Semiconductor Silicon Sputtering Target Purity Analysis: Testing Methods, Standards, and Quality Con

Semiconductor silicon sputtering target purity analysis is a critical quality control process that verifies the elemental purity of silicon targets used in physical vapor deposition (PVD) for advanced chip manufacturing. With purity grades ranging from 5N (99.999%) to 7N (99.99999%), even trace metallic impurities at parts-per-billion (ppb) levels can degrade thin-film performance, device reliability, and overall fabrication yield. This article provides a comprehensive technical overview of sample types, impurity characterization items, analytical methods, applicable standards, and industrial applications.

Technical Overview

Silicon sputtering targets serve as the source material in magnetron sputtering and other physical vapor deposition processes used to deposit silicon-based thin films onto semiconductor wafers, display panels, and photovoltaic substrates. During sputtering, energetic ions bombard the target surface, ejecting atoms that form amorphous or polycrystalline silicon films used for gate electrodes, passivation layers, barrier layers, and interlayer dielectrics.

Because every atom ejected from the target is transferred into the deposited film, the chemical purity of the target directly determines the purity of the resulting film. Trace contamination from transition metals such as iron, chromium, nickel, and copper can create deep-level recombination centers in silicon, increasing junction leakage and degrading minority carrier lifetime. Alkali metal ions such as sodium and potassium cause mobile ion contamination that shifts threshold voltages in MOS devices, while excessive oxygen and carbon content alter film stress, refractive index, and etch behavior.

The semiconductor industry typically classifies silicon target purity using the "N" notation, where 5N denotes 99.999%, 6N denotes 99.9999%, and 7N denotes 99.99999% total purity. Advanced logic and memory applications, as well as high-end display manufacturing, generally demand 6N or higher bulk purity with strict individual limits on critical impurity elements. Reliable purity verification therefore requires ultra-trace analytical techniques with detection limits in the ppb to sub-ppb range, performed by accredited third-party laboratories under cleanroom-compatible sample handling protocols.

Testing Samples

Monocrystalline Silicon Sputtering Targets

Czochralski (CZ) grown monocrystalline targets with controlled crystallographic orientation, typically supplied as discs, rectangular plates, or bonded planar targets. Purity analysis focuses on bulk trace elements, dopant residuals (boron and phosphorus), and interstitial oxygen and substitutional carbon inherited from the growth process.

Polycrystalline Silicon Sputtering Targets

Sintered or cast polycrystalline targets used extensively in TFT-LCD and photovoltaic coating lines. In addition to bulk trace metal analysis, evaluation includes oxygen and carbon pickup from sintering atmospheres, grain structure uniformity, and density verification.

Raw Materials and Process Control Samples

High-purity polysilicon granules, chunks, ingot sections, sawn wafers, machining offcuts, and recycling feedstock. Incoming raw material certification and lot-to-lot consistency checks ensure that impurities do not accumulate through the target manufacturing chain.

Failure Analysis and Incoming Inspection Samples

Film-side witness coupons, target erosion-track samples after burn-in, and customer-received lots requiring independent incoming inspection against supplier certificates of analysis.

Testing Items

A complete purity assessment quantifies all major impurity groups and reports both individual element concentrations and the calculated total purity. Typical testing items include:

• Total bulk purity calculation (100% minus the sum of all quantified impurities), reported on a weight basis
• Transition metal impurities: Fe, Cr, Ni, Co, Cu, Zn, Mn, Ti, V, Mo, W, Ta, Zr, and Nb
• Alkali and alkaline earth elements: Na, K, Li, Ca, Mg, and Ba
• Dopant and Group III/V elements: B, P, Al, Ga, As, and Sb
• Noble and heavy metals: Au, Ag, Pt, Pb, Sn, and Bi
• Light element impurities: total oxygen, total carbon, nitrogen, and hydrogen
• Radioactive trace elements U and Th, relevant for soft-error considerations in memory devices
• Supplementary physical properties: resistivity (four-point probe), grain size and crystal orientation (XRD), relative density (Archimedes method), and surface cleanliness

Acceptance specifications are typically defined element-by-element, with the most stringent limits applied to lifetime-killers (Fe, Cr, Ni, Cu, Au) and mobile ions (Na, K), often required below 1 to 10 ppb for semiconductor-grade material.

Testing Methods

Glow Discharge Mass Spectrometry (GDMS)

GDMS is the industry reference technique for bulk ultra-trace analysis of conductive and semiconductive solids. A glow discharge sputter-etches the sample surface, and the released ions are analyzed by high-resolution mass spectrometry. GDMS quantifies more than 60 elements simultaneously with detection limits down to sub-ppb levels, requires minimal sample preparation, and is the primary method for certifying 6N and 7N silicon targets. Depth profiling capability also allows assessment of surface enrichment or depletion of impurities.

Inductively Coupled Plasma Mass Spectrometry (ICP-MS)

Following acid digestion of the silicon sample in high-purity reagents within a cleanroom environment, ICP-MS delivers exceptional sensitivity for dissolved trace metals, reaching ppt-level detection for elements such as Na, K, Fe, Cu, and Ni. Isotope dilution calibration further improves accuracy and is commonly used for certification-grade measurements.

Inductively Coupled Plasma Optical Emission Spectrometry (ICP-OES)

ICP-OES provides robust, high-throughput screening of impurity elements in the ppm to low ppb range. It is well suited for routine production control of raw polysilicon and for cross-validation of GDMS results at higher concentration levels.

Inert Gas Fusion and Combustion Infrared Analysis

Light elements cannot be measured reliably by GDMS or ICP techniques. Oxygen, nitrogen, and hydrogen are determined by inert gas fusion with infrared and thermal conductivity detection, while total carbon is measured by high-temperature combustion followed by infrared detection. These methods are essential because oxygen and carbon concentrations in silicon targets typically range from a few ppma to several hundred ppma and directly influence sputtered film properties.

Fourier Transform Infrared Spectroscopy (FTIR)

For monocrystalline targets, FTIR in accordance with ASTM F1188 quantifies interstitial oxygen, and ASTM F1391 quantifies substitutional carbon content, providing a non-destructive complement to fusion analysis and verifying consistency with crystal growth records.

Secondary Ion Mass Spectrometry (SIMS)

SIMS offers the lowest detection limits available for selected dopants and light elements, reaching ppba levels for boron, phosphorus, oxygen, and carbon in silicon. It is employed for advanced qualification, depth-resolved distribution studies, and correlation of target impurities with deposited film contamination.

Physical and Structural Characterization

Resistivity and conductivity type are verified by four-point probe methods consistent with ASTM F84 and related SEMI procedures. X-ray diffraction confirms preferred crystallographic orientation, which affects sputter yield and film uniformity, while metallographic microscopy and the Archimedes method evaluate grain size and densification of sintered targets.

All measurements are performed following validated test methods and internationally recognized documentation practices, including ISO/IEC 17025 quality management principles, SEMI material specifications, and applicable ASTM and GB/T standards for electronic-grade silicon materials.

Applications

• Gate electrode and polycrystalline silicon layer deposition in advanced CMOS logic and memory device fabrication
• Passivation, anti-reflection, and barrier silicon films for semiconductor packaging and power devices
• Thin-film transistor (TFT) arrays for LCD and OLED display manufacturing
• Amorphous and microcrystalline silicon absorber layers in thin-film photovoltaic modules
• MEMS structures, pressure sensors, image sensors, and microbolometer devices
• Optical coatings, hard masks, and etch-stop layers requiring precisely controlled silicon stoichiometry
• Target manufacturer R&D, process qualification, lot release certification, and end-user incoming inspection

FAQ

What purity grade is required for semiconductor-grade silicon targets?

Most semiconductor and high-end display applications require a minimum of 6N (99.9999%) bulk purity, with individual critical elements such as Fe, Cr, Ni, Cu, Na, and K controlled below a few ppb. Leading-edge logic and memory processes increasingly specify 7N-class material to minimize film contamination and protect device yield.

What is the detection limit of GDMS for silicon targets?

Depending on the element and matrix conditions, GDMS routinely achieves detection limits between approximately 0.001 and 0.05 ppb by weight for most metallic impurities, which is sufficient to certify materials up to 7N purity. Light elements such as oxygen, nitrogen, carbon, and hydrogen are excluded from GDMS quantification and require dedicated fusion and combustion techniques.

How is total purity calculated and reported?

Total purity equals 100% minus the sum of all individually quantified impurities, reported on a weight basis. The certificate of analysis lists each element concentration with its detection limit, clearly distinguishing measured values from less-than-detection-limit values so that customers can reconstruct the purity calculation transparently.

What sample preparation is required for testing?

GDMS samples are typically sectioned to approximately 2 to 3 mm pins or small blocks and lightly surface-cleaned to remove machining residue. ICP-MS analysis requires acid digestion in a cleanroom using sub-boiled high-purity acids, while fusion and combustion analyses need a few grams of crushed or chipped material. All handling is performed with powder-free tools and cleanroom packaging to prevent adventitious contamination.

Why are oxygen and carbon analyzed separately from metallic impurities?

Oxygen and carbon exist in silicon at concentration levels orders of magnitude higher than metallic trace elements, and their measurement principles differ fundamentally. Dedicated gas fusion and combustion infrared methods, supplemented by FTIR for crystalline material, provide the accuracy needed because even moderate variations in oxygen and carbon content measurably affect sputtered film stress, density, and refractive index.

How should a laboratory be selected for purity analysis?

Choose accredited third-party laboratories operating under ISO/IEC 17025 accreditation with documented experience in ultra-trace silicon analysis, cleanroom sample preparation capabilities, and detection limits matched to the purity grade being certified. Independent verification protects both target manufacturers and device makers against specification risk throughout the supply chain.