Float Zone Silicon Research Articles

Influencing Light and Elevated Temperature Induced Degradation and Surface-Related Degradation Kinetics in Float-Zone Silicon by Varying the Initial Sample State

Light and elevated temperature induced degradation (LeTID) kinetics in float-zone silicon are investigated by varying the initial sample state, composed of different base material, base doping, SiN x :H films, and subsequent firing, and/or annealing steps. The approach of deliberately changing the initial sample state is shown to allow for specific studies of influences of LeTID kinetics. Bulk- and surface-related degradations are examined separately and the influence on the kinetics of bulk- and surface-related degradation is illustrated by a four-state and three-state model, respectively. In case of bulk-related degradation, an increase in defect density because of the firing step is shown, whereas the annealing step has an inverse effect. Both temperature steps—individually and combined—influence the transition rates of bulk-related degradation and regeneration by presumably changing the distribution of a defect precursor. For surface-related degradation, the firing step reduces the transition rate from the initial to the degraded state. In addition, the influence of a comparably humid atmosphere and the absence of UV light are found to be negligible.

Quality control of silicon pixel wafers for the CMS Phase-1 pixel upgrade

The CMS detector at the CERN Large Hadron Collider features as its innermost component a silicon pixel detector. The original pixel detector was completely replaced during the 2016-2017 winter technical stop. One of the goals of this Phase-1 Upgrade of the pixel detector was to replace the sensors in the original CMS forward pixel detector with new, unirradiated sensors. The new CMS forward pixel detector must survive an integrated luminosity of 300 fb$^{-1}$ before being replaced again prior to the High-Luminosity LHC era. Just as in the original construction, the Phase-1 forward pixel sensors were made of n$^{+}$-in-n Diffusion Oxygenated Float Zone silicon. This note documents the quality spot-checking of the new sensors, comparing our results with those provided by the vendor. In general there was good agreement between the results.

3D numerical study of the asymmetric phenomenon in 200 mm floating zone silicon crystal growth

Abstract In this paper, we propose a three-dimensional model for a 200 mm floating zone silicon crystal growth process to investigate the fluid flow and solid–liquid interface. To study the effect of high-frequency (HF) electromagnetic (EM) heating on the melt flow and interface shape, HF-EM and heat transfer calculations were conducted in three dimensions. Through comparison of EM and Marangoni forces, EM force was found to have a larger effect than Marangoni force on the free surface flow. By considering 3D Marangoni and EM forces at the free surface, a more accurate melt flow distribution has been obtained. Moreover, the results showed that local growth rate became more inhomogeneous when the rotation speed of the crystal was increased. However, a more homogeneous three-phase line could be obtained with a high rotational crystal speed.

Development of integration-type silicon-on-insulator monolithic pixel detectors using a float zone silicon

In this paper, we describe the development of monolithic pixel detectors using silicon-on-insulator (SOI) technology for a rapid X-ray residual stress measurement system. Conventional two-dimensional X-ray detectors are not suitable for rapid X-ray residual stress measurement because of their large pixel size and slow readout. For this reason, we developed highly sensitive SOI monolithic pixel detectors that are made up of smaller pixels and can provide a more rapid X-ray residual stress measurement readout. The detectors are fabricated using a 0.2μm CMOS fully-depleted SOI process (Lapis Semiconductor Co., Ltd). The SOI wafer is made by directly bonding a thick, high-resistivity Si wafer and a low-resistivity Si CMOS wafer. The process does not make use of mechanical bump bonding. We developed an integration-type SOI pixel detector, INTPIX4, for a rapid X-ray residual stress measurement system; it uses a float zone (FZ) or Czochralski (Cz) silicon wafer. Cz SOI detectors have been in use since 2005. After 2011, FZ SOI detectors were successfully fabricated. In this paper, we state recent progresses and test results of the SOI monolithic pixel detector using a FZ silicon and compare them with the results obtained using the Cz detector.

Luminescence Properties of FZ Silicon Irradiated with Swift Heavy Ions

The optical properties of float-zone (FZ) silicon irradiated with swift heavy ions (SHI) are studied. In the low-temperature photoluminescence spectra, a broad peak in the range 1.3–1.5 μm is evident along with the well-known X, W, W', R, and C lines. In this case, it is found that, as the irradiation dose is increased in the range 3 × 1011–1013 cm–2, the photoluminescence peak falls and narrows and, at the same time, its maximum shifts to longer wavelengths.

Optical properties of the crystalline silicon wafers described using the universal dispersion model

The optical properties of a slightly boron doped float-zone crystalline silicon wafer are studied using ellipsometry and spectrophotometry in a wide spectral range from far IR to vacuum UV. One side of the wafer was cleaned in an argon plasma, which influenced the optical properties of silicon near the surface. The dielectric response of silicon was modeled using a simplified universal dispersion model which is constructed on the basis of parameterization of the joint density of states describing both the electronic and phonon excitations. Several variants of models describing phonon absorption and interband transitions are discussed. It was possible to accurately determine the optical constants of bulk silicon and the optical constants near the perturbed surface over a wide spectral range. These optical constants agree well with those found in other works. From the optical measurements, it was also possible to determine the thickness of the wafer and the static value of resistivity, and the determined values agreed with nominal values specified for the wafer.

Passivation of Detector‐Grade Float Zone Silicon with Atomic Layer Deposited Aluminum Oxide

Silicon radiation and particle detectors are traditionally passivated with thermal silicon dioxide. It has been shown that aluminum oxide (Al2O3) films provide better surface passivation due to their high negative charge, but studies on Al2O3 surface passivation are usually performed on low‐resistivity substrates. Herein, the passivation of high‐resistivity, detector‐grade float zone silicon (FZ‐Si) with Al2O3 is studied, with a specific emphasis on the effect of post‐annealing temperature on carrier lifetimes and film properties. It is confirmed that Al2O3 provides excellent surface passivation also on high‐resistivity FZ‐Si substrates, with a low interface defect density of around (2–4) × 1011 cm−2 eV−1 and a high negative oxide charge of 1 × 1012 to 3 × 1012 q cm−2, when post‐annealed at temperatures of up to 450–500 °C. In addition, high‐resistivity samples are studied for the phenomenon of bulk lifetime degradation occurring at typical post‐annealing or metal sintering temperatures, which has been reported for low‐resistivity FZ‐Si. At post‐annealing temperatures of >500 °C, reduced bulk lifetimes are observed if the substrates did not receive high‐temperature treatment prior to surface passivation. Furthermore, it is noticed that n‐type samples exhibit lower bulk lifetimes even when a high‐temperature treatment is performed, which indicates a connection between FZ‐Si bulk lifetime degradation and doping type.

Validation of a 3D mathematical model for feed rod melting during floating zone Si crystal growth

A mathematical model of global 3D heat transfer in floating zone silicon single crystal growth process is used to predict the shape of the open melting front of the feed rod. The model is validated using measurement data from research-scale growth experiments. Shape profiles of the open melting front are obtained from the feed rod leftover using a movable dial gauge. Azimuthal asymmetry of the rim of the open melting front is revealed in both simulations and measurements, quantitatively indicating the influence of the main slit of the inductor.

Easy-to-apply methodology to measure the hydrogen concentration in boron-doped crystalline silicon

Hydrogen is a highly relevant impurity in crystalline silicon associated with a variety of effects particularly important in solar cell technology such as defect and surface passivation. It has been the subject of countless studies. Much of this however relies to a certain degree on speculation due to limitations in hydrogen measurement capability, making a direct quantitative correlation of these phenomena with the hydrogen content largely impossible so far. In this contribution, we apply recent advances in the understanding of hydrogen introduction and behaviour in silicon - in particular the conversion of quenched in dimeric hydrogen to boron hydrogen pairs (BH) - and introduce an easy-to-apply methodology to determine quantitatively the hydrogen concentration in bulk boron-doped silicon. The technique involves the measurement and analysis of changes to the bulk resistivity, as recorded by contactless eddy-current measurements, due to the formation of BH pairs during annealing at temperatures between 140 and 180 °C. To demonstrate the method, we apply it to boron-doped float-zone silicon wafers with SiNx:H coatings of different compositions, introducing different total hydrogen concentrations between 5 × 1014 and 1.3 × 1015 cm−3 into the silicon bulk during a fast-firing step. Without firing, the hydrogen content is below the detection limit of 3 × 1014 cm−3 for the 1 Ω cm test sample used.

On-chip single-mode high-Q terahertz whispering gallery mode resonator

The chip-scale terahertz (THz) devices and systems are helpful to enable actual applications in THz communication and sensing. Here, we demonstrate an on-chip THz whispering gallery mode resonator (WGMR) delivering a high-quality (Q) factor TE single mode, fabricated on a high-resistivity float zone silicon wafer by complementary metal–oxide–semiconductor compatible technology. For a TE mode in a WGMR, it is the first time to achieve a Q factor of 1210 at 0.4925 THz with 28 dB extinction ratio, to the best of our knowledge. With this excellent WGMR, we have demonstrated a stable and highly sensitive sensor for particle distance estimation. Our investigation provides an effective approach to fabricate high-performance integrated THz-WGMRs for chip-scale systems.

3D modeling of growth ridge and edge facet formation in [formula omitted] floating zone silicon crystal growth process

A 3D quasi-stationary model for crystal ridge formation in FZ crystal growth systems for silicon is presented. Heat transfer equations for the melt and crystal are solved, and an anisotropic crystal growth model together with a free surface shape solver is used to model the facet growth and ridge formation. The simulation results for 4″ and 5″ crystals are presented and compared to experimental ridge shape data.

Numerical simulations of two-photon absorption time-resolved photoluminescence to extract the bulk lifetime of semiconductors under varying surface recombination velocities

We investigate the limitations of two-photon absorption time-resolved photoluminescence to measure the low-injection bulk lifetime of different semiconductor materials under varying surface recombination. The excitation source is assumed to be a sub-bandgap pulsed laser and the localized absorption and carrier generation was modeled using a focused TEM00 Gaussian beam under the assumption of diffraction-limited performance. The subsequent carrier kinetics were simulated by applying the finite-difference time-domain method to the continuity equation. Three typical semiconductor materials were modeled: direct bandgap low-mobility material (such as CZTS), direct bandgap high mobility (such as GaAs), and indirect bandgap high mobility (such as float-zone silicon). The extracted effective lifetime as a function of surface recombination velocity was compared to the bulk lifetime and the effective lifetime calculated using an analytical 1D approximation. For the direct bandgap materials, focusing inside the material yields an effective lifetime within a few percent of the bulk lifetime, regardless of the surface recombination velocity, while for excitation close to the surface it is up to 30% lower than the bulk lifetime at high surface recombination velocities (>104 cm/s). For the indirect bandgap material, the effective lifetime is dominated by the surface, making the bulk lifetime inaccessible, even at surface recombination velocities of 100 cm/s. Finally, we use the 1D approximation to find under what conditions the bulk lifetime can be extracted by this method and determine that both the bulk diffusion length and the product of the bulk lifetime and surface recombination velocity must be much less than twice the device thickness.

A comparison between rad-hard float zone silicon diodes as gamma dosimeter in radiation processing

In this work, we report on the results obtained with rad-hard Standard Float Zone (STFZ) and Diffused Oxygenated Float Zone (DOFZ) silicon diodes in radiation processing dosimetry. The dosimetric probes were designed to operate in the direct current mode, as on-line radiation dosimeter. The irradiation of the samples was performed using a 60Co source with a dose rate of almost 2.4 kGy/h. The current response of each diode was measured as a function of the exposure time in steps from 5 kGy up to 50 kGy to achieve a total absorbed dose of 275 kGy. In this dose range it is observed a significant decrease in the photocurrent generated in both devices due to gamma radiation defects produced in their active volumes. To mitigate this effect, the samples were pre-irradiated with Co-60 gamma rays at 700 kGy. Despite of being less sensitive, these devices presented stable and reproducible current signals with a relative sensitivity decrease of about 19% within the whole range of dose studied. The dose-response curves of the pre-irradiated diodes showed quadratic behavior with correlation coefficient higher than 0.9999 for total absorbed dose up to 275 kGy. The comparison of the FZ and DOFZ responses evidenced that the latter was slightly superior to the first. However, it is important to note that all pre-irradiated diodes can be used as gamma dosimeters in radiation processing applications.

Diagnostic x-ray dosimeters using standard float zone (FZ) and XRA-50 commercial diodes

The results obtained with a standard float zone (FZ) silicon diode, processed at the Helsinki Institute of Physics, used as on-line diagnostic X-ray dosimeter are described in this work. The device was connected in the short-circuit current mode to the input of an integrating electrometer. The response repeatability and the current sensitivity coefficient of the diode were measured with diagnostic X-ray beams in the range of 40-80 kV. The dose-response of the device, evaluated from 10 mGy up to 500 mGy, was linear with high charge sensitivity. Nevertheless, significant energy dependence was observed in the charge sensitivity of FZ device for energies below 70 kV. The dosimetric characteristics of this FZ diode were compared to those of an XRA-50 commercial Si diode, specially designed to X-ray dosimetry. The results obtained with the FZ diode evidenced that it can be an alternative choice for diagnostic X-ray dosimetry, although it needs to be calibrated for individual X-ray beam energies. The studies of long-term stability and the radiation hardness of these diodes are under way.

Люминесцентные свойства высокоомного кремния, облученного тяжелыми ионами высоких энергий

The float zone (high-resistance) silicon irradiated with swift heavy ions was studied by photoluminescence (PL) spectroscopy method. In addition to well-known X, W, W’, R and C lines, a broad peak (1.3 – 1.5 μm) appears in PL spectra at low-temperatures. As the irradiation dose increases from 3•10_11 to 10_13 cm-2, the PL intensity decreases by about an order of magnitude, also a narrowing of the peak with red-shift is observed.

Scalable Techniques for Producing Field-Effect Passivation in High-Efficiency Silicon Solar Cells

This paper presents techniques to tailor and optimize the field-effect passivation of silicon surfaces using the deposition and field-assisted migration of potassium ions. While field-effect passivation of this nature has previously been demonstrated using laboratory scale techniques, in this paper, it is shown how it can be realized using fast and scalable ion deposition and migration methods. Together, the deposition and migration processes are seen to produce excellent improvements to the surface passivation of oxidized 1-Ω·cm n-type float-zone silicon. Effective lifetimes as high as 2.1 ms are observed, in the best case yielding a surface recombination velocity as low as 3.3 cm/s with a corresponding surface dark saturation current density of 8.4 fA/cm2.

Controlling Surface Carrier Density via a PEDOT:PSS Gate: An Application to the Study of Silicon‐Dielectric Interface Recombination

This communication reports a technique to control the surface carrier population of silicon during photo‐conductance decay measurements, by using a semi‐transparent PEDOT:PSS gate. The potential of this technique has been demonstrated by characterizing carrier‐dependent surface recombination of 1 cm n‐type float zone silicon, passivated with dielectric stack layers of either SiO2, SiO2/SiNx, a‐Si/SiOx, a‐Si/SiOx/SiNx, AlOx, or AlOx/SiNx. Carrier density at the Si‐dielectric interface has been controlled from heavy inversion to heavy accumulation regimes despite leakage currents. This has provided insightful information into the recombination activity at the silicon surface.

(Invited) Ionic Surface Passivation to Extend the Performance of Crystalline Silicon

High efficiency silicon solar cells require substrates with carrier lifetimes well into the millisecond range. High bulk lifetimes can be extremely difficult to measure accurately because of recombination at the sample’s surfaces. Although it is possible, in principle, to passivate the surface well by oxidation or by deposition of a dielectric, such surface passivation approaches often change the bulk lifetime which is being measured. This can be because: (i) the passivation occurs at elevated temperature hence creating or destroying bulk defects; (ii) hydrogen enters the bulk and passivates recombination centres; or (iii) impurities are gettered to the passivating film at the sample’s surface. It has recently been discovered that the thin film which forms on the silicon surface when samples are immersed in ionic solutions containing superacids can provide exceptional surface passivation [1-3]. As this passivating layer is formed at room temperature, the bulk lifetime of the silicon is not modified and hence the artefacts identified above are avoided. In this talk it will be shown how surface recombination velocities of 0.7 cm/s can be achieved using room temperature ionic passivation, and it will be demonstrated how this can be used to measure extremely long bulk lifetimes (>43 ms) [3]. Results of experiments which aim to establish the mechanism by which passivation occurs will be reported, and factors such as the solvent polarity, humidity and solution storage conditions will be reviewed. The talk will then highlight three applications of the ionic passivation for improving the performance of silicon materials. The first will be in the development of high efficiency interdigitated back contact (IBC) solar cells, in which ionic surface passivation is used to understand process-induced bulk lifetime degradation [4]. The second will be in the understanding of light-induced degradation which has recently been found to occur in high performance p-type float-zone silicon [5]. The final example will involve using the superacid-based passivation to measure bulk lifetimes up to and beyond the current intrinsic lifetime limit [6]. It will be shown that the true limit of the bulk lifetime of silicon must be higher than previously assumed, and results from a study which aims to re-evaluate the fundamental limit of silicon’s performance will be presented.

Thermally activated defects in float zone silicon: Effect of nitrogen on the introduction of deep level states

Float zone silicon (FZ-Si) is typically assumed to be an extremely high quality material, with high minority carrier lifetimes and low concentrations of recombination active defects. However, minority carrier lifetime in FZ-Si has previously been shown to be unstable following thermal treatments between 450 and 700 °C, with a range of unidentified deep level states being linked to reduced carrier lifetime. There are suspicions that nitrogen doping, which occurs from the growth atmosphere, and intrinsic point defects play a role in the degradation. This study aims to address this by using deep level transient spectroscopy (DLTS), minority carrier transient spectroscopy, Laplace DLTS, and photoluminescence lifetime measurements to study recombination active defects in nitrogen-doped and nitrogen-lean n-type FZ-Si samples. We find that nitrogen-doped samples experience increased degradation due to higher concentrations of deep level defects during thermal treatments compared to nitrogen-lean samples. In an attempt to explain this difference, in-diffusion of nickel has been used as a marker to demonstrate the existence of higher vacancy concentrations in the nitrogen-doped samples. The origin of the recombination active defects responsible for the thermally induced lifetime degradation in FZ-Si crystals is discussed.

History, status and prospects of producing silicon sensors for HEP experiments at Infineon Technologies

Within this project we are developing single-sided AC-coupled silicon strip sensors and DC-coupled hexagonal pad sensors for applications in High Energy Physics (HEP) experiments. The project started with a feasibility study to re-produce the state-of-the art of silicon sensors of that time, which were based on high-resistivity n-type float-zone 6-inch silicon wafers. Later, the first silicon strip sensors on 8-inch p-type wafers were produced within this project. This influenced the decision of another project to build a silicon sensor based sampling calorimeter. This paper presents the history, current status, and prospects of this sensor development project, jointly conducted by the Institute for High Energy Physics (HEPHY) of the Austrian Academy of Sciences and Infineon Technologies AG.