Emission of Volatile Organic Compounds from Wood and Wood-Based Materials

3. Emission Testing

3.1 Experimental Investigations
3.2 Results
3.2.1 Qualitative Screening
3.2.2 Quantitative Emission Chamber Measurements

The emission testing of wood and wood-based materials is based on the determination of all the chemical substances in the emissions to evaluate the emissions from the individual materials by comfort and health evaluation and to be able to make proposals for product modifications/substitutions.

3.1 Experimental Investigations

As basis of the planning of the chemical analyses a theoretical evaluation was carried out to presume which substances based on knowledge on the enclosed materials (wood, glue, oils and lacquers) in principle might be expected to be emitted. By this evaluation special weight was laid on substances, which have carcinogenic, reprotoxic, immunologic or neurotoxic effects according to lists from the Danish working environment authorities, chapter 4. Substances, which were selected on this basis, appear from "All chemicals List", cf. chapter 5 and Appendix 5.

The selected wood and wood-based materials according to clause 2.6, were investigated by an initial qualitative screening analysis designed to lay down an analysis programme for the quantitative test chamber measurements for each wood-based material.

The qualitative screening was carried out by headspace analysis, as the material to be examined was placed in a diffusion tight bag with a relatively low amount of clean air and was heated to 120°C for 1 hour. An air sample of 1 ml was sampled with a gas tight syringe without using a collection medium and analysed by headspace by capillary column gas chromatography combined with mass spectrometric detection (GS-MS) cf. Appendix 6.

By the initial headspace analyses it was possible to identify all the substances, which might emit from wood and wood-based materials (according to "All Chemicals List") if they were present except from the substances: Carbendazim and tetramethylthiuramsulfid (fungicide).

The quantitative determinations were carried out by emission chamber measurement according to the standard test method for determination of emission of volatile organic compounds especially adjusted to wood-based products, cf. Appendix 3.

The method has been co-ordinated according to pr ENV 717-1: "Wood-based panels. Determination of formaldehyde release. Part 1: formaldehyde release by the chamber reference method", April 1996 and proposal pre-sently subject to preparation within CEN/TC 264/WG7: "Building products. Determination of Volatile Compounds. Emission Test Chamber Method", October 1995 - March 1997.

The method has furthermore been co-ordinated between the Nordic wood research institutes and the Nordic wood industry (Larsen, A., 1998).

Generally, the test method has aimed at being close to practice and to reflect the use of the materials and products in buildings.

The size of the test specimens has been adjusted according to the chamber volume to obtain the desired material load. A relation of n/L=1 between the air exchange rate (n) and the material load (L) has been applied. By n/L=1 the same value of emission rate [m g/ m²× h] and of the concentration [m g/ m³] is obtained.

Test specimens of wood-based panels were edge sealed prior to the testing.

The moisture content was measured in all test specimens before and after testing. Moisture content, dimensions etc. of the individual materials appear from the test specimen description in Appendix 6.

Test conditions:

The test specimens have been stored at standard test conditions during the entire test period.

Emission chamber measurements have been carried out in 225 l test chambers made of stainless steel. The general principle of emission measurements in emission chambers is that the test specimen to be examined, are placed in a test chamber under standard conditions. Gases and vapours emitted from the test specimen are then mixed with the chamber air. Air samples are sampled at fixed times and are analysed by chemical analysis techniques.

In this investigation the measurement times were fixed at:

The analysis programme for the quantitative analyses comprised determination of all individual volatile organic compounds, VOC’s, of the emissions.

The emissions from all the investigated wood-based materials have been analysed by capillary column gas chromatography combined with mass spectrometric detection (GC-MS) cf. Appendix 6 and by aldehyde analysis by means of dinitrophenylhydrazin tubes, acetonitril extraction and subsequent liquid chromatography on equipment with diode detector, HPLC-DNPH, (Possanzini, M., 1995).

Determination of formaldehyde by photometry by the acetylacetone method according to prENV 717-1 was carried out as part of all tests of wood-based panels.

Quantitative gas chromatographic analyses have been carried out by collecting gases and vapours on adsorption tubes of charcoal and tenax and extracted by solvent desorption with carbondisulfid and diethylether with isotop labelled internal standard. The eluates were thereafter analysed by GC-MS.

By all analyses the quantifications were carried out according to calibration standards of the detected substance or closely related chemical substances. Totally, 39 substances (comprising 9 aldehydes, 1 ketone, 6 alcohols, 5 esters, 5 glycols and glycol esters, 5 hydrocarbons and 8 terpenes and terpene-alcohols) were used for quantification of the emission collected on charcoal and tenax tubes.

Sensory determination of the perception of odour was carried out on 3 of the 23 selected materials: spruce , beech veneered particleboard with polyvinylacetate glue for veneering and UV-curing lacquer on beech veneered par-ticleboard. The sensory determinations have been carried out on materials of different degrees of complexity, as they in this investigation are directory only and primary serve to control the evaluation of the chemical analysis according to odour threshold values.

The sensory determination was carried out by emission chamber testing (Danish Society of Indoor Climate, Standard Test Method, 1994) in a Climpaq chamber with an odour funnel diffuser allowing sensory evaluation of the odour perception. By this method the perception of the air quality of the air that has been into contact with the test material and the air from an empty reference chamber is evaluated by an untrained panel of approx. 20 persons reflecting the composition of the society. The odour perception is evaluated regarding intensity (scale from no odour to overwhelming odour) and acceptability (scale from clearly acceptable to clearly unacceptable). The evaluations are carried out to the time 1 day and by a material load of approx. 0.12 m² surface per m³ room air.

3.2 Results

The emission results for the investigated wood and wood-based materials are stated generally for the qualitative screenings in clause 3.2.1 and the quantitative emission chamber measurements including the directory sensory evaluations in clause 3.2.2. Results in detail are tabled in Appendix 6.

With a view to the health and comfort evaluations the concentrations measured in an emission chamber were converted into the concentrations to which people are exposed in the indoor air.

Standard Room

A standard room generally used for calculation of the indoor-relevant concentrations reflects the critical case, as it has a relatively large surface in relation to the volume of the room (Danish Standards Association DS/INF 90, 1994). For example a flooring material makes up 7 m² of the standard room of 17.4 m³, and a material for lining all the surfaces of the room makes up 38 m² in the standard room of 17.4 m³.

A part of a piece of furniture, e.g. edges of a table top constitutes as little as 0.1 m² or less.

Material Load

In the result tables cf. Appendix 6, a standard room concentration of the broad area interval of 0.1 - 38 m² corresponding to material loadings of 0.06-2.2 m²/m³ is stated. For most of the examined wood-based materials the whole of this interval is not relevant.

Solid ash, beech, oak, wood-based panels and beech veneered particleboards are examples of materials that in practice are seldom used without surface treatment. It appears a.o. from material prepared by the Ministry of Housing and Building in connection with the latest revision of the Danish Building Code (regarding the requirements of Danish Control Organization for Wood-Based Panels to formaldehyde content in wood-based panels) that wood-based panels in the smallest room in a house constitute up till 0.42 m²/m³. Examples of usual material load for wood and wood-based material appear from Table 3.1. In connection with evaluation according to clause 5.3.2, Table 5.3, a need for measurement in excess of 28 days is seen.

Table 3.1

Examples of Material/Product Load in Buildings

3.2.1 Qualitative Screening

By the initial screenings by head space analysis of the 23 selected materials 144 different chemical substances were identified. The 144 substances were divided as follows:

If the identification of headspace of solid wood and uncoated wood-based panels are considered alone, most of the identified substances were: aldehydes and terpenes, it was likewise these types that formed part of most of the emissions.

3.2.2 Quantitative Emission Chamber Measurements

By the quantitative emission chamber measurements of the emissions from the 23 investigated wood and wood-based materials 84 different chemical substances were quantified cf. the result tables ranked according to chemical type of substance in Appendix 6 and "Project Specific List" in Appendix 5. List of Synonyms, which likewise comprises emissions quantified by emission chamber measurements of the 23 examined wood and wood-based materials, is given in Appendix 10.

The measured emissions can not be considered adequate for the investigated type of material nor as generally representative for emissions from wood and wood-based materials.

The predominant emissions from untreated wood and untreated wood-based panels were aldehydes and terpenes as by the headspace analyses.

In the following the results from the quantitative test chamber measurements are illustrated in bar charts. It should be noted that the scales of the emission rate varies in the different bar charts.

Solid Wood
For the investigated types of solid wood:

Ash, beech and oak
Spruce, narrow growth rings
Pine, northern Finland, approx. 84% of heartwood share
Pine, northern Finland, approx. 96% of sapwood share
Pine, southern Sweden, approx. 67% of heartwood share
Pine, southern Sweden, approx. 98% of sapwood share

the emissions of aldehydes and terpenes (expressed as emission rates at the three measurement times) are shown on page 27.

The predominant emissions from solid wood were aldehydes, above all acetaldehyde, propanal, butanal, pentanal and hexanal, as well as the ketone acetone. The aldehydes varied in amount in the different materials.

Very low indoor-relevant concentrations were found for solid ash, beech and oak when calculating at a material load of 0.4 m²/m³ e.g. corresponding to a flooring or a table and six chairs.

Solid pine gave considerably larger emissions than the other tested solid wood species including spruce. In the emission from pine also terpenes primarily a-pinene, 3-carene and limonene were found in large emission rates. The four pine variants (that vary what regards origin (northern Finland and southern Sweden) and content of heart- and sapwood) showed considerable differences in emission rates of especially 3-carene and a-pinene.

The broken line on bars for the terpene emission from solid pine states that over exposure of adsorption tubes has taken place and that the emission rates to the time 4 and 11 days are minimum emission rates.

Wood-Based Panels
For the six investigated wood-based panels:

the aldehyde emissions (expressed as emission rate at the three measurement times) are shown on page 28.

The predominant emissions from untreated wood-based panels were aldehydes, mainly acetaldehyde, propanal, butanal, pentanal and hexanal, as well as the ketone acetone. The aldehydes varied in amount in the different materials. Formaldehyde was the predominant individual emittant in the urea-formaldehyde glued wood-based panels.

The emissions from the investigated wood-based panels varied as expected dependent on the applied glues.

Veneered Particleboard
For the two investigated examples of beech veneered particleboard:

	Beech veneered particleboard, veneered with PVAc-glue
	Beech veneered particleboard, veneered with UF-glue

and the materials forming part: beech and particleboard.

the predominant emissions of aldehydes and benzylalcohols (expressed as emission rate at the three measurement times) are shown on page 28.

The emissions from the investigated veneered wood-based materials reflected the different glues used by veneering and consisted mainly of aldehydes and benzylalcohol from the polyvinylacetate glue. In the emission from beech veneered particleboard glued with urea-formaldehyde-glue considerable amounts of formaldehyde were seen.

Surface Treated Wood-Based Materials
For the two examined flooring oils:

	Flooring oil based on urethanealkyd/linseed oil applied on solid beech
	Flooring oil based on natural resin/linseed oil applied on solid beech

and solid beech the predominant emissions of aldehydes and hydrocarbons (expressed as emission rate at the three measurement times) are shown on pages 28 and 29.

Diagrams   [Click here]

For the five investigated lacquers :

	Nitrocellulose lacquer on beech veneered particleboard
	UV-curing lacquer on beech veneered particleboard
	Acid-curing lacquer on beech veneered particleboard
	Waterborne acrylic lacquer on beech veneered particleboard
	Polyurethane lacquer on beech veneered particleboard

as well as the beech veneered particleboard (veneered with urea-formaldehyde glue)

the predominant emissions of aldehydes and esters and glycolethres and esters respectively (expressed as emission rate at the three measurement times) are shown on page 29.

The emissions from the investigated treated wood-based materials originated primarily from the oils and the lacquers and comprised mainly alcohols, unsaturated aldehydes, esters, glycolethers and -esters. The emission of most of the aldehydes from the surface treated materials can be traced back to lacquer or oil and wood.

Results from the directory sensory analysis of the odour perception carried out for spruce, beech veneered particleboard with polyvinylacetate-glue for veneering and UV-curing lacquer on beech veneered particleboard confirm the result from the chemical testing. The acceptability varied for the three investigated materials between 0.3 and 0.7 (0 = acceptable and 1 = clearly acceptable) and the intensity between 1.0-1.7 (0 = no odour and 1 = slight odour and 2 = moderate odour) already when measuring the day one. Regarding the results in detail cf. Appendix 6.

Comparison of time-values determined by emission chamber measurement accompanied by chemical analysis and time-values based on emission chamber measurement with sensory evaluation appears from Table 3.2.

Table 3.2
Comparison of Time-Values

*By a material load of 0.12 m² surface per m³ room volume