Spreading Resistance Probe Research Articles

The lower boundary of the hydrogen concentration required for enhancing oxygen diffusion and thermal donor formation in Czochralski silicon

Hydrogen-enhanced thermal donor formation is achieved in p-type Czochralski silicon after exposure to hydrogen plasma and posthydrogenation annealing. Hydrogen diffusivities for the temperatures between 350 and 450 °C are determined based on spreading resistance probe measurement. The hydrogen diffusion is found to be trap limited. Two relationships (for different temperature ranges) are established to describe the lower boundary of the hydrogen concentration required for enhancing oxygen diffusion and thermal donor formation in silicon. The result reveals that hydrogen atoms both in free and in trapped states can enhance the oxygen diffusion.

Research on the effect of nitrogen implantation doses on the structure of separation byimplantation of oxygen and nitrogen

In our work, separation by implantation of oxygen and nitrogen (SIMON) wafers werefabricated with different nitrogen implantation doses and post-annealing. Secondary ion massspectrometer (SIMS) analysis showed that for the samples with low nitrogen dose somenitrogen ions were distributed in the buried oxide layers and some others were collected at theSi/SiO2 interface after annealing, and for the samples with large nitrogen dose distinct delaminationappeared between the layer containing the nitrogen element and that containing the oxygenelement. The results of the spreading resistance probe (SRP) suggested that a buriedinsulator was formed for wafers with large nitrogen implantation dose. The results of crosssection transmission electron microscopy (XTEM) confirmed the analysis above. Theresults show that the quality of SIMON materials is closely related to nitrogenimplantation parameters.

Suppression of hydrogen diffusion at the hydrogen-induced platelets in p-type Czochralski silicon

Hydrogen diffusion in p-type Czochralski silicon is investigated by combined Raman spectroscope, scanning electron microscope, and spreading resistance probe measurements. Exposure of silicon wafers to rf hydrogen plasma results in the formation of platelets. The increase of hydrogenation duration leads to the growth of the platelets and the reduction of the hydrogen diffusivity. The large platelets grow faster than the small ones. The growth of the platelets is based on the capture of hydrogen. The dependence of the hydrogen diffusivity upon the average size of the platelets suggests that the indiffusion of hydrogen is suppressed by the platelets.

TCAD Modeling of Metal Induced Lateral Crystallization of Amorphous Silicon

AbstractIn our previous publications [1, 2] nickel diffusion and spreading resistance probe (SRP) measurements for quality control of metal induced lateral crystallization (MILC) of amorphous silicon (a-Si) were studied. Now we present TCAD modeling and an explanation of experimental results. By using ISE TCAD the Ni concentration distributions were calculated and compared with results obtained by experiments using SIMS analysis.

Characterization of electrically active dopant profiles with the spreading resistance probe

Since its original conception in the 1960s, the spreading resistance probe (SRP) has evolved into a reliable and quantitative tool for sub-micrometer, electrically active dopant, depth profiling in silicon. Its application limit has in recent years even been pushed down to ultra-shallow (sub-50 nm) structures. In this review, a systematic discussion is presented of all issues of importance for a high quality raw data collection and subsequent high accuracy data analysis. The main focus will be on the new developments over the last two decades. The qualification requirements to be fulfilled for 10% profile accuracy (in the absence of carrier spilling) and some of the main fields for SRP application in today's industry will be covered. Finally, a critical assessment of the technique will be made, and its future roadmap will be discussed.

Hydrogen diffusion at moderate temperatures in p-type Czochralski silicon

In plasma-hydrogenated p-type Czochralski silicon, rapid thermal donor (TD) formation is achieved, resulting from the catalytic support of hydrogen. The n-type counter doping by TD leads to a p-n junction formation. A simple method for the indirect determination of the diffusivity of hydrogen via applying the spreading resistance probe measurements is presented. Hydrogen diffusion in silicon during both plasma hydrogenation and post-hydrogenation annealing is investigated. The impact of the hydrogenation duration, annealing temperature, and resistivity of the silicon wafers on the hydrogen diffusion is discussed. Diffusivities of hydrogen are determined in the temperature range 270–450°C. The activation energy for the hydrogen diffusion is deduced to be 1.23eV. The diffusion of hydrogen is interpreted within the framework of a trap-limited diffusion mechanism. Oxygen and hydrogen are found to be the main traps.

Accurate electrical activation characterization of CMOS ultra-shallow profiles

Understanding dopant diffusion and activation mechanisms is a key issue for future sub-45 nm CMOS technologies. This understanding requires the availability of accurate chemical and electrically active dopant profiles. In this work we will focus on the accurate characterization of the electrical active portion of ultra-shallow junction (USJ) profiles, including a precise sheet resistance determination. Here, we will discuss respectively, sheet resistance measurements with conventional and low weight four point probe (FPP) systems, electrical depth profiling by the spreading resistance probe (SRP), alternative solutions based on probe-spacing experiments with an SRP-tool, and electrical characterization by non-contact, non-destructive optical tools, such as surface voltage based resistance and leakage (RsL) measurements, carrier illumination (CI) and infra-red spectroscopic ellipsometry (IR-SE). The comparison will mainly be based on state-of-the-art, low temperature, 1–2 nm/decade, sub-50 nm depth, chemical vapor deposition (CVD) layers with different thicknesses and dopant levels. Furthermore the activation of boron in solid phase epitaxially regrown (SPER) source-drain structures will be discussed. It will be illustrated that the combination of electrical characterization tools can supply essential information not otherwise obtainable through other means such as secondary ion mass spectrometry (SIMS) chemical profiling.

Formation of silicon-on-aluminum nitride using ion-cut and theoretical investigation of self-heating effects

We propose to replace the buried SiO 2 layer in silicon-on-insulator (SOI) with a plasma-synthesized AlN thin film to mitigate the self-heating penalty. The AlN films synthesized on Si by metal plasma immersion ion implantation–deposition (Me-PIIID) exhibit outstanding surface topography and excellent insulating characteristics. Using direct bonding process and the hydrogen-induced layer transfer method, a silicon-on-AlN (SOAN) structure has been successfully fabricated. Cross-sectional high-resolution transmission electron microscopy (HRTEM), X-ray photoelectron spectroscopy (XPS) depth profiles and spreading resistance probe (SRP) reveal that a uniform buried AlN layer is under a single crystal Si overlayer. The interfaces between the top Si layer, buried AlN layer, and Si substrate are smooth and sharp. In addition, the new SOAN device has been verified in two-dimensional device simulation and demonstrates that the self-heating penalty of SOI can indeed be reduced using SOAN substrates.

Boron-enhanced diffusion in excimer laser annealed Si

The effect of excimer laser annealing (ELA) and rapid thermal annealing (RTA) on B redistribution in B-implanted Si has been studied by secondary ion mass spectrometry (SIMS) and spreading resistance probe (SRP). B has been implanted with an energy of 1 keV and a dose of 10 16 cm −2 forming a distribution with a width of 20–30 nm and a peak concentration of ∼5 × 10 21 cm −3. It has been found that ELA with 10 pulses of the energy density of 850 mJ/cm 2 results in a uniform B distribution over the ELA-molten region with an abrupt profile edge. SRP measurements demonstrate good activation of the implanted B after ELA, with the concentration of the activated fraction (∼10 21 cm −3) exceeding the solid solubility level. RTA (30 s at 1100 °C) of the as-implanted and ELA-treated samples leads to a diffusion of B with diffusivities exceeding the equilibrium one and the enhancement is similar for both of the samples. It is also found that RTA decreases the activated B in the ELA-treated sample to the solid solubility limit (2 × 10 20 cm −3). The similarity of the B diffusivity for the as-implanted and ELA-treated samples suggests that the enhancement of the B diffusivity is due to the so-called boron-enhanced diffusion (BED). Possible mechanisms of BED are discussed.

Electrical activation of B and As implants in Silicon On Insulator (SOI) wafers

MOS devices fabricated on Silicon On Insulator (SOI) substrates show several advantages with respect to those realised on silicon bulk. However, some aspects of the dopant diffusion and activation in SOI are not conclusively clarified yet. With this aim, we have investigated the mechanisms of electrical activation of dopants in very thin (100 or 60 nm) SOI materials. The samples have been doped with As or B, and then spike annealed in the temperature range 450–1125 °C. For comparison, selected samples have been pre-amorphised. Spreading resistance probe and four point probe (FPP) measurements have been used as characterisation techniques. The sheet resistance measured in SOI substrates implanted with As is similar to that measured in bulk Si. In contrast, the R S values measured in BF 2 implanted SOI samples are higher than that in bulk Si regardless of the temperature. Electrical profiles measured in SOI and bulk Si substrates implanted with As and BF 2 are in agreement with the R S measurements.

Excimer laser annealing of shallow As and B doped layers

Excimer laser annealing (ELA) of As-, B- and BF 2-implanted Si has been studied by secondary ion mass spectrometry (SIMS), spreading resistance probe (SRP) and transmission electron microscopy (TEM). The implantations have been performed in the energy range from 1 to 30 keV with doses of 10 15–10 16 cm −2. ELA has been carried out with the energy densities in the range of 600–1200 mJ/cm 2 and the number of laser pulses from 1 to 10. It is shown that ELA results in a more uniform dopant distribution over the doped region with a more abrupt profile edge as compared to those after rapid thermal annealing (RTA). Besides, in contrast to RTA, ELA demonstrates a highly confined annealing effect, where the distribution of dopants below the melting region is not affected. SRP measurements demonstrate almost complete activation of the implanted dopants after ELA, and TEM does not reveal extended defects in the ELA-treated samples. The depth of the doped layers, abruptness of the profiles and the total doping dose as a function of ELA energy density and number of laser pulses are investigated. Computer simulations of ELA show a good agreement with the experimental data.

Electrical properties and microstructure of buried oxide (BOX) of SIMOX studied by conducting atomic force microscopy (C-AFM)

The nanometer-scale interface morphology of top Si/buried oxide in SIMOX (separation by implantation of oxygen) and the influence of nano-size Si clusters to the electronic properties of the buried oxide (BOX) layer were studied by Conducting Atomic Force Microscopy (C-AFM). With C-AFM, the interface morphology and the current–voltage ( I– V) curves of the BOX can be obtained simultaneously. The results found that the interface of Si/BOX in SIMOX had a very smooth surface morphology and there were some nanometer-size Si clusters distributed in the BOX, forming conducting channels. Furthermore, the high-resolution transmission electron microscope (HRTEM) and spreading resistance probe (SRP) were used to characterize the BOX.

Metal film characterization with qualified spreading resistance

The spreading resistance probe (SRP) tool is widely used for obtaining the electrically active dopant profile in silicon. In this work, the capabilities of SRP for thin metal film (Cu/Al/TaN) characterization are investigated. It is well known that quite large correction factors (∼500) are involved in sub-100 nm silicon SRP profiling to extract the underlying carrier depth profiles from the measured raw resistances. These large correction factors also allow, in principle, for the characterization of sub-100 nm metal layers, as their measured resistances will fall in the 1–100 ohm range. Structures consisting of Cu or Al metal layers, respectively, with different thicknesses (25–400 nm) on a Ta/TaN barrier layer (∼20 nm) on SiO2 have been characterized with SRP. For calibration, the characteristics of the probe imprints have been measured with atomic force microscopy. Present results indicate that for the thinnest Cu films, it is possible to obtain a resistivity depth profile in good agreement with available theoretical predictions, based on the impact of an electron mean-free-path length of 39 nm.

Characterisation of oxygen and oxygen-related defects in highly- and lowly-doped silicon

In this paper, an overview will be given about analytical techniques which are suitable for the study of oxygen and oxygen precipitation in highly- and lowly-doped silicon. It will be shown that in the case of highly-doped silicon, the application of Fourier Transform Infrared (FT-IR) absorption spectroscopy requires the use of ultra-thinned or high-fluence irradiated samples and a dedicated data analysis. This sample preparation is necessary to reduce the free carrier absorption in the mid-IR region. It is shown that besides the interstitial oxygen concentration [O i] and the amount of precipitated oxygen, it is possible to determine the stoichiometry of oxygen precipitates from the study of the corresponding absorption bands. Oxygen precipitation in p + silicon can also be investigated by the D1–D2 lines in photoluminescence (PL) on as-grown or heat–treated material without special sample preparation. In oxygen-doped high-resistivity float-zone silicon, standard FT-IR analysis can be applied to determine [O i]. The presence of oxygen-related shallow donors can be probed by a combination of electrical (spreading resistance probe, SRP; capacitance–voltage, C– V) and (quasi-)spectroscopic techniques (deep-level transient spectroscopy, DLTS).

Carrier spilling revisited: On-bevel junction behavior of different electrical depth profiling techniques

It is well known that the electrical junction depth position measured along a beveled surface, as is routinely done in the spreading resistance probe (SRP) technique, is shallower than the corresponding metallurgical junction as seen by secondary ion mass spectrometry. The amount of on bevel junction shift (i.e., the difference in electrical on bevel versus metallurgical junction depth) in SRP has previously been attributed to a combination of material removal during the beveling (i.e., one-dimensional zero-field Poisson model) and pressure enhanced carrier spilling (enhanced permittivity). Recently the interest in the application of two-dimensional electrical characterization techniques such as scanning capacitance microscopy, with virtually zero pressure, and scanning spreading resistance microscopy, with a much smaller contact, on beveled surfaces has emerged in order to meet the needed resolutions. The data from these techniques, however, indicate that our present understanding of the carrier-spilling phenomenon is incomplete. Based on a detailed intercomparison of the depth profiles on well-calibrated, junction isolated samples, the shapes of the calibration curves, current–voltage curves, and device simulations, an improved carrier spilling model is proposed that incorporates a uniform surface state density of 5×1012/eV/cm2, with a neutral (Fermi-pinning) level of 0.28 eV above mid-band gap.

Direct Electrical Characterization of Metal Induced Lateral Crystallization Regions by Spreading Resistance Probe Measurements

AbstractMaterial characterization of metal induced lateral crystallization (MILC) process of amorphous silicon (a-Si) has been performed by using the spreading resistance probe (SRP) measurements. It was found that carrier mobility in boron ion implanted layer, formed in MILC region, is up to 65 % in comparison with mobility in boron ion implanted layer, formed in single crystalline silicon. It was also observed in this work that prolongation of MILC process from 1 hour to 2 hours had induced the increasing of mobility from 24 cm2/Vs to 34 cm2/Vs.

Two-dimensional dopant concentration profiles from ultrashallow junction metal-oxide-semiconductor field-effect transistors using the etch/transmission electron microscopy method

The etch/transmission electron microscopy (TEM) method for obtaining one dimensional (1D) and two dimensional (2D) dopant concentration profiles has been developed and applied to n+/p silicon metal-oxide-semiconductor field-effect transistor (MOSFET) structures fabricated using 0.25 μm device technology with ultrashallow source/drain junctions obtained by 5-keV arsenic ion implantation. Cross-section thin-foil specimens were dopant selectively etched with 0.5%HF/99.5%HNO3. Local etch rates were determined from TEM thickness fringes and converted to dopant concentrations using an independently determined calibration curve. The profiles indicated a vertical junction depth of ∼51 nm, a lateral junction distance under the gate edge of ∼27 nm, and an effective gate length of ∼161 nm. The etch/TEM 1D dopant concentration profiles agreed well with spreading resistance probe and secondary ion mass spectroscopy 1D dopant concentration profiles obtained from blanket silicon wafers similarly processed.

Impact of three-dimensional lateral current flow on ultrashallow spreading resistance profiles

The spreading resistance probe (SRP) is a widely used measurement tool for electrical characterization of Si structures. From the measured spreading resistance depth profile, the underlying electrically active dopant profile can be extracted through the application of a series of specific software corrections. In this letter, it is shown that for sub-100 nm deep structures a new type of correction is needed in order to account for the impact of three-dimensional lateral current flow through the conductive surface sublayers located above the actual measurement depth.

Improved extraction of carrier concentration and depletion width from capacitance–voltage characteristics of silicon n+–p-well junction diodes

An accurate method is proposed for the extraction of the carrier concentration profile (pp well) and the depletion width (Wp well) in a p-well region from high-frequency capacitance measurements by accounting for the series resistance and the capacitance of the n+ region. Wp well was calculated from the capacitance in the p-well region (Cp well), while pp well was derived from the slope of the plot 1/Cp well2 versus reverse bias. The pp well extracted was compared with profiles obtained from spreading resistance probe results. The differences between the two techniques are within 15% in the accessible depletion width, which will be discussed in view of the depth resolution anticipated.

Impact of probe penetration on the electrical characterization of sub-50 nm profiles

The spreading resistance probe (SRP) and four point probe (FPP) use, respectively, two or four metal probes with loads in the range of 5–100 g to obtain indispensable information on the electrical characteristics (sheet resistance, junction depth, activation degree, profile shape) of impurity depth profiles in silicon. Recently, however, it has been reported that FPP sheet values are becoming irreproducible and that SRP depth profiles can be significantly shallower than the corresponding secondary ion mass spectrometry dopant profiles for ultrashallow structures. In this work we analyze the impact of probe penetration on the accuracy of SRP and FPP measurements for a series of sub-50 nm profiles with different steepness and substrate level. The probe penetration has been measured by atomic force microscopy and ranged from 5 to 130 nm. The FPP sheet resistance errors of up to 300%, as found for the higher loads, can be correlated with the penetration of the probes through the electrical junction taking into account material removal effects. Experimental data and simulations indicate that SRP raw data are relatively insensitive to probe penetration for source/drain implantation and diffusion profiles. Dependent on the structure involved, however, the SRP depth scale must be corrected for an offset due to probe penetration.