Phosphorus In Silicon Research Articles

Single-Electron Structures Based on Solitary Dopant Atoms of Arsenic, Phosphorus, Gold, and Potassium in Silicon

Here we present CMOS compatible fabrication methods and the results of an experimental study of single-atom single-electron transistors made from silicon on insulator and based on various dopant atoms. Transistors with channels doped with arsenic (As), phosphorus (P), gold (Au) and potassium (K) atoms were fabricated and studied. Two methods for fabricating of experimental transistor structures are presented. The first method (As, P transistors) used a inhomogeneously doped in depth silicon layer and controlled reduction of the size of the transistor channel in several cycles of isotropic reactive-ion etching. The second method (Au and K transistors) used an undoped silicon layer and the subsequent implantation of dopant atoms into a preformed transistor channel. Dopant electron and hole levels of Au and K atoms in silicon are located near the middle of the silicon band gap, which provides a small effective size of the dopant charge center and, as a result, a high value of the charge energy and operating temperature of the transistor compared to the traditional dopants (P, As, Sb, B). The values of the charge energy of the Au and K transistors, which were estimated from the measurements (Ec ≥ 150 meV), are much higher than those of the As and P transistors (Ec < 30 meV). Important advantages of the proposed methods are: controlled implantation of various impurities and possibility to combine etching and implantation cycles during sample preparation.

Distribution of Phosphorus in n-Type Multicrystalline Silicon Produced by Directional Solidification

n-type multicrystalline silicon ingot with a doping element of phosphorus was produced by directional solidification under industrial conditions. The evaporation and distribution of phosphorus were studied by content analysis in different positions. The results show that it is mainly controlled by melting and solidification processes. During the melting process, the content of phosphorus reduced from 16.6 and 6.7 ppmw to 11.1 and 6.0 ppmw under different melting times, respectively. The evaporation of phosphorus is mainly affected by the temperature. During the solidification process, the segregation of phosphorus was promoted by evaporation. A formula similar to the Scheil equation was established to calculate the longitudinal distribution of phosphorus in silicon. The results show that its longitudinal distribution is influenced by the evaporation coefficient, the solidification rate, and the temperature gradient. The effective segregation coefficient of phosphorus in ingots a and b was 0.31 and 0.33, respectively, which could be obtained from the fitting result. During the whole process, the phosphorus removal efficiency in ingots a and b was 60.8% and 50.5%, respectively. Its horizontal distribution is related to the morphology of the solid–liquid interface.

Effects of oxygen-inserted layers and oxide capping layer on dopant activation for the formation of ultrashallow p-n junctions in silicon

The effects of oxygen-inserted (OI) layers and a low-temperature-deposited oxide (LTO) capping layer on rapid thermal activation of ultrashallow implanted boron, phosphorus, and arsenic atoms in silicon (Si) are investigated using sheet resistance (Rsh) measurements, secondary ion mass spectrometry analyses, and technology computer-aided design simulations. The experimental findings suggest that the electrical activation of dopants in Si is not significantly affected by the presence of OI layers so that they can be effective for achieving lower Rsh along with shallower junction depth, thanks to reduced dopant loss and diffusion during thermal annealing. On the other hand, an LTO capping layer is found to result in larger Rsh associated with the lower peak active dopant concentration as a result of dopant segregation and/or reduced uphill diffusion. The presence of OI layers is found to mitigate these detrimental effects.

Experimental Study on Phosphorus Removal from Ferrosilicon Alloys under Electromagnetic Levitation

Along with the growing demand of green energy, the study of solar energy continues to deepen and the production of solar cells is increasing ceaselessly. However, the cost of solar power has become an economic constraint to the large-scale production of solar power equipment. In order to meet the demand of solar power generation, metallurgical grade silicon may be used to replace solar grade silicon to generate electricity. The challenge is that metallurgical grade silicon contains phosphorus impurities. In this paper, electromagnetic levitation is recommended to remove the phosphorus in metallurgical grade silicon. The results show that when the mass of levitated droplets is 0.59-0.64g, the input current is 300-400A, the frequency is 280kHz, and the input power range is 3000-4500W, the removal rate of phosphorus in the ferrosilicon sample is as high as 41%. Therefore, it is feasible to use metallurgical grade silicon to replace solar grade silicon.

Effects of oxygen-inserted layers on diffusion of boron, phosphorus, and arsenic in silicon for ultra-shallow junction formation

The effects of oxygen-inserted (OI) layers on the diffusion of boron (B), phosphorus (P), and arsenic (As) in silicon (Si) are investigated, for ultra-shallow junction formation by high-dose ion implantation followed by rapid thermal annealing. The projected range (Rp) of the implanted dopants is shallower than the depth of the OI layers. Secondary ion mass spectrometry is used to compare the dopant profiles in silicon samples that have OI layers against the dopant profiles in control samples that do not have OI layers. Diffusion is found to be substantially retarded by the OI layers for B and P, and less for As, providing shallower junction depth. The experimental results suggest that the OI layers serve to block the diffusion of Si self-interstitials and thereby effectively reduce interstitial-aided diffusion beyond the depth of the OI layers. The OI layers also help to retain more dopants within the Si, which technology computer-aided design simulations indicate to be beneficial for achieving shallower junctions with lower sheet resistance to enable further miniaturization of planar metal-oxide-semiconductor field-effect transistors for improved integrated-circuit performance and cost per function.

Enhanced separation of phosphorus from metallurgical grade silicon by CaAl2Si2 phase reconstruction

Abstract To find a high-efficiency approach for removing phosphorus from metallurgical grade silicon, a novel method of silicon-aluminum-calcium alloy refining was investigated through phase reconstruction to form CaAl 2 Si 2 phase. The distribution of phosphorus between purified solid silicon and liquid silicon-aluminum-calcium alloy was determined to investigate the possibility of removing phosphorus from the silicon by alloy refining with silicon-aluminum-calcium melt. It was found that the CaAl 2 Si 2 phase had a high affinity for phosphorus resulting in dissolving a large quantity of phosphorus. Decreasing the amount of getter materials was not good for the phosphorus removal from the silicon, while adopting twice alloy refining could strengthen the removal with obtaining as high as 99.99 wt.% purity silicon. In addition, the purity of refined silicon depends on the quenching temperature and the cooling rate. Moreover, to evaluate efficiency of phosphorus removal from silicon by silicon-aluminum-calcium alloy refining, the thermodynamic properties of phosphorus in solid silicon and liquid alloy phase at 1273-1473 K were further evaluated by the chemical equilibrium method.

Removal of Phosphorus in Silicon by the Formation of CaAl2Si2 Phase at the Solidification Interface

To fully understand the role of CaAl2Si2 phase in concentrating the non-metallic impurity phosphorus, an experiment of directional solidification of Al-70 at. pct Si alloy with extreme small lowering rate 0.05 mm min−1 was carried out. With good dynamic condition for the diffusion of impurity (Ca, Al, P) from silicon to the S/L interface, the CaAl2Si2 phase with 0.6-0.7 at. pct P was successfully observed by Electron Probe Micro Analyzer (EPMA), and its distribution character was originally presented. This impurity phase was widely detected in the refined sample but only at the interface of silicon crystal and Al-Si alloy which contributed to the deep removal of impurity P. The formation mechanism of CaAl2Si2-P phase was thus explored, in which the microsegregation and concentration of element P, Ca, Al in front of S/L interface were crucial. After acid leaching, the P content decreased from the original 23 ppm to below 5 ppm. Compared with normal solidification, a 16 pct higher removal efficiency of P was obtained in this study.

Thermodynamics of Phosphorus in Solvent Refining of Silicon Using Ferrosilicon Alloys

The thermodynamics of phosphorus distribution between solid silicon and iron-silicon melt was studied to examine the degree of phosphorus removal from silicon by solvent refining with ferrosilicon alloys. The experiments were performed on silicon-iron-phosphorus alloys with ~80 wt pct silicon and ~20 wt pct iron. A phosphorus distribution coefficient, which is defined as the ratio of the mole fraction of phosphorus in solid to that of liquid is as follows: 0.22 ± 0.02 [1583 K (1310 °C)], 0.29 ± 0.02 [1533 K (1260 °C)], and 0.33 ± 0.02 [1483 K (1210 °C)]. The corresponding removal percentages of phosphorus were 86 pct [1583 K (1310 °C)], 75 pct [1533 K (1260 °C)], and 67 pct [1483 K (1210 °C)]. The average phosphorus content of the refined silicon in the current process would be more than two times less than that of the conventional solidification refining techniques. The values of interaction coefficient of phosphorus on iron \( (\varepsilon_{\text{Fe}}^{P} ) \) at different temperatures were obtained as −3460 ± 155 [1583 K (1310 °C)], −3595 ± 159 [1533 K (1260 °C)], and −3694 ± 119 [1483 K (1210 °C)]. The self-interaction parameters of phosphorus \( \left( {\varepsilon_{P}^{P} } \right) \)at different temperatures are as follows: 68 ± 4 [1583 K (1310 °C)], 78 ± 10 [1533 K (1260 °C)], and 103 ± 19 [(1483 K 1210 °C)]. The calculated values for the distribution coefficients of phosphorus at infinite dilution are 0.22 ± 0.00 [1583 K (1310 °C)], 0.30 ± 0.00 [1533 K (1260 °C)], and 0.34 ± 0.00 [1483 K (1210 °C)]. Considering the solid (red phosphorus) standard state for solid silicon, the activity coefficient of phosphorus in solid silicon is estimated as \( {{ln\gamma }}_{\text{P \, in \, solid \, Si}}^{^\circ } = - 17395\left( {\frac{1}{\text{T}}} \right) + 10 \).

Instrumental determination of phosphorus in silicon for photovoltaics by β spectroscopy: a new approach

In this study, we report on the investigation of a β-γ-anticoincidence set up for the determination of phosphorus in silicon for photovoltaics by Instrumental Neutron Activation Analysis. For the suppression of disturbing β/γ radiation emitted by impurities, a plastic scintillator for β-detection is surrounded by a well type NaI(Tl) γ-detector. A suppression of 40 % for the impurities 60Co and 124Sb could be achieved. The limit of detection was determined to be less than 0.1 ppm. In order to correct different β absorption, dedicated Geant4 simulations were used. With first quantitative measurements the phosphorus concentration in silicon could be determined.

Evaporation Behavior of Phosphorus from Metallurgical Grade Silicon via Calcium-Based Slag Treatment and Hydrochloric Acid Leaching

Phosphorus removal from metallurgical grade silicon by CaO-SiO2-CaCl2 slag treatment, HCl leaching, and vacuum refining was investigated. The effect of different compositions of slag was evaluated. The calcium concentration in slag-treated silicon increased with increasing CaO/SiO2 mass ratio of slag, decreasing the evaporation efficiency of phosphorus in molten silicon. The total phosphorus removal efficiency changed from 93.0% to 98.3% when the slag-treated silicon was treated with HCl before vacuum refining. The final concentration of phosphorus in silicon was 0.43 ppmw. This is because phosphorus was removed from metallurgical-grade silicon as follows: Phosphorus reacts with slag at the silicon/slag interface and forms Ca3(PO4)2 and Ca3P2, most of which diffuse from the interface to the slag phase. The remaining Ca3(PO4)2 and Ca3P2 reduce the phosphorus removal efficiency by altering the activity coefficient of phosphorus in molten silicon. HCl leaching enhanced the phosphorus removal efficiency by removing the remaining Ca3(PO4)2 and Ca3P2. Therefore, the mass transfer of phosphorus from metallurgical-grade silicon was accelerated.

Terahertz lasers based on intracentre transitions of group V donors in uniaxially deformed silicon

This paper presents a brief overview of available experimental data on the characteristics of stimulated terahertz emission from optically excited neutral group V donors (phosphorus, antimony, arsenic and bismuth) in crystalline silicon subjected to uniaxial compressive strain along the [100] axis. Strain is shown to have a significant effect on the characteristics in question. Optimal strain depends on the dopant and may reduce the threshold pump intensity and improve lasing efficiency. We discuss possible mechanisms behind this effect and estimate the limiting output emission parameters.

Efficient dynamic nuclear polarization of phosphorus in silicon in strong magnetic fields and at low temperatures

Efficient manipulation of nuclear spins is important for utilizing them as qubits for quantum computing. In this work we report record high polarizations of ${}^{31}\mathrm{P}$ and $^{29}\mathrm{Si}$ nuclear spins in P-doped silicon in a strong magnetic field (4.6 T) and at temperatures below 1 K. We reached ${}^{31}\mathrm{P}$ nuclear polarization values exceeding $98%$ after 20 min of pumping the high-field electron spin resonance (ESR) line with a very small microwave power of 0.4 $\ensuremath{\mu}\mathrm{W}$. We evaluate that the ratio of the hyperfine-state populations increases by three orders of magnitude after 2 hours of pumping, and an extremely pure nuclear spin state can be created, with less than 0.01 ppb impurities. A negative dynamic nuclear polarization has been observed by pumping the low-field ESR line of ${}^{31}\mathrm{P}$ followed by the flip-flip cross relaxation, the transition which is fully forbidden for isolated donors. We estimate that while pumping the ESR transitions of ${}^{31}\mathrm{P}$ also the nuclei of $^{29}\mathrm{Si}$ get polarized, and polarization exceeding $60%$ has been obtained. We performed measurements of relaxation rates of flip-flop and flip-flip transitions which turned out to be nearly temperature independent. Temperature dependence of the ${}^{31}\mathrm{P}$ nuclear relaxation was studied down to 0.75 K, below which the relaxation time became too long to be measured. We found that the polarization evolution under pumping and during relaxation deviates substantially from a simple exponential function of time. We suggest that the nonexponential polarization dynamics of ${}^{31}\mathrm{P}$ donors is mediated by the orientation of $^{29}\mathrm{Si}$ nuclei, which affect the transition probabilities of the forbidden cross-relaxation processes.

Optimization of a plasma immersion ion implantation process for shallow junctions in silicon

A plasma immersion ion implantation (PIII) process has been developed for realizing shallow doping profiles of phosphorus and boron in silicon using an in-house built dual chamber cluster tool. High Si etch rates observed in a 5% PH3 in H2 plasma have been ascribed to high concentration of H(α) radicals. Therefore, subsequent work was carried out with 5% PH3 in He, leading to much smaller etch rates. By optical emission spectroscopy, the radical species H(α), PH*2, and PH* have been identified. The concentration of all three species increased with pressure. Also, ion concentrations increased with pressure as evidenced by Langmuir data, with a maximum occurring at 0.12 mbar. The duty cycle of pulsed DC bias has a significant bearing on both the implantation and the etching process as it controls the leakage of positive charge collected at the surface of the silicon wafer during pulse on-time generated primarily due to secondary electron emission. The P implant process was optimized for a duty cycle of 10% or less at a pressure of 0.12 mbar with implant times as low as 30 s. Secondary ion mass spectroscopy showed a P dopant depth of 145 nm after rapid thermal annealing (RTA) at 950 °C for 5 s, resulting in a sheet resistance of 77 Ω/◻. Si n+/p diodes fabricated with phosphorus implantation using optimized PIII and RTA conditions exhibit Jon/Joff &amp;gt; 106 with an ideality factor of nearly 1.2. Using similar conditions, shallow doping profiles of B in silicon have also been realized.

Cat-doping: Novel method for phosphorus and boron shallow doping in crystalline silicon at 80 °C

Phosphorus (P) or boron (B) atoms can be doped at temperatures as low as 80 to 350 °C, when crystalline silicon (c-Si) is exposed only for a few minutes to species generated by catalytic cracking reaction of phosphine (PH3) or diborane (B2H6) with heated tungsten (W) catalyzer. This paper is to investigate systematically this novel doping method, “Cat-doping”, in detail. The electrical properties of P or B doped layers are studied by the Van der Pauw method based on the Hall effects measurement. The profiles of P or B atoms in c-Si are observed by secondary ion mass spectrometry mainly from back side of samples to eliminate knock-on effects. It is confirmed that the surface of p-type c-Si is converted to n-type by P Cat-doping at 80 °C, and similarly, that of n-type c-Si is to p-type by B Cat-doping. The doping depth is as shallow as 5 nm or less and the electrically activated doping concentration is 1018 to 1019 cm-3 for both P and B doping. It is also found that the surface potential of c-Si is controlled by the shallow Cat-doping and that the surface recombination velocity of minority carriers in c-Si can be enormously lowered by this potential control.

High-field impurity magneto-optics of Si:Se

Just as phosphorus in silicon produces a hydrogenic defect, the double donor selenium in silicon is an analog of helium. We have measured the impurity absorption spectrum at high magnetic field, and we show that the odd-parity excited states of Si:Se behave identically to those of Si:P. This fact allows us to isolate the electron-electron interactions (exchange and correlation) in the ground state from the quadratic Zeeman effect. The field tuning allows us to put upper limits on the strength of some of these interactions (e.g., at 30 T the electron-electron correlation interaction in the ground state of Se is less than about 40 \ensuremath{\mu}eV; at 30 T the quadratic Zeeman energy in the ground state of P is less than about 200 \ensuremath{\mu}eV).

Removal of Phosphorus in Metallurgical Silicon by Rare Earth Elements

Removal of phosphorus in metallurgical silicon is one of the crucial steps for the production of solar grade Si feedstock. The possibility of doping rare earth elements for phosphorus removal has in this work been studied both theoretically and experimentally. Thermochemical properties of Ce, Nd, and Pr monophosphides have first been estimated by ab initio thermodynamic simulations based on density functional theory and the direct phonon method. The reliability of the first principles calculations was assessed by coupling with the phase diagram data of the Pr-P system. Equilibrium calculations confirmed the existence of stable rare earth monophosphides in solid silicon. Experimental investigations were then carried out, employing a high temperature resistance furnace. The Ce-doped silicon samples were examined by electron probe micro analyzer and inductively coupled plasma analysis. The efficiency of phosphorus removal by means of rare earth doping was discussed in detail in the paper.

Mapping of band-bending for doped C60 films

We used a Kelvin probe to map band-bending occurring at interfaces between indium tin oxide electrodes and fullerene (C60) films doped with either cesium carbonate (Cs2CO3) as a donor or molybdenum oxide (MoO3) as an acceptor. The carrier concentrations determined from the depletion layer widths coincided well with those obtained from the capacitance–voltage characteristics. An ionization efficiency of 10% was determined for the molecularly dispersed dopant (Cs2CO3) in C60, which is significantly lower than the corresponding value of 100% for phosphorus in silicon.

Investigation into the diffusion of boron, phosphorus, and arsenic in silicon during annealing in a nonisothermal reactor

The influence of the temperature gradient on the diffusion of boron, phosphorus, and arsenic during annealing of silicon in a nonisothermal lamp reactor in the second and minute ranges is investigated experimentally and theoretically. Parameters of the thermodiffusion process are determined: for the boron diffusion in the second range, the effective diffusivity Deff ∼ 10−12 cm2/s and the effective measured heat of transport Qmeff* ∼ 103–104 eV. The results are interpreted based on the equations of nonequilibrium thermodynamics.

Quantum error correction in a solid-state hybrid spin register

Error correction is important in classical and quantum computation. Decoherence caused by the inevitable interaction of quantum bits with their environment leads to dephasing or even relaxation. Correction of the concomitant errors is therefore a fundamental requirement for scalable quantum computation. Although algorithms for error correction have been known for some time, experimental realizations are scarce. Here we show quantum error correction in a heterogeneous, solid-state spin system. We demonstrate that joint initialization, projective readout and fast local and non-local gate operations can all be achieved in diamond spin systems, even under ambient conditions. High-fidelity initialization of a whole spin register (99 per cent) and single-shot readout of multiple individual nuclear spins are achieved by using the ancillary electron spin of a nitrogen-vacancy defect. Implementation of a novel non-local gate generic to our electron-nuclear quantum register allows the preparation of entangled states of three nuclear spins, with fidelities exceeding 85 per cent. With these techniques, we demonstrate three-qubit phase-flip error correction. Using optimal control, all of the above operations achieve fidelities approaching those needed for fault-tolerant quantum operation, thus paving the way to large-scale quantum computation. Besides their use with diamond spin systems, our techniques can be used to improve scaling of quantum networks relying on phosphorus in silicon, quantum dots, silicon carbide or rare-earth ions in solids.

SIMS depth profiling analysis of P‐doped n‐type Si layer to develop the Si QD solar cell

Doping of phosphorus in silicon is important for the formation of n‐type semiconducting material for the silicon‐based electronic devices. In this study, the diffusion behavior of P in a SiO2/InP‐doped Si multilayer film was investigated by SIMS depth profiling. It showed a clear preferential diffusion of P to the Si/SiO2 interface regions. This behavior will act as a negative factor for the active doping of P in silicon quantum dot (QD) solar cells. As a result, the conversion efficiency of a heterojunction Si QD solar cell with a P‐doped Si QD layer was much lower than that with a B‐doped Si QD layer. Copyright © 2014 John Wiley &amp; Sons, Ltd.