Determination of the Standard Curve for Atmospheric Carbon Monoxide: Principles, Methods, and Qualit

The standard curve determination for atmospheric carbon monoxide (CO) is the foundation of accurate quantitative analysis in ambient air quality monitoring, linking analyzer response to certified, metrologically traceable gas concentrations. This article explains the measurement principles, applicable standards such as GB 9801, ISO 4224, and EN 14626, sampling requirements, step-by-step calibration procedures, and the quality control criteria applied by accredited third-party laboratories. Understanding standard curve methodology is essential for producing defensible CO monitoring data that satisfies regulatory limits under GB 3095-2012 and comparable international frameworks.

Technical Overview

Carbon monoxide is a colorless, odorless, and tasteless gas produced primarily by the incomplete combustion of fossil fuels and biomass. In urban atmospheres, motor vehicle exhaust, industrial boilers, residential heating, and open biomass burning are the dominant sources. Because inhaled CO binds strongly to hemoglobin and forms carboxyhemoglobin, elevated ambient concentrations directly reduce the oxygen-carrying capacity of blood, which is why strict regulatory limits apply.

Under the ambient air quality standard GB 3095-2012, atmospheric CO is limited to 4 mg/m³ as a 24-hour average and 10 mg/m³ as a 1-hour average. Quantitative determination at these low concentrations depends on establishing a standard curve, also called a calibration curve, that relates instrument response to a series of certified reference gas concentrations. The curve is generated by analyzing zero air together with at least four certified CO standard gas mixtures spanning the expected working range, followed by least-squares linear regression. The resulting slope, intercept, and correlation coefficient define the calibration function used to convert raw instrument signals into concentration values.

Unit conversion is a routine part of data reporting: at 0 °C and 101.325 kPa, 1 ppm of CO equals 1.25 mg/m³, while at 25 °C and 101.325 kPa, 1 ppm corresponds to approximately 1.15 mg/m³. Correct temperature and pressure compensation is therefore an integral element of the standard curve determination.

Testing Samples

Ambient Air from Continuous Monitoring Stations

Automated analyzers installed at air quality monitoring stations draw ambient air through a particulate filter, typically a PTFE membrane, and a conditioning sampling manifold. All tubing in contact with the sample should be made of inert materials such as PTFE, FEP, or borosilicate glass, with short residence times to prevent adsorption losses or contamination. Flow rates must exceed the analyzer demand so that the instrument samples directly from the manifold stream.

Grab Samples in Inert Gas Bags and Canisters

For off-site analysis by accredited third-party laboratories, grab samples may be collected in aluminum-lined inert sampling bags or evacuated passivated canisters using calibrated flow controllers. Bags should be analyzed within a validated holding time, commonly 24 to 48 hours, while canisters can preserve sample integrity for longer validated periods provided they are cleaned, leak-checked, and passivated beforehand. Samples must be protected from sunlight and temperature extremes during transport.

Certified Reference Gases and Zero Air

The standard curve requires certified reference materials, namely CO-in-nitrogen or CO-in-air mixtures traceable to a national metrology institute, typically available at concentrations from sub-ppm levels up to 50 ppm or higher. Intermediate calibration points are generated by dynamic dilution using calibrated mass flow controllers. Zero air with a CO content below 0.1 ppm, preferably below 0.05 ppm, is used to establish the zero point and to verify the purity of the dilution system.

Testing Items

When determining the standard curve for atmospheric CO, the following parameters are characteristically evaluated and reported:

• CO mass concentration expressed in mg/m³ or ppm, at 1-hour and 24-hour averaging periods
• Calibration curve slope, intercept, and correlation coefficient derived from linear regression
• Linearity deviation at each individual calibration point relative to the certified value
• Zero drift and span drift of the analyzer over the operating period
• Method detection limit and lower limit of quantification
• Repeatability, expressed as the relative standard deviation of replicate readings
• Interference response from carbon dioxide, water vapor, and light hydrocarbons
• Temperature and pressure compensation accuracy affecting unit conversion

Testing Methods

Non-Dispersive Infrared Absorption Spectrometry

According to GB 9801, CO molecules selectively absorb infrared radiation in the wavelength band near 4.6 to 4.7 micrometers. A non-dispersive infrared analyzer directs infrared energy through the sample cell to a detector, and the attenuation follows the Beer-Lambert relationship between absorbance and concentration. The analyzer is calibrated across the full working range using the multi-point standard curve, which corrects any residual nonlinearity of the optical system and defines the definitive concentration-response relationship.

Gas Filter Correlation Infrared Analysis

The gas filter correlation technique specified in ISO 4224 and EN 14626, and reflected in analyzer specifications such as JIS B 7953, employs a rotating filter wheel containing a CO-filled cell and a neutral reference cell. Because only gas molecules with the same spectral signature as CO modulate the signal, cross-interference from carbon dioxide and water vapor is strongly suppressed. This method dominates modern continuous monitoring networks and is also addressed in the system performance requirements of HJ 654.

Gas Chromatography with Catalytic Conversion and FID

For laboratory-based determination, CO can be separated on a molecular sieve column, catalytically reduced to methane over a nickel catalyst heated to approximately 350 to 400 °C, and quantified with a flame ionization detector. This approach offers excellent sensitivity and selectivity, with detection limits reaching the tens-of-ppb level, making it suitable for background and research applications.

Standard Curve Establishment Procedure

Accredited third-party laboratories follow a disciplined sequence when establishing the standard curve:

• Allow the analyzer to warm up and stabilize according to the manufacturer specification, then perform leak checks and confirm stable operating conditions.
• Introduce zero air and record the stabilized zero response until readings vary by less than the specified noise threshold.
• Introduce at least four certified CO standards from low to high concentration, for example 1, 5, 10, and 20 ppm, or dynamically diluted equivalents covering up to roughly 80 percent of the measurement span.
• At each point, record the average of at least three consecutive stable readings once the response has equilibrated.
• Perform least-squares linear regression of instrument response against certified concentration, with acceptance criteria typically requiring a correlation coefficient of at least 0.999 and a point-by-point deviation within plus or minus 2 percent of the calibration gas value, or as specified by the governing method standard.
• Verify the curve with an independent mid-range check standard obtained from a different cylinder lot, confirming recovery within the specified tolerance.
• Document gas certificate numbers, cylinder pressures, dilution flow settings, ambient temperature and pressure, and the date and operator identity for full traceability.

Quality control does not end with curve establishment. Continuous monitoring networks perform routine zero and span checks, commonly daily or at minimum weekly, with a full multi-point recalibration at least quarterly, and immediately after analyzer maintenance, relocation, replacement of key components, or whenever drift exceeds the acceptance limit. Calibration gases must remain within their certification validity period, and regulators and flow controllers must be dedicated to corrosive-free, high-purity service.

Applications

Standard curve determination for atmospheric CO underpins a wide range of regulatory, industrial, and scientific activities:

• Urban and regional ambient air quality networks reporting compliance against GB 3095-2012 and regional air quality indices
• Roadside and traffic-related monitoring near tunnels, intersections, highways, and enclosed parking structures
• Industrial park surveillance and fence-line monitoring around combustion-related facilities
• Indoor air quality investigations in offices, residences, and underground garages
• Occupational hygiene assessments against workplace exposure limits such as those in GBZ 2.1
• Atmospheric research using CO as a tracer for combustion emissions, photochemical studies, and air quality model validation
• Ground-truthing and comparability exercises supporting remote sensing and satellite retrieval programs
• Instrument verification and calibration services delivered by accredited third-party laboratories

FAQ

Why is a multi-point standard curve necessary when analyzers are factory-calibrated?

Factory calibration establishes general instrument performance, but it does not account for the specific gas standards, dilution system, tubing, and environmental conditions present at each site. A multi-point curve verifies linearity across the actual working range, quantifies any intercept offset, and provides documented traceability required for regulatory data.

What concentration range should the standard curve cover?

The curve should span from zero to at least 80 percent of the instrument span and fully bracket expected ambient concentrations. For typical urban monitoring, points between roughly 1 and 20 ppm are common, with the span adjusted upward when elevated concentrations from traffic or industrial plumes are anticipated.

How often must the standard curve be re-established?

Zero and span checks are performed daily or weekly in continuous networks, while full multi-point calibration is generally repeated at least quarterly. Recalibration is mandatory after maintenance, spare-part replacement, relocation, transport, or whenever zero or span drift exceeds the tolerance defined by the method standard or network operating procedures.

What are the typical linearity acceptance criteria?

Most method standards and network specifications require a correlation coefficient of at least 0.999, with each calibration point agreeing with its certified value within approximately plus or minus 2 percent. An independent verification standard must also be recovered within the specified tolerance before the curve is accepted for data reporting.

How are ppm and mg/m³ concentrations converted?

At 0 °C and 101.325 kPa, 1 ppm of CO equals 1.25 mg/m³, and at 25 °C the factor becomes approximately 1.15 mg/m³. Reporting must state the reference temperature and pressure used, and the analyzer data system should apply consistent compensation throughout the measurement period.

What interferences affect CO measurement and how are they controlled?

Water vapor and carbon dioxide are the principal spectral interferents in infrared methods. Gas filter correlation optics, sample conditioning, pressure compensation, and interference specifications verified during calibration jointly control these effects, while periodic checks with interference gases confirm that cross-sensitivity remains within the limits of the applicable standard.