N. I. M. Nor

High Power LED Thermal and Stress Simulation on Copper Slug

High power LED are captivating attention due to its cogent impacts on lighting industry in terms of efficacy, low power consumption, long lifetime and miniature physical size. Nonetheless, the efficiency and reliability of the LED is signified by the junction temperature. This work demonstrates the thermal and stress simulation of single chip LED package with 1mm x1mm x 1mm copper heat slug. The simulation was performed using Ansys version 11. The GaN LED chip was powered with input power of 0.1 W and 1 W. The simulation outcome exhibited that at the maximum junction temperature and stress of the LED chip were 115.81°C and 221.56MPa correspondingly for input power of 1W.

Remazol orange dye sensitized solar cell

Water based Remazol Orange was utilized as the dye sensitizer for dye sensitized solar cell. The annealing temperature of TiO2 working electrode was set at 450 °C. The performance of the device was investigated between dye concentrations of 0.25 mM and 2.5 mM at three different immersion times (3, 12 and 24 hours). The adsorption peak of the dye sensitizer was evaluated using UV-Vis-Nir and the device performance was tested using solar cell simulator. The results show that the performance was increased at higher dye concentration and longer immersion time. The best device performance was obtained at 0.2% for dye concentration of 2.5 mM immersed at 24 hours.

Few-layers graphene oxide for NO2 gas sensor on plastic

Vacuum filtration method was used in order to fabricate a homogeneous and uniform thin film of multi-layer graphene oxide on plastic substrate. This self-regulating technique allows the number of graphene oxide layer to be controlled thus controlling the film thickness by simply varying either the concentration of the graphene oxide in the suspension or the filtration volume. Measurement of the sheet resistance as a function of graphene oxide concentration in solution shows the percolation behavior of multi-layer films. The device was then exposed in nitrogen dioxide (NO2) environment at room temperature and 200 °C, under atmospheric pressure. Results demonstrate that the graphene oxide film shows good sensitivity and excellent recovery time using plastic substrate.

Fundamental study of reduction graphene oxide by sodium borohydride for gas sensor application

The efficient reduction of graphene oxide (rGO) was performed using Sodium Borohydride (NaBH4). These reduction approaches remove the majority of the oxygen-containing functional groups at the basal plane and surface of graphene oxide sheets. Structural and physiochemical properties of the GO were investigated with help of Atomic Force Microscopy (AFM), Transmission Electron Microscopy (TEM), Fourier Transform Infrared Spectroscopy (FTIR), and Ultraviolet-Visible-Near infrared (UV-Vis-NIR). The effects of the chemical reduction on a GO surface were analyzed using a Semiconductor Parameter Analyzer (SPA) in order to obtain the electrical resistance measurement. It was found that the resistance of reduced graphene oxide was greatly reduced when compared to the condition of before reduction process. Then, the formation of uniform thin film of rGO sheets was produced using vacuum filtration method in order to fabricate a gas sensor. In this project, plastic was used as a substrate. The sensor was then being e...

Multi-walled carbon nanotubes plastic NH3 gas sensor

Multi-walled Carbon Nanotubes (MWCNTs) functionalized -COOH was used as the sensing material. The MWCNTs suspension was then deposited on the plastic substrate using vacuum filtration method, hence created uniform thin film carbon nanotubes network. Plastic membrane was chosen as the substrate in order to produce flexible, lightweight, wearable and low cost sensor. This device was exposed to ammonia gas (NH3) at two different concentrations; 19.2 and 231.4 ppm. The device shows high sensitivity at 23.4 % when exposed to 231.4 ppm NH3 and less sensitivity at 4.39 % for 19.2 ppm NH3 exposure.

The enhancement of high speed pHEMT device for IoT applications

We report the enhancements on pHEMT epilayers to suite the high speed applications for IoT. Presented here are the DC and RF comparisons between XMBE #109 as the baseline of high speed pHEMT and VMBE #2100 as the enhanced epitaxial layer. The Hall data from Van der Pauw measurement shows higher sheet carrier concentration is observed on the improved epitaxial layer devices. The DC comparisons between the devices from both epilayers verified that the enhanced epitaxial layer demonstrates low leakage-high breakdown characteristics. Improvement of transconductance and output current density also been observed from the VMBE #2100 epitaxial layer. The RF characteristics, however, show comparable fT of 25 GHz. In short, this will facilitates the implementation of high speed circuits, including high speed broadband systems for IoT applications.

Comprehensive study on gate recess step for the fabrication of high-speed InGaAs/InAlAs/InP pHEMT

We report a comprehensive etching study on the gate recess step for the fabrication of the novel high speed pHEMT devices. The experiments focused on the elimination of “hump” structure as a result of an incomplete etching process at the InGaAs cap layer. In this work, two types of test samples were used, namely bulk InGaAs and epitaxial structure together with an etch stop layer. The result showed that the etch rate of bulk InGaAs is about 360 A/min and the percentage of dome height is consistent at approximately 25%. Meanwhile, the study on pHEMT epitaxial layer showed that the etching time of 3 minutes is sufficient in order to completely remove the cap layer. Gate leakage current of magnitude more than 10 times lower is observed on the devices that engaging Succinic Acid as the gate recess etching agent. The optimized processing steps will tailor for highly reproducible pHEMT fabrication process for high speed applications.

The design and optimisation of the square and circular inductor using MEMS technology

This paper presents the design and optimization of high quality (Q) factor inductors using Micro Electro-Mechanical Systems (mEMs) technology for 10GHz to 20GHz frequency band. Two inductors have been designed with square and circular topologies. Inductors are designed on Silicon-on-Sapphire (SOS) because of its advantages including high resistivity and low parasitic capacitance. The effects of various parameters such as the width of metal traces (W), the thickness of the metal (T) and the air gap (AG) on the Q-factor are thoroughly investigated. Results indicate that the optimization of the geometrical parameters has minimized the resistive, capacitive and inductive losses of the inductor.

Linear modelling of novel InGaAs/InAlAs/InP pHEMT for low noise applications

Linear modelling of novel InGaAs/InAlAs/InP pHEMT for low noise applications is substantial to the future transistors that will operate in high speed and low noise conditions. The novel pHEMT is constructed by sandwiching two different materials together with different lattice constants, for instance InGaAs and InAlAs in order to form a heterojunction in between. However, InP is only utilised to be the substrate base of pHEMT. In the modelling process, extrinsic and intrinsic parameters need to be extracted. Briefly, a high accuracy transistor modelling enables designers to predict the real output of a circuit before it can be fabricated onto an actual chip.

Small signal modelling of novel InGaAs/InAlAs/InP pHEMT for high frequency applications

HEMT is a GaAs based field effect transistor that retains higher cutoff frequency compared to silicon based transistors. Alternatively, pHEMT enhance the performance of the HEMT in term of leakage, current conduction and the cutoff frequency of the device. The heterostructure of pHEMT improve the performance two-dimension electron gas (2DEG) in the channel layer. With these, pHEMT is believed could be perfectly used in the most of the high frequency application. In this project, small signal models of InGaAs/InAlAs/InP pHEMT with 7 extrinsic parameters and 8 intrinsic parameters are modelled.