This PDF file contains the front matter associated with SPIE Proceedings Volume 8782, including the Title Page, Copyright information, Table of Contents, Introduction, and Conference Committee listing.

PQ/PMMA and PVA/acrylamide photopolymer are fabricated in our Lab. We investigate the holographic characteristics of SiO₂, Zinc methacrylate (ZnMA) doped and the methacrylate (MAA) linked PQ/PMMA photopolymer. By optimizing the doping content, the diffraction efficiency, photosensitivity and temperature stability have increased. Moreover, the holographic properties of PVA/acrylamide photopolymer have also been improved. The response time decreases by 55.7% at the optimized SiO₂ concentration of 0.4wt%. The photosensitivity increases by 23.1% at the optimal preillumination energy of 72 μJ. Multilayer photopolymers with thicknesses over 500 μm were fabricated, exhibiting better Bragg selectivity.

A new kind of multi-level recordable disc is presented in this paper. By improving the recording control signal in DVD-R/RW burning process, a Multi-level disc with signal waveform modulation can be achieved. The multi-level recording method increases the recording density but also the channel noise. The optimally designed partial response maximum likelihood (PRML) equalizer is effective in recovering channel symbol, and the original error rate reach 10E-4. This kind of multi-level method can be used for the purpose such as increasing the capacity of disc, hiding data and burning disc identification information etc.

Photopolymer composite materials doped with different kinds of nanoparticles (such as SiO₂ TiO₂, Mg(OH)₂, etc.) were fabricated. Photopolymerization processes in the materials and some holographic performance parameters were measured and analyzed. The experimental results show that the maximum diffraction efficiency, maximum refractive index modulation and some other holographic performance parameters increase with the increase in the nanoparticles concentration until it arrives the optimum concentration. Especially, the Bragg-mismatch angle of the material films decreases obviously and the dimensional stability is more remarkably improved, too. For example, the shrinkage rate of the sample films undoped and or doped with TiO₂ nanoparticles, decreases from 1.12% to 0.28%. The surface structure of the two dimensional holographic gratings was investigated by Scanning Electron Microscopy. Two dimensional digital images generated by Spatial Light Modulator were stored in the thick films (about 200μm), and the reconstructed data pages have good fidelity. The results show that the nanoparticle-doped photopolymer materials have a better potential in holographic storage.

In this work, light beam is simulated through nonlinear super-resolution films by the finite element method. The results indicate that the transmitted beam can be reshaped and the light spot can be smaller than the diffraction limit due to the reshaping effect of strong nonlinear saturatable absorption. However, the size of the transmitted light spot cannot be decreased continuously with the increase of the film thickness, and the light spot size then decreases with the increase of the film thickness. The simulated results mean that the carrier-to-noise-ratio first increases, and then decreases with the film thickness, and there is an optimum film thickness for the read-only nonlinear super-resolution optical disks, which is consistent with the reported experimental results.

The present paper describes a method for decreasing the pit size in optical lithography by using combination of a four-zone annular binary phase filter and the phase transition material. The binary phase filter was designed by vector diffraction theory when linearly polarized light is focused under high numerical aperture objective lens ( NA=0.95), the figures of merit produced by this filter are as follows: compared with the diffraction limited spot, Strehl ratio S is 0.254, the spot size in the short axis direction is reduced down to 77.3%, the depth of focus is elongated to 317% for the super-resolved spot,. Then a phase transition material is placed in the focal plane of the objective lens, according to the threshold effect of the material, the groove linewidth and pit size can decrease down to about 0.2λ, which is about 90nm at the wavelength of 405nm. Therefore, nanoscale phase transition optical lithography is realized, and the capacity and density of the optical memory devices can be increased up to 2-3 times the blu-ray disks.

Super-resolution near-field structure (super-RENS) is a functional structure which can overcome the optical diffraction limit and play an important role in nano scale optical data storage. The resolution enhancement of the scatter-type super- RENS optical disk is related to the localized surface plasmon of silver particles dissociated from the AgO_x layer and its near-field interaction with the recording pits in the phase change layer. Recently, a new method for optically synthesizing silver nanoparticles in a phase change matrix has been proposed by our group [Mater. Chem. Phys. 135, 467-473(2012)], which provides a potential approach to forming a more simple plasmonic recording structure than the traditional AgO_x-type structure. In this paper, field distribution of silver-nanoparticles-embedded Ge₂Sb₂Te₅ phase change recording pits is numerically analyzed by the finite-difference time-domain (FDTD) method. The optical contrast enhancement capability is confirmed for the optimized recording structures when the pit size is smaller than the optical diffraction limit.

As the demand for ultrahigh density information storage continues to grow, recording mark size of several tens nanometer which is smaller than the optical diffraction limit is required in optical memory. Functional film super-resolution technique is one of practical approaches to overcome the optical diffraction limit. Optical constants are important parameters to optical films as super-resolution masks. In this paper, the influence of film thickness on optical constants of antimony-based bismuth-doped super-resolution mask layer is investigated. The structure of the samples with different thickness was studied by X-ray diffraction. The transmission spectrum was measured by spectrophotometry. The optical constants of the films in the range of 300-800 nm were measured by spectroscopic ellipsometry. The results show that the structure of the film transforms from amorphous state to crystal state when the thickness increases from 7 nm to 300 nm. In the range of 300-800 nm, the refractive index and extinction coefficient increase with increasing wavelength. The transmission decreases rapidly when the thickness increases from 7 nm to 30 nm. The influences of film thickness on optical constants are more significant in the thickness range of 7-50 nm than that in the thickness above 50 nm.

A nickel(II)-azo dye was synthesized in order to obtain a suitable optical recording medium for write-once blu-ray recording with low-to-high signal polarity. Smooth thin film of the nickel(II)-azo dye was prepared by spin-coating method. Absorption, reflectance spectra and optical constants (complex refractive indices N=n+ik) of the thin film were investigated in the wavelength region of 300-700 nm. To demonstrate the physical basis of low-to-high polarity, the relationships between the absorption, reflectivity and optical constants for the nickel(II)-azo dye in spin-coated films were discussed. It is found that the thin film has a strong absorption band in the wavelength region of 370-500 nm and a strong absorbance at 405 nm. The reflectance spectra show that a low reflectivity of the unrecorded thin film at 405 nm can be obtained. The thin film of the nickel(II)-azo dye gives a relatively low n value of 1.36 and a relatively high k value of 0.51 at 405 nm. In addition, in order to examine its possible use as a blu-ray recording medium, the spin-coated film of the nickel(II)-azo dye was studied by a static optical recording testing system. The results demonstrate that high reflectivity contrast (85%) with low-to-high polarity can be obtained at an optimum laser writing power and pulse width using the nickel(II)-azo dye thin film as the recording layer. Moreover, the recording marks on the film are clear, circular, regular, and their size is as small as 200 nm or even less. These results indicate that the nickel(II)-azo dye is promising for applications in high-performance write-once blu-ray recording systems at 405 nm.

Anthracene derivative is one of the important two photon absorption materials. In this work, novel symmetric-type D- π-D anthracene derivative, 2, 6 - di- substituted anthracene chromophore with 4-methylstyrene has been synthesized via Heck reaction. The two-photon absorption (TPA) fluorescence emission property and TPA absorption coefficient are investigated by Sapphire femtosecond laser pulses at the wavelength of 800nm, repetition rate of 1 KHz and pulse width of 120fs, in an attempt to obtain the TPA cross-sections of this compound (σ_film=695.8×10^–50cm⁴s/photon). Finally, Three-dimensional (3D) optical data storage experiments are carried out and realized by two-photon absorption with 800nm-fs laser in the compound films.

Preparation of recording materials with high two-photon absorption activities is one of the important issues to superhigh- density two-photon absorption (TPA) three-dimensional (3D) optical data storage. In this paper, three new carbazole derivatives containing nitrogen heterocyclic ring with symmetric and asymmetric structures are prepared using ethylene as the π bridge between the carbazole unit and nitrogen heterocyclic ring, namely, 9-butyl-3-(2-(1,8- naphthyridin)vinyl)-carbazole (material 1), 9-butyl-3,6-bis(2-(1,8-naphthyl)vinyl)-carbazole (material 2) and 9-butyl-3,6- bis(2-(quinolin)vinyl)-carbazole (material 3). Their one photon properties including linear absorption spectra, fluorescence emission spectra, and fluorescence quantum yields are studied. The fluorescence excited by 120 fs pulse at 800 nm Ti: sapphire laser operating at 1 kHz repetition rate with different incident powers of 9-butyl-3-(2-(quinolin) vinyl)-carbazole (material 3) was investigated, and two-photon absorption cross-sections has been obtained. It is shown that material 3 containing quinoline rings as electron acceptor with symmetric structure exhibit high two-photon absorption activity. The result implies that material 3 (9-butyl-3-(2-(quinolin) vinyl)-carbazole) is a good candidate as a promising recording material for super-high-density two-photon absorption (TPA) three-dimensional (3D) optical data storage. The influence of chemical structure of the materials on the optical properties is discussed.

Holographic gratings were recorded in spirooxazine doped PMMA films by blue-violet laser (405nm), Nd:YAG laser (532nm) and He-Ne laser (632.8nm), respectively. It was found that the photo-dynamics of the holographic grating was dependent on recording wavelength for the photochromic characteristics of spirooxazines. A theoretical description of orientation and isomerization gratings agrees well with the experimental results. It was also found that a pure isomerization grating is formed by two interferential 405nm beams while a mixed grating consisting of isomerization and orientation components is formed by the visible interferential beams (532nm or 632.8nm). Due to the long life of the isomerization gratings, the holographic interference fringes with different periods were clearly observed by Confocal Laser Scanning Microscope.

Hydrophilic photopolymers with AA as the monomer are versatile materials for use as holographic recording media. It has demonstrated a possibility to improve the refractive index modulation by doping inorganic nanoparticles into the polymer matrix. But the agglomerate effect of inorganic nanoparticles always deteriorate the holographic storage characteristics of the inorganic nanoparticle dispersed photopolymer compound materials. monodisperse hydrophilic gold nanoparticles are fabricated by the method of sodium citrate reduction, and then were dispersed into the PVA/AA photopolymer. Our results indicate that citrate ions coated on the surface of gold nanoparticles can effectively prevent the agglomerate effect, so that the gold nanoparticles can uniformly dispersed in the sol-gel photopolymer. The permanent refractive index grating can be formed owing to a spatial distribution of the effective refractive index with periodically distributed nanoparticles and polymerized monomers during the holographic recording. Furthermore, the citrate ion modified gold nanoparticles and polymer chain can induce the formation of hydrogen bonds which can prevent the material from distortion and at the same time improve the stability of the material by the coordination effect of carboxylate. Therefore, the incorporation of gold nanoparticles leads to a significantly improvement in the holographic properties of the materials, so that the volume shrinkage during the holographic exposure can be well suppressed. Additionally, there exists the optimum concentration of gold nanoparticles, at which the maximum diffraction efficiency can be as high as 90% and the volume shrinkage can be reduced to 0.8%.

Ni-doped SrTi_0.75Sn_0.25O₃ (STS) thin films have been successfully prepared for the resistive random access memory (RRAM) cells in TiN (top electrode)/STS/ (bottom) Pt and TiN/STS/LNO sandwich structures, respectively. The reproducible resistive switching (RS) characteristics of the two types of structures are studied systematically for RRAM applications. The former structure shows a filament type of RS behaviors mainly controlled by the top interface between the TiN and STS films. On the contrary, the latter structure presents a Schottky-like barrier modulation type of RS behaviors mainly confined at the bottom interface layers between STS and LNO films. Both of the two types of RS behaviors can be attributed to the movement of the oxygen vacancies in the STS films under external voltage.

An 8Mb phase change random access memory (PCRAM) has been developed by a 130nm 4-ML standard CMOS technology based on the Resistor-on-Via-stacked-Plug (RVP) storage cell structure. This phase change resistor is formed after CMOS logic fabrication. PCRAM can be embedded without changing any logic device and process. The memory cell selector is implemented by a standard 1.2V NMOS device. Aimed at the resistance distributions, lowering the operation current and improving the bit yield, some methods are used to optimize the design of the chip.

A serial peripheral interface (SPI) 16-Kbit phase change memory chip based on 0.13μm CMOS process is designed. It contains a parallel error correcting code (ECC) circuit, which can correct 2 bits in every 8 bits without clock delay, enabling the write and read operations performed at bus speed. All the data transfers in 8-bit groups and can be read or written with write protection scheme by unlimited cycle, in which address can automatically increase one by one. Simulation results show that the chip can work correctly in SPI mode and with ECC scheme. It is now under testing.

Sb-rich Sb-Te films with different composition were investigated by temperature-dependent resistance measurement, crystal structure characterization, and in situ crystallization behavior study. The results show that when the Sb content is more than 80 at.%, Sb-rich Sb-Te films cannot be used as phase change material due to their low crystallization temperature and small resistance contrast. Sb-rich Sb-Te films with the Sb content being between 80~67 at.% can be used as phase-change-material and they have similar properties because of their similar growth-dominant crystallization behaviors.

^{A phase change memory cell based on Ge_0.5Sb₂Te₃/Ti_0.6Sb₂Te₃ double-layer structure is proposed for 3-level storage. The fabricated cell can realize 3-level storage ability by both current and voltage operation. Cycling ability has been proved better than 2×103. Thermal simulation shows that the resistivity difference between the two materials can greatly affect the temperature distribution in the cell. More heat will be generated in the amorphous Ge Ge_0.5Sb₂Te₃/ film when the current flow through due to the higher resistivity. And the lower crystallization temperature of Ge_0.5Sb₂Te₃/compared to that of Ti0.6Sb2Te3 ensures its priority of crystallization, which makes the 3-level storage feasible.}

New phase change materials development has become one of the most critical modules in the fabrication of low power consumption and good data retention phase change memory (PCM). Among various candidates of new phase change materials, SiSbTe (SST) is one of the most promising materials due to its benefits of low RESET current, high crystallization temperature, good adhesion and small volume shrinkage during phase change from amorphous to crystalline state. However, the oxidization of SST film was found when exposing to the atmosphere. By analyzing the depth profile of chemical states, we found oxygen more easily penetrated into the SST film and bonded with Si and Sb compared to GeSbTe (GST) film. The oxidization mechanism between SST and GST was briefly discussed. We achieved 80% improvement of oxidization issue by nitrogen and argon surface treatment. We proposed a manufacturing solution of SST for PCM.

Crystallization behavior of the Ge₂Te₃-TiO₂ films prepared by the co-sputtering using Ge₂Te₃ and TiO₂ targets was investigated by in situ resistance-temperature measurement and transmission electron microscopy. The crystallization kinetic parameters including rate factor and kinetics exponent were obtained by the non-isothermal change in resistance using Kissinger’s plot and Ozawa’s method. The average kinetics exponent was estimated by the nonisothermal change in resistance. Compared with other studied compositions, the composition with TiO2 concentration of 5 at.% exhibited shorter crystallization time which was calculated by the Johnson-Mehl-Avrami equation. The crystallization behavior of Ge₂Te₃-TiO₂ film was verified by the transition electron microscopy at different annealing temperature. With the short crystallization time and high crystallization temperature, the compositions with TiO2 concentration of 5-15 at.% may be one of the competing candidates for phase change memory application.

In the fabrication of phase change memory devices, HBr/He gas is employed in patterning Ge₂Sb₂Te₅ (GST) because it is damage free to GST sidewall. Accurate and reproducible endpoint detection methods are necessary in this etching process. In-situ optical emission spectroscopy (OES) is collected and analyzed to control the GST etching process due to its non-invasiveness. By analyzing the light emitted from plasma, we report an effective etch endpoint detection method for GST etching process is developed and the results are also confirmed using scanning electron micrographs.

The crystallization properties of as-deposited amorphous AgInSbTe thin films irradiated by single-shot picosecond laser pulses were studied using in-situ transient optical reflectance and electrical resistance measurements with nanosecond resolution. It was found that the real-time optical and electrical signal responses were different under the same pumping conditions. The optical signals showed a multistage crystallization process with a total time of approximately 150 ns, while the electrical signals showed a negative exponential trend decreasing to the final stable state within about several microseconds. A resistor–capacitor model was constructed to explain this delayed electrical response. The fluencedependent evolution dynamics maybe implied a non-fully crystallization process under ultra-short pulse stimulation.

Phase change lithography has pretty potential applications for high density optical data storage mastering and micro/nano structure patterning because it is not restricted by optical diffraction limitation and at relatively low cost. GeSbTe, as an initially investigated material for phase change lithography, its mechanism of selective etching in inorganic or organic alkaline aqueous solutions, such as NaOH and tetramethylammonium hydroxide (TMAH), is still unknown. In this paper, XPS measurement is used to study the selective wet etching mechanism of GeSbTe phase change thin films with TMAH solution, and the results show that oxidization played an important role in the etching process. Ge, Sb and Te are oxidized into GeO₂, Sb₂O₅ and TeO₂, respectively, and then as the corresponding salts dissolved into the etchant solution. Ge-X (X is Ge, Sb or Te) bonds are first broken in the etching, then Sb-X bonds, and finally Te-Te bonds. To confirm the effect of oxidization in the etching, H₂O2 as an oxidant is added into the TMAH solution, and the etching rates are increased greatly for both amorphous and crystalline states. The selective etching mechanism of Ge₂Sb₂Te₅ phase change films is discussed by the difference of bonds breakage between the amorphous and crystalline states.

Te-based phase-change materials (such as Te, Ge₂Sb₂Te₅, and Ag₈In₁₄Sb₅₅Te₂₃) are employed now in optical data storage and non-volatile electrical memory due to the acceptable optical and electrical contrast between the crystalline and amorphous states. Thermal properties of the materials have an important influence on the device performance. In this work, the dependence of thermal properties of Te-based phase-change materials on temperature is measured and analyzed. Comparisons among these three Te-based phase-change materials are made, and the origins of thermal conductivity in Te, Ge₂Sb₂Te₅ and AgInSbTe are also discussed. One can see that the thermal properties of crystalline Ge₂Sb₂Te₅ are more similar to those of Te with respect to AgInSbTe due to containing higher percentage of Te element. These experimental data can be helpful for optimizing the performance of phase change memory devices.

Two series of FePt/[Co/Pt]_N (FPCP-N) and FePt/[Co/Ni]_N (FPCN-N) exchange coupled composite (ECC) films were prepared and studied. Well isolated island-like grains with an average grain size of about 20nm are obtained. Both FePt hard layer and Co based soft multilayer are of perpendicular anisotropy. Large reduction of the switching field is observed when the Co based multilayer is deposited on the top of FePt films. When the thickness of Co based multilayers increase, thermal stability of the composites decreases slightly in both two series. Therefore, low switching field and high thermal stability are both obtained in these perpendicular ECC films.

Ultrafast dynamics of spin waves and coercivity of a FePt/FeNi exchange coupling film is studied by femtosecond laser Kerr spectroscopy. Magnetization-precession spin waves are observed. Time-delayed magneto-optical Kerr hysteresis loop (TDMOKHL) technique is used to study the dynamics of coercivity of the magnetic film under laser excitations as done usually. However, we find that the dynamics of coercivity revealed by TDMOKHL technique does not reflect the evolution of initial coercivity given by the initial hysteresis loop with no pump excitation, but does that of final coercivity under laser excitation. The final coercivity is given by the hysteresis loop measured at a negative pump-probe delay time under pump excitation. The evolution of initial coercivity can be observed only under a weak pump excitation in which magnetic system behaves reversible magnetically. In general, TDMOKHL technique can not track the evolution of coercivity of a magnetic system because TDMOKHLs do not reflect the time evolution of initial hysteresis loop with no laser excitation. The correctness of all previous reports on the dynamics of coercivity and magnetic-order transition of a laser-excited magnetic system needs reexamining if those reports were dependent on TDMOKHL technique. The oscillatory recovery behavior of final coercivity is observed under pump excitation and ascribed to metastability of zero magnetization state.

Cloud storage technology has become a research hotspot in recent years, while the existing cloud storage services are mainly designed for data storage needs with stable high speed Internet connection. Mobile Internet connections are often unstable and the speed is relatively low. These native features of mobile Internet limit the use of cloud storage in portable consumer electronics. The Network Attached Flash File System (NAFFS) presented the idea of taking the portable device built-in NAND flash memory as the front-end cache of virtualized cloud storage device. Modern portable devices with Internet connection have built-in more than 1GB NAND Flash, which is quite enough for daily data storage. The data transfer rate of NAND flash device is much higher than mobile Internet connections^[1], and its non-volatile feature makes it very suitable as the cache device of Internet cloud storage on portable device, which often have unstable power supply and intermittent Internet connection. In the present work, NAFFS is evaluated with several benchmarks, and its performance is compared with traditional network attached file systems, such as NFS. Our evaluation results indicate that the NAFFS achieves an average accessing speed of 3.38MB/s, which is about 3 times faster than directly accessing cloud storage by mobile Internet connection， and offers a more stable interface than that of directly using cloud storage API. Unstable Internet connection and sudden power off condition are tolerable, and no data in cache will be lost in such situation.

The efficiency of Service composition optimal solution algorithm depends on the structure of the user requirement model. MLQEM (Multi-level Hybrid QoS Extended Model) provides a multi-level hybrid QoS structure. This structure allows users to actively give the specific constraint conditions about the professional field of web service. The number of user constraint conditions increase that increase the calculation of the current global optimization algorithm. In this paper we propose a new kind of algorithm MLHSA based on MLQEM. We integrate local optimal strategy and global optimal strategy in different levels of QoS. At last our experimental result indicate that MLHSA algorithm are improved obviously in efficiency and accuracy.

With the development of HD video, the protection of copyright becomes more complicated. More advanced copyright protection technology is needed. Traditional digital copyright protection technology generally uses direct or selective encryption algorithm and the key does not associate with the video content ^[1]. Once the encryption method is cracked or the key is stolen, the copyright of the video will be violated. To address this issue, this paper proposes a Sensitive Data Extraction Algorithm (SDEA) based on the content associated encryption technology which applies to the Internet Certification Service (ICS). The principle of content associated encryption is to extract some data from the video and use this extracted data as the key to encrypt the rest data. The extracted part from video is called sensitive data, and the rest part is called the main data. After extraction, the main data will not be played or poorly played. The encrypted sensitive data reach the terminal device through the safety certificated network and the main data are through ICS disc. The terminal equipments are responsible for synthesizing and playing these two parts of data. Consequently, even if the main data on disc is illegally obtained, the video cannot be played normally due to the lack of necessary sensitive data. It is proved by experiments that ICS using SDEA can destruct the video effectively with 0.25% extraction rates and the destructed video cannot be played well. It can also guarantee the consistency of the destructive effect on different videos with different contents. The sensitive data can be transported smoothly under the home Internet bandwidth.

An architecture and a kind of method are proposed for designing Viterbi Detector of NVD (Next-genertion Versatile Disc) system based on specific partial-response channels by us. It offers a standard design framework and process, leading to a simple and convenient design. The architecture and method are verified in both simulation test and hardware description with complete function and good performance. The schema and method are used to accomplish a partial response channel for PR[1 2 2 2 1] Viterbi detector in NVD systems, and to achieve the goal of the software simulation and hardware description.

This paper presents a novel adaptive signal quality evaluation method for Signal Waveform Modulation Multi-Level (SWM-ML) disc. The experimental results show the adaptive signal quality evaluation method is useful and proper for SWM ML disc.

Lens actuator is one of the most important components in an optical pickup system, which decides the performance of the disc readout system. A significant advance in technical capability has recently been achieved in the fabrication of integrated micro lens actuators of optical pickup by microelectromechanical systems (MEMS) technology. A comb-drive tracking and focusing integrated lens actuator fabricated on a silicon-on-insulator (SOI) wafer has been reported. Twodimensional tuning of the objective lens is generated by the integrated comb structures. Large displacements of about ±24.6μm in tracking direction and 5.7μm in focusing direction are demonstrated. The device has a high sensitivity and an ignorable coupling between the two dimensional driving movements. The small-form-factor device provides an excellent performance and size reduction. Furthermore, high quality polymer micro-lenses with high numerical aperture (NA) are fabricated on a pre-patterned hydrophobic glass substrate by liquid dispensing. The surface profiles are adjusted by the patterned diameter and the volume of the dispensed polymer, which is controlled by the dispensing time. This extremely low cost, high NA and easily fabricated lens represents an important step for further integration of the pickup system, thus expands the application area of optical storage.

A new jitter measurement method for multi-level discs is presented in this paper. This method, with merits such as high precision, flexible and low cost, combines the traditional pulse counting method and interpolation algorithm. The principle of this method is explained in detail, and its precision is analyzed. The analysis result is shown to be very effective. It has a broad application prospect and practical value in quality evaluation of multi-level discs.

Amorphous Ge₂Sb₂Te₅ (a-GST) chemical mechanical planarization (CMP) using KClO4 as the oxidizer in an acidicslurry is investigated in the present work. It is shown that the removal rate (RR ) of the a-GST firstly increases and thentends to saturate when the KClO4 concentration is greater than 0.8 wt%, but the static etch rate (SER) linearly increasesfrom low to high KClO₄ concentration. To understand the oxidation-reaction capability of Ge, Sb and Te, depth profilesof composition of elements and etch morphology of a-GST immersed in the slurry for some time are measured,respectively. It is found that selective corrosion occurs among Ge, Sb and Te, and an accumulation of Te and loss of Gein a-GST surface region are obvious observed, especially at high KClO₄ concentrations. Temperature dependent sheetresistance measurements of all the samples pre- and post-CMP reveal a similar trend, which implies a-GST CMP is ableto keep its characteristic well.

A novel phase change material, Si₂Sb₂Te₃ has been reported to show good phase change abilities. Etching of this material is a critical step in the fabrication of phase change memory devices. In this paper, the characteristics of Si₂Sb₂Te₃ etched in CF₄/Ar atmosphere are investigated. The influence of the etching rate and surface roughness with different CF₄/Ar ratio, pressure, and power are systematically studied. Furthermore, our X-ray photoelectron spectroscopy test results show that Te is the bottleneck to accelerating the etching rate.

Power on reset circuit provides a reset signal for the entire system into normal working condition. When power reaches normal operating voltage and remains stable, power on reset circuit provides one rectangular pulse signal which has a steep edge for the digital baseband part of the tag. After reset, POR circuit is isolated with the follow-up circuits, having good stability. In this paper, we designed two new static ultra-low power consumption power-on reset circuits. The first circuit uses Schmitt triggers for signal threshold detection and a certain delay. The peak current of the first circuit is 31uA, with a static current being 33 pA. The second circuit is based on a clamp circuit and PMOS gate cross-coupled circuit which greatly reduces the static current (190 pA). And the peak current of the second circuit is 21 μA.

This paper presents an optimized write driver used in Phase change memory (PCM). To pursue fast RESET/SET operation, the proper clock scheme is applied, with the maximum frequency at about 200MHz. The write driver uses current pulses at a fixed frequency to successfully write into memory cells. Compared with the traditional SET operation, the novel dual-pulse SET operation divides the program pulse into 2 periods: pre-programming period provides large energy to cross the threshold-switching fast, programming period quenches the phase change resistance to that of the crystalline state. The optimization of the write operation decrease the program time and improves the resistance distribution.

Resistance distributions of the crystalline (SET) state and amorphous (RESET) state for phase change memory (PCM) are experimentally investigated at the array level. The RESET distribution shows a low resistance tail, which potentially affects the reading margin of the chip. These tail cells are divided into two types by resistance programming current (R-IP) and current voltage (I-V) characteristics. Finally, approaches of improving the integration process to remove the Type-1 tail cells and optimizing the programming operation to repair the Type-2 tail cells are proposed.

A high density design of Schottky-barrier diode array with self-aligned nickel-silicidation under 40nm technology node fabricated on epitaxial layer for low power phase-change memory application is proposed. According to N-type doping profile from simulation, large ON/OFF current ratio, the lower barrier height of ФB and series resistance RS are all determined by the dosage of buried N+ layer, epitaxial layer thickness. In addition, the temperature effect of the Schottky diode array is demonstrated by I-V electrical characteristics. From the optimal silicon-based results, a 9F² 16 × 16 diode array with the ideality factor of 1.21~1.40 shows a drive current density of ~14.9 mA/μm², a J_on/J_off ratio of ~5.17×103, and crosstalk immunity. Furthermore, this calibrated physical model makes it possible to predict and improve the performance of accessing device array next generation for non-volatile memory application.

Up to now, there is no direct test system of current pulse in phase change memory (PCM). The traditional test system uses direct current or voltage pulses to do the set operation and voltage pulses to do reset operation. In this work, a new test system is introduced. This system can give current source pulses to the PCM device to do set operation. The test results are presented and analyzed.

A write driver for PCM is designed to improve reliability and bit yield in the write operation, due to the distributions during the phase change process. And the PCM cell can be injected by current or voltage respectively. Meanwhile, owing to the possible variations of the SET process parameters, the designed circuit can generate either multiple stepdown current pulse or multiple step-down voltage pulse. The circuit is developed based on SMIC 130 nm CMOS standard technology. Compared to the traditional constant current pulse programming, the test results show that the proposed multiple step-down current generator for SET operation can improve the uniformity of resistance and bit yield.

Circuit design of an adaptable pulse current source chip is presented in this paper. The pulse current source is supposed to be used to supply Reset and Set current in the phase change memory chip testing system. The value and width of the pulse current source are variable, with the maximum value of 10mA and minimum width of 50ns. Two pulse currents output simultaneously with the same width but different values. A voltage pulse input is used to control the width of pulse current output. This high frequency voltage pulse could induce noise jamming to the inner circuits. To avoid this, a novel ESD and bonding structure is proposed.