAs well layers (gray layers) are clearly defined, with no segregation observed.
The center image in Figure 3 shows the distribution of Al elements, and it can be observed that they are evenly distributed in the wider barrier layers while absent in the narrower well layers (appearing black). The interface between the well layers and barrier layers is very clear, with no signs of segregation. This, from another perspective, indicates the excellent interface quality of GaAs/AlGaAs. In summary, high-resolution transmission electron microscopy and energy dispersive spectrometer analysis both demonstrate the excellent epitaxial quality and interface quality of the microstructures in the non-uniform quantum well epitaxy.
Figure 3: EDS Image of Sample A
In order to obtain further information about the composition of each layer in the epitaxial material and to investigate another characteristic of the non-uniform quantum well, which is non-uniform doping, secondary ion mass spectrometry (SIMS) was used for an in-depth analysis of the Si doping process in the GaAs well layer of sample A. This analysis aimed to compare the actual doping concentration in the well with the designed values. Figure 4 presents the SIMS test data for sample A, providing a visual representation of the structural characteristics of the non-uniform quantum well material. Along the epitaxial growth direction, it is evident that the barrier width of the quantum well gradually narrows, while the doping concentration in the well increases sequentially.
Figure 4: SIMS Test Results for Sample A
Photovoltaic Performance Study
NUQWIP vs. Conventional QWIP Photovoltaic Performance
A non-uniform quantum well (NUQW) is a quantum well structure with continuously changing barrier width and well doping concentration along the epitaxial growth direction. This alteration in the electric field distribution of the quantum well affects the performance of quantum well detectors. Dark current refers to the current generated due to thermal excitation or tunneling effects in the absence of illumination. It is a critical factor influencing parameters such as noise and detection rate in photodetectors. Therefore, analyzing the dark current characteristics of non-uniform quantum wells is of significant importance for optimizing photodetector designs and enhancing their performance. Figure 5 compares the dark current characteristics as a function of bias voltage for non-uniform structure (Sample C) and conventional structure (Sample D) quantum wells in the temperature range of 50 K to 70 K.
Figure 5: Relationship between Dark Current and Bias Voltage at Different Temperatures, Solid Line represents Non-Uniform Quantum Well, Dashed Line represents Conventional Quantum Well
Figure 6 shows the photoconductive (PC) spectrum of the quantum well detector. It can be observed that the full width at half maximum (FWHM) of the PC spectrum for the non-uniform QWIP is significantly reduced compared to that of the conventional QWIP, with Δλ/λ decreasing from 16% to 8%. This is attributed to the gradual inward shift of the first excited state from outside the well to inside, providing additional evidence of a change in the transition mode. Furthermore, Figure 7 compares the curves of blackbody response rate as a function of bias voltage for non-uniform structure (Sample C) and conventional structure (Sample D) QWIPs at 50 K and 60 K temperatures.
Figure 6: Photocurrent Response Spectrum of Non-Uniform Quantum Well and Conventional Quantum Well at 50 K Temperature
Figure 7: Variation of Blackbody Response Rate with Bias Voltage for Non-Uniform Quantum Well and Conventional Quantum Well at Different Temperatures
Different Well Widths of NUQWIP Photovoltaic Performance
To investigate the influence of well width variation on the electrical performance of non-uniform quantum wells, samples A, B, and C were grown with the only difference being the well widths (6.1 nm, 6.3 nm, and 6.5 nm for samples A, B, and C, respectively), while keeping all other parameters constant. Figure 8 displays the photocurrent spectra of samples A, B, and C at 50 K.
Figure 8: Photocurrent Response Spectra of Samples A, B, and C at 50 K
Figure 9 illustrates the relationship between the dark current and bias voltage of non-uniform quantum wells with different well widths. From the graph, it can be clearly observed that the dark current increases rapidly with increasing device bias voltage. This is due to the increasing bias voltage causing band tilting in the quantum well, which in turn increases the collision ionization energy for electrons, leading to an increase in the dark current. As shown in Figure 10, it displays the curves of blackbody response rate as a function of bias voltage for non-uniform quantum well infrared detectors with different well widths at different temperatures.
Figure 9: Dependency of Dark Current on Bias Voltage for Samples A, B, and C at Different Temperatures
Figure 10: Variation in Blackbody Response Rate with Bias Voltage for Samples A, B, and C at Different Temperatures
Furthermore, there are some differences in the temperature dependency of the blackbody response rate between the non-uniform quantum well structure and the conventional quantum well structure. For the traditional quantum well structure, the response rate remains relatively constant at different temperatures. The temperature dependence of the response rate, as observed, was previously only found in QWIPs with a single quantum well period. As shown in Figure 10, for the non-uniform quantum well structure, under negative bias, the response rate exhibits a weak temperature dependence. However, under positive bias (above 1.5 V), the response rate increases with increasing temperature. This is because the injection current into each quantum well in the non-uniform quantum well structure is affected by temperature. To explain this based on the internal electric field distribution of the quantum well, at lower temperatures, the injection current into the low-doped wells is lower, and the electric field distribution is divided into two parts: a uniform high-field region and a longer low-field region, nearly zero. Due to the low quantum efficiency and high capture probability in this region, the response rate is lower. As the temperature increases, this low-field region decreases, gradually increasing the response rate.
Conclusion In this study, non-uniform GaAs/AlGaAs quantum well infrared detector materials were successfully grown using molecular beam epitaxy (MBE) technology. Detailed characterization and analysis of the non-uniform epitaxial microstructure were conducted using high-resolution transmission electron microscopy (HRTEM) combined with energy dispersive spectrometer (EDS), and the doping in the non-uniform well was characterized using secondary ion mass spectrometry (SIMS). A comparative study of the performance of non-uniform GaAs/AlGaAs quantum well infrared detectors and conventional quantum well infrared detectors was conducted, and the influence of different well widths on the performance of the non-uniform quantum well infrared detector was analyzed. The results indicate that high-quality non-uniform quantum well epitaxial materials were successfully grown using the MBE system. By altering the doping concentration and barrier width of each well, the electric field distribution can be changed, resulting in a decrease in dark current by an order of magnitude. Different well widths can alter the transition mode of the non-uniform quantum well, with the bound-to-quasi-bound transition mode (B-QB) devices exhibiting higher blackbody response rates and lower dark currents. This work contributes to the improvement of focal plane quantum well infrared detector performance and lays the foundation for the development of non-uniform quantum well infrared detectors for long-wave infrared imaging applications.
Comentarios