• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2009.11a
• /
• pp.55-55
• /
• 2009

Silicon Carbide (SiC) is a material with a wide bandgap (3.26eV), a high critical electric field (~2.3MV/cm), a and a high bulk electron mobility (${\sim}900cm^2/Vs$). These electronic properties allow high breakdown voltage, high frequency, and high temperature operation compared to Silicon devices. Although various SiC DMOSFET structures have been reported so far for optimizing performances. the effect of channel dimension on the switching performance of SiC DMOSFETs has not been extensively examined. In this paper, we report the effect of the interface states ($Q_s$) on the transient characteristics of SiC DMOSFETs. The key design parameters for SiC DMOSFETs have been optimized and a physics-based two-dimensional (2-D) mixed device and circuit simulator by Silvaco Inc. has been used to understand the relationship with the switching characteristics. To investigate transient characteristic of the device, mixed-mode simulation has been performed, where the solution of the basic transport equations for the 2-D device structures is directly embedded into the solution procedure for the circuit equations. The result is a low-loss transient characteristic at low $Q_s$. Based on the simulation results, the DMOSFETs exhibit the turn-on time of 10ns at short channel and 9ns at without the interface charges. By reducing $SiO_2/SiC$ interface charge, power losses and switching time also decreases, primarily due to the lowered channel mobilities. As high density interface states can result in increased carrier trapping, or recombination centers or scattering sites. Therefore, the quality of $SiO_2/SiC$ interfaces is important for both static and transient properties of SiC MOSFET devices.

• Journal of the Korean Institute of Telematics and Electronics D
• /
• v.36D no.10
• /
• pp.9-16
• /
• 1999

When the conventional IPD (inter-poly-dielctrics) layer with ONO(oxide-nitride-oxide) structure was used in the Flash EEPROM cell, its data retention characteristics were significanfly degraded because the top oxide of the ONO layer was etched off due to the cleaning process used in the gate oxidation process for the peripheral MOSFETs. When the IPD layer with the ONON(oxide-nitride-oxide-nitride) was used there, however, its data retention characteristics were much improved because the top nitride of the ONON layer protected the top oxide from being etched in the cleaning process. For the modelling of the data retention characteristics of the Flash EEPROM cell with the ONON IPD layer, the decrease of the threshold voltage cue to the charge loss during the bake was here given by the empirical relation ${\Delta}V_t\; = \;{\beta}t^me^{-ea/kT}$ and the values of the ${\beta}$=184.7, m=0.224, Ea=0.31 eV were obtained with the experimental measurements. The activation energy of 0.31eV implies that the decrease of the threshold voltage by the back was dur to the movement of the trapped electrons inside the inter-oxide nitride layer. On the other hand, the results of the computer simulation using the model were found to be well consistent with the results of the electrical measurements when the thermal budget of the bake was not high. However, the latter was larger then the former in the case of the high thermal budger, This seems to be due to the leakage current generated by the extraction of the electrons with the bake which were injected into the inter-oxide niride later and were trapped there during the programming, and played the role to prevent the leakage current. To prevent the generation of the leakage current, it is required that the inter-oxide nitride layer and the top oxide layer be made as thin and as thick as possible, respectively.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.20 no.2
• /
• pp.16-23
• /
• 2006

This paper presents a novel single-stage high frequency resonant inverter link type DC-DC converter using zero voltage switching with high power-factor. The proposed topology is integrated half-bridge boost rectifier as power factor corrector(PFC) and half-bridge high frequency resonant converter into a single-stage. The input stage of the half-bridge boost rectifier works in discontinuous conduction mode(DCM) with constant duty cycle and variable switching frequency. So that a boost converter makes the line current follow naturally the sinusoidal line voltage waveform. Simulation results have demonstrated the feasibility of the proposed high frequency resonant converter. Characteristics values based on characteristics analysis through circuit analysis is given as basis data in design procedure. Also, experimental results are presented to verify theoretical discussion. This proposed inverter will be able to be practically used as a power supply in various fields as induction heating applications, fluorescent lamp and DC-DC converter etc.

• Journal of the Korean Institute of Telematics and Electronics D
• /
• v.36D no.11
• /
• pp.56-62
• /
• 1999

All the cells on the whole memory array or a block of the memory array in the Flash EEPROM's are erased at the same time using Fowler-Nordheim (FN) tunneling. some of the cels are often overerased since the tunneling is not a self-limited process. In this paper, the optimum doping concentration of the floating gate solve the overerase problem has been studied. For these studies, N-type MOSFETs and MOS capacitors with various doping concentrations of the gate polysilicon have been fabricated and their electrical characteristics have been measured and analyzed. As the results of the experiment, it has been found that the overerase problem can be prevented if the doping concentration of the floating gate is low enough (i.e. below $1.3{\times}10^{18}/cm^3$). It is because the potential difference between the floating gate and the source is lowered due to the formation of the depletion layer in the floating gate and thus the erasing operation stops by itself after most of the electrons stored in the floating gate are extracted. On the other hand, the uniformity of the Vt and the gm has been significantly poor if the coping concentration of the floating, gate is too much lowered (i.e. below $1.3{\times}10^{17}/cm^3$), which is believed to be due to nonuniform loss of the dopants from the nonuniform segregation in the floating gate. Consequently, the optimum doping concentration of the floating gate to suppress the overerase problem and get the uniform Vt and has been found to range from $1.3{\times}10^{17}/cm^3$ to $1.3{\times}10^{18}/cm^3$ in the Flash EEPROM.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.45 no.10
• /
• pp.7-14
• /
• 2008

In this paper, we designed the HVIC(High Voltage Gate Driver IC) which has improved noise immunity characteristics and high driving capability. Operating frequency and input voltage range of the designed HVIC is up to 500kHz and 650V, respectively. Noise protection and schmitt trigger circuit is included in the high-side level shifter of designed IC which has very high dv/dt noise immunity characteristic(up to 50V/ns). And also, rower dissipation of high-side level shifter with designed short-pulse generation circuit decreased more that 40% compare with conventional circuit. In addition, designed HVIC includes protection and UVLO circuit to prevent cross-conduction of power switch and sense power supply voltage of driving section, respectively. Protection and UVLO circuit can improve the stability of the designed HVIC. Spectre and Pspice circuit simulator were used to verify the operating characteristics of the designed HVIC.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.45 no.4
• /
• pp.57-63
• /
• 2008

In this paper, the dependence of MOSFET performance on the channel stress is characterized in depth. The tensile and compressive stresses are applied to CMOSFET using a nitride film which is used for the contact etch stop layer (CESL). Drain current of NMOS and PMOS is increased by inducing tensile and compressive stress, respectively, due to the increased mobility as well known. In case of NMOS with tensile stress, both decrease of the back scattering ratio ($\tau_{sat}$) and increase of the thermal injection velocity ($V_{inj}$) contribute the increase of mobility. It is also shown that the decrease of the $\tau_{sat}$ is due to the decrease of the mean free path ($\lambda_O$). On the other hand, the mobility improvement of PMOS with compressive stress is analyzed to be only due to the so increased $V_{inj}$ because the back scattering ratio is increased by the compressive stress. Therefore it was confirmed that the device performance has a strong dependency on the channel back scattering of the inversion layer and thermal injection velocity at the source side and NMOS and PMOS have different dependency on them.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2014.02a
• /
• pp.363-363
• /
• 2014

Ruthenium (Ru) has attractive material properties due to its promising characteristics such as a low resistivity ($7.1{\mu}{\Omega}{\cdot}cm$ in the bulk), a high work function of 4.7 eV, and feasibility for the dry etch process. These properties make Ru films appropriate for various applications in the state-of-art semiconductor device technologies. Thus, it has been widely investigated as an electrode for capacitor in the dynamic random access memory (DRAM), a metal gate for metal-oxide semiconductor field effect transistor (MOSFET), and a seed layer for Cu metallization. Due to the continuous shrinkage of microelectronic devices, better deposition processes for Ru thin films are critically required with excellent step coverages in high aspect ratio (AR) structures. In these respects, atomic layer deposition (ALD) is a viable solution for preparing Ru thin films because it enables atomic-scale control of the film thickness with excellent conformality. A recent investigation reported that the nucleation of ALD-Ru film was enhanced considerably by using a zero-valent metallorganic precursor, compared to the utilization of precursors with higher metal valences. In this study, we will present our research results on the synthesis and characterization of novel ruthenium complexes. The ruthenium compounds were easy synthesized by the reaction of ruthenium halide with appropriate organic ligands in protic solvent, and characterized by NMR, elemental analysis and thermogravimetric analysis. The molecular structures of the complexes were studied by single crystal diffraction. ALD of Ru film was demonstrated using the new Ru metallorganic precursor and O2 as the Ru source and reactant, respectively, at the deposition temperatures of $300-350^{\circ}C$. Self-limited reaction behavior was observed as increasing Ru precursor and O2 pulse time, suggesting that newly developed Ru precursor is applicable for ALD process. Detailed discussions on the chemical and structural properties of Ru thin films as well as its growth behavior using new Ru precursor will be also presented.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.7
• /
• pp.1657-1662
• /
• 2014

In this paper, the impact of substrate bias($V_{BS}$) on the zero capacitor RAM(Z-RAM) in n-channel junctionless multiple gate MOSFET(MuGFET) has been analyzed experimentally. Junctionless transistors with fin width of 50nm and 1 fin exhibits a memory window of 0.34V and a sensing margin of $1.8{\times}10^4$ at $V_{DS}=3.5V$ and $V_{BS}=0V$. As the positive $V_{BS}$ is applied, the memory window and sensing margin were improved due to an increase of impact ionization. When $V_{BS}$ is increased from 0V to 10V, not only the memory window is increased from 0.34V to 0.96V but also sensing margin is increased slightly. The sensitivity of memory window with different $V_{BS}$ in junctionless transistor was larger than that of inversion-mode transistor. A retention time of junctionless transistor is better than that of inversion-mode transistor due to low Gate Induced Drain Leakage(GIDL) current. To evaluate the device reliability of Z-RAM, the shifts in the Set/Reset voltages and current were measured.

• The KIPS Transactions:PartA
• /
• v.11A no.2
• /
• pp.115-122
• /
• 2004

In this paper, the multiple-valued adders and multipliers are implemented by current-mode CMOS. First, we implement the 3-valued T-gate and the 4-valued T-gate using current-mode CMOS which have an effective availability of integrated circuit design. Second we implement the circuits to be realized 2-variable 3-valued addition table and multiplication table over finite fields $GF(3^2)$, and 2-variable 4-valued addition table and multiplication table over finite fields $GF(4^2)$ with the multiple-valued T-gates. Finally, these operation circuits are simulated under $1.5\mutextrm{m}$ CMOS standard technology, $15\mutextrm{A}$ unit current, and 3.3V VDD voltage Spice. The simulation results have shown the satisfying current characteristics. The 3-valued adder and multiplier, and the 4-valued adder and multiplier implemented by current-mode CMOS is simple and regular for wire routing and possesses the property of modularity with cell array. Also, since it is expansible for the addition and multiplication of two polynomials in the finite field with very large m, it is suitable for VLSI implementation.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.45 no.7
• /
• pp.37-43
• /
• 2008

In this paper, the dependence of electrical characteristics of Silicon-$Al_2O_3$-Nitride-Oxide-Silicon (SANOS) memory cell transistors and program/erase (P/E) speed, reliability of memory device on interface trap between Si substrate and tunneling oxide and bulk trap in nitride layer were investigated using charge pumping method which has advantage of simple and versatile technique. We analyzed different SANOS memory devices that were fabricated by the identical processing in a single lot except the deposition method of the charge trapping layer, nitride. In the case of P/E speed, it was shown that P/E speed is slower in the SANOS cell transistors with larger capture cross section and interface trap density by charge blocking effect, which is confirmed by simulation results. However, the data retention characteristics show much less dependence on interface trap. The data retention was deteriorated as increasing P/E cycling number but not coincides with interface trap increasing tendency. This result once again confirmed that interface trap independence on data retention. And the result on different program method shows that HCI program method more degraded by locally trapping. So, we know as a result of experiment that analysis the SANOS Flash memory characteristic using charge pumping method reflect the device performance related to interface and bulk trap.