Atmospheric Compound Pollution Cause Analysis Experiment: Technical Framework, Methods, and Applicat

The atmospheric compound pollution cause analysis experiment is a systematic technical program that combines field observation, high-precision laboratory characterization, and receptor-model-based source apportionment to determine how and why PM2.5 and ozone accumulate simultaneously in the atmosphere. By quantifying the chemical composition of particulate matter, speciating gaseous precursors, and resolving source contributions, the experiment delivers the scientific basis for targeted emission control and coordinated PM2.5–O3 management. This article explains the experimental principle, sample types, analytical items, testing methods, and typical applications of the procedure.

Technical Overview

Atmospheric compound pollution refers to the co-occurrence of elevated fine particulate matter (PM2.5) and ground-level ozone (O3), driven by the interaction of primary emissions with secondary formation from precursors such as sulfur dioxide (SO2), nitrogen oxides (NOx), ammonia (NH3), and volatile organic compounds (VOCs). Unlike single-pollutant episodes, compound pollution involves multiphase chemistry, strong atmospheric oxidation capacity, and rapid secondary aerosol production, which makes cause identification far more technically demanding.

The cause analysis experiment is therefore designed as a closed-loop workflow: representative sampling, multi-dimensional chemical speciation, local source profile construction, quantitative receptor modeling, process-level validation, and uncertainty evaluation. The typical deliverable is a quantified source-contribution inventory — secondary nitrate, secondary organics, coal combustion, motor vehicles, dust, biomass burning, and industrial processes — accompanied by a mechanism-level explanation of how each pollution episode formed, evolved, and which precursor controls would be most effective.

• Observation objective: simultaneous characterization of PM2.5/PM10, O3, precursors, and meteorology at monitoring super-station resolution.
• Analytical objective: full chemical mass closure of particulate composition (water-soluble ions, carbon, elements, water, and unresolved fraction).
• Modeling objective: quantitative source apportionment with complete diagnostic validation (bootstrap, displacement, residual analysis).
• Application objective: design of emission-reduction pathways for coordinated PM2.5 and ozone control.

Testing Samples

Ambient Particulate Matter Samples

PM2.5, PM10, and PM1 samples are collected on quartz-fiber and polytetrafluoroethylene (PTFE) membranes using mid-volume or high-volume samplers that meet the reference-equivalence requirements of EN 12341 and the US EPA PM2.5 reference method. Sampling is typically performed for 23 ± 1 hours per day at fixed rooftop, roadside, industrial, or background sites, with parallel channels allocated for ion, carbon, element, and organic-marker analysis. Size-segregated samples are additionally obtained with multi-stage cascade impactors when accumulation-mode formation mechanisms are a research focus.

Gaseous Pollutant and Precursor Samples

VOCs are collected in pre-evacuated passivated canisters and in DNPH-silica cartridges for carbonyl compounds; acidic gases and ammonia are trapped using annular denuder systems to support gas–particle partitioning analysis. Continuous data streams for SO2, NO–NO2–NOx, O3, CO, and NH3 are recorded in parallel with online analyzers conforming to the performance specifications of HJ 653 and HJ 654, and ambient levels are assessed against the limit values of GB 3095.

Source Profile Samples

Receptor modeling requires locally representative source profiles. Typical samples include road and construction dust, resuspended soil, coal fly ash from different boiler classes, vehicle exhaust by fleet and fuel type, biomass burning smoke for regional fuel varieties, cooking aerosol, and stack emissions from key industrial processes. Each profile is characterized with the same analytical suite as the ambient samples to keep the receptor-model input internally consistent.

Event-Based and Special Samples

Precipitation, dry deposition, and high-frequency filter samples captured during heavy-pollution, dust-storm, and biomass-burning episodes are collected to quantify wet scavenging, regional transport, and episodic contributions. When online instruments detect a rapid concentration rise, sampling resolution is shortened to 1–6 hours to preserve the evolution signature of the episode.

Testing Items

The analytical program covers the physical, chemical, and optical dimensions required for mass closure and source apportionment:

• Mass and water content: gravimetric PM mass at constant humidity, with aerosol liquid water content estimated through thermodynamic modeling.
• Water-soluble inorganic ions: sulfate (SO4), nitrate (NO3), ammonium (NH4), chloride (Cl), potassium (K), sodium (Na), calcium (Ca), and magnesium (Mg) by ion chromatography.
• Carbonaceous components: organic carbon (OC) and elemental carbon (EC) by thermal-optical analysis, with secondary organic carbon estimated using EC-tracer and minimum-OC/EC approaches.
• Inorganic elements: more than twenty major and trace elements (Si, Al, Fe, Ca, Ti, Mn, Cu, Zn, Pb, Cd, Cr, Ni, As, Se, V, and others) serving as crustal, traffic, and industrial tracers.
• Organic markers: levoglucosan and mannosan for biomass burning; hopanes and steranes for vehicle lubricating oil; fatty acids and cholesterol for cooking; PAHs for combustion sources; and isoprene and alpha-pinene oxidation products for biogenic secondary organics.
• VOC speciation: PAMS hydrocarbons, TO-15 halogenated compounds, and oxygenated VOCs, from which ozone formation potential and OH reactivity indicators are computed.
• Physical and optical properties: particle number and size distribution from 10 nm to 10 µm, light scattering and absorption coefficients, hygroscopic growth factor, and mixing-state indicators relevant to haze formation.

Testing Methods

Laboratory Reference Methods

• Gravimetric mass: 24-hour equilibration at 20 ± 1 °C and 50 ± 5% relative humidity, weighed on a microbalance with 1 µg readability (EN 12341; US EPA reference method equivalence).
• Water-soluble ions: ultrasonic water extraction followed by suppressed ion chromatography, cross-checked against online wet-chemistry analyzers for hourly resolution.
• OC/EC: thermal-optical analysis following the IMPROVE_A or EUSAAR_2 temperature protocol with charring correction.
• Elements: non-destructive energy-dispersive X-ray fluorescence per EPA IO-3.3, verified by acid digestion and ICP-MS per EPA IO-3.5 for toxic-element quantification.
• Organic markers: solvent extraction with derivatization followed by GC-MS in selected-ion monitoring mode.
• VOCs: canister preconcentration with GC-MS/FID per EPA TO-15 and HJ 759; carbonyls by DNPH derivatization and HPLC analysis per EPA TO-11A.

Online High-Time-Resolution Techniques

Hourly and sub-hourly data are obtained with aerosol chemical speciation monitors for non-refractory PM1 composition, online ion monitors for water-soluble gases and PM2.5 ions, multi-wavelength absorption photometers for equivalent black carbon source attribution, and scanning mobility and optical particle sizers for number size distributions. Vertical profiling by ceilometer and multi-axis differential optical absorption spectroscopy, combined with boundary-layer observation, distinguishes local accumulation from regional transport and quantifies the mixing-layer dynamics that drive episode onset.

Source Apportionment and Mechanism Modeling

Receptor models form the quantitative core of the experiment. Positive matrix factorization (EPA PMF 5.0) is applied to the speciation dataset with constrained factor profiles, followed by bootstrap, displacement, and residual diagnostics recommended in the model guidance. Chemical mass balance is used where robust local profiles exist, and multilinear engine (ME-2) methods resolve rotational ambiguity between traffic and cooking factors. Results are cross-validated with process-level tools: brute-force sensitivity and tagged-source experiments in WRF-Chem or CMAQ, HYSPLIT backward trajectories with potential source contribution function analysis, and smog-chamber or oxidation-flow-reactor simulation of secondary organic aerosol formation. Observation-based box models and EKMA-curve analysis determine whether ozone formation is VOCs-limited or NOx-limited, which directly informs control strategy design.

Applications

• Heavy-pollution episode diagnosis: real-time identification of the dominant driver — secondary nitrate explosion, dust intrusion, or stagnant accumulation — to support emergency response decisions.
• PM2.5 attainment planning: seasonal and annual source-contribution inventories that prioritize reduction measures and estimate the effectiveness of control programs.
• Coordinated PM2.5–O3 control: prioritization of VOC species by ozone formation potential and secondary organic aerosol yield to guide solvent, industrial, and mobile-source strategies.
• Policy evaluation: ex-post assessment of coal substitution, fleet renewal, and VOC treatment programs using meteorology-normalized receptor results.
• Scientific research: mechanism studies of heterogeneous chemistry, new-particle formation events, and multiphase oxidation pathways, often conducted with accredited third-party laboratories for independent data verification.

FAQ

What is atmospheric compound pollution?

It is a pollution pattern in which PM2.5 and ozone occur together or alternately at elevated concentrations, produced by the coupled interaction of primary particle emissions and secondary formation from SO2, NOx, NH3, and VOCs under strong atmospheric oxidation conditions.

What does source apportionment mean in this experiment?

Source apportionment is the quantitative attribution of measured ambient concentrations to specific emission sources. It combines the chemical fingerprints of ambient samples with local source profiles in receptor models such as EPA PMF and CMB, reporting the percentage contribution of each source category together with uncertainty bounds.

How long does a typical cause analysis campaign take?

A scientifically robust campaign normally covers at least one full year with four intensive observation periods, because source contributions and formation regimes differ sharply between the heating and photochemical seasons. A short diagnostic campaign targeting a single episode typically runs from two to eight weeks.

Which standards govern the sampling and analysis?

Sampling and analysis follow EN 12341 and the US EPA PM2.5 reference method for mass, EPA Compendium methods IO-3.3, IO-3.5, TO-11A, and TO-15 for speciation, GB 3095 for ambient air quality assessment, and HJ technical specifications for automatic monitoring, with quality assurance and quality control executed by accredited third-party laboratories where independent verification is required.

How is data quality assured?

Field blanks, duplicate samplers, standard reference materials, inter-laboratory comparisons, ion-balance and mass-closure checks, and full model diagnostics — bootstrap factor mapping, displacement error estimation, and residual analysis — are applied at every stage so that the reported source contributions remain stable, reproducible, and defensible.