SYTTY
INDOOR AIR QUALITY CONTROL
Project leader: Kristina Saarela, VTT Building and Transport,
P.O.Box 1401, FIN-02044 VTT, Finland, tel. +358-9-456 5292, e-mail: Kristina.Saarela@vtt.fi
| PUBLICATIONS |
| TIIVISTELMÄ SUOMEKSI |
Researchers:
Kirsi Villberg, VTT Building and Transport, P.O. Box 1401, 02044 VTT,
Finland, tel. +358-9-456 5233, e-mail: Kirsi.Villberg@vtt.fi
Tiina Tirkkonen, VTT Building and Transport, P.O. Box 1401, 02044 VTT,
Finland, tel. 358-9-456 5287, e-mail: Tiina.Tirkkonen@vtt.fi
Helena Mussalo-Rauhamaa, Helsinki University Central Hospital,
Departments of Dermatology and Allergic diseases, e-mail: Mussalo@hotmail.com
Anna-Liisa Pasanen, University of Kuopio, Department of Environmental
Sciences, P.O. Box 1627, 70211 Kuopio, Finland, tel. +358-17-163 580, e-mail:
AnnaL.Pasanen@uku.fi
Jukka-Pekka Kasanen, University of Kuopio, Department of Environmental
Sciences, tel. +358-17-163 220, e-mail: Kasanen@uku.fi
Pertti Pasanen, University of Kuopio, Department of Environmental Sciences,
P.O. Box 1627 70211 Kuopio, Finland, tel. +358-17-163 157, e-mail: Pertti.Pasanen@uku.fi
Pentti Kalliokoski, University of Kuopio, Department of Environmental
Sciences, P.O. Box 1627 70211 Kuopio, Finland, tel. +358-17-163 150, e-mail:
Pentti.Kalliokoski@uku.fi
Financing SYTTY organization: Tekes, The Ministry of Environment
Funding from SYTTY / Total funding of project (€): 356693
/ 587898
Person-months of work funded by SYTTY / Total person-months of work:
70,7 / 77,7
KEY WORDS: Indoor air, material emissions, VOC, sensory evaluation,
statistical correlations, irritation, mouse bioassay, modelling
EXTENDED ABSTRACT
1 Introduction
Task I: This task was aimed at establishing measurable criteria for healthy and comfortable in indoor air and finding out the connection between indoor air quality in a space, perceived comfort and diagnosed health effects. The indoor air quality was measured with methods used today in the Finnish classification, but complementary new methods were applied and tested for their relevance in attaining a better coverage of the spectrum of different chemical substances in indoor air. The indoor air and material emission data collected in different tasks of the project were saved and combined with respective health, odour threshold and other relevant data in VTT’s material emission and indoor air databases. These databases contain at this moment detailed information on over 1500 materials and 1100 indoor spaces into which 250 indoor environments have been added from the EU Expolis project. This data was used as additional information in drafting conclusions and recommendations for the improvement of characterising indoor air quality and the classification procedure.
Task II: The aim of the second task was to develop procedures to evaluate the irritating and odorous chemical compounds of material emissions and the perceived air quality. The causative relationships between sensory assessment method used in the present Finnish Classification of Building Materials, olfactometry and emission measurements in chemical terms were determined.
Another objectives of this project was to investigate irritation properties of building material emissions and chemical mixtures by the mouse bioassay, to clarify the indicator value of human sensory evaluations to estimate irritancy of building material emissions, to study an impact of ageing of materials on odour and irritation responses, and finally to develop a model for estimating irritancy of chemical mixtures based on physico-chemical properties and previous knowledge on irritation potency of individual chemicals without animal experiment.
Task III: Studying the emissions from indoor materials and their sorption behaviour with a variety of materials in an indoor space.
2 Methods
Task I. Building related symptoms and indoor air quality
The health and comprehensive indoor air data were collected from subjects,
which were chosen among the patients treated in Helsinki University Central
Hospital because of building related symptoms. The number of the selected
families was 120. Additionally 30 control families were randomly
selected from Helsinki area. All participants were interviewed for their
residential conditions and any building related problems, such like odour
annoyances, using modified Örebo and Tuohilampi questionnaires. Basically
the questions consisted of four parts: 1) living environment, which describes
e.g. building type, building year, ventilation system, used interior materials,
water damages etc., 2) profiling the patient/control case; the questions
asked of smoking habit, pets, sex, age, living habits, used detergents
in cleaning etc., 3) living related symptoms, which characterise the conditions
in homes and different symptoms related to living in homes, 4) medical
history of the patients, which describes previous and present illnesses
and different allergies based on a diagnosis on a medical doctor. Clinical
data have only been collected from the patients in medical examination.
Indoor climate and quality measurements (volatile organic compounds (VOCs),
ammonia and formaldehyde) were carried out in homes of patients and respective
the control subjects. For collecting more comprehensive data, a new method
was utilised in order to detect very volatile organic compounds (VVOCs).
These very volatile compounds includes many carbonyl compounds with low
molecular weight, which are known to be main reason for-off odours and
irritation compounds as well in indoor air as in material emissions.
Task II. Procedures for material emission control
Chemical mixtures representing existing building materials emissions
were selected to the determination of the odour threshold by an olfactometry
and estimation of irritation responses by the mouse bioassay (ASTM E981-84).
Also some materials with high emissions were tested by the mouse bioassay
over a period of two months. To investigate the modelling of irritation
potency of VOC mixtures, the mixture of two non-reactive chemicals as well
as the mixture with formaldehyde and two non-reactive chemicals with different
concentrations were tested with the bioassay. By comparing irritation responses
of single chemicals to that of their mixtures, the aim was to verify, whether
the irritation potency of chemical mixtures could be estimated based on
its components. If so, the irritation potency of different VOC mixtures
(e.g. material emissions) might be possible to predict based on physico-chemical
properties and previous knowledge on irritancy of individual compounds.
Task III. Sorption phenomena
The third task was focused on a procedure for testing sorption effects
in laboratory scale and material emission together with sorption data is
produced. Very little data of the sorption behaviour between volatile organic
compounds and building materials is currently available. During this examination
sorption phenomena between solid materials and gaseous compounds have been
studied both as a surface effect (adsorption and desorption) and inside
the material (absorption and diffusion). The main aim of these empirical
examinations is collect more data about sorption phenomena, which can be
used for example in considerations of utilisation of sorption phenomena
in Finnish Classification.
3 Results and Discussion
Task I. Results of health data
A wide range of symptoms was reported from patients. Control families
had approximately very few symptoms or not at all. Thus the control selections
was obviously succeeded. The most reported symptoms were stuffy air and
unpleasant odour, different nose symptoms (rhinitis, stuffy nose, irritation,
sinusitis and phlegm), eye symptoms, throat symptoms (headache, eczema,
asthma, hoarseness, dyspnea and cough) and mucous membrane symptoms. Using
the statistical correlations the dominating chemical groups were detected
in cases with above-mentioned symptoms. In concerning the odour dependant
symptoms such as stuffy air and unpleasant odour, the obvious chemical
groups were basically carbonyl compounds (aldehydes, ketones and carboxylic
acids) which are well known odour-causing compounds. In addition
some aromatic compounds took part of a responsibility of off-odour. The
most common compound correlating with a wide range of symptoms was 2-ethyl-1-hexanol,
which was connected to the nose symptoms (in every listed one), throat
symptoms and also mucous membrane symptoms. Another significant compound
was TXIB, which correlated to nose symptoms and eye symptoms. Ammonia had
a statistical connection with mucous membrane symptoms and eye symptoms.
However, terpenes were the dominating group only in the residents reported
of stuffy air. Noteworthy, limonene concentration was much higher in homes
of the control families with no specific symptoms with inhabitants. The
concentration level of single compounds will be determined using e.g. odds
ratio when some variables will not taken into account (sex, age, smoking
habits, allergic nose symptoms).
Task II. Results of determination of detection concentration
Selected material emissions (analysed earlier in VTT; data existing
in VTT’s databases) and VOC mixtures were chosen and sent to UKu for determinations
of irritancy. The emissions of product were classified as follows: 1) glue,
2) flooring material (PVC), 3) paint, 4) levelling agent. The selected
VOC-mixtures were mixtures of accepted material and not-accepted material.
The detection concentrations for the mixtures were determined by VTT. The
concentration of the sample gas flow is increased step by step, i.e. the
dilution is decreased until the test person who functions as a detector,
perceives odour in the sample gas flow. Olfactomat-n works according to
the Forced choice method in which a panel of selected persons are forced
to make a choice out of two or more air flows, one of, which is the diluted
sample, even if no difference observed. From the determina-tions of several
test people the olfactometer ca-lculates the odour concentration (Z50)
in odour units/m3 (O.U./m3). The odour concentra-tion signifies the number
of dilu-tions required in order that only 50 % of the me-mbers of the odour
panel will perceive odour in the sample gas flow. The odour- panel consisted
of personnel of VTT Chemical Technology. The odour threshold of the mixture
was calculated based on the odour concentration. It is possible that odorants
interact or react with each other in a mixture. The odour threshold of
a mixture has to be determined, because the perceived intensity from a
mixture of several odorants can be less than, equal to or greater than
the sum of the individual intensities. The detection concentrations were
as follows: glue (+) (0,001 mg/m3); glue (-) (0,030 mg/m3); PVC (+) (0,040
mg/m3); PVC (-) (0,100 mg/m3); levelling agent (+) (0,005 mg/m3) and levelling
agent (-) (0,010 mg/m3).
Results of determination of irritating potency of VOC-mixtures and
material emissions
The results of the experiments with different chemical mixtures showed
that no clear and unambiguous relationship between the odour threshold,
odour intensity and the irritation response exists. Regarding irritation
responses, especially formaldehyde and some other compounds, like ammonia,
higher alcohols, and compounds with double bonds, were the most dominating
irritants in the mixtures. The mouse bioassay was too insensitive to detect
differences in the irritation potency between emissions from four-week
old materials, and thus the assay is not suitable for routine testing of
materials. A theoretical model for mixtures of so-called non-reactive compounds
already exists, and the results of the animal experiments with two non-reactive
compounds in this project indicated that this model works sufficiently
well in practice. Similarly, preliminary results with mixture of reactive
(formaldehyde) and non-reactive chemicals corresponded predictable outcomes
based on the irritancy of single compounds of the mixture. Thus, it proved
to be possible to develop a preliminary model for estimating the irritation
potency of different VOC mixtures based on physico-chemical properties
and previous knowledge on irritancy of individual compounds.
Results of material emission control
Additionally, a new pilot chamber (5 m3) for simultaneous chemical
and sensory analysis of materials was constructed by VTT. This chamber
operates identically with chemical emission chamber (CEN ENV13419-1), the
airflow fulfilling the requirements both for sensory and chemical analysis
simultaneously. Co-operation with other laboratories has been done (Karolinska
Institutet, Sweden) to find out the good construction of the pilot test
chamber. For choosing the chamber materials, properties of different kinds
of steel and anodised aluminium have been studied in sensory terms since
the background odour is the major requirement for a sensory testing chamber.
Task III. Results of sorption (ended in 2000)
Two different methods have been used for studying sorption behaviour
of the building materials. The cup method have been modified and developed
to be suitable for evaluating the diffusion capacity of VOCs through the
building materials and the chamber method have been used for define adsorption
and desorption capacity of the building materials. The results have been
very significant and characteristic for transmission of VOCs in the building
materials. The fact remains that the sorption behaviour can be very different
for one porous medium or another. Before it is possible to set building
materials into common categories according to their capability of sorption,
more tests should be done.
4 Conclusions
The project will be continued till the end of June 2002 and thus the final conclusions cannot be drawn yet.
Task I. As noticed of the first basic correlation between building related VOCs and symptoms there is many obvious correlation between building based VOCs and symptoms experienced in homes. Calculating the odds ratios will highlight the concentration level of significant compounds connecting to different symptoms.
Task II. Based on the present results it seems to be possible
in the future to develop a computer model for estimating the irritation
potency of different VOC mixtures based on physico-chemical properties
and previous knowledge on irritancy of individual compounds. This model
would be useful for Research and development of new building materials
as well as interpretation of indoor air VOC measurements and their relations
to occupants' complaints
and symptoms.