Three-Channel Chromatographic Simultaneous Analysis of Routine Impurities in Ethylene

Literature Overview

The paper by Sun Leili, Li Miao, Chen Binggang, Wang Teng, Wang Dexiang, Chu Jinwei, and Zhang Yan (2020), published in Guangzhou Chemical Industry, presents a gas chromatography (GC) method for simultaneous analysis of routine impurities in ethylene using a single instrument with three valves, three channels, and three hydrogen flame ionization detectors (FID). The method enables one-injection analysis of hydrocarbon impurities, trace CO, CO2, and oxygenated compounds. The reported detection limits are 0.05 mL/m³ for hydrocarbon impurities, 0.03 mL/m³ for CO and CO2, and 0.3 mL/m³ for oxygenated compounds. Precision and accuracy studies confirm relative standard deviations below 2.0% and recovery rates between 95% and 105%.

Core Technical Viewpoints

The innovation in this work is the integration of three independent analytical channels into a single GC system, allowing simultaneous determination of multiple impurity classes in a single injection cycle. This approach significantly improves analytical throughput compared to sequential single-channel methods while maintaining the sensitivity and specificity required for trace-level impurity analysis in ethylene streams.

Interpretation of Key Technical Points

System Configuration

The three-channel configuration uses three valves to route the sample through three separate analytical pathways:

1. Channel 1 - Hydrocarbon impurities: Uses a non-polar or weakly polar column (e.g., PoraPLOT Q or equivalent) to separate trace hydrocarbons such as methane, ethane, propane, propylene, and higher hydrocarbons from the ethylene matrix.
2. Channel 2 - CO and CO2: Uses a molecular sieve column (e.g., PoraPLOT N or Carboplot B) for separation of carbon oxides, which are critical impurities in ethylene streams as they can poison polymerization catalysts.
3. Channel 3 - Oxygenated compounds: Uses a polar column (e.g., PoraPLOT U or Carboxen) to separate oxygen-containing impurities such as water, methanol, acetaldehyde, and other oxygenates.

Central Cut Technique

The paper mentions central cut technology, which is essential for analyzing trace impurities in a high-concentration ethylene matrix. The central cut technique involves:

1. Injecting the sample into a pre-column or guard column.
2. Allowing the ethylene peak to pass through without being trapped.
3. Transferring only the trace impurity fraction to the analytical column.

This technique prevents column overload from the major ethylene component and allows trace impurities to be concentrated and separated with high sensitivity.

Quality Assurance Parameters

The method validation results demonstrate:

• Precision: RSD < 2.0% for all components, indicating excellent reproducibility.
• Accuracy: Recovery rates of 95–105%, confirming method accuracy.
• Detection limits: Sufficient sensitivity for trace-level analysis.

These parameters meet the requirements for routine quality control in ethylene production and polymerization feedstock preparation.

Integration with Engineering Practice

In the steel pipe and process piping industry, the analysis of gas stream composition is critical for:

• Natural gas pipeline integrity: Trace impurities such as H2S, CO2, and water affect corrosion rates and material selection for pipeline steel grades (e.g., API 5L X70, X80, X100).
• Hydrogen pipeline systems: Trace oxygen and moisture in hydrogen streams affect embrittlement behavior of carbon steel and low-alloy steel pipelines.
• LNG pipeline systems: Methane purity specifications require trace-level analysis of nitrogen, CO2, and water.
• Process plant piping: Catalyst-grade ethylene requires stringent impurity control, and the piping materials must be compatible with the purified gas stream.

The three-channel GC method described here could be adapted for monitoring impurity levels in various gas streams transported through steel pipelines. For example, the same multi-channel approach could be used to simultaneously monitor:

• Hydrocarbon composition in natural gas pipelines
• Trace oxygen and moisture in hydrogen pipelines
• Corrosive species (H2S, CO2) in sour gas pipelines

Material Selection Implications

The impurity levels determined by this method directly influence material selection for piping systems. For instance:

• CO2 levels above 1% require consideration of carbonic acid corrosion in carbon steel pipelines
• Water content above 100 ppm requires consideration of internal corrosion and hydrogen blistering
• Oxygen levels above 50 ppm may require consideration of stress corrosion cracking susceptibility

Key Questions and Reflections

Several practical questions emerge from this work. First, the three-channel configuration requires careful valve timing and flow control to ensure that the sample is distributed correctly among the channels. How sensitive is the method to valve timing errors, and what is the impact on quantification accuracy? Second, the use of three FID detectors increases system cost and complexity. Could a single FID detector with a multi-column configuration achieve similar results? Third, the method is validated for ethylene streams, but how well does it perform for other hydrocarbon streams with different matrix compositions?

Study Insights and Implications

The most valuable contribution of this paper is the demonstration that multi-channel GC analysis can achieve simultaneous determination of multiple impurity classes with trace-level sensitivity in a single injection. This approach significantly reduces analysis time and sample consumption, which is critical for real-time process monitoring in petrochemical plants. For piping engineers, the ability to rapidly determine gas stream composition enables more informed decisions about material selection, corrosion allowance, and inspection intervals. The method also highlights the importance of analytical capability in supporting piping integrity management programs, as accurate impurity data is essential for predicting corrosion rates and remaining life of pipeline assets.