Carrier Concentration Determination for Monocrystalline Silicon Wafers: Principles, Methods, and Sta

Carrier concentration determination is a fundamental electrical characterization test for monocrystalline silicon wafers used in semiconductor, power device, and photovoltaic manufacturing. This article explains the measurement principles, the principal test methods—including four-point probe resistivity with dopant-density conversion, Hall effect measurement, non-contact eddy current, spreading resistance profiling, and electrochemical capacitance-voltage profiling—along with applicable ASTM, SEMI, and GB standards, typical sample requirements, and industrial applications.

Technical Overview

Carrier concentration refers to the density of free charge carriers (electrons in n-type material or holes in p-type material) per unit volume in a monocrystalline silicon wafer, typically expressed in atoms per cubic centimeter (cm⁻³ is not used here; values are written as, for example, 1.0 × 10^15 cm^-3). It is governed primarily by the intentionally introduced dopant species—boron for p-type, and phosphorus, arsenic, or antimony for n-type silicon—and, at typical device operating temperatures, the free carrier density closely tracks the electrically active dopant density in non-compensated material.

The three key electrical parameters are linked by the fundamental relationship: resistivity (ρ) = 1 / (q · n · μ), where q is the elementary charge, n is the carrier concentration, and μ is the carrier mobility. Because mobility itself depends on doping level, standards such as ASTM F723 (SEMI MF723) provide validated conversion tables between resistivity and dopant density for boron-doped and phosphorus-doped silicon.

Carrier concentration spans many orders of magnitude in practice: high-purity floating-zone silicon can exhibit concentrations below 10^12 cm^-3 (resistivity above several kΩ·cm), standard IC-grade substrates fall around 10^14 to 10^16 cm^-3, photovoltaic wafers typically range from 10^15 to 10^16 cm^-3, and heavily doped substrates reach 10^18 to 10^19 cm^-3. Accurate determination is essential because this parameter directly controls breakdown voltage in power devices, series resistance and contact behavior, threshold characteristics in MOS technology, and minority-carrier lifetime and conversion efficiency in solar cells.

Testing Samples

Prime Monocrystalline Silicon Wafers

Standard Czochralski (CZ) or float-zone (FZ) wafers in as-cut, lapped, etched, or polished condition, in diameters from 50 mm to 300 mm. Bulk resistivity measurement combined with dopant-density conversion yields the volume carrier concentration of the substrate.

Epitaxial Wafers

Wafers carrying a lightly or heavily doped epitaxial layer on a contrasting substrate. Layer-specific techniques such as spreading resistance profiling (SRP) and electrochemical capacitance-voltage (ECV) profiling are used to separate the epi layer concentration from that of the underlying substrate.

Diffused and Ion-Implanted Layers

Surface regions created by diffusion or ion implantation followed by annealing. Junction depth, peak concentration, and the full depth profile of electrically active dopants are characterized.

Photovoltaic Monocrystalline Wafers

As-cut pseudo-square or full-square wafers (125 mm to 210 mm class) with thicknesses of roughly 130 to 200 μm. Non-contact eddy-current measurement is well suited to rough, saw-damaged surfaces, with resistivity typically specified between 0.3 and 6 Ω·cm for p-type boron-doped material.

Sample Requirements

• Minimum sample size of approximately 10 × 10 mm for Hall effect measurement; full wafers or quarters for four-point probe mapping.
• Clean, crack-free surfaces free of contamination, oxide smut, or coatings that affect probe or electrode contact.
• Accurate wafer thickness data, required for converting sheet resistance to bulk resistivity and concentration.
• Known or declared dopant type (P or N), if available, to select the correct conversion model.

Testing Items

• Carrier concentration (volume dopant density), reported in cm^-3.
• Resistivity (Ω·cm) and sheet resistance (Ω/sq).
• Conductivity type (N or P), determined from the sign of the Hall coefficient or hot-probe measurement.
• Hall coefficient (cm³/C) and Hall mobility (cm²/V·s).
• Uniformity parameters: center-to-edge variation, radial variation percentage, and within-wafer standard deviation.
• Depth-resolved carrier concentration profiles for epitaxial, diffused, and implanted structures, including junction depth.

Testing Methods

Four-Point Probe Resistivity with Dopant-Density Conversion

The collinear four-point probe method per ASTM F84 (SEMI MF84) and GB/T 1552 is the most widely used routine technique. A known current passes through the outer probes and the voltage is measured between the inner probes, eliminating contact resistance errors. Bulk resistivity is calculated using geometric correction factors, then converted to dopant or carrier concentration using the standardized resistivity-to-density relationships of ASTM F723 (SEMI MF723). Measurements are performed at a controlled temperature of 23 ± 1 °C with appropriate temperature correction. For epitaxial or diffused layers, sheet resistance methods per ASTM F374 and uniformity mapping per ASTM F1529 are applied. The effective range extends from roughly 0.0008 Ω·cm to several thousand Ω·cm.

Hall Effect Measurement (Van der Pauw Configuration)

Hall measurement per ASTM F76 (SEMI MF76) is the most direct method for carrier concentration. A magnetic field is applied perpendicular to a current-carrying sample, and the resulting Hall voltage yields the Hall coefficient. The carrier concentration is calculated as n = 1 / (q · |R_H|), where R_H is the Hall coefficient, and the sign of R_H identifies the carrier type. Combined with resistivity data, the Hall mobility is obtained directly. This method covers an exceptionally wide range, from below 10^12 cm^-3 in high-purity FZ silicon up to heavily doped material, and enables analysis of compensation in partially compensated wafers.

Non-Contact Eddy Current Measurement

Per ASTM F673 (SEMI MF673), a non-contact eddy-current gauge induces currents in the wafer and derives resistivity from the damping response. Because no mechanical contact is required, it is ideal for high-speed, one-hundred-percent screening of production wafers, including rough as-cut photovoltaic slices. Resistivity is subsequently converted to carrier concentration using the same standardized dopant-density relationships, with calibration against certified resistivity standards.

Spreading Resistance Profiling (SRP)

SRP measures resistance between two precisely conditioned probe steps scanned along a small-angle beveled surface. Calibrated against reference standards, it produces a high-resolution depth profile of resistivity, from which carrier concentration versus depth is calculated. SRP is the method of choice for epitaxial layer characterization, buried layer evaluation, and junction depth determination, with depth resolution in the nanometer range.

Electrochemical Capacitance-Voltage (ECV) Profiling

ECV profiling uses an electrolyte as a Schottky contact. By measuring capacitance-voltage characteristics while electrochemically etching the semiconductor in controlled increments, the net active dopant concentration (N_D minus N_A) is obtained as a function of depth. It is particularly effective for compound structures, steep epi transitions, and heavily doped layers in the 10^14 to 10^19 cm^-3 range.

Infrared Plasma Resonance Methods

For heavily doped silicon above approximately 5 × 10^18 cm^-3, where free carriers shift the infrared reflectance spectrum, plasma-resonance-based infrared techniques provide an independent determination of carrier concentration, complementing electrical methods near the solubility limit of dopants.

Applications

• Incoming inspection and lot acceptance of monocrystalline wafers by device manufacturers, supported by test reports from accredited third-party laboratories.
• Crystal growth and wafering quality control, including ingot axial grading assessment and radial resistivity uniformity verification.
• Power semiconductor development (diodes, IGBTs, thyristors), where starting-wafer concentrations near 10^13 to 10^14 cm^-3 determine breakdown voltage ratings.
• CMOS and memory substrate and epitaxial layer specification control, including EPI resistivity and transition width verification.
• Photovoltaic cell manufacturing, where base doping influences open-circuit voltage, minority-carrier lifetime, and cell efficiency.
• Research and process development: dopant activation studies after implant annealing, neutron transmutation doping verification, and compensation analysis in upgraded metallurgical-grade silicon.

FAQ

What is the difference between carrier concentration, dopant density, and resistivity?

Resistivity is the quantity measured directly by most instruments. Dopant density is the concentration of incorporated dopant atoms, while carrier concentration is the density of free charge carriers actually available for conduction. In fully activated, uncompensated material the two are essentially equal; in compensated or incompletely activated silicon, Hall effect measurement reveals the difference.

Which measurement method should be selected?

Four-point probe with standardized conversion is the standard choice for routine bulk wafer characterization. Hall effect is preferred when carrier type, mobility, or compensation must be quantified. Eddy-current testing suits rapid full-surface production screening, while SRP and ECV are required whenever depth-resolved profiles of epitaxial, diffused, or implanted layers are needed.

What accuracy and range can be expected?

Typical four-point probe resistivity reproducibility is better than ±2% over the range of about 0.0008 to 3000 Ω·cm. Hall measurement extends from below 10^12 to above 10^19 cm^-3. ECV profiling generally covers 10^14 to 10^19 cm^-3, and infrared methods cover heavily doped material above roughly 5 × 10^18 cm^-3. Concentration uncertainty is usually dominated by mobility and calibration assumptions, typically within ±5% to ±10%.

Can as-cut photovoltaic wafers with saw damage be tested?

Yes. Non-contact eddy-current instruments measure through the rough, textured, or saw-damaged surface without probe penetration, making them the preferred method for as-cut slices. Four-point probe measurement is generally applied after surface etching or polishing to avoid errors from damaged surface layers.

What information should accompany a sample for testing?

Clients should provide wafer thickness, dopant type if known, expected concentration range, wafer diameter or dimensions, and the applicable standard or specification to be reported against. Testing is conducted at a controlled ambient of 23 ± 1 °C, and test reports from accredited third-party laboratories normally include measurement method, standard reference, raw resistivity or Hall data, calculated concentration, and measurement uncertainty.