Formaldehyde (HCHO), the most abundant carbonyl compound in the atmosphere, is generated as an intermediate product in the oxidation of nonmethane hydrocarbons. Proton transfer reaction mass spectrometry (PTR-MS) has the capability to detect HCHO from ion signals at m/z 31 with high time-resolution. However, the detection sensitivity is low compared to other detectable species, and is considerably affected by humidity, due to back reactions between protonated HCHO and water vapor prior to analysis. We performed a laboratory calibration of PTR-MS for HCHO and examined the detection sensitivity and humidity dependence at various field strengths. Subsequently, we deployed the PTR-MS instrument in a field campaign at Mount Tai in China in June 2006 to measure HCHO in various meteorological and photochemical conditions; we also conducted intercomparison measurements by Multi-Axis Differential Optical Absorption Spectroscopy (MAX-DOAS). Correction of interference in the m/z 31 signals by fragments from proton transfer reactions with methyl hydroperoxide, methanol, and ethanol greatly improves agreement between the two methods, giving the correlation [HCHO]_MAX-DOAS=(0.99±0.16) [HCHO]_PTR-MS+(0.02±0.38), where error limits represent 95% confidence levels.

An observation-based box model approach was undertaken to estimate concentrations of OH, HO₂, and RO₂ radicals and the net photochemical production rate of ozone at the top of Mount Tai, located in the middle of Central East China, in June 2006. The model calculation was constrained by the measurements of O₃, H₂O, CO, NO, NO₂, hydrocarbon, HCHO, and CH₃CHO concentrations, and temperature and J values. The net production rate of ozone was estimated to be 6.4 ppb h⁻¹ as a 6-h average (09:00–15:00 CST), suggesting 58±37 ppb of ozone is produced in one day. Thus the daytime buildup of ozone recorded at the mountain top as ~23 ppb on average is likely affected by in situ photochemistry as well as by the upward transport of polluted air mass in the daytime. On days with high ozone concentrations (hourly values exceeding 100 ppb at least once), in situ photochemistry was more active than it was on low ozone days, suggesting that in situ photochemistry is an important factor controlling ozone concentrations. Sensitivity model runs for which different NO_x and hydrocarbon concentrations were assumed suggested that the ozone production occurred normally under NO_x-limited conditions, with some exceptional periods (under volatile-organic-compound-limited conditions) in which there was fresh pollution. We also examined the possible influence of the heterogeneous loss of gaseous HO₂ radicals in contact with aerosol particle surfaces on the rate and regimes of ozone production.

The emission characteristics of refractory black carbon (rBC) from biomass burning are essential information for numerical simulations of regional pollution and climate effects. We conducted combustion experiments in the laboratory to investigate the emission ratio and mixing state of rBC from the burning of wheat straw and rapeseed plants, which are the main crops cultivated in the Yangtze River Delta region of China. A single particle soot photometer (SP2) was used to measure rBC-containing particles at high temporal resolution and with high accuracy. The combustion state of each burning case was indicated by the modified combustion efficiency (MCE), which is calculated using the integrated enhancement of carbon dioxide and carbon monoxide concentrations relative to their background values. The mass size distribution of the rBC particles showed a lognormal shape with a mode mass equivalent diameter (MED) of 189 nm (ranging from 152 to 215 nm), assuming an rBC density of 1.8 g cm⁻³. rBC particles less than 80 nm in size (the lower detection limit of the SP2) accounted for ∼ 5 % of the total rBC mass, on average. The emission ratios, which are expressed as ΔrBC ∕ ΔCO (Δ indicates the difference between the observed and background values), displayed a significant positive correlation with the MCE values and varied between 1.8 and 34 ng m⁻³ ppbv⁻¹. Multi-peak fitting analysis of the delay time (Δt, or the time of occurrence of the scattering peak minus that of the incandescence peak) distribution showed that rBC-containing particles with rBC MED = 200 ± 10 nm displayed two peaks at Δt = 1.7 µs and Δt = 3.2 µs, which could be attributed to the contributions from both flaming and smoldering combustion in each burning case. Both the Δt values and the shell / core ratios of the rBC-containing particles clearly increased as the MCE decreased from 0.98 (smoldering-dominant combustion) to 0.86 (flaming-dominant combustion), implying the great importance of the rapid condensation of semi-volatile organics. This laboratory study found that the mixing state of rBC particles from biomass burning strongly depends on its combustion processes, and overall MCE should be taken carefully into consideration while the climate effect of rBC particles from open biomass burning is simulated.