THE RELATIONSHIP BETWEEN AEROSOL CONCENTRATIONS IN INDOOR AND OUTDOOR AIR AND TRANSPORT FROM OUTDOOR TO INDOOR

Project leader: Markku Kulmala, University of Helsinki, Department of Physical Sciences, P.O.Box 64, 00014 University of Helsinki, Finland, tel. +358-9-191 50756,
e-mail: Markku.Kulmala@helsinki.fi
 
 

PUBLICATIONS

TIIVISTELMÄ SUOMEKSI

Researchers:
Pasi Aalto, University of Helsinki, tel. +358-9-191 8320, e-mail: Pasi.Aalto@helsinki.fi
Ari Asmi, University of Helsinki, tel. +358-9-191 8689, e-mail:  Ari.Asmi@helsinki.fi
Tareq Hussein, University of Helsinki, tel. +358-9 -91 50709, e-mail: Tareq.Hussein@helsinki.fi
Kaarle Hämeri, University of Helsinki, tel +358-9 -91 8717, e-mail: Kaarle.Hameri@helsinki.fi
Petri Keronen, University of Helsinki, tel. +358-9-191 8320, e-mail: Petri.Keronen@helsinki.fi
Ismo K. Koponen, University of Helsinki, tel. +358-9-191 8689, e-mail: Ismo.k.Koponen@helsinki.fi
Risto Pesonen, Finnish Meteorological Institute, tel. +358-9-1929 5430,e-mail: Risto.Pesonen@fmi.fi
Harri Pietarila, Finnish Meteorological Institute, tel. +358-9-1929 5432,e-mail: Harri.Pietarila@fmi.fi
Liisa Pirjola, University of Helsinki, tel. +358-9-191 8718, e-mail: Liisa.Pirjola@helsinki.fi
Katri Puhto, Finnish Meteorological Institute
Erkki Rantakrans, Finnish Meteorological Institute, tel. +358-9-1929 5433, e-mail: Erkki.Rantakrans@fmi.fi

Consortium: Urban air particles and environmental health
Financing SYTTY organisation: The Academy of Finland, The Ministry of Transportation and Communications
Funding from SYTTY / Total funding of project (€): 178884 / 178884
Person-months of work funded by SYTTY / Total person-months of work: 40 / 40

KEY  WORDS: particle size distribution, office building, indoor/outdoor ratio, aerosol, modelling, characterisation, traffic density, meteorological conditions
 

EXTENDED ABSTRACT

1  Introduction

Recent studies suggest correlation between number concentration of ultra-fine aerosols outdoors and adverse health effects. These studies correlate the concentration of a centrally located measurement site and the health effects of population on surrounding area. People spend however very little of their time generally in outdoors, and hence the observed effect of outdoor particulate pollution could be a result of pollutant transport from outdoors to indoors. The relationship between outdoor and indoor particle concentration is important to determine the actual dose of the general population.

2  Methods

The measurements were concentrated on aerosol number concentrations and particle size distributions, but many other parameters and pollutants were also measured in both indoors and outdoors. Measurements were carried out in three different locations in the Helsinki metropolitan.

The first Measurement period was from 7th of January to 1st of February 1999. The measurements took place in an office room and on the rooftop near ventilation inlet point. The measurement site was in central Helsinki (Pasila) in a large governmental office building near heavily used traffic lines and other office and residential buildings. The aerosol size distribution measurement instrument was Differential Mobility Particle Sizer (DMPS). Concentrations of several inorganic gases were monitored continuously with commercial instruments. Several VOCs were analysed from air samples taken indoors and outdoors during the measurement period.

The second measurement period was from 1st of November 1999 to 30th of June 2000. The measurement site was located about 5 km north of downtown Helsinki (Viikki). The site is typical of a suburban background area with minor local anthropogenic aerosol sources except for traffic. One of the major highways leading out of the Helsinki area is located about 100 m from the building, providing a nearby source of aerosol particles with probable significant source rates and high temporal variability. The office building itself was a two-storey construction with clean air intake about 2 m above the ground level. The room was located in the basement, and the fresh air was filtrated and led to the room. The aerosol measurement included high time resolution of total number concentration, and it was performed with a condensation particle counter (CPC) placed in a storage room next to the office. The particle size distributions measurements were carried out during the period from May 15 to June 30, 2000 with the differential mobility particle sizer (DMPS) system. The size range covered with the DMPS setups was between 7 and 600 nm. A laser particle counter was used to measure (during discontinues and short periods) fine and coarse mode particles inside the room. The optical particle counter (OPC) counts particles in the size range 0.3-25 ?m.

The third measurement period was from the 1st to the 22nd of February 2001. The measurement site was a family house located in Espoo (Friisilä). The particle size distributions [3-400 nm] were measured with the differential mobility particle sizer (DMPS) system. The indoor air sampling was carried out at 1.5 m from the ground inside the main living room in the ground floor of the house. The outdoor air sampling was done outside as close as possible to shorten the sampling lines length. Ventilation system of the building was combined gravitationally and mechanically with an exhaust system and a possibility to use special kitchen exhaust fan for cooking fumes.

Measurements of indoor air concentrations give information of the measurement site during the measurement period. One method to increase the usefulness of the measurements is modelling of the data with physically consistent models. Modelling of indoor aerosol dynamics requires the use of some simplification to the aerosol composition. The aerosols used in our model are all composed of some very soluble material, similar in all aspects to sulphuric acid (H2SO4). All aerosols in the model are assumed to belong to one of discrete size (diameter) classes. In the current version of the model aerosols inside one class are identical in composition. The actual calculation of concentration is done by calculating set of differential equations numerically. The equations to be solved are

   (1)

where Ni is the number concentration in size class i, f is the ventilation rate, Oi is the outdoor number concentration in size class i, pi is the penetration coefficient of filter, So is indoor source term and Si is the indoor sink term. Source and sink terms in the model include indoor deposition, re-emission, indoor sources, nucleation, coagulation and condensation.

3  Discussion and conclusions

The results indicate that due the absence of actual indoor sources for particles within measured room, indoor particle concentration followed the same trend as outdoor concentrations. However partly due filter properties the indoor/outdoor concentration ratio was not uniform in all particle sizes. I/O ratios of different sized aerosols were not similar and the maximum of the ratio was usually encountered with aerosols of diameter greater than 90 nm. The outdoor concentrations were typically relatively low during the night-time. After 6:00 in the morning the aerosol number concentrations increased sharply to the range of 104-105 particles per cubic centimetre. The indoor concentrations clearly follows the outdoor concentration especially as the ventilation rate increases from night-time rate to day time rate. The basic behaviour of the aerosol size distribution was similar in all working days. Weekends usually had lot smaller concentration in both outdoors and indoors.

The first measurement location (Pasila) was chosen to be an office for several reasons: (1) the pollution levels are usually highest during the working hours when large portion of people are inside office buildings, (2) offices have usually good ventilation system, which makes the approximation of ventilation parameters in analysis easy, (3) offices do not seem to have many sources for aerosols of the measured size range (diameter 7-500 nm), (4) office buildings are usually well-maintained, (5) the activities inside the office were easier to control than in residential building and (6) largest pollution concentrations are often detected in dense urban areas, where offices are common. Also, the second measurement site was chosen to be an office for the same reasons except that the high pollution level is due to the near high way. The third measurement site was a family house to monitor the indoor activities effects on the number concentration. Figures 1 and 2 present the difference in the particle size distributions in the measurement sites.

The three measurement sites were different in the ventilation operation, type and filter class. This, indeed, gave the opportunity to study the effect of the filtration operation and type, filter and penetration on indoor-to-outdoor aerosol relationship. Figures 3 and 4 present the Indoor/outdoor concentration ratios and the penetration efficiency in every measurement site.

Coupled measurements of indoor-outdoor fine particle aerosols were performed with two identical differential mobility particle sizer (DMPS) systems. The measurements were carried out in three different locations: an urban site influenced directly by traffic emissions and located in the Helsinki downtown, a suburban site located near a freeway, and a rural site influenced by long range transported aerosols. All located in the Helsinki metropolitan. We characterized the aerosol particle size distributions in connection with the traffic emissions and the site location. We also investigated the penetration of fine particles (FP, PM2.5) and ultrafine particles (UFP, PM0.1) in connection with the ventilation operation (mechanically and gravitationally) and the filter-class type (EU3-class, EU7-class, and gravitational penetration) installed in the ventilation system.

Three modes (nucleation 9-25 nm, Aitken 25-90 nm, and accumulation 90-500 nm) were observed in the urban region. In the suburban region there were three main modes (nucleation 7-25 nm, Aitken 25-100 nm, and accumulation 100-600 nm), but during the weekdays the nucleation and Aitken modes were divided into two submodes for each. The rural region showed the lowest aerosol number concentration with four modes (a nucleation 3-20 nm, two Aitken submodes 20-40 nm and 40-100 nm, and an accumulation 100-400 nm). The effect of traffic emissions was clear by shifting the geometrical mean diameter of the Aitken mode towards a smaller value between 20-30 nm. Another effect of the traffic emissions was a sub-Aitken mode with a geometrical mean diameter between 20-30 nm. In general, the effect of traffic emissions was clear in aerosol particles < 100 nm in diameter. Similar particle size characterization was observed indoors with the same number of particle modes, even though, it was hard to distinguish the two overlapped Aitken submodes, so they were considered as one wide Aitken mode. The indoor aerosol concentration was decreased as a function of the particle size due to filtration.

The channel-to-channel indoor/ outdoor (I/O) concentration ratio showed the effect of filtration and ventilation effect on the indoor-to-outdoor aerosol relationship. In the absence of ventilation, the ultrafine particles (especially < 60 nm in diameter) were accumulating indoors and those larger than that size were deposited. While, the accumulation mode particles were accumulated indoors and the ultrafine particles were filtered efficiently when the mechanical ventilation system was on-operation.

The filtration efficiency was estimated in the particle diameter size ranges [7-300 nm], [7-320 nm] and [8-515 nm] for gravitational ventilation and the two filter classes (EU7 and EU3), respectively. Extrapolation of the estimated filtration efficiency was performed to an upper limit of 6.3 ?m in diameter, at which the filtration efficiency was assumed 100%. As the deposition rate of particles was not known indoors, this was not taken into account in our filtration efficiency analysis, and hence, we observe higher filtration efficiency due to deposition of aerosols on indoor surfaces. The best filtration efficiency was obtained for EU7-class filters with minimum filtration efficiency 68%. Surprisingly, the filtration efficiency for gravitational ventilation (35% minimum) was better than that for EU3-class filters (8% minimum). In average, maximum penetration occurred in particles diameter size ranges [0.1-1.0 ?m] for EU7-class filters and [0.03-3.0 ?m] for EU3-class filters and gravitational ventilation. The indoor aerosol concentration was decreased as a function of the particle size, and depended on the filtration type and ventilation operation.

In measurement site (1), monitoring of the indoor inorganic gases proved to be a good addition to the aerosol measurements. Typically the concentration of SO2 increased in the morning as the ventilation went to daytime rate. The SO2 concentrations varied between 0.5 and 8.0 ppb. Ozone concentrations were usually near the detection limit of the instruments, but the a few episodes of increased concentrations to as far as 17ppb were observed. The NO and NOx behaved very similarly and they had a strong anticorrelation with ozone levels. A simple cluster analysis showed that combustion related pollutants (NO, NOx, SO2 and outdoor and indoor accumulation mode particles) have a good correlation with each other. Outdoor and indoor correlation of largest measured particles (90-500nm) is good, but they seem not to correlate well with other pollutants. The smallest measured particles do not seem to correlate with any other measured quantity.

The wind was the main factor to affect the pollution level in the measurement site by transporting pollutants into the site location. The traffic density was characterized by two different daily patterns according to weekdays and weekends. As a result, the traffic activity influenced the total number concentration characterization to have two different daily patterns. The measurement period covered the winter, spring and summer seasons, and accordingly the variation of the meteorological conditions affected the daily pattern of the aerosol total number concentration. In total, and according to the traffic density variation (weekdays and weekends) and the seasonal variation (three seasons), (2x3) different daily patterns for the total number concentration of aerosols were observed outdoors.

The measurements gave good information on the indoor/outdoor connection of a typical office room. The results clearly show that indoor concentrations can be heavily depended on outdoor air quality. With the model used in this work indoor concentrations can be relatively well approximated. Models can be used to analyse the effect of changes in indoor environment and different ventilation or air cleaning systems and it can also be used to help the estimations of the pollution risks to general population. The results also showed that some of the aerosol dynamical processes can be neglected in normal situations.

The results of model studies show a very good agreement between measured indoor size distributions and modelled distributions in normal situations. The model was used to help analyse measurement results and to study the importance of different environmental factors to the I/O ratio of different sized aerosols.
 

[ Projects | Main Page ]