GIPO newsletter

Complementary Oxide-TFT-Based Circuits with N-Channel ZnO and P-Channel SnO Thin-Film Transistors

Professor I-Chun Cheng

Graduate Institute of Photonics and Optoelectronics, National Taiwan University

Complementary thin-film transistor (TFT) circuitry consisting of inverters is the building block for realizing integrated logic circuits on large-area and/or flexible substrates. Although numerous oxide-TFT circuits have been reported previously, most of them were based on n-channel MOSFET (NMOS) technology, implying that the development of p-channel oxide-TFTs with sufficient satisfactory performance via a low-temperature process is crucial. In this study, we report, for the first time, fully oxide-TFT-based complementary metal-oxide-semiconductor (CMOS) ring oscillators by using large–area-compatible sputtering processes. The p-channel tin monoxide (SnO) and n-channel zinc oxide (ZnO) TFTs used in the CMOS inverter have inverted-staggered bottom-gate structures. The SnO TFT exhibits a threshold voltage (V_th) of 3.5 V, field-effect mobility of 0.33 cm²/V-s, sub-threshold swing (S.S.) of 2.5 V/decade, and on/off current ratio of ~10³. The corresponding values for the ZnO TFT are 6.22 V, 3.5 cm²/V-s, 350 mV/decade, and >10⁶. The achieved voltage gain of the CMOS inverters is ~17 at a supplied voltage (V_DD) of 10 V when the geometric aspect ratio is 5. An oscillation frequency of 2 kHz is obtained from a five-stage oxide-based CMOS voltage control oscillator at V_DD of 14 V. The result shows the feasibility of realizing fully oxide-TFT-based CMOS circuits.



Fig. 1. (a)(b) Transfer and (c)(d) output characteristics of the p-channel SnO TFT (W/L = 60 μm/30 μm) and ZnO n-channel TFT (W/L = 30 μm/30 μm), respectively. The linear field-effect mobilities as a function of gate voltage for SnO and ZnO TFTs are plotted in (a) and (b), respectively.


Fig. 2 (a) The static voltage transfer characteristics and voltage gain curves at different supply voltages (V_DD) of the complementary oxide-TFT inverter with a geometric aspect ratio, (W/L)_p/(W/L)_n, of 5. (b) Output waveform of a five-stage ring oscillator based on complementary oxide-TFT inverters at V_DD = 14 V.

Development of numerical modeling program for organic/inorganic hybrid solar cells by including tail/Interfacial states models

Professor Yuh-Renn Wu

Graduate Institute of Photonics and Optoelectronics, National Taiwan University

A simulation software is developed for modeling the organic/inorganic hybrid solar cells. Firstly, the optical field in the solar cell device is calculated by using the two-dimensional finite-difference time-domain (2D-FDTD) method. By solving the Poisson’s and drift-diffusion equation which utilizes the finite element method (FEM), the electrical properties of the device can be obtained. A Gaussian distribution function of tail states is addressed to simulate the carrier transportation in the organic material. These tail states located around the lowest unoccupied molecular orbital (LUMO) and the highest occupied molecular orbital (HOMO) of the organic material provides a path way for carriers to hop in and transport through the organic material. In this research, a poly(3,4-ethylenedioxythiophene):poly(4-styrenesulfonate) (PEDOT:PSS)/silicon nanowire (SiNW) hybrid solar cell is set up as the simulation model. The validity of the software is verified by comparing the simulated J-V curve with experimental J-V curves. The electrical and optical properties are analyzed for further optimization.

Fig. 1. The simulated band diagrams of the PEDOT:PSS/SiNW hybrid solar cells.

Fig. 2. The generation rate of the PEDOT:PSS/SiNW hybrid solar cells calculated by 2D-FDTD.

Polymer-Stabilized Blue-Phase Liquid Crystal Display Using Distributed Floating electrode

Professor Wing-Kit Choi

Graduate Institute of Photonics and Optoelectronics, National Taiwan University

We have recently proposed a new electrode design for a Polymer Stabilized Blue Phase Liquid Crystal (PS-BPLC) display based on floating electrode [1]. We name it “distributed floating electrode design”. In this new design, we can use less area of floating electrode while still achieving the similar increase of the overall light transmission as in the previously proposed structure [2]. Our new design may have the potential to help reduce the image sticking problem that may be caused by the use of floating electrode and also further improve the light efficiency due to the use of less ITO area. Due to its attractive features such as i) fast response time, ii) no need of alignment layers and iii) wide viewing angle property, PS-BPLC has been proposed and actively researched as the next generation display technology in recent years.

Reference: [1] Y.C. Chang, S.H. Wei, C.L. Chang and W.K. Choi, “Polymer-Stabilized Blue-Phase Liquid Crystal Display Using Distributed Floating electrode”, Society for Information Display (SID), 2016, [2] Y. Liu, Y. Li, D. Lai, J.W. Shiu and S.T. Wu, “High Transmittance Blue-phase LCD with a Floating Electrode” Society for Information Display (SID), 2013

1.1 μm InAs/GaAs quantum-dot light-emitting transistors grown by molecular beam epitaxy

Professor Chao-Hsin Wu

Graduate Institute of Photonics and Optoelectronics, National Taiwan University

We report the enhanced radiative recombination output from an AlGaAs/GaAs heterojunction bipolar transistor with InAs quantum dots embedded in the base region to form a quantum-dot light-emitting transistor (QDLET) grown by molecular beam epitaxy systems. For the device with a 100×100 μm2 emitter area, we demonstrate the dual output characteristics with an electrical output and an optical output when the device is operating in the common-emitter configuration. The quantum-dot light-emitting transistor exhibits a base recombination radiation in the near-infrared spectral range with a dominant peak at λ of 1100 nm. This work has been published in Optics Letters, Vol. 40, No. 16, pp. 3747, (2015).

Fig. 1. The atomic force microscopic image (inset) and photoluminescence (PL) spectrum of the InAs/GaAs QDLET at 300 K with 4 mW/cm² excitation power. The PL peak at 1116 nm indicates the transitions of carriers on predominant InAs QDs.


Fig. 2. (a) Electrical I-V and (b) optical L-I-V of the InAs/GaAs QDLET with a 100×100 μm² emitter area. The blue circles indicate the V_BC = 0 V points.	Fig. 3. (a) Top view layout of the InAs/GaAs QDLET and (b) CCD image of the infrared light emission when transistor operates in the common-emitter configuration (I_B = 5 mA, V_CE = 4 V).

Fig. 4. Electroluminescence spectrum of the InAs/GaAs QDLET at different base currents. The wavelength peaks are at 1100 nm corresponding to the radiative recombination contributed from the InAs QDs.

Dual-functional Memory and Threshold Resistive Switching Based on the Push-Pull Mechanism of Oxygen Ions

Professor Si-Chen Lee’s laboratory

Graduate Institute of Photonics and Optoelectronics, National Taiwan University

The resistive random access memory (RRAM) becomes more and more important as the candidate to overcome the technological limitations for next-generation non-volatile memory. We have developed a special device structure (a-TiO_x)/(Ag NPs)/(poly-TiO_x) sandwiched between indium tin oxide (ITO) and fluorine-doped tin oxide (FTO) electrode as shown in Fig. 1 (a) which performs dual functions, i.e., memory switching (RRAM device) and threshold switching (Selector device). When used for memory switching, the device is forming-free with low operation voltage (< ±1 V) and self-compliance to current up to 50 μA as shown in Fig. 1(b). This reduced the damage of forming process on the device endurance. When the device is used for threshold switching as shown in Fig 1(c), the reduction of the threshold currents to -2 μA and 0.1 μA in negative and positive voltage regions, respectively, is beneficial for improving the device selectivity to solve the sneak current problem in the crossbar memory arrays. It is found that the push and pull actions of the oxygen ions in the top amorphous and bottom polycrystalline TiO_x films due to voltage sweeps can account for the dual-functional properties in a single device structure, as illustrated in Fig. 2.

Fig. 1. Device structure and electrical properties of the dual functional switching device.

Fig. 2. Push-pull mechanism of oxygen ions migration in the memory switching and threshold switching modes.

Y. J. Huang, S. C. Chao, D. H. Lien, C. Y. Wen, J. H. He and S. C. Lee, “Dual-functional Memory and Threshold Resistive Switching Based on the Push-Pull Mechanism of Oxygen Ions”, Nature Sci. Rep. 6, 23945, 2016.

Multi-Bent-Section Nano-Antennas

Professor Hung-chun Chang

Graduate Institute of Photonics and Optoelectronics, National Taiwan University

Nano-antennas are a class of nano-structures that can produce “hot spots” in a subwavelength gap of the structure due to plasmonic effects under incident electromagnetic fields and have been extensively investigated in the visible and infrared wavelength range both theoretically and experimentally. Bowtie nano-antennas or bowtie antennas (BAs), as depicted in Fig. 1, have been a popular structure. The resonant wavelength λ_res of the BA is proportional to the antenna length. For a given footprint size, achieving longer λ_res is useful in applications within the near- or mid-infrared range. We have recently proposed multi-bent-section nano-antennas (MBSNAs) which are dipole nano-antennas with multi-bent-sections and demonstrate that such simple structures can achieve long λ_res’s with high gap fields. The study was based on our in-house developed 3D finite-difference time-domain (FDTD) method [IEEE Photon. J. 7, 4501108 (2015)]. Referring to the BA of Fig. 1 as the basic structure, the bowtie-shaped gold film with thickness d is supported by a silica substrate with relative permittivity of 2.25, and it is air above the structure. The function of the gold is described by the Drude model [Phys. Rev. B. 71, 085416 (2005)]. The antenna footprint area is kept the same to be 450 nm × 450 nm, with the thickness d fixed to be 36 nm and the antenna gap area fixed to be 30 nm × 30 nm. Four MBSNAs with different shapes of the gold region as shown in the top of Fig. 2, each with the same d and the gap area as aforementioned, have been simulated. Every section of each antenna is with the same width of 30 nm. It can be seen the antenna structure is very simple even for the fourth one. The results of Fig. 2 is to be compared with the reference BA for which the resonance occurs at l = 1260 nm and the corresponding gap-field enhancement is 31. The proposed structures are also with wide bandwidth in the enhancement spectrum. By numerically solving the modal effective indices of the plasmonic waveguide having its cross-section as that of the gold line with the substrate of the MBSNA, the λ_res’s for the bent gold-line waveguide length were determined and found to be in good correlation with those of the peaks in Fig. 2. (Yu-Tsung Chen and Hung-chun Chang, in Proc. OPTIC 2015 .)

Figure 1. Schematic of the 3-D representation for a bowtie nano-antenna.

Figure 2. Upper part: four different shapes of the gold region of the MBSNAs. Lower part: Gap enhancement factor versus the wavelength for the four MBSNAs. Each MBSNA has the same footprint size of 450 nm × 450 nm.