Britt-Inger Andersson, Finn Englund
Selection of Building Materials
for Healthy Buildings with Special
Reference to Wood Products
Trätek
Britt-Inger Andersson. Finn Englund
SELECTION OF B U I L D I N G M A T E R I A L S FOR H E A L T H Y B U I L D I N G S , W I T H SPECIAL REFERENCE TO W O O D PRODUCTS
Trätek, Rapport I 9411057 ISSN 1 1 0 2 - 1071 ISRN T R Ä T E K - R - - 94/057 - - SE Nyckelord building materials emissions environmental effects
Manuscript previously published in the Proceedings of the 3rd International Conference Healthy Buildings '94, pp. 259-263, International Council for Building Research, Studies and Documentation (CIB) - International Society for Indoor Air Quality and Climate (ISIAQ) - Hungarian Academy of Sciences (HAS), Budapest, 22-25 augusti 1994.
Rapporter från Trätek — Institutet för träteknisk forskning — är kompletta sammanställningar av forskningsresultat eller översikter, utvecklingar
och studier. Publicerade rapporter betecknas med
I eller P och numreras tillsammans med alla ut-gåvor från Trätek i löpande följd.
Citat tillätes om källan anges.
Reports issued hy the Swedish Institute for Wood Technology Research comprise complete accounts for research results, or summaries, sur\eys and
studies. Published reports hear the designation I or P and are numbered in consecutive order together with all the other publications from the Institute. Extracts from the fe.xt max be reproduced provided the source is acknowledged.
Trätek — Institutet för iräteknisk forskning — be-tjänar de fem industrigrenarna sågverk, trämanu-faktur (snickeri-, trähus-, möbel- och övrig träför-ädlande industri), träfiberskivor, spånskivor och ply-wood. Ett avtal om forskning och utveckling mellan industrin och Nutek utgör grunden för verksamheten som utförs med egna, samverkande och externa re-surser. Trätek har forskningsenheter i Stockholm, Jönköping och Skellefteå.
The Swedish Institute for Wood Technology Re-search serves the five branches of the industry: sawmills, manufacturing (joinery, wooden hous-es, furniture and other woodworking plants), fibre board, particle board and plywood. A research and development agreement between the industry and the Swedish National Board for Industrial and Technical Development forms the basis for the Institute's activities. The Institute utilises its own resources as well as those of its collaborators and other outside bodies. Our research units are located in Stockholm. Jönköping and Skellefteå.
Sammanfattning
Materialvalsprocessen i samband med projektering och byggande har förändrats så att andra egenskaper hos byggnadsmaterialen än de rent tekniska har fått ökad tyngd. Förändringen, som fortfarande pågår, innebär framför allt att en större uppmärksamhet ges åt materialens miljöpåverkan, både gällande yttre miljö och inomhusmiljö ( l A Q , Indoor Air Quality). Intres-set ökar för materialens eventuella inverkan på människors hälsa, och speciellt för deras av-givning till luften av flyktiga organiska ämnen (VOC, volatile organic compounds).
Under de senaste åren har mycket intresse och forskning ägnats åt V O C i inomhusluften. I många fall uttrycks de uppmätta ämnenas sammanlagda halt (totalhalten TVOC), oavsett sammansättning. Det bör uppmärksammas att det föreligger stora skillnader mellan olika provtagnings och analysmetoder varför resultat från enskilda studier som regel inte kan j ä m -föras. Begreppet T V O C är inte heller entydigt definierat.
I Sverige är det endast några få byggnadsmaterial som uppmärksammats från ett hälso-perspektiv. De mest kända exemplen är avjämningsmassor (flytspackel) innehållande kasein samt äldre spånskivor med hög avgivning av formaldehyd. För de flesta byggnadsmaterial finns det inte något konstaterat samband med SBS-symptom {Sick Building Syndrome). I en Trätek-studie (Englund, Andersson 1994) ges en överblick av ett stort antal veten-skapliga rapporter och artiklar inom lAQ-området. Emissioner av flyktiga organiska ämnen från trä och träprodukter behandlas bara i ett fåtal vetenskapligt dokumenterade studier. Obe-handlad furu och gran avger främst olika terpener. Uppgifter om typiska emissionsfaktorer för olika träslag och träprodukter saknas i stort sett. Tillgängliga resultat medger inte generalise-rade slutsatser. Emissionsfaktom är i hög grad beroende av bland annat virkets halt av extrak-fivämnen, av den tid som har passerat (dagar, månader, år) efter sågning/hyvling samt av yt-behandling. För träbaserade skivor har emissionsfaktorer mellan 20 och 2400 ng/m h rappor-terats, beroende på bl a skivtyp. ålder, tjocklek samt mätmetodik.
Internationellt standardiseringsarbete pågår beträffande mätmetodik och värdering. För närvarande medger dock inte kunskapsläget att en vetenskapligt grundad allmän klassificering av byggnadsmaterial kan göras med avseende på emissionerna av VOC. Det är inte heller visat vilken nytta en sådan klassificering av byggnadsmaterial skulle ha för folkhälsan.
S E L E C T I O N OF B U I L D I N G M A T E R I A L S FOR H E A L T H Y B U I L D I N G S . W I T H SPECIAL REFERENCE TO W O O D PRODUCTS.
Britt-Inger Andersson, M.Sc. and Finn Englund, Ph.D. The Swedish Institute for Wood Technology Research
Box 5609, S-114 86 Stockholm, Sweden
I N T R O D U C T I O N
The prescribed and practised ways of selecting building materials during the design phase of a building are changing. Several parameters other than strictly technical properties rightfully claim their place in the selection process. Formal requirements are raised for the inclusion of new, and in many cases "softer", selection criteria among those traditionally used.
Environmental issues, including recyclability and energy consumption and other aspects of total life cycle assessments, are becoming of primary concern. Health issues are also brought up as important criteria. The emphasis on the implications of building materials on human health is justified insomuch as the potential impact of a poor choice may become large for the individual inhabitant, as well as for the economically responsible contractor. On the other hand, these health implications are generally much more difficult to elucidate than most
"hard" technical facts.
A few notable examples of materials with an undisputed detrimental influence on the indoor climate have been recorded i n Sweden, such as the casein-containing self-levelling compounds used for a short period in the late 70's and early 80's. For most building materials, though, it has not been possible to state that adverse health effects have appeared in a building. The quantification of such health effects in measurable terms, which is necessary for making direct comparisons, remains an elusive goal. This can also be said about the relative contributions of different materials and products to poor indoor air quality.
DISCUSSION
Regardless of source, building-related illness symptoms or discomfort are due to physical factors of the indoor climate or to air-mediated causative agents. The discussion is thereby somewhat narrowed by setting aside e.g. direct physical contact with building materials. Nevertheless, we are left with a set of different mechanisms that could act separately or in concert. Mould growth, bacterial colonisation, mites and dust are factors that may have a very strong influence, but at least the first-mentioned ones are dependent on unfavourable moisture conditions and construction solutions and can be minimized by good building practices. They
are all associated with maintenance and cleaning routines but to some extent also with the choice of materials.
The most general influence f r o m all kinds of surfaces and materials on the indoor air quality is exerted by their emissions of volatile organic compounds (VOC). During the last years, much interest and quite a lot of efforts have been centered around V O C concentrations, both in field measurements and in smdies of properties of individual materials and products. In many cases V O C are expressed as a sum or total concentration (TVOC).
There are, however, large discrepancies in the methods used for sampling and analysis of VOC. The most common way to express T V O C is to use an integrated area as given by a F I D detector o f a gas chromatograph. One then has to assume a single response factor f o r all compounds, normally that o f toluene, hexane, or decane is used. This is a simplification that can be fully acceptable when similar samples are compared and relative changes in concentrations may be of greater interest than absolute values. When samples with quite different compositions are compared, though, large errors can be introduced, as has been pointed out by many, e.g. Otson and Fellin (1993) and Cottica et al (1993). A more correct method is to quantify individually calibrated components identified in the gas mixture (cf. for instance Molhave and Nielsen 1992 or Seifert 1990). Such analyses are unformnately much more expensive and time-consuming, which for practical purposes prevents them f r o m being used routinely i n indoor air monitoring. Today, the lack of a uniform definition o f T V O C is a great hindrance for comparison of separately published data.
I n all branches of science, many analytical methods previously regarded as too cumbersome to be practically useful, have later gotten new life as the methods are refined and the rapid development of computers grants us with new possibilities. GC-MS may soon prove to have reached a stage where it can be more universally employed, but another obstacle then still remains: Where do we draw the line for how far we must carry our analyses?
Several suggested definitions of the T V O C concept are built on a speciation and quantification of a number of the major components i n the separated mixture. It is hard to overlook the fact, though, that indoor air is normally characterized by being rather complex mixtures, and also minor constituents may be among those chiefly responsible for SBS symptoms. Toxicological experience also points out that combinations rather than single compounds may be most important, as well as short-lived and therefore not easily detected transient reaction products formed in the air. This serves to say that contributions f r o m medical science are indispensable in the future work.
I n a recent survey of the large body of published results f r o m research in the field l A Q (Englund and Andersson 1994), we have reached the conclusion that investigations o f emissions of VOC from wood and wood-based products are far too sparse to f o r m a basis for generalizations. Untreated pine and spruce wood predominantly emits terpenes. The emission rate depends, among other things, on the original content of extractives in the lumber and the time that has passed after machining of the surface. Surface treatments have a strong influence. Reliable data on emission rates are largely missing. The emission factor for wood-based boards is normally found in the range 2 0 - 2 400 fxg/m^ h, depending on e.g. the type of board, its age and thickness.
Data f r o m field measurements
In a study of 44 residential timber-frame single-family houses built in Sweden during the period 1982-1989 by the same contractor, constituents of indoor air, ventilation rates and other climatic parameters were measured (Nilsson, Rosell and Thorstensen 1993). This study
enabled comparisons of houses being relatively concordant in age, construction and choice of materials. This set-up would appear to be ideal for discerning variations o f e.g. V O C concentrations as a function of some characteristic parameters.
The ventilation rates were compared with those recorded at the time when the houses were completed, 3-10 years earlier. On an average the ventilation rates have decreased by 25 %, and as a result a little less than half the houses now fall below the limit o f 0.5 air changes per hour (here corresponding to 0.35 l/s,m-) set by the building code. The T V O C concentrations were found to lie in the range o f approximately 100 to 800 /xg/m-* (average value 300). No clear correlation could be established between TVOC concentrations and ventilation rates, age, or the predominant flooring material.
We have had the opportunity to smdy these data in somewhat greater detail than published. The contents of identified aldehydes and terpenes, respectively, were summed up separately. The observations were pooled into two groups, emanating f r o m houses with ventilation rates above and below 0.5 air changes per hour, respectively. The expected reciprocal relationship of T V O C and ventilation rates could be faintly seen in a total plot of the data. A n analysis of variance, however, revealed that there was no statistically significant difference between the two pooled groups at the 5 % level. The aldehyde and the terpene data of the two groups were even less statistically different. Moreover, the varying proportions o f chosen materials in floors (parquet, PVC, ceramic tiles) and on other surfaces did not show a significant influence on the emissions, nor did the age o f the house.
The statistical basis is admittedly not broad enough to permit far-reaching generalizations, but by and large, these results supports the idea that other emission sources than building materials are of prime importance. At early stages of the life of a house, emissions f r o m building materials may dominate the V O C pattern, but these emissions decline so rapidly during a breaking-in period that they are superseded by other sources (such as ftimimre) before long. The mean T V O C was 300 Mg/^n^ which is lower than, but not very far f r o m , the mean value of 380 /xg/m-* found for 100 single-family dwellings in a larger nation-wide study (Norlén and Andersson 1993). It should be borne in mind that the measurement techniques are not identical.
Even i f great caution must be used in comparing studies where different techniques have been used, some studies performed in other countries have indicated clearly higher VOC levels in indoor air than in Sweden (Wallace et al 1991, Rothweiler et al 1990). On the whole, the VOC levels in Swedish houses are probably generally low.
The declination behaviour o f the emission during the first year after completion was established in a study of new, uninhabited timber-frame single-family houses reported in these conference proceedings (Andersson and Salö 1994). These houses were furnished with a carefully planned selection of standard materials and fittings, designed to promote a good indoor air quality for allergic persons. After initial levels o f T V O C (determined as in the study by Nilsson et al) of 300-400 /xg/m\ they came down to 100-150 p.g/nv' within a year. Also the importance of the inhabitants themselves and their furniture and activities is indicated here. Emissions f r o m wood
As has already been said, there is a demand for a more solid bank of data regarding emissions from wood and wood products. Especially the span of wood samples with different origin and history has to be further studied. We intend to contribute to some small extent to this in the near fumre through both field and laboratory measurements.
conditions of two pine samples, one of pure sapwood and one of pure heartwood. Small chips were placed in a closed vial and equilibrated. A i r (1 ml) was taken with out with a syringe and injected directly onto a gas chromatograph with a D B 5 column where the sample was cryofocused. Mass spectra were taken on a V G 70-250 SE high resolution MS with E I ionization at 70 eV. Only some peaks were identified. The dominating peaks were a-pinene (47 % ) , 3-carene (18 % ) , and p-cymene (8 % ) . Another ten monoterpenes were identified with reasonable certainty, together constituting about 85 % of the total V O C mixture. The same sample was run again after equilibrating the vial at 70 °C. The total concentration had approximately doubled, and now the three dominating peaks were the same but with the percentages of 30, 2 1 , and 9 %, respectively. A slight bias towards compounds in the higher boiling-point range could be seen, as expected. It was mostly a surprise to see how small a difference it was (cf. Terauchi et al 1993). The sapwood sample was dominated by a-pinene, 3-carene, and 1-acetylcyclohexen. This very brief account will be published in more detail with the experiments planned for the future.
Concluding remarks
Even i f the need for standardization is large, it is foreseeable that we will have to deal with a variety o f methods for V O C measurements for a long time ahead. Recently, a few very interesdng approaches for coupling emission data to selection process o f building materials have been published (0ie et al 1993, Nielsen and W o l k o f f 1993). However, it still seems as though the situation is not mamre for a general classification of materials into emission classes. There is not yet enough evidence what the real benefits would be for the indoor air quality.
REFERENCES
[1] Andersson, B-I. and Salö, S. "Timber-frame houses for especially allergic families". Proc. Healthy Buildings '94, CIB-ISIAQ-HAS, Budapest, 1994.
[2] Cottica, D . , Cavallo, D . , Alcini, D . , Schlitt, C. and Maroni, M . "Comparison of two
methods for the determination of total volatile organic compounds". Proc. of Indoor
Air '93, 2, 239-244, 1993.
[3] Englund, F. and Andersson, B - I . "Emissions of volatile compounds from wood and
wood products. A survey of current international research". Swedish Institute for
Wood Technology Research, Stockholm, Report I 94, 1994
[4] Molhave, L . and Nielsen, G . D . "Interpretation and Limitations of the Concept
'Total Volatile Organic Compounds' ( T V O C ) as an Indicator of Human Responses to Exposures of Volatile Organic Compounds (VOC) in Indoor A i r " . Indoor A i r , 2,
65-77, 1992.
15] Nielsen, P.A, Wolkoff, P. "Indeklimamaerkning af byggevarer. Del 1: Beskrivelse
af en prototypeordning", SBI-rapport 232, 1993 and "Indeklimamaerkning af byggevarer. Del 2: Faglig og teknisk dokumentation af en prototypeordning",
[6] Nilsson, I . , Rosell, L . and Thorstensen, E. "Indoor air quality in airtight Swedish
dwellings". Swedish National Testing and Research Institute, SP Report 1993:18, 1993.
(In Swedish)
[7] Norlén, U . and Andersson, K . "Bostadsbeståndets inneklimat". ELIB Report No 7, T N : 3 0 , Statens Institut för Byggnadsforskning, Gävle, 1993. (In Swedish)
[8] 0 i e , L . , Morck, H . I . , Borresen, B . A . , Hersoug, L . G . and Madsen, J.O. "Selection of
building materials for good indoor air quality". Proc. of Indoor A i r '93, 2, 629-633,
1993.
[9] Otson, R. and Fellin, P. " T V O C measurement: Relevance and limitations". Proc. of Indoor A i r '93, 2, 281-286, 1993.
[10] Rothweiler H . , Wäger P., Schlatter C . "Volatile organic compounds and very volatile
organic compounds in new and freshly renovated buildings". Proc. of Indoor A i r '90,
Toronto, 2, 747-752, 1990
[11] Seifert, B. "Regulating Indoor Air". Proc. of Indoor A i r '90, 5, 35-49, 1990. [12] Terauchi, F . , Ohira, T . , Yatagai, M . , Ohgama, T . , A o k i , H . and Suzuki, T.
"Headspace Volatiles from Woods and Essential Oils". Mokuzai Gakkaishi, 39(12).
1431-1438, 1993.
[13] Wallace, L . , Pellizarri, E. and Wendel C . "Total Volatile Organic Concentrations in
2700 Personal, Indoor, and Outdoor Air Samples collected in the U S E P A Team Studies". Indoor A i r , 4, 465-477, 1991.
Detta digitala dokument skapades med anslag från
Stiftelsen Nils och Dorthi Troedssons forskningsfond
Trätek
I N S T I T U T E T FÖR T R Ä T E K N I S K F O R S K N I N G
Box 5609, 114 86 STOCKHOLM
Besöksadress: Drottning Kristinas väg 67 Telefon: 08-762 18 00 Telefax: 08-762 18 01 Åsenvägen 9, 553 31 JÖNKÖPING Telefon: 036-3065 50 Telefax: 036-3065 60 Skeria 2, 931 77 SKELLEFTEÅ Besöksadress: Laboratorgränd 2 Telefon: 0910-65200 Telefax: 0910-65265
