Mitigation of parasitic contact resistances require processing techniques that yield highly doped junctions such as ultra-shallow ion implantation. The depth of these junctions is so shallow that the role of the native oxide warrants further study. Investigation of an in-situ native oxide clean prior to amorphizing the surface of Si with 3x10(14) cm(-2) Si+ ions at an energy of 5 keV has been studied. HRTEM confirmed that, when the native oxide is present under these conditions, the amorphization process is incomplete, leaving near-surface crystallites embedded in the amorphous layer. However, when the oxide is removed in-situ, the samples had a continuous amorphous layer with no crystallites for the exact same implant conditions. This is the opposite of what is expected from damage simulations. Experiments with a plasma treatment and regrowth of the oxide prior to implantation show the complete amorphization process still occurs. It is proposed that hydrogen incorporation during plasma treatment may be reducing I-V recombination during the dynamic annealing stage of the implant leading to more effective amorphization. SRIM and FLOOXS simulations will be shown that support this model.

Polytypism in SiC has created interest and opportunity for device heterostructures and bandgap engineering in power electronic applications. As each SiC polytype possesses a different bandgap, electron mobility, and degree of anisotropy, unique interfaces can be created without changing its chemical composition. The 4H polytype is commonly used, but the 3C polytype offers high surface electron mobility with isotropic properties as the only cubic polytype. This has driven research on heteroepitaxy with limited success in traditional chemical vapor deposition chambers. Discussion on polytype control and stability has been restricted to bulk and epitaxial crystal growth, despite numerous reports of polytypic transformations occurring during other processing steps. This study revealed the polytypic transformation of 4H-SiC to 3C-SiC after high temperature annealing using high resolution cross-sectional transmission electron microscopy (TEM) and scanning transmission electron microscopy (STEM). Above 1750 °C, the surface significantly roughened under a reduced pressure of Ar, whereas surface planarity was maintained under Ar atmospheric pressure. The formation of 3C-SiC islands occurred adjacent to large surface pits through an epitaxial growth process for the reduced pressure condition only. Loss of SiC stoichiometry at the surface with Si enrichment and availability of on-axis terraces enabled 3C nucleation. 3C-SiC growth was retarded using a protective carbon cap (C-cap) where defect-free single crystal 3C-SiC has a coherent interface with the 4H-SiC substrate underneath. These findings demonstrate that the 3C polytype can be stable at high temperatures, encouraging the need for a better understanding of polytype stability and control.

Reducing ion beam damage from the focused ion beam (FIB) during fabrication of cross sections is a well-known challenge for materials characterization, especially cross sectional characterization of nanostructures. To address this, a new method has been developed for cross section fabrication enabling high resolution transmission electron microscopy (TEM) analysis of 3-D nanostructures free of surrounding material and free of damage detectable by TEM analysis. Before FIB processing, nanopillars are encapsulated in a sacrificial oxide which acts as a protective layer during FIB milling. The cross sectional TEM lamella containing the nanopillars is then mounted and thinned with some modifications to conventional FIB sample preparation that provide stability for the lamella during the following wet-chemical dip etch. The wet-chemical etch of the TEM lamella removes the sacrificial oxide layer, freeing the nanopillars from any material that would obscure TEM imaging. Both high resolution TEM and aberration corrected scanning TEM images of Si/SiGe pillars with diameters down to 30 nm demonstrate the successful application of this approach.

Helium implantation into silicon may offer the ability to alter the point-defect population of silicon in favor of vacancies. Helium can stabilize small vacancy clusters into HemVn complexes below the threshold for cavity formation. This work characterizes the effects of helium implantation in a unique range that has not been studied before. Transmission electron microscopy shows that cavity formation is suppressed at lower energies for elevated temperature implantation. However, extended defects are visible at low energies for a range of doses. Thermal helium desorption spectrometry shows that ~98% of the helium dose dynamically desorbs during the implant and confirms the presence of small sub-microscopic HemVn clusters. Finally, positron Doppler broadening spectroscopy indicates the presence of additional vacancy clusters.

There is a growing interest in using high dose helium implants to alter point defect populations in silicon. Previous reports have shown that the interaction between helium and vacancies leads to the formation of cavities for medium energy (e.g., 20–100 keV) implants. However, the role of certain factors, such as the proximity of the surface, the damage created by the implant, and the effect of the implant temperature, is not well understood for low energy implants. This study explored a new regime of ultralow energy, elevated temperature implants in order to offer an insight into the effect of these parameters. Transmission electron microscopy (TEM) showed that cavity formation was avoided for 0.5 keV, 450 °C implants up to a dose of 8 × 1016 cm−2. However, extended defects in the form of {311} ribbon-like defects and stacking faults were observed. Quantitative TEM showed that the number of interstitials in these defects was less than 0.2% of the implant dose. In addition, thermal helium desorption spectrometry suggested that only 2% of the implanted He dose was retained in interstitial He and HemVn complexes. A first-order dissociation kinetic model was applied to assess desorption from HemVn, which closely matched energies predicted by density functional theory. This population of excess vacancies and excess interstitials was possibly formed because of incomplete Frenkel pair recombination. Raman spectroscopy showed that the stress from the implant was dominated by the stress from the interstitial-type defects. The evolution of the stress and defects was also explored as a function of post-implant annealing.

The oxidation of silicon is known to inject interstitials, and the presence of silicon–germanium (SiGe) alloys at the Si/SiO2 interface during oxidation is known to suppress the injection of silicon self-interstitials. This study uses a layer of implantation-induced dislocation loops to measure interstitial injection as a function of SiGe layer thickness. The loops were introduced by a 50 keV 2 × 1014 cm−2 P+ room-temperature implantation and thermal annealing. Germanium was subsequently introduced via a second implant at 3 keV Ge+ over a range of doses between 1.7 × 1014 cm−2 and 1.4 × 1015 cm−2. Results show that upon oxidizing at 850 °C for 3 h or 900 °C for 70 min to condense the germanium at the Si/SiO2 interface, where if forms a Si0.5Ge0.5 alloy. Upon subsequent oxidations of 850 °C for 6 h or 900 °C for 2 h, partial suppression of interstitial injection can be observed for sub-monolayer doses of germanium, and more than three monolayers of Si0.5Ge0.5 (1.4 × 1015 cm−2) are necessary to suppress interstitial injection below the detection limit during oxidation. These results show that low-energy implantation of germanium can be used to eliminate or modulate injection of oxidation-induced interstitials.

Molybdenum disulfide (MoS2) is a promising potential replacement for Si in future microelectronic devices. Integration in electronic devices will likely involve the growth or transfer of large-area MoS2 films onto substrates and subsequent isolation of devices. In this paper, the effect of ion implantation on the electrical properties of MoS2 is reported. Large-area ∼4 layer MoS2 films were implanted by low energy phosphorus plasma at biases of 100, 200, and 300 V and a dose of 1 × 1014 cm−2. Electrical measurements using patterned Ni/Au contacts show that after implantation, independent of bias, there is greater than a 104 increase in resistivity. TEM and Raman spectroscopy suggest that the film is crystalline prior to and after ion implantation and annealing and that there is no measurable sputtering following implantation. This suggests that the increase in resistivity is likely the result of radiation damage in the MoS2. The thermal stability of the increase in electrical resistivity was assessed by a series of 15 min anneals beginning at 325 °C in a sulfur overpressure and progressing up to 525 °C under an Al2O3 ALD cap. The resistivity increase remained unchanged after annealing. These results suggest that implant isolation could provide a preferable alternative to reactive ion etching or chemical etching for electrical isolation of MoS2.

This letter demonstrates vertical GaN junction barrier Schottky (JBS) rectifiers fabricated with novel ion implantation techniques. We used two different methods to form the lateral p-n grids below the Schottky contact: 1) Mg implantation into n-GaN to form p-wells and 2) Si implantation into p-GaN to formn-wells. Specific differential ON-resistances (RON) of 1.5-2.5 m Omega.cm(2) and 7-9 m Omega.cm(2) were obtained in the Mg-implanted and Si-implanted JBS rectifiers, respectively. A breakdown voltage of 500-600 V was achieved in both devices, with a leakage current at high reverse biases at least 100-fold lower than conventional vertical GaN Schottky barrier diodes. The impact of n-well and p-well widths on the R-ON and BV was investigated. Fast switching capability was also demonstrated. This letter shows the feasibility of forming patterned p-n junctions by novel ion implantation techniques, to enable high-performance vertical GaN power devices.

This Letter reports on the unusual diffusion behavior of Ge during oxidation of a multilayer Si/SiGe fin. It is observed that oxidation surprisingly results in the formation of vertically stacked Si nanowires encapsulated in defect free epitaxial strained SixGe1-x. High angle annular dark field scanning transmission electron microscopy (HAADF-STEM) shows that extremely enhanced diffusion of Ge occurs along the vertical Si/SiO2 oxidizing interface and is responsible for the encapsulation process. Further oxidation fully encapsulates the Si layers in defect free single crystal SixGe1-x (x up to 0.53), which results in Si nanowires with up to -2% strain. Atom probe tomography reconstructions demonstrate that the resultant nanowires run the length of the fin. We found that the oxidation temperature plays a significant role in the formation of the Si nanowires. In the process range of 800-900 °C, pure strained and rounded Si nanowires down to 2 nm in diameter can be fabricated. At lower temperatures, the Ge diffusion along the oxidizing Si/SiO2 interface is slow, and rounding of the nanowire does not occur, while at higher temperatures, the diffusivity of Ge into Si is sufficient to result in dilution of the pure Si nanowire with Ge. The use of highly selective etchants to remove the SiGe could provide a new pathway for the creation of highly controlled vertically stacked nanowires for gate all around transistors.

MoS2 is a promising two-dimensional material that is being considered as a replacement for Si in sub-5 nm device technologies. Integration could include the growth or transfer of large areas of MoS2 films onto Si wafers. Subsequent isolation of devices fabricated on MoS2 films is an open research area that has received little attention. Although reactive ion etching or chemical etching can be used to create isolated channels, these techniques are problematic due to the weak van der Waals interaction between the two-dimensional films and their substrates. Subsequent wet processing during lithography can cause delamination from the substrate, resulting in mechanical damage to the films or even displacement of the channels. An alternative to physical isolation of MoS2 was the focus of this study. It is known that, in many semiconductors, radiation damage can be used for isolation. In this study, large-area 3-5 layer MoS2 films were grown on sapphire and subsequently transferred onto SiO2/Si wafers. The MoS2 was then exposed to low energy phosphorus plasma implantation at biases of 100, 200, and 300 V and a dose of 1 x 1014 cm-2. Electrical measurements using patterned Ni/Au contacts show that after implantation, independent of bias, there is a 105 increase in resistivity and a similar increase in specific contact resistivity of the MoS2. TEM shows that the film is still crystalline and there is no measurable etching of the films after implantation, suggesting that the increase in resistivity is likely the result of radiation damage in the MoS2. The thermal stability of the increase in electrical resistivity was assessed by a series of 15 minute anneals beginning at 325°C in a sulfur overpressure and progressing up to 525°C under an Al2O3 ALD cap. The resistivity increase remained unchanged after annealing, suggesting that this is a stable alternative to physical isolation in MoS2 devices.

The presence of Silicon-Germanium (SiGe) alloys at the Si/SiO2 interface during oxidation is known to suppress the injection of silicon self-interstitials that normally accompanies silicon oxidation and lead to observed effects such as Oxidation Enhanced Diffusion (OED) and stacking fault growth. This study uses a layer of implantation induced dislocation loops to measure interstitial injection as a function of SiGe layer thickness. The loops were introduced by implanting phosphorus and thermal annealing, and Germanium was subsequently introduced via a second implant at 3 keV over a range of doses between 1.7 x10(14) cm(-2) and 1.4 x10(15) cm(-2). Results show that partial suppression of interstitial injection can be observed for sub-monolayer doses of germanium, and that more than three monolayers of SiGe are necessary to fully suppress interstitial injection below our detection limit during oxidation. They further show that low energy implantation of germanium opens up possibilities to eliminate or modulate injection of interstitials during thermal processing of future devices.

Studies of implant activation in InAs have not been reported presumably because of challenges associated with junction leakage. The activation of 20 keV, Si+ implants into lightly doped (001) p-type bulk InAs performed at 100 °C as a function of annealing time and temperature was measured via Raman scattering. Peak shift of the L+ coupled phonon-plasmon mode after annealing at 700 °C shows that active n-type doping levels ≈5 × 1019 cm−3 are possible for ion implanted Si in InAs. These values are comparable to the highest reported active carrier concentrations of 8–12 × 1019 cm−3 for growth-doped n-InAs. Raman scattering is shown to be a viable, non-contact technique to measure active carrier concentration in instances where contact–based methods such as Hall effect produce erroneous measurements or junction leakage prevents the measurement of shallow n+ layers, which cannot be effectively isolated from the bulk.

The evolution of implant damage in InGaAs is studied for electrically active Si+ and isoelectronic P+ implants. Extrinsic loops formed by excess interstitials are shown to be less stable upon annealing for n-type Si+ implants relative to isoelectronic P+ implants. Damage created by P+ implants into heavily n-doped InGaAs is also shown to be less stable than damage created by P+ implants into unintentionally doped InGaAs indicating that the background doping concentration can significantly effect the evolution of implant damage upon annealing. Previous results have suggested that the electrical activation and diffusion behavior of n-type dopants, like Si in InGaAs, may be strongly influenced by vacancy concentration. TEM results in this study also suggest that heavy n-type doping in InGaAs results in the formation of a large population of vacancy defects that enhance the dissolution or inhibit formation of interstitial loops. (c) 2015 The Electrochemical Society. All rights reserved.

An overview of various processing and dopant considerations for the creation of heavily-doped n-InGaAs is presented. A large body of experimental evidence and theoretical prediction point to dopant vacancy-complexing as the limiting mechanism for electrical activation in heavily Si doped InGaAs and GaAs. Dopant incorporation techniques which require thermal treatment steps to move dopants onto lattice sites like ion implantation and monolayer doping exhibit stable activation up to a limit of approximate to 1.5 x 10(19) cm(-3). Growth-based dopant incorporation methods have shown much higher (5 x 10(19) cm(-3)) active concentrations but these activate concentrations are shown in multiple studies to be metastable. Other device specific process-flow constraints with respect to modern CMOS devices which may make some means of dopant incorporation method, or species selection more appropriate for a given application are also discussed. (C) The Author(s) 2016. Published by ECS. All rights reserved.

The use of ultra low energy ion implantation is investigated as a doping method for MoS2. 200 eV Cl and Ar implants at doses between 1 x 10(13) and 1 x 10(15) cm(-2) were introduced into exfoliated MoS2 flakes. XPS results for Cl implants show a decrease in core peak binding energies for Mo3d and S2p with increasing dose, implying a p-doping effect. Implantation of MoS2 device channel regions is shown to reduce the channel's conductivity. However, isolated implantation of the contact region with low doses (1 x 10(13) cm(-2)) of Cl and Ar are shown to improve output characteristics by linearizing the I-DS-V-DS curves and by increasing current through the device. Cl was shown to be more effective than Ar at increasing the current, implying there is a potential chemical effect as well as damage effect. For higher doses (>= 1 x 10(14) cm(-2)), the current through the device is reduced with increasing dose for both implant species. This report presents the as-implanted, unannealed results. Post-implantation anneals may be necessary to activate the dopants and fully realize the potential of this doping method. (C) The Author(s) 2016. Published by ECS. All rights reserved.

Continued effort has been placed on maximizing activation while controlling the diffusion of silicon doping in InGaAs for present and future complementary metal-oxide semiconductor devices. In order to explore the diffusion and activation behavior, Si marker layers were grown in InGaAs on InP by molecular beam epitaxy. The nature of Si diffusion was explored using a series of isoelectronic implants to introduce excess point defects near the layer. It was observed that excess interstitials reduce the Si diffusion consistent with a vacancy-driven diffusion mechanism. A diffusion and activation model implemented in the Florida object oriented process simulator has been developed to predict silicon diffusion behavior over a variety of temperatures and times. Using current and previous experimental data and complimentary density functional theory results, the diffusion model employs the SiIII–VIII pair as the primary mechanism for silicon diffusion in InGaAs.

There is a renewed interest in integrating high mobility III-V channel materials into sub 10 nm nMOS devices but the continued scaling of devices has resulted in the need to create contacts and source/drains with ultra low contact resistivities in order to reduce current losses. This study investigates thermal stability of Si dopants incorporated via MBE growth and ion implantation as potential methods to create heavily-doped, low resistance source/drain regions in III-V channel devices. For this study, the electrical activation and diffusion of Si active layers in In 0.53 Ga 0.47 As formed by a 10 keV, 5×10 14 cm -2 Si implant and MBE growth doping with a peak Si concentration of 7×10 19 cm -3 were investigated as a function of post growth and post-implant isochronal annealing. While most previous studies conclude that MBE doping can achieve higher active Si concentrations than ion implantation, the results of this study show conclusively that electrically active Si concentrations above 1.4×10 19 cm -3 formed by MBE doping are prone to deactivation upon thermal treatment after growth whereas ion-implantation shows no metastable activation behavior. Significant Si deactivation in MBE doped substrates is shown to occur before the onset of Si diffusion whereas saturated activation in ion implanted substrates does not occur until diffusion is observed. Upon annealing at sufficiently high temperatures to cause Si diffusion, the electrical activation of Si in MBE doped substrates and ion implanted Si are shown to converge to a stable activation limit of 1.4×10 19 cm -3 . The common activation limit upon Si profile motion for both Ion implanted and growth doped Si active layers after thermal annealing at 750°C suggests that the maximum stable electrical activation of Si is an intrinsic property of In 0.53 Ga 0.47 As. Si diffusivity has also been calculated from SIMS results in both ion-implanted substrates and growth-doped substrate and Si diffusion in MBE doped substrates was observed to be nearly three times as fast as ion implanted substrates presumably due to the observed concentration dependent diffusion effects. The mechanism of Si diffusion and its relation to the observed concentration dependent diffusion and electrical activation behavior in both materials are also discussed. Figure 1

Elevated temperature, nonamorphizing implants of Si+, and a second co-implant of either Al+, P+, or S+ at varying doses were performed into In0.53Ga0.47As to observe the effect that individual co-implant species had on the activation and diffusion of Si doping after postimplantation annealing. It was found that Al, P, and S co-implantation all resulted in a common activation limit of 1.7 × 1019 cm−3 for annealing treatments that resulted in Si profile motion. This is the same activation level observed for Si+ implants alone. The results of this work indicate that co-implantation of group V or VI species is an ineffective means for increasing donor activation of n-type dopants above 1.7 × 1019 cm−3 in InGaAs. The S+ co-implants did not show an additive effect in the total doping despite exhibiting significant activation when implanted alone. The observed n-type active carrier concentration limits appear to be the result of a crystalline thermodynamic limit rather than dopant specific limits.