This paper describes a method for direct analysis of methanol, ethanol, isopropanol, acetoaldehyde, acrolein, acetone, C₂-C₆ hydrocarbons in ambient air by gas chromatography.
The flow diagrams for sampling and calibration are demostrated in [A] and [B] in Fig. 1. Water vapor in samples was removed by a trap containing 10 ml of slightly moisten Molecular Sieve 3 A. The samples of about 12.5l were condensed in the trap chilled in liquid oxygen. The standard gases were prepared in 2 l-dilution flask and the calibration was carried out under the similar conditions and procedures to the sampling.
The C₁-C₆ aliphatic hydrocarbons whose peaks overlap those of oxgenates in the chromatograms obtained with the analytical column (Porapak Q) were pre-fractionated from oxgenates and aromatic hydrocarbons with pre-column (ODPN). In order to pre-fractionate effectively, a temperature of the pre-column was 0°C at a separation step and was increased to 80-85°C for a back-flush step. Second pre-column of 35 cm-long Porapak Q was employed to cut aromatic hydrocarbons and others which prolonged the analytical time into a few hours. The column system and conditions are shown in Fig. 2 and Table 1, respectively.
The concentrations of samples taken on March 4 and 5 in 1974 are listed in Table 2 and one of the chromatograms is exemplified in Fig. 3. The correlations of each other are poor in oxgenates and oxgenates-hydrocarbons, but good in hydrocarbons, which are grouped as shown in Fig. 4.

View full abstract

Absolute calibration of a commercially available chemiluminescent ozone analyzer and U. V. absorption ozone analyzers was performed against infrared photometry using an evacuable smog chamber as a large ozone reservoir. Ozone concentrations (0.5-5 ppm) determined simultaneously by longpath infrared photometry and ultraviolet photometry gave the following least square linear regression fit;
(O₃) _UV= (0.974±0.001) (O₃) _IR+ (0.058±0.004)
Given errors are twice of standard deviation (2σ). Gas phase titration was also carried out to check the consistency of these three independent methods. When a U. V. absorption ozone analyzer was used as a transfer standard, the following relationship was obtained in the ozoneconcentration range of 0.05-0.8 ppm.
(O₃) _GPT= (0.954±0.004) (O₃) _IR+ (0.020±0.004)
Thus the consistency of I. R. photometry and U. V. photometry was confirmed within 3%, and I. R.photometry and gas phase titration was confirmed within 5%. Preference among these methods for practical ozone calibration was discussed. It is proposed to adopt the ozone concentration determined by one of the photometric methods as an absolute standard for the calibration of continuous ozone and oxidant analyzers in this country.

View full abstract

Simultaneous measurements of gaseous and particulate sulfur concentrations in the atmosphere were made at one hour sampling during 42 days in winter, and the observational results were discussed in connection with meteorological conditions, comparing with the observational results in summer which were reported in the preceding paper. Striking differences between the observational results in summer and winter were as follows.
(1) Wind direction dependences of both gaseous and particulate sulfur concentrations in summer and winter were considerably different from each other.(Against this, wind direction dependences of the concentration ratio of particulate sulfur to gaseous sulfur in summer and winter were quite similar.)
(2) Variation coefficients of both gaseous and particulate sulfur concentrations were larger in summer than in winter.
(3) Correlation between gaseous and particulate sulfur concentrations was generally stronger in winter than in summer.
(4) Concerning the 24 hour periodicity appearing in the variation with time of sulfur concentrations, appreciable differences were found between the observational results in summer and winter.
(5) Duration of high particulate sulfur concentration was sometimes observed in summer after continuation of a few hot and fine days. Such a phenomenon was never observed in winter.
It was attempted to explain these differences between the observational results in summer and winter in connection with meteorological conditions in summer and winter at the site of observation.

View full abstract

Atmospheric particulate matters usually give a bimodal size distribution separated at about 2 μm in diameter. Each group of particles has a different chemical composition and physical characteristics. The coarse particles are originated from soils, sea salts and mechanically formed dusts by human activities. Then the fine particles are thought to be emitted mainly from the sources of high temperature parts of combustion or burning processes.
In this investigation, particulate samples were collected at five sites in Mizushima district by Andersen type cascade impactors composed of two impaction stages and a back-up filter. Samples were analyzed by neutron activation analysis and atomic absorption spectrometry for elemental constituents. Seasonal variations of concentrations of particulates and trace elements were observed over eight months, from July 1976 through February 1977. The concentrations of coarse and fine particulates varied independently owing to weather conditions. Unusual high concentration of total suspended particulates observed in winter appears to attribute only to the coarse particles. The coarse particles are rather affected by natural phenomena over a wide area, while the fine particles less than 2 μm in size seem to be reflected by local man-made sources.

View full abstract

On 19 and 20 May, 1973 and on 24 and 25 Februay, 1977, when abnormally high concentration of loess aerosol was observed in Osaka district, airborne dust samples were collected with an Andersen air sampler (a multistage impaction type size separation aerosol sampler). These samples were subjected to multielement analysis by instrumental neutron activation method and the analytical results were examined in detail.
Variations with time of monthly average atmospheric aluminium concentrations during four years at six stations of NASN (National Air Surveillance Network) were also examined by comparing with one another.
As the result of the examinations, the following information was obtained.
(1) Loess aerosol appears to show a size distribution having its peak at an aerodynamic diameter of 4-5 μm.
(2) The elemental composition of loess aerosol estimated from the results of multielement analyses was in fairly good agreement with the previously reported elemental compositions of some loess (yellowsand) samples.
(3) Seasonal variations of atmospheric aluminium concentration at four cities in southern Japan, namely, Kawasaki, Matsue, Osaka and Omuta, were quite similar with one another, not only in March to May when loess phenomena are often observed in Japan, but also in the other months when appreciable loess phenomena are rarely observed. This fact is considered to suggest that continental aerosol contributes seriously to natural background aerosol over Japan.
(4) The size distribution of aerosol over Japan is in general assumed to consist of the following component distributions.
(1) Distribution due to aerosol generated by industrial activities, which shows its peak below 1 μmand has a wide bottom beyond 1 μm.
(2) Distribution due to continental aerosol, which has its rather sharp peak around 4-5 μm.
(3) Distribution due to locally generated soil particles, which has its peak at a larger size than 4-5 μm.

View full abstract