Survey of Chemical Substances in Consumer Products, 64 – Emission of chemical substances from products made of exotic wood

Emission of chemical substances from products made of exotic wood

4 Method

4.1 Initial Qualitative Screening
4.2 Quantitative Determination of the Emission in Climate Chamber
4.3 Determination of Natural Rubber Latex Allergen
4.4 Determination of Content of Fungicide
- 4.4.1 Organic Compounds
- 4.4.2 Elements
4.5 Determination of the Emission of Compounds by Migration into artificial saliva
4.6 Evaluation of the Allergic Potential of the Wood Species
4.7 Principles for Evaluation of Chemical Compounds

As described in paragraph 3.2 ten products were selected for initial qualitative screening of the emission of chemical compounds. Of these 10 products 5 were selected for further quantitative analysis.

In excess of analyses of the emission of chemical compounds to the air, samples of belalu, Albiz(z)ia falcata, have been analysed for the emission of chemical compounds by migration into artificial saliva and samples of rubber tree, Hevea brasiliensis, have been analysed for the emission of chemical compounds by migration into artificial saliva, content of latex allergens and fungicide. Description of these analyses has been treated later in this chapter.

4.1 Initial Qualitative Screening

The qualitative screening is carried out at 40°C, where samples of the product to be examined, are placed in a glass and heated to 40°C for 4 hours. An air sample of 1 ml is taken with a gastight syringe and analysed by capillary column gas chromatography combined with mass spectrometric detection (GC-MS).

The components are identified by comparison of the respective mass spectres with mass spectres from NIST 98 library. The percentage part of the total VOC content is stated as area percentage of the spectre assuming that all detected components have the same response to the same amount. The detection limit is on this theory 5-10 ng/g test material. The degree of accuracy of the relative area percentages constitutes 10-15%.

Brief results of the static headspace analysis are stated in paragraph 5.1. Results in detail are stated in Appendix C.

4.2 Quantitative Determination of the Emission in Climate Chamber

The quantitative determination was carried though by climate chamber measurement according to ENV 13419-1 Building Products. Determination of the emission of volatile organic compounds. Part 1: Emission test chamber method (CEN, 2001).

4.2.1 Preparation of Test Specimens

The test specimens were prepared in size in relation to the chamber volume to obtain the desired material load. A relation of n/L = 1 between air change (n) and material load (L) is used.

4.2.2 Test Conditions

Climate chamber	225± l polished stainless steel
Temperature	23± 0.5°C
Relative humidity	45± 3% RH
Air change in climate chamber	1± 0.05 h^-1
Air velocity	0.15± 0.05 m/s.
Material load	0.225 m²
n/L	1

The test specimens were placed in the climate chamber during the entire testing period.

4.2.3 Measuring Method

The general principle of emission measurements in climate chambers is that the test specimen to be examined is placed in a climate chamber at standard testing conditions. Gases and vapours emitted from the test specimen are mixed with the chamber air. Air samples are taken at fixed intervals and are analysed at different analysis techniques.

In the examination the measurement intervals were fixed at 3, 10 and 28 days.

4.2.4 Chemical Analysis

The emissions from the examined products were collected on Tenax TA, desorbed thermally and subsequently analysed by capillary column gas chromatography combined with mass spectrometric detection, GC-MS-SCAN (29-450 amu-screening analysis) according to ISO/DIS 16000-6.2 (2002). The components were identified by comparison of the respective mass spectres with mass spectres from the NIST 98 library.

Aldehydes were collected on dinitrophenylhydrazine (DNPH) filters and subsequently analysed by liquid chromatography with UV detection (HPLC-UV).

By the analyses quantification has been carried out in relation to the external calibration standards of the detected compound of closely related chemical compounds.

Detection limit for VOC's on Tenax 0.3-1 μg/m³

Detection limit for aldehydes on DNPH 1.2 μg/m³

Inaccuracy of analysis results 10-15 %

Blind values have been analysed for the empty chamber prior to testing, and unexposed tubes have been analysed together with test tubes.

4.3 Determination of Natural Rubber Latex Allergen

In the literature there are innumerable descriptions of allergy towards natural rubber latex extracted from Hevea brasiliensis. Proteins cause the allergy. In the literature there are no studies described, in which the content of natural rubber latex allergens in the wood itself can cause reaction in natural rubber latex allergic persons.

The dining table made of rubber tree, which is one the selected products, is lacquered on legs and upper side of the table top. Samples of the tabletop have been analysed for content of natural rubber latex allergens in the wood itself. Most natural rubber latex allergens are water-soluble and in case they are able to penetrate the surface treatment, they may cause allergic symptoms at contact with the surface.

The tabletop consists of laminboards. Totally 3 test specimens have been sampled from different staves. The specimens have been sampled in the middle of the material by means of a metal drill, so that neither glue nor wax forms part of the bored material. The material was subsequently pulverised in a metal grinder. The sampled test material has thus not been in contact with materials, which might contain/emit natural rubber latex allergens.

The material is extracted with phosphate-buffered salt water in the relation 1 g to 5 ml buffer according to ASTM D5712. Particles have subsequently been removed by centrifuging for minimum 15 minutes at minimum 500G. The supernatants were then tested by an immunological method for each of the allergens Hev b 1, Hev b 3, Hev b 5 and Hev b 6.02 by application of specific antibodies (FIT-kit^TM) (FIT Biotech; Palusou et al., 2002) both undiluted as diluted 1:10, if a high allergen level was expected. The amount of allergen is stated for each allergen in μg/l. The total amount is stated as μg/g test material.

The detection limit of the analyses for the allergens Hev b 1, Hev b 3, Hev b 5 and Hev b 6.02 is given in Table 12.

Table 12 Detection limit for latex-allergens

	Hev b 1	Hev b 3	Hev b 5	Hev b 6.02
Detection limit [μg/l]	1.2	2.3	0.5	0.1

The samples were analyses by FIT Biotech Oyj Plc., Tampere, Finland. The results are stated in paragraph 5.3.

4.4 Determination of Content of Fungicide

Among the selected wood species rubber tree is especially exposed to attack of mould and insects from the wood is felled and until it has been dried. Timber of rubber tree is due to this always treated with a fungicide/insecticide to prevent these attacks.

The treatment takes place shortly after felling and cutting and prior to drying, thereby it is carried out while the wood is still wet. The treatment takes in some cases place by dipping the timber into a container with a liquid containing different active ingredients. In other cases an actual preservative treatment takes place. As the treatment is carried out while the wood is wet, the penetration of fungicide only takes place to a limited depth and will predominantly be on/in the surface.

After drying the wood is often further processed implying that the surface is planed, sanded or smoothed. This process entails that the parts of the wood, which might contain the largest amount of fungicide, are removed. It was, however, deemed relevant to perform analyses of potential content of fungicide in the wood.

Test specimens from more staves of the rubber tree in the tabletop of the dining table were sampled (specimen no. 1, Table 10). The sampling was carried out by means of a metal drill with subsequent pulverisation of the material in a metal grinder.

The sample was analysed for content of numerous organic active ingredients: tebuconazole, propiconazole, tolylfluanid, dichlofluanid, IPBC (3-iodo-2-propynyl-butyl-carbamat) and pentachlorophenol. In excess, the sample was analysed for a number of elements (silver, arsenic, boron, bismuth, cadmium, cobalt, chromium, copper, quicksilver, manganese, nickel, lead, antimony, selenium, tin, thallium, vanadium and zinc).

4.4.1 Organic Compounds

For all analyses, except for pentachlorophenol a weighed amount of test material was extracted with acetone. For analyses to determine content of pentachlorophenol a weighed out amount was extracted with dichlormethan added deuterium marked internal standard of C13-PCP by ultrasound.

The extracts were analysed with capillary gas chromatography combined with mass spectrometry (GC-MS). A double determination was carried out.

Tebuconazole, propiconazole, tolylfluanid, IPBC and pentachlorophenol were quantified in relation to standards of the respective compounds. Dichlofluanid was searched by means of the NIST 98 library.

Detection limit for the compounds is stated in Table 13. The degree of accuracy of the analysis results is approx. 10%.

Table 13 Detection limit for organic components

Component	CAS no.	Detection limit [μg/g]
IPBC	55406-53-6	0.5
Tolylfluanid	731-27-1	0.5
Tebuconazole	80443-41-0	0.35
Propiconazole	60207-90-1	0.35
Dichlofluanid	1085-98-9	0.5
Pentachlorophenol	87-86-5	0.1

The result of the analyses is stated in paragraph 5.4.

4.4.2 Elements

The test material was destroyed by nitric acid in telflon cylinders by microwave heating. Subsequently, the extracts were analysed for content of the elements by ICP-MS and ICP-AES (for boron).

For the analysis for boron the sample was spiked with boron and re-analysed. The retrieval was 104%.

The detection limits for the compounds are stated in Table 14. The degree of accuracy is stated together with the results in paragraph 5.4.1.2.

Table 14 Detection limit for elements

Component	Detection limit [mg/kg]	Component	Detection limit [mg/kg]
Argent (Ag)	0.5	Manganese (Mn)	0.5
Arsenic (As)	0.5	Nickel (Ni)	0.5
Boron (B)	10	Lead (Pb)	0.05
Bismuth(Bi)	0.5	Antimony (Sb)	0.5
Cadmium (Cd)	0.05	Selenium (Se)	0.5
Cobalt (Co)	0.5	Tin (Sn)	0.5
Chromium (Cr)	0.5	Thallium (Tl)	0.05
Copper (Cu)	0.5	Vanadium (V)	0.5
Quicksilver (Hg)	0.05	Zinc (Zn)	1

4.5 Determination of the Emission of Compounds by Migration into artificial saliva

Two products were analysed for emission of chemical compounds by migration into artificial saliva: Lacquered dining table made of rubber tree (Specimen no. 1) and ink treated figure made of belalu (Specimen no. 10). The determinations were carried out by assessing the intake of compounds, which might happen by children sucking and biting at the products.

For the analyses the surface of the test specimens were scraped and pulverised, then 1 g test material was firstly added internal standards and subsequently 25 ml simulated saliva (DIN 53160-1). The sample was then placed in an incubator at 37°C; it was initially shaken for 1 hour and was then left for 1 hour.

The samples were centrifuged and the saliva simulant were extracted by "solid phase extraction" SPE (IST Isolute, C18/ENV). The SPE-tubes were dried by a flow of nitrogen and were eluted with dichlormethan. The dichlormethane evaporated to 250 l and was analysed by GC-MS (SCAN).

4.6 Evaluation of the Allergic Potential of the Wood Species

As a part of the assessment of all 5 wood species sampled for climate chamber analyses, a literature survey was carried out to evaluate the tendency of the wood species to cause allergic reactions from skin or airways.

The literature studies are based on information from numerous textbooks (Woods, 1976; Mitchell, 1979; Hausen, 1981; Lovell, 1993; Bendtzen, 2000; Hausen 2000; Krant and Cohen, 2000) and a systematic survey of "Contact Dermatitis" and searching on Medline.

4.7 Principles for Evaluation of Chemical Compounds

4.7.1 Lowest Concentration of Interest to the Indoor Climate

Health effects caused by the indoor environment are normally very unspecified and may comprise symptoms such as headache, fatigue, mucous membrane irritation and dry skin. In excess malaise caused by odour may occur. These symptoms do not necessarily imply critical health effects, but can be of decisive importance to the general well being when staying indoors.

When evaluating, whether a chemical compound or a product consisting of numerous chemical compounds can constitute a health risk, it would be natural to evaluate the result of epidemiological examinations on the reaction of humans to the exposure in question. These examinations are, unfortunately, rare and they in many cases impossible to carry out. It is, therefore, necessary to use other methods to be able to carry out a toxicological evaluation of compounds emitted from e.g. exotic wood.

In a former examination of emission from wood and wood-based materials, furniture and interior furnishings a method for health evaluation (Jensen et al., 2001) is described. This method has been applied in this project.

The main principles of the health evaluation appear from Figure 1.

Figure 1 Fundamental principles of health evaluation of emissions

Figure 1 Fundamental principles of health evaluation of emissions

The evaluation of the individual compounds found in the emissions by the climate chamber measurements have, when possible, included the following: Toxicological effects (cancer, allergy, congenital malformations, nervous system effects and other health effects) and irritation and odour. As a superior principle the emission of carcinogenic compounds, allergens and feto-toxic compounds from the products examined are considered as undesirable, and it is recommended to avoid import of these articles.

For all compounds an assessment for 1) effects, where no observed effect level is expected (NOEL) expected, 2) effects, where no NOEL effect level is present (e.g. carcinogenic compounds), 3) sensory irritation and odour have been carried out. The principles of this stipulation are thoroughly described in the Guidelines for Air Emission Regulation (The Danish Environmental Protection Agency, 1990).

Information about individual compounds has been retrieved by searching in the databases TOXLINE, RTECS, NIOSHtic, and Medline, and by application of existing criteria documents.

The Danish Environmental Protection Agency has carried out toxicological evaluation of chemical compounds related to air pollution, in drinking water and by soil pollution. The toxicological principles, which the Danish Environmental Protection Agency has used, were in accordance with the principles used in this project in the cases, where the assessment of the Danish Environmental Protection Agency was carried out on basis of the effects of a compound and not on basis of odour. In the cases where a toxicological assessment by the Danish Environmental Protection Agency of a chemical compound with a fixed B-value was available (The Danish Environmental Protection Agency, 1990, 1996) this was used. If the assessed value (LCI-value) deviates from this, it has been stated in the individual compound assessment.

As part of a former report on emission of wood and wood-based materials (The Danish Environmental Protection Agency, 1999) a toxicological assessment was carried out on a part of the measured compounds in this project. For these compounds the former assessment and fixed LCI-value applied, as according to our knowledge, there are no new data on the compounds that might affect the assessment.

4.7.2 Determination of LCI- and S-Values

LCI-values (Lowest Concentration of Interest) are a term, which was introduced in the report on emission from wood and wood-based materials (The Danish Environmental Protection Agency, 1999). We have defined the LCI-value as the lowest concentration of a certain compound, which not – according to our present knowledge – at permanent exposure to the indoor air would imply risk of hazardous effects to the human beings. LCI-values are considered as a quality criterion to the indoor air and not as fixed limit values.

There was only limited information about a lot of the chemical compounds and their effects, which emitted from the exotic wood products or from the surface treatment of the products, the LCI-value determination is, therefore, for a part of the compounds determined on basis of poor knowledge of the effects. This has implied that it has been necessary to work with considerable safety margins when determining the LCI-value.

For most of the emissions irritation was the effect, which was decisive for the determination of the LCI-values. More severe effects were for most compounds not found until concentrations were reached, which were way over the level, which implied irritation.

In more cases LCI-values were determined on basis of analogous considerations. In case data were missing for the compounds, and where the most essential effect was assessed to be irritation, the LCI-values were fixed on basis of reduction of the respiratory frequency by 50% in mice (RD₅₀-value).

The values converted into LCI-values by introduction of a safety factor of 10 to protect especially sensible communities and by calculating with an influence of 24 hours 7 days per week (Nielsen et al., 1997).

One of the limitations of the LCI-values is that they very often are concluded from limited knowledge about the individual compounds. The LIC-values can, therefore, only be used by comparison of materials with a uniform emission profile and with uniform determination of the LCI-values. They can, therefore, only be used as an initial indication of potential comfort and health effects in the indoor environment.

In this report the evaluation for all compounds is according to the same principles except for formaldehyde. Formaldehyde follows WHO's recommendations for an indoor environment value, which is essentially less restrictively determined than for the other individual compounds.

Concentrations of the compounds, which were measured in the climate chamber, were calculated by means of conversion factor so that they became relevant in relation to the indoor environment. The calculation appears from formula 1, where C_m [mg/m³] is the calculated equilibrium concentration in the indoor environment, n is the air change in the indoor environment [times per hour]; V is the volume of the actual room [m³], R_s is the specific emission rate [mg x h^-1 x m²] determined by climate chamber tests and A is the area of the specimen in the actual room in m².

(1)

The calculations were carried out by application of a volume, V, corresponding to a standard room of 17.4 m³ c.f. DS/INF 90 (1994). For all products a material load of 0.4 m²/m³ was used, which corresponds to e.g. a table and 6 chairs or a floor area.

For all compounds identified by the climate chamber measurements a calculation for day 3, 10 and 23 was carried out. The concentration in the standard room, c, for the compound in question was divided by the fixed LCI-value. An S-value was calculated by adding the contribution of c_i/LCI_i for all individual compounds (formula 2).

(2)

In principle this should be done for compounds with comparable effects. For the majority of LCI-values in question the irritative effect was decisive. On basis of this and as the other compounds at the same time only contributed very poorly to the total S-value, we have taken the liberty to make a total addition for all compounds.

The S-values can be used for a quick comparison of emission from the different wood species to select the products, which emit the least. The lower the S-value the more acceptable the emission from the exotic wood and/or its surface treatment. By S-values below 1 no health effects are expected.

4.7.3 Indoor-Relevant Time-Values

The indoor-relevant time-value for a product or material is an expression of the time, which passes until the concentration of all individual, emitted compounds have dropped to 50% for the odour and irritation thresholds of the products. All concentrations have been converted into indoor-relevant concentrations by using standard room conditions, cf. the definition in 4.7.2.

For the majority of the products examined this loading is, however, to be considered as worst-case.

Regarding irritation it is estimated that the effect from more compounds is more extensive than the irritative contribution for the individual compound. In case of more compounds with irritative effects, requirements for the sum of irritative compounds in the emission are made (formula 3).

(3)

C_i/CL_i states the concentration of the i'th compound in relation to the acceptable concentration of the i'th compound, CLi, in the indoor climate. CL is calculated as 50% of the irritation threshold for the individual compound.

The sum is determined by adding the contribution of the indoor-relevant concentration divided by the CL-value for all individual compounds with irritative effect in the emission (formula 3).

The indoor-relevant time-value is normally based on both chemical determination and sensory evaluation of the emission. In this project the indoor-relevant time-value of a product is, however, solely based on chemical measurements of the emission in the climate chamber.

The indoor-relevant time-value stated in days, which should be as low as possible, is a direct expression of how long time passes from installation of a product until the emission from the product no longer is expected to cause odour or irritation of mucous membranes in eyes, nose and upper airways.

Application of indoor-relevant time-values enables - like LCI-values - indoor-related comparison of materials and products.

The most extensive limitation is that threshold values for odour and irritation only are available for a limited number of compounds, and that there are distinct deviations between the published threshold values. Odour does not signify that the emission causes health effects, just like no odour do not imply that the emission does not cause health effects.

Indoor-relevant time-values form the basis of indoor climate labelling of building products, furniture and interior furnishings (Danish Society of Indoor Climate, 2003).