In the rapid development of high-tech industries such as semiconductor and photovoltaic, electronic-grade silane gas, as a key raw material, its purity directly affects the performance and reliability of downstream products. Traditional offline detection methods can no longer meet the stringent requirements of modern production for real-time, safety, and precision.

This paper will introduce the application of online gas chromatography (GC) system in silane analysis, and show how this technology can upgrade the production quality control by means of automation and intelligence, and provide more safe and efficient solutions for the industry.

silane properties

It is highly reactive and exhibits strong reducing property at room temperature.

It will react violently with oxygen, halogen and other oxidizing substances.

The explosive limit range of the gas is wide (0.8%-98%).

It has the characteristic of spontaneous combustion and can be ignited without open flame.

It has strong irritation effect on human body, and high concentration exposure may cause acute injury.

These characteristics make the detection of silane challenging:

High security risk

In the traditional manual sampling, the direct contact of silane gas will cause serious explosion hazard because of its self-igniting property.

Inefficiency

Laboratory analysis is a tedious process, including sampling, transportation, pretreatment, analysis and other links, and a single test usually takes more than 2 hours.

Data lag

It cannot reflect the component fluctuations in real time during production, and quality issues are detected with a delay.

Advantages of the Application of Online Gas Chromatography (GC) System

This customized online gas chromatography (GC) system, developed through a deep collaboration between Gao (Beijing) Intelligent Technology and Agilent Technologies—a global leader in analytical instruments—specifically addresses the demand for electronic-grade silane gas detection. Utilizing fully automated analysis technology, it enables real-time, precise monitoring of key components at multiple sampling points, effectively resolving industry challenges in traditional silane detection such as high safety risks and low detection efficiency.

test automation

The system automatically completes the sampling and analysis of samples without manual intervention.

You can configure the detection cycle (e.g., every 15 minutes) to cover the entire production period.

high efficiency pretreatment

The rapid loop design ensures efficient sample circulation, while the combination of pipeline optimization and low-volume valve structure significantly reduces pipeline dead volume.

The helium purge and vacuum system are used to eliminate the dead volume residue in the pretreatment process, effectively prevent cross-contamination and avoid the retention of dangerous gas, and ensure the safety of use and the accuracy of detection.

real-time data interconnection

The test results are transmitted directly to the control system, and abnormal data automatically triggers an alarm.

You can process and view data remotely.

The historical data storage supports the characterization and traceability of MSA/SPC.

AI Decision Engine

By machine learning analysis of historical data, the system can predict the trend of impurity change and give early warning.

Safety design

The pretreatment unit is made of EP grade corrosion-resistant material with VCR welding and sealed treatment, which is suitable for silane medium.

The station building is designed with positive pressure explosion-proofing and equipped with multiple detectors for 24-hour real-time monitoring, automatically triggering alarms when abnormalities occur.

application case

The on-line gas chromatography system can meet the on-line determination of H2, (Ar+O2), N2, CH4, CO, CO2, SiH3Cl, SiH2Cl2, SiHCl3, SiCl4, Si2H6, hydrocarbon (C2-C4) and other impurities in electronic grade silane gas.

detector: helium ion detector

Detection capability:

Detection limit ≤1ppbv (CH4 basis), repeatability error <5%, single sample analysis time ≤15 minutes (including vacuum extraction, purging, and replacement time).

Column temperature: For light components, a constant temperature of 50°C is recommended as the sample has a lower boiling point and elutes faster. For heavy components, a gradient heating from 40°C to 100°C is used. It is advisable to run the column after the column box and to bake the column at high temperature.

Continuous analysis of data stability in an online gas chromatography system

heavy constituent, heavy ends

light component

The Relative Standard Deviation (RSD) is consistently below 3%, demonstrating the system's excellent repeatability!