Project Title Performance Analysis of Full Adders on CMOS Decimation Filter for Delta-Sigma ADC
Student Name Yeoh Poay Zheng
Supervisor Dr. Lee Lini
Year of Completion 2018
Description This project aims to develop, analyze and implement a low power 3rd order cascaded integrator-comb (CIC) filter for delta-sigma ADC. CIC decimation filter is widely used due to its simplicity, low power and area efficiency if compared to others. Nonetheless, delay element and full adder are the two main components to construct a CIC filter. In this project, four different types of full adder architecture were studied for its performance when integrated into CIC filter. There are 28T conventional CMOS full adder, pseudo-NMOS full adder, 16T hybrid full adder and 14T modified hybrid full adder as shown in figures below. This project was simulated by using Silterra 130nm CMOS technology in Mentor Graphic.

Figure 1: Block diagram of CIC decimation filter

(a) 28T conventional

(b) pseudo-NMOS

(c) 16T hybrid

(d) 14T modified hybrid

Figure 2: Transistor level diagrams

Results / Conclusion In pre-layout simulation, 16T hybrid static full adder have the best performance output in terms of power consumption, delay propagation and power-delay product. In overall, the performance of 16T hybrid full adder had improved 38% if compared to 28T static full adder in 1-bit CIC filter configuration. Hence, a complete single bit 3rd order of 16T hybrid full adder CIC filter is implemented with size area of 118.23μm × 22.38μm.

Figure 3: Comparison of power consumption

Figure 4: Comparison of propagation delay

Figure 5: Comparison of PDP

Figure 6: Layout design of 1-bit 3rd order CIC filter


Project Title Synthesis of Titanium Oxide For The Application of Water Treatment
Student Name Joshua Soo ZheYan
Supervisor Dr. Sin Yew Keong
Year of Completion 2016
Description Titanium dioxide (TiO2) as a photocatalyst produces oxidizing radicals for water treatment. Treatment performance of TiO2 in natural light will be improved when doped with nitrogen. This is because the bandgap energy of TiO2 will be lowered while its visible light absorbance will be increased. TiO2 and nitrogen-doped TiO2 (N-TiO2) powders were synthesized via sol-gel method using titanium isopropoxide (TTIP) and urea as titanium and nitrogen precursors respectively. Calcination at 773 K followed by ball milling was done to obtain the TiO2 powders. The effect of nitrogen concentration on the photocatalytic performance of TiO2 was studied by varying the ratio of TTIP to urea. A filtration system was set up by incorporating TiO2 powders with filter paper to obtain the optimum parameters for degrading methylene blue and treating bacteria-contaminated water. Characterizations of the synthesized powders include ultraviolet-visible light spectroscopy, dynamic light scattering measurement and scanning electron microscopy.

Illustration of synthesis steps to obtain TiO2 and N-TiO2 gel

Setup of water treatment

Results / Conclusion Nitrogen doping in TiO2 changes the appearance of powders from white to pale yellow. N-TiO2 with TITP:urea of 1:2 has the lowest bandgap energy of 2.740 eV. However, TTIP:urea of 1:3 yields the highest visible light absorbance and the smallest average particle size at 425 nm with the least polydispersity. In this study, the optimum setup for removing methylene blue is by using N- TiO2 with TTIP:urea ratio of 1:2 with 16 layers of filter paper which gives 98.42% methylene blue reduction. However, N-TiO2 with TTIP:urea of 1:3 gives the optimum antibacterial performance with 65.26% less bacterial coverage.

SEM images of (a) TiO2 (b) 2N-TiO2 and (c) 3N-TiO2 thin films at magnification of 10000x

Obtained water sample of
(a) Pre-treatment, (b) Control, (c) TiO2 (Dark), (d) TiO2 (UV)
(e) TiO2 (f) 2N- TiO2 (g) 3N- TiO2


Project Title Fabrication of Gas Sensor Using Aluminium Doped Zinc Oxide
Student Name Low Soon Wei
Supervisor Dr. Sin Yew Keong
Year of Completion 2016
Description The rapid industrialization has caused air pollution that greatly affects the environment. This project studies the thin film based gas sensors fabricated using Zinc Oxide (ZnO) and Aluminium-doped ZnO (AZO) colloids. The effect of aluminium (Al) doping concentrations and the device’s operating temperatures on the sensitivity of the gas sensors are also researched. ZnO and AZO colloids are synthesized through hydrolysis method and coated on glass substrates using spin coating. Silver (Ag) paste is used to form the metal contacts on the gas sensor. The surface morphology of the samples are characterized using scanning electron microscope (SEM). The electrical characterizations are carried out using four point probe and Hall Effect measurement.

Setup of resistance measurement for fabricated ZnO and AZO gas sensor

Gas chamber set-up

Results / Conclusion Particle sizes of 395 nm is recorded for ZnO colloids while smaller colloidal sizes of 147 nm and 142 nm are recorded for AZO 6% and AZO 3% colloids, respectively. Four point probe tests show the resistance of ZnO colloids decreases after doping with Al. Hall effect measurements show AZO 6% colloids have highest conductivity of 6.0898 x 10-4 S/cm. AZO 6% gas sensor recorded the highest sensing response with a value of 77.08 % at 250 °C. The response of AZO 3% and AZO 6% gas sensors are optimum at the operating temperature of 250 °C. Among ZnO, AZO 3% and AZO 6% colloids, AZO 6% is concluded to be the best sample for gas sensing purposes due to good uniformity, high conductivity and high gas sensing response.

SEM images of a) ZnO b) AZO3 c) AZO6 at magnification of 10000x

Sensor response for ZnO, AZO3 and AZO6 gas sensors


Project Title Simulation of Silicon Nanowire Transistor Using Sentaurus
Student Name Kong Kok Hong
Supervisor Dr. Sin Yew Keong
Year of Completion 2015
Description As Moore’s Law is approaching its limit, strategy such new device architectures are needed. In this project, a junctionless n-type silicon nanowire transistor has been designed and simulated using Sentaurus. The effects of changing the nanowire parameters are investigated and the final optimized structure is determined. The process flow is relatively simple and epitaxy-free. Four masking layout and ten steps of fabrication steps are involved. The parameters variation includes the nanowire height and width variation from 100nm to 1nm, nanowire length variation from 400nm to 100nm as well as gate oxide thickness from 10nm to 1nm.

Silicon nanowire transistor process and device simulation flow chart

Results / Conclusion The simulated device shows excellent properties such as low threshold voltage and subthreshold swing. Changing the nanowire parameters has shown a great impact on the threshold voltage and other electrical parameters. At nanowire height and width lower than 20nm, the threshold voltage increases exponentially. Nanowire lengths shorter than 200nm cause a sharp increase in subthreshold slope because of the short channel effect. Gate leakage occurs at gate oxide thicknesses smaller than 4nm and a sharp increase in drain current is observed. From this project, the optimized parameters for the device are nanowire height and width at= 20nm, nanowire length at= 200nm and gate oxide thickness at= 4nm.

Simulated structure of silicon nanowire transistor from
(a) horizontal and (b) vertical cross section


Project Title Synthesis and Optical Characterisation of Aluminium-doped Zinc Oxide Colloids
Student Name Saw Min Jia
Supervisor Dr. Sin Yew Keong
Year of Completion 2015
Description In this project, aluminium doped zinc oxide nanoparticles in colloidal form were synthesized using hydrolysis method. The hydrolysis method was chosen due to its low cost, low temperature and fast process, where the samples could be mass produced. Zinc precursor was added to solvent and the mixture solution was then heated under reflux. The reaction solution was allowed to heat up to 10 °C higher than temperature when it turned milky. Then, it was aged and centrifuged. The sample was then undergone washing and storing processes.

Schematic diagram of synthesis

Results / Conclusion The effect of aluminium as dopants on the optical properties of zinc oxide colloids was studied. The average transmittances of aluminium doped zinc oxide colloids was reported to be over 80 % for solution-based samples and over 65 % for thin film-based samples that had 2.4 % aluminium doping concentration onwards. The absorbance peak edge of zinc oxide colloids is blue shifted when doped with aluminium.

The effect of varying aluminium doping concentration on the structural and optical properties of zinc oxide colloids was also investigated. Structural properties were characterized using scanning electron microscopy and energy-dispersive X-ray spectroscopy. Optical properties were characterized using ultraviolet-visible spectroscopy. Aluminium doped zinc oxide colloids with aluminium concentrating concentration of 1.2%, 2.4%, 3.6%, 4.8% and 6% have diameter in the range of 656 nm, 383 nm, 185 nm, 116 nm, 119 nm and 128 nm, respectively. Smaller particle size was obtained with increasing aluminium doping concentration. Energy-dispersive X-ray results confirmed the presence of aluminium doped zinc oxide. Higher aluminium doping concentration also resulted in the absorption peak edge being more blue shifted, thus indicating the broadening of optical band gap. The solubility limit of aluminium in zinc oxide obtained was 2.4 %.

SEM images of (a) zinc oxide and aluminium-doped zinc oxide colloids with aluminium concentrating concentration of
(b) 1.2%, (c) 2.4%, (d) 3.6%, (e) 4.8% and (f) 6%.

Transmittance spectrums (intensity vs wavelength) of zinc oxide and aluminium-doped zinc oxide colloids with aluminium concentrating concentration of
1.2% (AZO1), 2.4% (AZO2), 3.6% (AZO3), 4.8% (AZO4) and 6% (AZO5).


Project Title Synthesis and Characterisation of Co-precipitated Magnetite Nanoparticles for Biomedical Applications
Student Name Ong Chin Kuan
Supervisor Dr. Ong Boon Hoong
Year of Completion 2009
Description In this project, the chemical synthesis of the magnetite, Fe3O4, magnetic nanoparticles for biomedical application was investigated. Parameters of the synthesis method were manipulated to produce different outcomes of magnetic nanoparticles. Various chemical synthesis methods were studied and the simplest, most promising and yet energy efficient method chosen was the co-precipitation method. Besides that, properties of magnetite, magnetisms, surface engineering and surfactants were studied to show the applicability for different biomedical applications. One of the most frequently used precursor combination, ferric chloride and ferrous chloride, were used with the alkaline sodium hydroxide solution in order to obtain magnetite nanoparticles with particle sizes below 15nm and exhibit superparamagnetic magnetic effect.
Results / Conclusion Two different particle sizes were obtained namely, 9.51±0.59nm and 14.25±0.9nm. Different sizes of magnetite nanoparticles were synthesized by varying the reaction time and stirring speed. It was found that higher stirring speed and shorter reaction time would result in smaller particle size and conversely, larger particle size was obtained with lower stirring speed and longer reaction time. These results were verified by using Transmission Electron Microscopy (TEM) characterization which gave particle sizes in agreement with the calculated particle size from the X-Ray Diffraction (XRD) result using Scherer equation. By controlling the heating temperature, magnetite phase obtained was verified by using XRD characterization without transforming into other iron oxide phase like maghemite or hematite. The superparamagnetic response of the synthesized nanoparticles was characterized using Vibrating Sample Magnetometer (VSM) and showed saturation magnetization from the range of 2.3emu/g till 10.1emu/g.


Project Title Process and Device Simulation for 70nm NMOS
Student Name Siew Hon Yin
Supervisor Dr. Ong Boon Hoong
Year of Completion 2009
Description The transistors count per chip has increase from times to times in every latest product manufactured by the chip makers. The increment in the transistor density for the chip has improved in the performance of the chip. In other words, in order to improve the performance in their products, larger amount of transistors is needed to be built into the chip. However, the size of the chip will become very large if the transistor is large in scaled. As a result, the down-scaling technology plays an important role to solve this problem, where the size of the transistor has scaled down to smaller size, thus producing a higher transistor density of chip. Care must be taken in scaling down the devices since it might involve the effect of quantum properties when it became too small. This project focuses mainly in the design of a nano-scaled NMOS device. A 70nm NMOS design has been developed according to the recipe from other researchers by using the TCAD Silvaco Simulation Tool. There exist differences between the results for simulated design and the experimental design has as shown in this project. Further analyze need to be performed to study the effects of the variation in the fabrication parameters on the performance of the fabricated device.
Results / Conclusion In this project, a 70 nm NMOS device has been successfully fabricated using the Silvaco TCAD simulation tools. The simulated device structure is quite similar to the experimented device structure. For example, both device structures consist of the polysilicon layer, TEOS layer, a 70 nm gate channel, arsenic doped source and drain regions, and boron doped silicon substrates. Besides this, the output I-V curves for both devices are showing the same trend or shape of the curve, where both consist of the linear region, triode region, and the saturation region.