Ulf Berglund

Olfactory and chemical characterization of indoor air. Towards a psychophysical model for air quality

Abstract The relationship between the odor strength of total air samples and the odor strengths of the constituents was investigated in three field experiments in an office building and a new preschool. The odor strength was scaled by magnitude estimation according to a master scale principle which results in comparable values for the total and the constituent odors. Between 60 and 120 chemical components were detected by GC/FID in the indoor air samples ( N = 66). Most (81%) of the detected components in an air sample were odorous, even though most of them were of the low concentrations. By a method of pattern analysis, chemical as well as odor patterns of indoor air were found to be characteristics of different buildings. From the odor patterns (POG), the “odor print” of the outdoor air associated with the buildings was also recognized in the indoor air. Thus, the “odor print” of an air sample is different from its “chemical print”. A model was found that predicts the overall odor strength of an air sample from the number of FID-detected components most frequently reported to have a strong odor.

Joint representation of physical locations and volatile organic compounds in indoor air from a healthy and a sick building

Abstract Many modern, energy-efficient buildings have been labeled “sick”, in view of the fact that their occupants display an unusually large number of sensory symptoms (e.g. eye irritation, dry skin and perception of persistent odors). Air samples were taken from locations (17) in two Stockholm preschools, one considered healthy, the other, sick. The samples (170) were analyzed by gas chromatography/mass spectrometry (GC/MS) and 33 volatile organic compounds were identified and quantified by GC peak heights. The relationship between chemicals and locations in the two preschools was represented spatially by correspondence analysis and other statistical techniques. The analyses clearly distinguish among the buildings and among locations within each building, including whether the sample was collected outdoors or indoors. Within the sick preschool, concentrations of light aromatic hydrocarbons increased with distance from the air supply, but there was no similar gradient in the healthy preschool. The chief results obtained by the comparatively new method of correspondence analysis are supported further by a discriminant analysis and a principal components analysis.

Separate and joint scaling of perceived odor intensity ofn-butanol and hydrogen sulfide

To determine which combination of scaling method and experimental procedure is optimal for measuring perceived odor intensity, the intensity perception of one odorous substance was scaled both separately and together with a second odorous substance in the same session. The methods used were magnitude estimation with and without a standard reference and cross-modal matching with finger span. The results show that separate and joint scalings ofn-butanol and H2S give identical scales, regardless of scaling method. Different results are obtained in magnitude estimations with a homoquality, a heteroquality, and no standard. From these results, one would not expect to find a single true power function independent of method. Cross-modal matching with finger span may be the best choice for odor intensity scaling, since it results in the widest response range, thereby giving the best resolution.

Formaldehyde odor and its interaction with the air of a sick building

Abstract Apparently the most typical effects of formaldehyde on humans in a nonindustrial indoor environment are the sensory reactions. This paper reports on the absolute odor threshold and the perceived odor strength of formaldehyde as well as on the interaction effects between formaldehyde and the indoor air of a “sick building” (a detached preschool). The absolute odor threshold of formaldehyde was determined by a modified method of limits with forced choice responses. The odor strength of formaldehyde and the mixtures of formaldehyde and the indoor air was determined by magnitude estimation according to a master scale principle, using pyridine as master stimuli. The individual odor thresholds for formaldehyde were shown to range over two powers of ten and the distribution to be extremely positively skewed. The median value of the individual odor thresholds (ED 50s) was 0.06 mg/m 3 (0.05 ppm). The psychophysical relationship for formaldehyde odor strength is a power function (exponent = 0.76). the results from mixing formaldehyde with indoor air is mainly a relative increase in the perceived odor strength of the stimulus mixture at low concentrations while the perceived odor strength at high concentrations remains largely unchanged.

Symptom patterns as an early warning signal of community health problems

Abstract A theoretical argument is presented to justify the use of symptom patterns as early warning indicators of health deterioration in a community. The basic assumption is that people (individuals) and the physical environment (settings) can be treated as probability density functions along a hypothetical health continuum. In practice, the interaction of people and settings results in a bivariate distribution within which a cutoff is placed by decision makers in order to separate the community at large into diagnostic categories of “healthy” and “sick”. It is assumed further that changes in the mix of settings and individuals over time can be tracked by health officials as a change in the nature of symptom patterns. Such indicators can give valuable information about the relative health of a community, complementing the conventional use of the incidence rate of specific diseases. Several suggestions are made about how such a theory can be realized through the systematic collection of symptom data that are subsequently treated by multivariate analysis.

Distinguishing between healthy and sick preschools by chemical classification

Abstract The Swedish building code of 1975 emphasizes energy conservation and encourages the construction of tightly insulated structures with adequate ventilation systems. Some of the new buildings constructed along these guidelines have been labeled “sick,” because people working in them report an unusual number of health problems—e.g., eye irritation, skin rashes, and fatigue. One possible indicator of whether a building is “healthy” or “sick” may exist in the pattern of chemicals present in the air. This article outlines an approach designed to find sets of chemicals that can be used to separate buildings according to their designation as sick and healthy. Air samples were taken from a healthy and sick preschool and subsequently analyzed by gas chromatography. The prevalence of different chemical substances was determined and compared across 16 different locations from which air samples were obtained, including outdoor, supply, room, and exhaust air. In all, 158 different chemical substances were found; more in the healthy building than in the sick one. Cluster analysis, based on the pattern of presence and absence of chemicals, was able to separate locations within and between the two types of building. A large number of chemicals (approx. one-half the total) were effective in distinguishing among locations. The ten most critical chemicals in this respect were subjected to more complete statistical analysis in order to highlight further possible differences between the buildings. The general approach described may prove useful in recognizing the environmental conditions associated with the “sick building syndrome.”

Research equipment for sensory air quality studies of nonindustrial environments

Abstract A mobile laboratory has been constructed consisting of three units, namely an environment chamber, an olfactometer, and sampling and HVAC equipment. The net yield of stimuli in the exposure hood of the olfactometer showed a high degree of correspondence with theoretical values. Only a small dispersion between concentrations in different points in the hood was shown. When dosing air from a study object into the environment chamber, the rise as well as the fall times are satisfactorily short and should permit the use of efficient and time-saving experimental designs for human whole-body exposures. Also in the chamber no differences in concentrations between different sites were found. There were only small losses in the sampling, dosing and exposure hood systems of the mobile laboratories.

On the Scaling of Annoyance Due to Environmental Factors

A recurrent problem in sociological annoyance surveys is to scale the environmental variables quantitatively as they are perceived. Data from two surveys on malodors and noise are discussed within a framework of scaling theory. A Thurstonian scaling technique was applied to the category annoyance data. In the investigation of malodors a rather stable picture of the results of the annoyance reports was obtained independently of the assumptions made in data treatment. On the other hand, the results of the noise investigation indicate that paired comparisons may furnish a better procedure than verbal category scaling for data collection in surveys. Scales of annoyance from different populations will give different units of measurement, and annoyance scales cannot be compared adequately unless the scales are calibrated. A possible calibration procedure is to introduce a defined psychological unit of measurement into the data.

Can sick buildings be assessed by testing human performance in field experiments

Abstract The present paper is devoted to the Sick Building Syndrome and describes an experiment comparing a diagnosed “sick” with a “healthy” Swedish preschool. The indoor air quality of both buildings were nearly the same and the concentrations of total separated volatile organic compounds were low according to suggested guidelines for indoor air in nonindustrial buildings. Forty-eight previously unexposed subjects were exposed to each of the two buildings for one day, and the effect of the exposure was assessed with a battery of diverse psychological tests (reaction-time, short-term memory, vigilance, and steadiness). Despite a favorable experimental situation of utilizing a building with a record of producing the Syndrome, the results of psychological tests of mental and motor performance, and therefore the answer to the question raised by the title above, were in the negative. This failure raises questions both regarding the choice of subjects (workers or visitors) and experimental methods including the selection of tests, the duration of exposure, and the environmental setting (a natural vs. laboratory setting). Several combinations of experimental method and subjects which must be considered in future research on indoor pollution are discussed.