Issue Browser
Volume 37, Issue 1, Jan 2016
INVITED PAPERS
Advances and prospects in visible light communications
Hongda Chen, Chunhui Wu, Honglei Li, Xiongbin Chen, Zongyu Gao, Shigang Cui, Qin Wang
J. Semicond.  2016, 37(1): 011001  doi: 10.1088/1674-4926/37/1/011001

Visible light communication (VLC) is an emerging technology in optical wireless communication (OWC) that has attracted worldwide research in recent years. VLC can combine communication and illumination together, which could be applied in many application scenarios such as visible light communication local area networks (VLANs), indoor localization, and intelligent lighting. In recent years, pioneering and significant work have been made in the field of VLC. In this paper, an overview of the recent progress in VLC is presented. We also demonstrate our recent experiment results including bidirectional 100 Mbit/s VLAN or Li-Fi system based on OOK modulation without blue filter. The VLC systems that we proposed are good solutions for high-speed VLC application systems with low-cost and low-complexity. VLC technology shows a bright future due to its inherent advantages, shortage of RF spectra and ever increasing popularity of white LEDs.

Visible light communication (VLC) is an emerging technology in optical wireless communication (OWC) that has attracted worldwide research in recent years. VLC can combine communication and illumination together, which could be applied in many application scenarios such as visible light communication local area networks (VLANs), indoor localization, and intelligent lighting. In recent years, pioneering and significant work have been made in the field of VLC. In this paper, an overview of the recent progress in VLC is presented. We also demonstrate our recent experiment results including bidirectional 100 Mbit/s VLAN or Li-Fi system based on OOK modulation without blue filter. The VLC systems that we proposed are good solutions for high-speed VLC application systems with low-cost and low-complexity. VLC technology shows a bright future due to its inherent advantages, shortage of RF spectra and ever increasing popularity of white LEDs.
SEMICONDUCTOR MATERIALS
Synthesis and electroluminescence characterization of a new aluminum complex,[8-hydroxyquinoline] bis[2, 2'bipyridine] aluminum Al(Bpy)2q
Rahul Kumar, Ritu Srivastava, Punita Singh
J. Semicond.  2016, 37(1): 013001  doi: 10.1088/1674-4926/37/1/013001

We have synthesized and characterized a new electroluminescent material,[8-hydroxyquinoline] bis[2,2'bipyridine] aluminum. A solution of this material Al(Bpy)2q in toluene showed absorption maxima at 380 nm, which was attributed to the moderate energy (π-π*) transitions of the aromatic rings. The photoluminescence spectrum of Al(Bpy)2q in the toluene solution showed a peak at 518 nm. This material shows thermal stability up to 300℃. The structure of the device is ITO/F4-TCNQ (1 nm)/α-NPD (35 nm)/Al(Bpy)2q (35 nm)/BCP (6 nm)/Alq3 (28 nm)/LiF (1 nm)/Al (150 nm). This device exhibited a luminescence peak at 515 nm (CIE coordinates, x=0.32, y=0.49). The maximum luminescence of the device was 214 cd/m2 at 21 V. The maximum current efficiency of OLED was 0.12 cd/A at 13 V and the maximum power efficiency was 0.03 lm/W at 10 V.

We have synthesized and characterized a new electroluminescent material,[8-hydroxyquinoline] bis[2,2'bipyridine] aluminum. A solution of this material Al(Bpy)2q in toluene showed absorption maxima at 380 nm, which was attributed to the moderate energy (π-π*) transitions of the aromatic rings. The photoluminescence spectrum of Al(Bpy)2q in the toluene solution showed a peak at 518 nm. This material shows thermal stability up to 300℃. The structure of the device is ITO/F4-TCNQ (1 nm)/α-NPD (35 nm)/Al(Bpy)2q (35 nm)/BCP (6 nm)/Alq3 (28 nm)/LiF (1 nm)/Al (150 nm). This device exhibited a luminescence peak at 515 nm (CIE coordinates, x=0.32, y=0.49). The maximum luminescence of the device was 214 cd/m2 at 21 V. The maximum current efficiency of OLED was 0.12 cd/A at 13 V and the maximum power efficiency was 0.03 lm/W at 10 V.
New method for thickness determination and microscopic imaging of graphene-like two-dimensional materials
Xudong Qin, Yonghai Chen, Yu Liu, Laipan Zhu, Yuan Li, Qing Wu, Wei Huang
J. Semicond.  2016, 37(1): 013002  doi: 10.1088/1674-4926/37/1/013002

We employed the microscopic reflectance difference spectroscopy (micro-RDS) to determine the layer-number and microscopically image the surface topography of graphene and MoS2 samples. The contrast image shows the efficiency and reliability of this new clipping technique. As a low-cost, quantifiable, no-contact and non-destructive method, it is not concerned with the characteristic signal of certain materials and can be applied to arbitrary substrates. Therefore it is a perfect candidate for characterizing the thickness of graphene-like two-dimensional materials.

We employed the microscopic reflectance difference spectroscopy (micro-RDS) to determine the layer-number and microscopically image the surface topography of graphene and MoS2 samples. The contrast image shows the efficiency and reliability of this new clipping technique. As a low-cost, quantifiable, no-contact and non-destructive method, it is not concerned with the characteristic signal of certain materials and can be applied to arbitrary substrates. Therefore it is a perfect candidate for characterizing the thickness of graphene-like two-dimensional materials.
Green preparation of Au nanoparticles for electrochemical detection of H2O2
Wenchao Wang, Ye Ji, Yong Zhang, Ziying Wang, Tong Zhang
J. Semicond.  2016, 37(1): 013003  doi: 10.1088/1674-4926/37/1/013003

A simple and green method for preparation of Au nanoparticles by reduction of HAuCl4 using carbon nanodots as the reducing agent is reported. The carbon nanodots were prepared by a green method as well, using hydrothermal treatment of grass. It is observed that Au nanoparticles show obvious electrochemical catalytic ability for reduction of hydrogen peroxide, leading to its application of a high-performance non-enzymatic hydrogen peroxide sensor. The hydrogen peroxide sensor based on Au nanoparticles was made, with the detection limit at 23 μ M and linear range between 0.1-160 mM.

A simple and green method for preparation of Au nanoparticles by reduction of HAuCl4 using carbon nanodots as the reducing agent is reported. The carbon nanodots were prepared by a green method as well, using hydrothermal treatment of grass. It is observed that Au nanoparticles show obvious electrochemical catalytic ability for reduction of hydrogen peroxide, leading to its application of a high-performance non-enzymatic hydrogen peroxide sensor. The hydrogen peroxide sensor based on Au nanoparticles was made, with the detection limit at 23 μ M and linear range between 0.1-160 mM.
Impact of Cu-rich growth on the Cu2ZnSnSe4 surface morphology and related solar cells behavior
Ding Sun, Yang Ge, Li Zhang, Shengzhi Xu, Ze Chen, Ning Wang, Xuejiao Liang, Changchun Wei, Ying Zhao, Xiaodan Zhang
J. Semicond.  2016, 37(1): 013004  doi: 10.1088/1674-4926/37/1/013004

In order to study the influence of Cu-rich growth on the performance of the Cu2ZnSnSe4 (CZTSe) thin film solar cells, a multi-stage co-evaporation process is applied. The CZTSe films are grown at a lower substrate temperature to reduce the existence time of CuxSey at the first period caused by the volatility of SnSex. This study examines the surface morphology and device performance in Cu-rich growth and close-to-stoichiometric growth. Although the grain size of Cu-rich growth film increases a little, the difference was not dramatic as the results of CIGS reported previously. A model based on the grain boundary migration theory is proposed to explain the experimental results. The mechanisms of Cu-rich growth between CZTSe and CIGS might be different.

In order to study the influence of Cu-rich growth on the performance of the Cu2ZnSnSe4 (CZTSe) thin film solar cells, a multi-stage co-evaporation process is applied. The CZTSe films are grown at a lower substrate temperature to reduce the existence time of CuxSey at the first period caused by the volatility of SnSex. This study examines the surface morphology and device performance in Cu-rich growth and close-to-stoichiometric growth. Although the grain size of Cu-rich growth film increases a little, the difference was not dramatic as the results of CIGS reported previously. A model based on the grain boundary migration theory is proposed to explain the experimental results. The mechanisms of Cu-rich growth between CZTSe and CIGS might be different.
Excellent performance of gas sensor based on In2O3-Fe2O3 nanotubes
Li Liu, Shouchun Li, Xin Guo, Yue He, Lianyuan Wang
J. Semicond.  2016, 37(1): 013005  doi: 10.1088/1674-4926/37/1/013005

In2O3-Fe2O3 nanotubes are synthesized by an electrospinning method. The as-synthesized materials are characterized by scanning electron microscope and X-ray powder diffraction. The gas sensing results show that In2O3-Fe2O3 nanotubes exhibit excellent sensing properties to acetone and formaldehyde at different operating temperatures. The responses of gas sensors based on In2O3-Fe2O3 nanotubes to 100 ppm acetone and 100 ppm formaldehyde are 25 (240℃) and 15 (260℃), and the response/recovery times are 3/7 s and 4/7 s, respectively. The responses of In2O3-Fe2O3 nanotubes to 1 ppm acetone (240℃) and formaldehyde (260℃) are 3.5 and 1.8, respectively. Moreover, the gas sensor based on In2O3-Fe2O3 nanotubes also possesses an excellent selectivity to acetone and formaldehyde.

In2O3-Fe2O3 nanotubes are synthesized by an electrospinning method. The as-synthesized materials are characterized by scanning electron microscope and X-ray powder diffraction. The gas sensing results show that In2O3-Fe2O3 nanotubes exhibit excellent sensing properties to acetone and formaldehyde at different operating temperatures. The responses of gas sensors based on In2O3-Fe2O3 nanotubes to 100 ppm acetone and 100 ppm formaldehyde are 25 (240℃) and 15 (260℃), and the response/recovery times are 3/7 s and 4/7 s, respectively. The responses of In2O3-Fe2O3 nanotubes to 1 ppm acetone (240℃) and formaldehyde (260℃) are 3.5 and 1.8, respectively. Moreover, the gas sensor based on In2O3-Fe2O3 nanotubes also possesses an excellent selectivity to acetone and formaldehyde.
SEMICONDUCTOR DEVICES
Ageing of GaN HEMT devices:which degradation indicators?
A. Divay, O. Latry, C. Duperrier, F. Temcamani
J. Semicond.  2016, 37(1): 014001  doi: 10.1088/1674-4926/37/1/014001

A following of diverse degradation indicators during the ageing in operational conditions of AlGaN/GaN HEMTs (high electron mobility transistors) is proposed. Measurements of pulsed I-V, Schottky barrier height, RF output power and gate current versus output power during the early phase of the ageing test (2000 h on a 6000 h total) are presented. These preliminary results give insight on some of the principal degradation indicators that are interesting to follow during an ageing test close to operational conditions on such components.

A following of diverse degradation indicators during the ageing in operational conditions of AlGaN/GaN HEMTs (high electron mobility transistors) is proposed. Measurements of pulsed I-V, Schottky barrier height, RF output power and gate current versus output power during the early phase of the ageing test (2000 h on a 6000 h total) are presented. These preliminary results give insight on some of the principal degradation indicators that are interesting to follow during an ageing test close to operational conditions on such components.
Cu2O-based solar cells using oxide semiconductors
Tadatsugu Minami, Yuki Nishi, Toshihiro Miyata
J. Semicond.  2016, 37(1): 014002  doi: 10.1088/1674-4926/37/1/014002

We describe significant improvements of the photovoltaic properties that were achieved in Al-doped ZnO (AZO)/n-type oxide semiconductor/p-type Cu2O heterojunction solar cells fabricated using p-type Cu2O sheets prepared by thermally oxidizing Cu sheets. The multicomponent oxide thin film used as the n-type semiconductor layer was prepared with various chemical compositions on non-intentionally heated Cu2O sheets under various deposition conditions using a pulsed laser deposition method. In Cu2O-based heterojunction solar cells fabricated using various ternary compounds as the n-type oxide thin-film layer, the best photovoltaic performance was obtained with an n-ZnGa2O4 thin-film layer. In most of the Cu2O-based heterojunction solar cells using multicomponent oxides composed of combinations of various binary compounds, the obtained photovoltaic properties changed gradually as the chemical composition was varied. However, with the ZnO-MgO and Ga2O3-Al2O3 systems, higher conversion efficiencies (η) as well as a high open circuit voltage (Voc) were obtained by using a relatively small amount of MgO or Al2O3, e.g., (ZnO)0.91-(MgO)0.09 and (Ga2O3)0.975-(Al2O3)0.025, respectively. When Cu2O-based heterojunction solar cells were fabricated using Al2O3-Ga2O3-MgO-ZnO (AGMZO) multicomponent oxide thin films deposited with metal atomic ratios of 10, 60, 10 and 20 at.% for the Al, Ga, Mg and Zn, respectively, a high Voc of 0.98 V and an η of 4.82% were obtained. In addition, an enhanced η and an improved fill factor could be achieved in AZO/n-type multicomponent oxide/p-type Cu2O heterojunction solar cells fabricated using Na-doped Cu2O (Cu2O:Na) sheets that featured a resistivity controlled by optimizing the post-annealing temperature and duration. Consequently, an η of 6.25% and a Voc of 0.84 V were obtained in a MgF2/AZO/n-(Ga2O3-Al2O3)/p-Cu2O:Na heterojunction solar cell fabricated using a Cu2O:Na sheet with a resistivity of approximately 10 Ω·cm and a (Ga0.975Al0.025)2O3 thin film with a thickness of approximately 60 nm. In addition, a Voc of 0.96 V and an η of 5.4% were obtained in a MgF2/AZO/n-AGMZO/p-Cu2O:Na heterojunction solar cell.

We describe significant improvements of the photovoltaic properties that were achieved in Al-doped ZnO (AZO)/n-type oxide semiconductor/p-type Cu2O heterojunction solar cells fabricated using p-type Cu2O sheets prepared by thermally oxidizing Cu sheets. The multicomponent oxide thin film used as the n-type semiconductor layer was prepared with various chemical compositions on non-intentionally heated Cu2O sheets under various deposition conditions using a pulsed laser deposition method. In Cu2O-based heterojunction solar cells fabricated using various ternary compounds as the n-type oxide thin-film layer, the best photovoltaic performance was obtained with an n-ZnGa2O4 thin-film layer. In most of the Cu2O-based heterojunction solar cells using multicomponent oxides composed of combinations of various binary compounds, the obtained photovoltaic properties changed gradually as the chemical composition was varied. However, with the ZnO-MgO and Ga2O3-Al2O3 systems, higher conversion efficiencies (η) as well as a high open circuit voltage (Voc) were obtained by using a relatively small amount of MgO or Al2O3, e.g., (ZnO)0.91-(MgO)0.09 and (Ga2O3)0.975-(Al2O3)0.025, respectively. When Cu2O-based heterojunction solar cells were fabricated using Al2O3-Ga2O3-MgO-ZnO (AGMZO) multicomponent oxide thin films deposited with metal atomic ratios of 10, 60, 10 and 20 at.% for the Al, Ga, Mg and Zn, respectively, a high Voc of 0.98 V and an η of 4.82% were obtained. In addition, an enhanced η and an improved fill factor could be achieved in AZO/n-type multicomponent oxide/p-type Cu2O heterojunction solar cells fabricated using Na-doped Cu2O (Cu2O:Na) sheets that featured a resistivity controlled by optimizing the post-annealing temperature and duration. Consequently, an η of 6.25% and a Voc of 0.84 V were obtained in a MgF2/AZO/n-(Ga2O3-Al2O3)/p-Cu2O:Na heterojunction solar cell fabricated using a Cu2O:Na sheet with a resistivity of approximately 10 Ω·cm and a (Ga0.975Al0.025)2O3 thin film with a thickness of approximately 60 nm. In addition, a Voc of 0.96 V and an η of 5.4% were obtained in a MgF2/AZO/n-AGMZO/p-Cu2O:Na heterojunction solar cell.
Performance analysis of STT-RAM with cross shaped free layer using Heusler alloys
Tangudu Bharat Kumar, Bahniman Ghosh, Bhaskar Awadhiya, Ankit Kumar Verma
J. Semicond.  2016, 37(1): 014003  doi: 10.1088/1674-4926/37/1/014003

We have investigated the performance of a spin transfer torque random access memory (STT-RAM) cell with a cross shaped Heusler compound based free layer using micromagnetic simulations. We have designed a free layer using a Cobalt based Heusler compound. Simulation results clearly show that the switching time from one state to the other state has been reduced, also it has been found that the critical switching current density (to switch the magnetization of the free layer of the STT RAM cell) is reduced.

We have investigated the performance of a spin transfer torque random access memory (STT-RAM) cell with a cross shaped Heusler compound based free layer using micromagnetic simulations. We have designed a free layer using a Cobalt based Heusler compound. Simulation results clearly show that the switching time from one state to the other state has been reduced, also it has been found that the critical switching current density (to switch the magnetization of the free layer of the STT RAM cell) is reduced.
Designing of 1 eV GaNAs/GaInAs superlattice subcell in current-matched four-junction solar cell
Haixiao Wang, Xinhe Zheng, Xinyuan Gan, Naiming Wang, Hui Yang
J. Semicond.  2016, 37(1): 014004  doi: 10.1088/1674-4926/37/1/014004

A reasonably-thick GaNAs/GaInAs superlattice could be an option as a roughly 1 eV subcell to achieve high-efficiency multi-junction solar cells on a lattice-matched Ge substrate. A detailed consideration of a high-efficiency design for a GaInP/GaAs/1 eV/Ge device is presented. Calculations have been done for this structure to obtain the confined energies of the electrons and holes by utilizing the Kronig-Penney model, as well as the absorption coefficient and thereby the external quantum efficiency. The effect of well layers, GaNAs or GaInAs, on the absorption and photocurrent density under the AM 1.5 condition is discussed in order to realize a requirement of current matching in the four-junction solar cells. The management of these considerations implies the feasibility of the GaNAs/GaInAs superlattice subcell design to improve the overall conversion efficiency of lattice matched GaInP/GaAs/1 eV/Ge cells.

A reasonably-thick GaNAs/GaInAs superlattice could be an option as a roughly 1 eV subcell to achieve high-efficiency multi-junction solar cells on a lattice-matched Ge substrate. A detailed consideration of a high-efficiency design for a GaInP/GaAs/1 eV/Ge device is presented. Calculations have been done for this structure to obtain the confined energies of the electrons and holes by utilizing the Kronig-Penney model, as well as the absorption coefficient and thereby the external quantum efficiency. The effect of well layers, GaNAs or GaInAs, on the absorption and photocurrent density under the AM 1.5 condition is discussed in order to realize a requirement of current matching in the four-junction solar cells. The management of these considerations implies the feasibility of the GaNAs/GaInAs superlattice subcell design to improve the overall conversion efficiency of lattice matched GaInP/GaAs/1 eV/Ge cells.
Sentaurus® based modeling and simulation for GFET's characteristic for ssDNA immobilization and hybridization
Yunfang Jia, Cheng Ju
J. Semicond.  2016, 37(1): 014005  doi: 10.1088/1674-4926/37/1/014005

The graphene field effect transistor (GFET) has been widely studied and developed as sensors and functional devices. The first report about GFET sensing simulation on the device level is proposed. The GFET's characteristics, its responding for single strand DNA (ssDNA) and hybridization with the complimentary DNA (cDNA) are simulated based on Sentaurus, a popular CAD tool for electronic devices. The agreement between the simulated blank GFET feature and the reported experimental data suggests the feasibility of the presented simulation method. Then the simulations of ssDNA immobilization on GFET and hybridization with its cDNA are performed, the results are discussed based on the electron transfer (ET) mechanism between DNA and graphene.

The graphene field effect transistor (GFET) has been widely studied and developed as sensors and functional devices. The first report about GFET sensing simulation on the device level is proposed. The GFET's characteristics, its responding for single strand DNA (ssDNA) and hybridization with the complimentary DNA (cDNA) are simulated based on Sentaurus, a popular CAD tool for electronic devices. The agreement between the simulated blank GFET feature and the reported experimental data suggests the feasibility of the presented simulation method. Then the simulations of ssDNA immobilization on GFET and hybridization with its cDNA are performed, the results are discussed based on the electron transfer (ET) mechanism between DNA and graphene.
Active multi-mode-interferometer broadband superluminescent diodes
Feifei Wang, Peng Jin, Ju Wu, Yanhua Wu, Fajie Hu, Zhanguo Wang
J. Semicond.  2016, 37(1): 014006  doi: 10.1088/1674-4926/37/1/014006

We report a new quantum dot superluminescent diode with a new device structure. In this device, a multi-mode-interferometer configuration and a J-bend structure were monolithically integrated. Owing to the multi-mode-interferometer structure, the superluminescent diode exhibits 60% increase in output power and 43% reduction in the differential resistance compared with the uniform waveguide width superluminescent diode fabricated from the same wafer. Our device produces an emission spectrum as wide as 103.7 nm with an output power of 2.5 mW at 600 mA continue-wave injection current. This broadband emission spectrum makes the axial resolution of the optical coherence tomography system employing the superluminescent diode to 6 μ m in theory, which is high enough for most tissue imaging.

We report a new quantum dot superluminescent diode with a new device structure. In this device, a multi-mode-interferometer configuration and a J-bend structure were monolithically integrated. Owing to the multi-mode-interferometer structure, the superluminescent diode exhibits 60% increase in output power and 43% reduction in the differential resistance compared with the uniform waveguide width superluminescent diode fabricated from the same wafer. Our device produces an emission spectrum as wide as 103.7 nm with an output power of 2.5 mW at 600 mA continue-wave injection current. This broadband emission spectrum makes the axial resolution of the optical coherence tomography system employing the superluminescent diode to 6 μ m in theory, which is high enough for most tissue imaging.
Suppression of electron leakage in 808 nm laser diodes with asymmetric waveguide layer
Xiang Li, Degang Zhao, Desheng Jiang, Ping Chen, Zongshun Liu, Jianjun Zhu, Ming Shi, Danmei Zhao, Wei Liu
J. Semicond.  2016, 37(1): 014007  doi: 10.1088/1674-4926/37/1/014007

Electron leakage in GaAs-based separately confined heterostructure 808 nm laser diodes (SCH LDs) has a serious influence on device performance. Here, in order to reduce the energy of electrons injected into the quantum well (QW), an AlGaAs interlayer with a smaller Al component is added between the active region and the n-side waveguide. Numerical device simulation reveals that when the Al-composition of the AlGaAs interlayer and its thickness are properly elected, the electron leakage is remarkably depressed and the characteristics of LDs are improved, owing to the reduction of injected electron energy and the improvement of QW capture efficiency.

Electron leakage in GaAs-based separately confined heterostructure 808 nm laser diodes (SCH LDs) has a serious influence on device performance. Here, in order to reduce the energy of electrons injected into the quantum well (QW), an AlGaAs interlayer with a smaller Al component is added between the active region and the n-side waveguide. Numerical device simulation reveals that when the Al-composition of the AlGaAs interlayer and its thickness are properly elected, the electron leakage is remarkably depressed and the characteristics of LDs are improved, owing to the reduction of injected electron energy and the improvement of QW capture efficiency.
Anomalous luminescence efficiency enhancement of short-term aged GaN-based blue light-emitting diodes
Kewei Cao, Binglei Fu, Zhe Liu, Lixia Zhao, Jinmin Li, Junxi Wang
J. Semicond.  2016, 37(1): 014008  doi: 10.1088/1674-4926/37/1/014008

The origin of anomalous luminescence efficiency enhancement of short-term aged GaN-based blue light-emitting diodes was studied. We found that the intensity of the electroluminescence and photoluminescence spectra were both increased in the very beginning period of aging. With the help of a rate-equation model, we concluded that this kind of luminescence efficiency enhancement is a joint effect of the defect reduction in active layers and the changes out of active layers, for example the Mg acceptor annealing.

The origin of anomalous luminescence efficiency enhancement of short-term aged GaN-based blue light-emitting diodes was studied. We found that the intensity of the electroluminescence and photoluminescence spectra were both increased in the very beginning period of aging. With the help of a rate-equation model, we concluded that this kind of luminescence efficiency enhancement is a joint effect of the defect reduction in active layers and the changes out of active layers, for example the Mg acceptor annealing.
Arbitrary magnetic field modulations to a semiconductor pump with two types of spin-orbit couplings
Yunchang Xiao, Changyong Zhu, Rixing Wang
J. Semicond.  2016, 37(1): 014009  doi: 10.1088/1674-4926/37/1/014009

Arbitrary magnetic field modulations to the semiconductor pump with both the Rashba and Dresselhaus spin-orbit couplings (SOC) are studied. The pump is driven by double time-dependent delta potentials, which are formed in the interfaces between the semiconductor region and two normal leads. Based on the Floquet scattering approach, our calculations show that various currents can be pumped by couplings of the magnetic fields and the SOCs. Pure spin currents modulated by the arbitrary magnetic fields are discussed in detail.

Arbitrary magnetic field modulations to the semiconductor pump with both the Rashba and Dresselhaus spin-orbit couplings (SOC) are studied. The pump is driven by double time-dependent delta potentials, which are formed in the interfaces between the semiconductor region and two normal leads. Based on the Floquet scattering approach, our calculations show that various currents can be pumped by couplings of the magnetic fields and the SOCs. Pure spin currents modulated by the arbitrary magnetic fields are discussed in detail.
Theoretical study on erbium ytterbium co-doped super-fluorescent fiber source
Wentao Guo, Feng Du, Manqing Tan, Jian Jiao, Xiaofeng Guo
J. Semicond.  2016, 37(1): 014010  doi: 10.1088/1674-4926/37/1/014010

Erbium ytterbium co-doped super-fluorescent fiber source (EYD-SFS) has been simulated by a theoretical model based on rate equations and power transfer equations. The output performances of four basic structures of EYD-SFS have been expressed, and it indicated that the DPF structure is a preferable structure. The dependence of output power, mean wavelength and bandwidth stability on the pump fiber length and the concentration of Er3+ and Yb3+ have also been studied. The results indicated with a proper doping concentration of Er3+ and Yb3+ of 6.0×1026 ions/m3 and 1.0×1027 ions/m3, the optimal gain fiber length is 3.6 cm. In this condition, good performances of DPF structure of EYD-SFS have been achieved.

Erbium ytterbium co-doped super-fluorescent fiber source (EYD-SFS) has been simulated by a theoretical model based on rate equations and power transfer equations. The output performances of four basic structures of EYD-SFS have been expressed, and it indicated that the DPF structure is a preferable structure. The dependence of output power, mean wavelength and bandwidth stability on the pump fiber length and the concentration of Er3+ and Yb3+ have also been studied. The results indicated with a proper doping concentration of Er3+ and Yb3+ of 6.0×1026 ions/m3 and 1.0×1027 ions/m3, the optimal gain fiber length is 3.6 cm. In this condition, good performances of DPF structure of EYD-SFS have been achieved.
SEMICONDUCTOR INTEGRATED CIRCUITS
A novel loss compensation technique analysis and design for 60 GHz CMOS SPDT switch
Zonghua Zheng, Lingling Sun, Jun Liu, Shengzhou Zhang
J. Semicond.  2016, 37(1): 015001  doi: 10.1088/1674-4926/37/1/015001

A novel loss compensation technique for a series-shunt single-pole double-throw (SPDT) switch is presented operating in the 60 GHz. The feed-forward compensation network which is composed of an NMOS, a couple capacitance and a shunt inductance can reduce the impact of the feed forward capacitance to reduce the insertion loss and improve the isolation of the SPDT switch. The measured insertion loss and isolation characteristics of the switch somewhat deviating from the 60 GHz are analyzed revealing that the inaccuracy of the MOS model can greatly degrade the performance of the switch. The switch is implemented in TSMC 90-nm CMOS process and exhibits an isolation of above 27 dB at transmitter mode, and the insertion loss of 1.8-3 dB at 30-65 GHz by layout simulation. The measured insertion loss is 2.45 dB at 52 GHz and keeps<4 dB at 30-64 GHz. The measured isolation is better than 25 dB at 30-64 GHz and the measured return loss is better than 10 dB at 30-65 GHz. A measured input 1 dB gain compression point of the switch is 13 dBm at 52 GHz and 15 dBm at 60 GHz. The simulated switching speed with rise time and fall time are 720 and 520 ps, respectively. The active chip size of the proposed switch is 0.5×0.95 mm2.

A novel loss compensation technique for a series-shunt single-pole double-throw (SPDT) switch is presented operating in the 60 GHz. The feed-forward compensation network which is composed of an NMOS, a couple capacitance and a shunt inductance can reduce the impact of the feed forward capacitance to reduce the insertion loss and improve the isolation of the SPDT switch. The measured insertion loss and isolation characteristics of the switch somewhat deviating from the 60 GHz are analyzed revealing that the inaccuracy of the MOS model can greatly degrade the performance of the switch. The switch is implemented in TSMC 90-nm CMOS process and exhibits an isolation of above 27 dB at transmitter mode, and the insertion loss of 1.8-3 dB at 30-65 GHz by layout simulation. The measured insertion loss is 2.45 dB at 52 GHz and keeps<4 dB at 30-64 GHz. The measured isolation is better than 25 dB at 30-64 GHz and the measured return loss is better than 10 dB at 30-65 GHz. A measured input 1 dB gain compression point of the switch is 13 dBm at 52 GHz and 15 dBm at 60 GHz. The simulated switching speed with rise time and fall time are 720 and 520 ps, respectively. The active chip size of the proposed switch is 0.5×0.95 mm2.
Wide band low phase noise QVCO based on superharmonic injection locking
Yalan Xu, Jinguang Jiang, Jianghua Liu
J. Semicond.  2016, 37(1): 015002  doi: 10.1088/1674-4926/37/1/015002

A wide band, injection-coupled LC quadrature voltage control oscillator is presented. In the proposed circuit, two oscillators are injection locked by coupling their second-order harmonics in anti-phase, forcing the outputs of two oscillators into a quadrature phase state. As the common-mode point sampling the second harmonic frequency, flicker noise of the tail current is suppressed, the phase noise is reduced.The proposed design accomplishes a wide tuning frequency range by a combination of using a 5-bit switch capacitor array (SCA) for discrete tuning in addition to linearly varying AMOS varactors for continuous tuning. The proposed design has been fabricated and verified in a 0.18 μ m TSMC CMOS technology process. The measurement indicates that the quadrature voltage controlled oscillator has a 41.7% tuning range from 3.53 to 5.39 GHz. The measured phase noise is 127.98 dBc/Hz at 1 MHz offset at a 1.8 V supply voltage with a power consumption of 12 mW at a carrier frequency of 4.85 GHz.

A wide band, injection-coupled LC quadrature voltage control oscillator is presented. In the proposed circuit, two oscillators are injection locked by coupling their second-order harmonics in anti-phase, forcing the outputs of two oscillators into a quadrature phase state. As the common-mode point sampling the second harmonic frequency, flicker noise of the tail current is suppressed, the phase noise is reduced.The proposed design accomplishes a wide tuning frequency range by a combination of using a 5-bit switch capacitor array (SCA) for discrete tuning in addition to linearly varying AMOS varactors for continuous tuning. The proposed design has been fabricated and verified in a 0.18 μ m TSMC CMOS technology process. The measurement indicates that the quadrature voltage controlled oscillator has a 41.7% tuning range from 3.53 to 5.39 GHz. The measured phase noise is 127.98 dBc/Hz at 1 MHz offset at a 1.8 V supply voltage with a power consumption of 12 mW at a carrier frequency of 4.85 GHz.
A high-linearity and high-resolution delay line structure with a calibration algorithm in delay-based LINC transmitters
Yue Han, Shushan Qiao, Yong Hei
J. Semicond.  2016, 37(1): 015003  doi: 10.1088/1674-4926/37/1/015003

In order to overcome the bottleneck of low linearity and low resolution, an improved delay line structure is proposed with a calibration algorithm to conquer PVT (process, voltage and temperature) variations for an all-digital design. The chip is implemented in 0.13 μ m CMOS technology. Measurement results show that the proposed structure with the calibration algorithm can evidently improve the linearity and resolution of the delay line. The delay resolution is 2 ps and the root mean square jitter of the delay is 4.71 ps, leading to an error vector magnitude enhancement of 1.32 dB.

In order to overcome the bottleneck of low linearity and low resolution, an improved delay line structure is proposed with a calibration algorithm to conquer PVT (process, voltage and temperature) variations for an all-digital design. The chip is implemented in 0.13 μ m CMOS technology. Measurement results show that the proposed structure with the calibration algorithm can evidently improve the linearity and resolution of the delay line. The delay resolution is 2 ps and the root mean square jitter of the delay is 4.71 ps, leading to an error vector magnitude enhancement of 1.32 dB.
A 10 bit 50 MS/s SAR ADC with partial split capacitor switching scheme in 0.18 μm CMOS
Dong Li, Qiao Meng, Fei Li
J. Semicond.  2016, 37(1): 015004  doi: 10.1088/1674-4926/37/1/015004

This paper presents a 10 bit successive approximation register (SAR) analog-to-digital converter (ADC) in 0.18 μ m 1P6M CMOS technology with a 1.8 V supply voltage. To improve the conversion speed, a partial split capacitor switching scheme is proposed. By reducing the time constant of the bit cycles, the proposed technique shortens the settling time of a capacitive digital-to-analog converter (DAC). In addition, a new SAR control logic is proposed to reduce loop delay to further enhance the conversion speed. At 1.8 V supply voltage and 50 MS/s the SAR ADC achieves a signal-to-noise and distortion ratio (SNDR) of 57.5 dB and spurious-free dynamic range (SFDR) of 69.3 dB. The power consumption is 2.26 mW and the core die area is 0.096 mm2.

This paper presents a 10 bit successive approximation register (SAR) analog-to-digital converter (ADC) in 0.18 μ m 1P6M CMOS technology with a 1.8 V supply voltage. To improve the conversion speed, a partial split capacitor switching scheme is proposed. By reducing the time constant of the bit cycles, the proposed technique shortens the settling time of a capacitive digital-to-analog converter (DAC). In addition, a new SAR control logic is proposed to reduce loop delay to further enhance the conversion speed. At 1.8 V supply voltage and 50 MS/s the SAR ADC achieves a signal-to-noise and distortion ratio (SNDR) of 57.5 dB and spurious-free dynamic range (SFDR) of 69.3 dB. The power consumption is 2.26 mW and the core die area is 0.096 mm2.
A reference voltage in capacitor-resister hybrid SAR ADC for front-end readout system of CZT detector
Wei Liu, Tingcun Wei, Bo Li, Lifeng Yang, Yongcai Hu
J. Semicond.  2016, 37(1): 015005  doi: 10.1088/1674-4926/37/1/015005

An on-chip reference voltage has been designed in capacitor-resister hybrid SAR ADC for CZT detector with the TSMC 0.35 μ m 2P4M CMOS process. The voltage reference has a dynamic load since using variable capacitors and resistances, which need a large driving ability to deal with the current related to the time and sampling rate. Most of the previous articles about the reference for ADC present only the bandgap part for a low temperature coefficient and high PSRR. However, it is not enough and overall, it needs to consider the output driving ability. The proposed voltage reference is realized by the band-gap reference, voltage generator and output buffer. Apart from a low temperature coefficient and high PSRR, it has the features of a large driving ability and low power consumption. What is more, for CZT detectors application in space, a radiation-hardened design has been considered. The measurement results show that the output reference voltage of the buffer is 4.096 V. When the temperature varied from 0 to 80℃, the temperature coefficient is 12.2 ppm/℃. The PSRR was-70 dB@100 kHz. The drive current of the reference can reach up to 10 mA. The area of the voltage reference in the SAR ADC chip is only 449×614 μm2. The total power consumption is only 1.092 mW.

An on-chip reference voltage has been designed in capacitor-resister hybrid SAR ADC for CZT detector with the TSMC 0.35 μ m 2P4M CMOS process. The voltage reference has a dynamic load since using variable capacitors and resistances, which need a large driving ability to deal with the current related to the time and sampling rate. Most of the previous articles about the reference for ADC present only the bandgap part for a low temperature coefficient and high PSRR. However, it is not enough and overall, it needs to consider the output driving ability. The proposed voltage reference is realized by the band-gap reference, voltage generator and output buffer. Apart from a low temperature coefficient and high PSRR, it has the features of a large driving ability and low power consumption. What is more, for CZT detectors application in space, a radiation-hardened design has been considered. The measurement results show that the output reference voltage of the buffer is 4.096 V. When the temperature varied from 0 to 80℃, the temperature coefficient is 12.2 ppm/℃. The PSRR was-70 dB@100 kHz. The drive current of the reference can reach up to 10 mA. The area of the voltage reference in the SAR ADC chip is only 449×614 μm2. The total power consumption is only 1.092 mW.
A low jitter supply regulated charge pump PLL with self-calibration
Min Chen, Yuntao Liu, Zhichao Li, Jingbo Xiao, Jie Chen
J. Semicond.  2016, 37(1): 015006  doi: 10.1088/1674-4926/37/1/015006

This paper describes a ring oscillator based low jitter charge pump PLL with supply regulation and digital calibration. In order to combat power supply noise, a low drop output voltage regulator is implemented. The VCO gain is tunable by using the 4 bit control self-calibration technique. So that the optimal VCO gain is automatically selected and the process/temperature variation is compensated. Fabricated in the 0.13 μ m CMOS process, the PLL achieves a frequency range of 100-400 MHz and occupies a 190×200 μ m2 area. The measured RMS jitter is 5.36 ps at a 400 MHz operating frequency.

This paper describes a ring oscillator based low jitter charge pump PLL with supply regulation and digital calibration. In order to combat power supply noise, a low drop output voltage regulator is implemented. The VCO gain is tunable by using the 4 bit control self-calibration technique. So that the optimal VCO gain is automatically selected and the process/temperature variation is compensated. Fabricated in the 0.13 μ m CMOS process, the PLL achieves a frequency range of 100-400 MHz and occupies a 190×200 μ m2 area. The measured RMS jitter is 5.36 ps at a 400 MHz operating frequency.
PMGA and its application in area and power optimization for ternary FPRM circuit
Pengjun Wang, Kangping Li, Huihong Zhang
J. Semicond.  2016, 37(1): 015007  doi: 10.1088/1674-4926/37/1/015007

Based on the research of population migration algorithms (PMAs), a population migration genetic algorithm (PMGA) is proposed, combining a PMA with a genetic algorithm. A scheme of area and power optimization for a ternary FPRM circuit is proposed by using the PMGA. Firstly, according to the ternary FPRM logic function expression, area and power estimation models are established. Secondly, the PMGA is used to search for the best area and power polarity. Finally, 10 MCNC Benchmark circuits are used to verify the effectiveness of the proposed method. The results show that the ternary FPRM circuits optimized by the PMGA saved 13.33% area and 20.00% power on average than the corresponding FPRM circuits optimized by a whole annealing genetic algorithm.

Based on the research of population migration algorithms (PMAs), a population migration genetic algorithm (PMGA) is proposed, combining a PMA with a genetic algorithm. A scheme of area and power optimization for a ternary FPRM circuit is proposed by using the PMGA. Firstly, according to the ternary FPRM logic function expression, area and power estimation models are established. Secondly, the PMGA is used to search for the best area and power polarity. Finally, 10 MCNC Benchmark circuits are used to verify the effectiveness of the proposed method. The results show that the ternary FPRM circuits optimized by the PMGA saved 13.33% area and 20.00% power on average than the corresponding FPRM circuits optimized by a whole annealing genetic algorithm.
SEMICONDUCTOR TECHNOLOGY
Numerical analysis of the non-equilibrium plasma flow in the gaseous electronics conference reference reactor
Bijie Yang, Ning Zhou, Quanhua Sun
J. Semicond.  2016, 37(1): 016001  doi: 10.1088/1674-4926/37/1/016001

The capacitively coupled plasma in the gaseous electronics conference reference reactor is numerically investigated for argon flow using a non-equilibrium plasma fluid model. The finite rate chemistry is adopted for the chemical non-equilibrium among species including neutral metastable, whereas a two-temperature model is employed to resolve the thermal non-equilibrium between electrons and heavy species. The predicted plasma density agrees very well with experimental data for the validation case. A strong thermal non-equilibrium is observed between heavy particles and electrons due to its low collision frequency, where the heavy species remains near ambient temperature for low pressure and low voltage conditions (0.1 Torr, 100 V). The effects of the operating parameters on the ion flux are also investigated, including the electrode voltage, chamber pressure, and gas flow rate. It is found that the ion flux can be increased by either elevating the electrode voltage or lowering the gas pressure.

The capacitively coupled plasma in the gaseous electronics conference reference reactor is numerically investigated for argon flow using a non-equilibrium plasma fluid model. The finite rate chemistry is adopted for the chemical non-equilibrium among species including neutral metastable, whereas a two-temperature model is employed to resolve the thermal non-equilibrium between electrons and heavy species. The predicted plasma density agrees very well with experimental data for the validation case. A strong thermal non-equilibrium is observed between heavy particles and electrons due to its low collision frequency, where the heavy species remains near ambient temperature for low pressure and low voltage conditions (0.1 Torr, 100 V). The effects of the operating parameters on the ion flux are also investigated, including the electrode voltage, chamber pressure, and gas flow rate. It is found that the ion flux can be increased by either elevating the electrode voltage or lowering the gas pressure.
Investigation of aluminum gate CMP in a novel alkaline solution
Cuiyue Feng, Yuling Liu, Ming Sun, Wenqian Zhang, Jin Zhang, Shuai Wang
J. Semicond.  2016, 37(1): 016002  doi: 10.1088/1674-4926/37/1/016002

Beyond 45 nm, due to the superior CMP performance requirements with the metal gate of aluminum in the advanced CMOS process, a novel alkaline slurry for an aluminum gate CMP with poly-amine alkali slurry is investigated. The aluminum gate CMP under alkaline conditions has two steps:stock polishing and fine polishing. A controllable removal rate, the uniformity of aluminum gate and low corrosion are the key challenges for the alkaline polishing slurry of the aluminum gate CMP. This work utilizes the complexation-soluble function of FA/O Ⅱ and the preference adsorption mechanism of FA/O Ⅰ nonionic surfactant to improve the uniformity of the surface chemistry function with the electrochemical corrosion research, such as OCP-TIME curves, Tafel curves and AC impedance. The result is that the stock polishing slurry (with SiO2 abrasive) contains 1 wt.% H2O2 ,0.5 wt.% FA/O Ⅱ and 1.0 wt.% FA/O Ⅰ nonionic surfactant. For a fine polishing process, 1.5 wt.% H2O2 , 0.4 wt.% FA/O Ⅱ and 2.0 wt.% FA/O Ⅰ nonionic surfactant are added. The polishing experiments show that the removal rates are 3000±50 Å/min and 1600±60 Å/min, respectively. The surface roughnesses are 2.05±0.128 nm and 1.59±0.081 nm, respectively. A combination of the functions of FA/O Ⅱ and FA/O Ⅰ nonionic surfactant obtains a controllable removal rate and a better surface roughness in alkaline solution.

Beyond 45 nm, due to the superior CMP performance requirements with the metal gate of aluminum in the advanced CMOS process, a novel alkaline slurry for an aluminum gate CMP with poly-amine alkali slurry is investigated. The aluminum gate CMP under alkaline conditions has two steps:stock polishing and fine polishing. A controllable removal rate, the uniformity of aluminum gate and low corrosion are the key challenges for the alkaline polishing slurry of the aluminum gate CMP. This work utilizes the complexation-soluble function of FA/O Ⅱ and the preference adsorption mechanism of FA/O Ⅰ nonionic surfactant to improve the uniformity of the surface chemistry function with the electrochemical corrosion research, such as OCP-TIME curves, Tafel curves and AC impedance. The result is that the stock polishing slurry (with SiO2 abrasive) contains 1 wt.% H2O2 ,0.5 wt.% FA/O Ⅱ and 1.0 wt.% FA/O Ⅰ nonionic surfactant. For a fine polishing process, 1.5 wt.% H2O2 , 0.4 wt.% FA/O Ⅱ and 2.0 wt.% FA/O Ⅰ nonionic surfactant are added. The polishing experiments show that the removal rates are 3000±50 Å/min and 1600±60 Å/min, respectively. The surface roughnesses are 2.05±0.128 nm and 1.59±0.081 nm, respectively. A combination of the functions of FA/O Ⅱ and FA/O Ⅰ nonionic surfactant obtains a controllable removal rate and a better surface roughness in alkaline solution.