0302009
nMPiPCDiRir
Jan Ekstedt
Bestämning av kvistgulning
på ytbehandlat trä
som EN-standard
Trätek
I N S T I T U T E T F O R T R A T E K N I S K F O R S K N I N GJanEkstedt
BESTÄMNING AV KVISTGULNING
PÅ YTBEHANDLAT TRÄ SOM EN-STANDARD
Trätek, Rapport P 0302009
ISSN 1102-1071
ISRN TRÄTEK - R — 03/009 - -SE
Nyckelord
coatings
EN-927'7
EN standards
discoloration
knot staining
tannin staining
wood
Stockholm februari 2003
Innehåll
Sid
Förord 3
Bakgrund 3
Industrinytta 4
Genomförande 4
Experimentellt arbete 4
Resultat och diskussion 5
Slutsatser 6
Resultatförmedling 6
Nätverk 6
Litteratur 7
Bilaga 1 9
Bilaga 2 17
Förord
Föreliggande rapport är en sammanställning av de arbeten som utförts inom ramen för
pro-jektet " Bestämning av kvistgulning på ytbehandlat trä som EN-standard" som finansierats
genom forskningsbidrag från NUTEK, Träforsk och Svenskt Trä (Projektnr 99055).
Resultaten från projektet har mer i detalj redovisats i följande rapporter:
Ekstedt, J., Bestämning av missfårgning av kvistar - Determination of discolouration of paints
on wood due to resin exudation fi-om knots. Trätek Rapport P 0006009.
Forslund, M., Gardeback, A. & von Schéele, C. Kvistgulning. Jämförelse mellan två
appara-ter för accelererad åldring. Trätek Rapport P 0007011 och L 0007012. Rapport HK
Trätekno-logi KTH (2000).
Vetter, S. & Ekstedt, J. Determination of discoloration of paints on wood due to resin
exudation fi"om knots. Report COST E 18 "Short Term Scientific Mission"(2001).
CEN/TC139AVG2 document N 303 Draft text for EN927-7 "Paints and vamishes. Coating
materials and coating systems for exterior wood. Assessment of knot and tannin stain
resistance of wood coatings" (2002). Bilaga 1.
Suttie E & Ekstedt J "Evaluation of a method to determine discolouration of paints on wood
due to resin exudation from knots". Paper 14. 3'^'^ International Woodcoating Congress, Haag,
NL, Oktober 2002. Bilaga 2.
Bakgrund
Vid övergång från lösningsmedelsbuma till vattenbuma färger har problemen med
missfärg-ning av den målade ytan på grund av kådflytaing och kvistgulmissfärg-ning accentuerats, särskilt på
snickerier som fönster och dörrar. I vissa fall har detta omöjliggjort användande av
miljö-mässigt fördelaktigare vattenbuma färger.
Träindustrin (fönster- och dörrtillverkama) och fargindustrin i Sverige uppdrog åt Trätek att ta
fram en testmetod som snabbt kunde bedöma färgsystemens kådflytnings- och
guhiingshäm-mande egenskaper. Denna testmetod har nu blivit en inofficiell branschstandard för de
nor-diska ländema. Den används bl a i P-märkningsreglema för fönster.
Metoden bygger på en accelererad åldring av ytbehandlade testpaneler med kvist. Beroende
på färgsystemens kådflytnings- och gulningshämmande egenskaper kommer testytan vid
kvisten att missfargas mer eller mindre. Denna missfårgning mäts instrumentellt med hjälp av
standardiserade fargmätningsinstrument. Man mäter skillnaden i kulör mellan färgen på
kvisten och färgen bredvid kvisten. Skilhiaden uttrycks i ett s k AE-värde. Genom att
under-söka hur stor skillnad i gulhet som kan accepteras av konsumenter kan ett max-värde på AE
fastställas. Man kan åsätta ett maxvärde för olika produkter, exvis fönster, dörrar, fasader som
ett kvalitetsmått. För fönster har man i den svenska fönsterindustrin och fargindustrin i stort
sätt enats om att AE-värde för en fönsterfarg skall understiga värdet 2 (två) för att vara
accep-tabelt för fönsterindustrin. Ett problem har tidigare varit att de mer miljöanpassade
ytbehand-lingama har haft svårt att komma under AE=2.
I den europeiska standardiseringskommitéen CEN TC 139AVG 2 "Coating materials and
coating systems for exterior wood" togs ett beslut 1999-10-15(CEN/TC 139 doc N 463) att
införa en testmetod för bedömning av missfargning på målade träytor (Assessment of knot and
tannin stain resistance of wood coatings) i europastandarden EN 927, med följande hålltider:
Maj 2001 "Working document circulated to Technical body" (stage 32),
Maj 2002 "Document available for CEN/CLC enquiry" (stage 40) och
Maj 2003 "Document available for formal vote" (stage 49).
Arbetet uppdrogs åt TC 139AVG 2 att ta fram standarden. Vid dess möte 1999-10-13 och
2000-05-11—12 beslöts att som underlag för den kommande standarden utgå från Träteks
kvistgulningsstandard med vissa modifieringar. Det praktiska arbetet uppdrogs åt en
arbets-grupp bestående av representanter från England (BRE), Holland (Akzo Nobel), Finland
(VTT), Danmark (Largo A/S), Schweiz (EMPA) och Sverige (Trätek), med Trätek som
koordinator.
Föreliggande projekt omfattar arbete med anpassning och j^erligare utveckling av
test-metoden.
Inom den amerikanska standardiseringen (ASTM) pågår ett arbete med en motsvarande
standard. Den föreslagna standarden har visats ha för dålig reproducerbarhet och
repeter-barhet.
Industrinytta
Projektets mål är att skapa en testmetodik som på ett invändningsfiitt sätt kan beskriva
egen-skaper hos ytbehandlingar som är avsedda att förhindra att träkomponenter missfargas p g a
kvistgulning och/eller missfargning av extraktivämnen ("tannin staining"). Färgindustrins
påståenden om färgproduktemas egenskaper måste kunna utvärderas av fönster- och
dörrtill-verkare för att dessa i sin tur kan lämna garantier till sina konsimienter. Det standardiserade
testförfarandet har även visat sig påskynda utvecklingen av de miljömässigt mer fördelaktiga
vattenburna fönsterfargssystemen.
Genomförande
Projektet har genomförts i följande tre delmoment.
1. Inventering av de europeiska ländemas önskemål på standardens innehåll och omfattning.
2. Praktiska försök och utvärdering av ev. förändringar/förbättringar av testmetoden.
3. Anpassning av provningsmetodiken och mätförfarandet till europeisk norm.
Experimentellt arbete
Det experimentella arbetet i projektet har utförts i huvudsak på Trätek, Stockholm.
Under-sökning av metodens reproducerbarhet och repeterbarhet har utförts i ett samarbete mellan
Trätek och Building Research Establishment (BRE) i England. Den engelska delen har
finansierats av "Construction Industry Directorate of the UK Government's Department of
Trade and Industry*' (DTI).
I det svensk/engelska samarbetet utfördes mätningar på testpaneler som tillverkats och
yt-behandlats på Trätek. Två olika ytbehandlingar testades. En ytbehandling med erkänt bra
skyddande egenskaper och en ytbehandling med väsentligt sämre prestanda. Dessa färger
målades på furupaneler med fiisk kvist. Ett stort antal replikat togs fi"am och analyserades for
bedömning av precision, reproducerbarhet och repeterbarhet. Paneler testades på Trätek och
på BRE med respektive instituts apparater, mätinstrument och operatörer.
Resultat och diskussion
1. Inventering av de europeiska ländemas önskemål på standardens innehåll.
Inventering av de olika ländemas önskemål gjordes genom kontakter med de olika
representanterna i TC 139/WG 2. Industrins önskemål togs fi-am genom forfi-ågningar
hos fönster- och dörrtillverkare och representanter for fargindustrin. I detta
samman-hang uppvisade såväl träindustrin som fargindustrin stor överensstämmelse om behovet
av och önskemålen om iimehållet i en europeisk standard. Två områden for forbättring
fi-amstod som starka önskemål.
• Byte av utrustning for accelererad åldring
Vid testning enligt Trätek-standarden används Atlas Weather-o-Meter^** (WoM) for
accelererad åldring och utveckling av missfargningen på testpanelema. Derma
apparatur är dyrbar och önskemål om billigare, mer vanligt forekorrmiande,
appara-tur var ett klart uttalat önskemål fi"ån industrin.
• Ökat operatörsoberoende i mätforfarandet
Det fi-amkom att det spektrofotometriska mätforfarandet måste standardiseras än
mer for att öka repeterbarheten och reproducerbarheten i mätningama. För
när-varande är förfarandet operatörsberoende i alltför stor utsträckning.
2. Praktiska försök och utvärdering av forändringar och forbättringar av testmetoden.
Med utgångspunkt fi-ån de önskemål som fi-amkom från standardiseringskonmiittéens
deltagare och från industrirepresentanter gjordes en jämförelse mellan två olika
apparater for accelererad åldring, QUV® (Q-Panel Co, USA) och Atlas
Weather-Ometer® Ci65 (Atlas Electric Devices Co, Chicago, USA). Projektet visar att en
över-gång till QUV® kan göras utan större olägenheter. Underlaget for ställningstagandet är i
detalj redovisat i rapporterna av Forslund et al (2000) och Suttie och Ekstedt (2002).
Industrin har välkonmat övergången till mindre kostnadskrävande utrustning och lägre
analyskostnader.
I den ursprungliga testmetoden (Ekstedt, J., Bestämning av missfärgning av kvistar
(2000)) används den normalt forekonmiande spektrofotometriska mätforfarandet ISO
7724/1-1984.1 detta mätforfarande uttrycks kvistgulningen i AE beräknat från
farg-koordinatema L, a och b. Mätning av gulheten kräver relativt dyra mätinstrument och är
även operatörsberoende. En möjlig ersättning till detta förfarande är att använda
digital-kamera och att bildema av '^kvistgulningen" analyseras i bildbehandlingsprogram där
kulören kan anges i CIELAB koordinater. Dessa kan sedan räknas om till L, a och b
koordinater och vidare till ett AE-värde. Utveckling av en metod med digitalkamera och
bildbehandlingsprogram har dock inte kunnat inrymmas i projektet. Ett sådant arbete
måste finansieras genom andra projekt. Ett förberedande diskussion om ett sådant
pro-jekt har tagits med Wilhelm Klauditz Institut (WKI) i Braunschweig, Tyskland.
Utvärdering av reproducerbarhet, repeterbarhet och precision har utförts på Trätek och
vid brittiska BRE. Resultaten har redovisas vid "3"* International Woodcoating
Congress". E. Suttie & J Ekstedt "Evaluation of a method to determine discolouration of
paints on wood due to resin exudation fi-om knots", Bilaga 2.
3. Anpassning av provningsmetodiken och mätförfarandet till europeisk norm.
De ändringar i testprocedur som har varit nödvändiga för att Träteks testmetod skall
kunna fastställas som EN-standard har genomförts baserat på de vetenskapliga resultat
som fi-amkommit i projektet. Ett slutgiltigt förslag kommer att presenteras vid TC
139AVG 2's möte i Braunschweig 23-24 juni 2003.
Slutsatser
En övergång från Atlas Weather-Ometer till QUV som apparat för accelererad utveckling av
kvistgulningen på testpanelema är möjlig. Dock är inte exponeringen identisk så att resultat
från tester med Atlas WoM direkt kan jämföras med resultat från tester med QUV. Rankingen
mellan bra, mindre bra och dåliga system är dock likartad mellan Atlas WoM och QUV. På
grund av det stora intresset för den billigare QUV apparaten som utrustning är en övergång till
QUV mer eller mindre en nödvändighet.
Vid de testningar som utfördes på Trätek och BRE kunde konstateras att vid "operating
temperatur" 60°C på QUV apparaten uppvisade även det bra systemet en viss kvistgulning.
Vid diskussioner med tillverkaren Q-Panel Co konstaterades att man kan sänka temperaturen
ner mot 40°C. Detta skulle medföra en lägre grad av transport av extraktivämnen (=
kvist-gulning). För vidareutveckling av standarden föreslås att temperaturen sätts till 40°C i stället
för 60°C.
Resultatförmedling
Resultaten i projektet har kontinuerligt förmedlats till CEN TC 139AVG 2 och den svenska
fönsterindustrin och fargtillverkare.
Nätverk
Ett stort nätverk bestående av, framför allt, företag från olika sektorer i "the wood chain" har
kontinuerligt informerats om de olika momentens framsteg och problem. Det fruktbara
samarbetet har lätt till att framstegen har haft kort väg till implementering. Nedan följer en
lista på aktörer som i större eller mindre utsträckning tagit aktiv del under projektets löptid.
Hyvlerier, byggare, småhusfabrikanter, fönster- och dörrtillverkare.
Sveriges Lantbruksuniversitet, Inst, för Trävetenskap, Uppsala. (SLU)
Building Research Establishment, UK. (BRE)
CEN TC 139/WG 2 "Coating materials and coating systems for exterior wood".
Färgindustrin (Alcro-Beckers; Becker Acroma KB; AKZO Nobel Industrial Coating; Engwall
& Claesson AB; Jotun A/S, Norge; Larco A/S, Danmark; Tikkurila Oy, Finland).
Bindemedelstillverkare (Perstorp AB, Zeneca Resins, Rohm & Haas, DSM, Solutia) samt
Målarmästamas Riksförening.
Litteratur
Ekstedt, J., Bestämning av missfårgning av kvistar - Determination of discolouration of paints
on wood due to resin exudation from knots. Trätek Rapport P 0006009 (2000)
Forslund, M., Gardeback, A. & von Schéele, C. Kvistgulning. Jämförelse mellan två
appara-ter för accelererad åldring.. Trätek Rapport P 0007011 och L 0007012.
Rapport HK Träteknologi KTH (2000)
Vetter, S. & Ekstedt, J. Determination of discoloration of paints on wood due to resin
exudation from knots. Report COST E 18 "Short Term Scientific Mission"(2001).
CEN/TC139/WG2 document N 303 Draft text for prEN927-7 Paints and vamishes. Coating
materials and coating systems for exterior wood. Assessment of knot and tannin stain
resistance of wood coatings" (2002)
Suttie E & Ekstedt J "Evaluation of a method to determine discoloration of paints on wood
due to resin exudation from knots". Paper 14. 3"^ Intemational Woodcoating Congress, Haag,
NL, Oktober 2002.
Ekstedt J "Studies on the barrier properties of exterior wood coatings". Doktorsavhandling,
KTH 2002.
Bilagal.
Assessment of knot and tannin stain
resistance of wood coatings
T C 1 3 9 / W G 2 N 3 0 3
Assessment of knot and tannin stain resistance of
wood coatings
Foreword
This European Standard has been prepared by Technical Committee C E N / T C 139 "Paint and varnishes", the secretariat of which is held by DIN.
This European Standard shall be given the status of a national standard, either by publication of an identical text or by endorsement, at the latest by (month, year), and conflicting national standards shall be withdrawn at the latest by (month, year)
EN 927 consists of the following parts under the general title "Paints and varnishes - Coating systems for exterior wood":
Parti Part 2 Parts Part 4 Parts Parte Part?
Classification and selection Performance specification Natural weathering test
Assessment of the water-vapour permeability Assessment of liquid water permeability
Assessment of knot and tannin staining
According to the C E N / C E N E L E C Internal Regulations, the national standards organisations of the following countries are bound to implement this European Standard: Austria, Belgium, Czech Republic, Denmark, Finland, France, Germany, Greece, Iceland, Ireland, Italy, Luxembourg, Netheriands, Norv/ay, Portugal, Spain, Sweden, Switzeriand and the United Kingdom.
Introduction
The treatment of exterior wood surfaces has both aesthetic and protective functions. A vital purpose of the coating system is to protect the wood against discoloration caused by the presence of wood extractives on the coating derived from the wood material.
This part of EN 927 provides a method for assessment of the discoloration of coating materials applied to exterior wood.
1 Scope
This part of EN 927 specifies a test method for assessing the discoloration of coatings on wood due to resin exudation from knots. The discoloration is measured as a colour difference between the coated surface on the knot and the coated surface beside the knot.
Establishing the magnitude of an acceptable colour difference is not covered by this method.
The colour difference is measured in accordance with ISO 7724.
The method of test specified requires to be completed by the following supplementary information.
a) Type of substrate and surface preparation of substrate.
b) Method of application of test coating to substrate.
c) The total dry film thickness (in pm) of coating system, including method of measurement, and the number of coating layers.
d) Duration and conditions of drying of the coated panels before testing (or conditions of stoving and ageing if applicable).
2 Normative references
ISO/DIS 7724/1 -1997 Paint and varnishes - Colorimetry - Part 1: Principles.
ISO/DIS 7724/2 -1997 Paint and varnishes - Colorimetry - Part 2: Colour measurement.
ISO/DIS 7724/3 -1997 Paint and varnishes - Colorimetry - Part 3: Calculation of colour differences.
ISO 11507 1997 Paint and varnishes Exposure of coatings to artificial weathering -Exposure to fluorescent UV and water.
ASTM G26-92. Standard Practice for Operating Light-Exposure Apparatus (Xenon-Arc Type) With and Without Water for Exposure of Nonmetallic Materials.
EN ISO 15528 Paints, varnishes and raw materials for paints and varnishes -Sampling (ISO 15528:2000)
EN 23 270 (1991) Paint and varnishes and their raw materials - Temperatures and humidities for conditioning and testing. (=ISO 3270:1984).
INSTA 141 Timber - Sawn and planed timber of softwood - Moisture content. (1991)
A coating system, according to this procedure, can consist of one or more separate products.
3 Terms and definitions
For the purposes of this part of EN 927, the following terms and definitions apply:
3.1 Tannin staining
The appearance of discoloration on coated surfaces due to the presence of wood extractives and/or other chromophores derived from substances in the wood.
3.2 Knot staining
The appearance of discoloration on coated surfaces due to the presence of wood extractives and/or other chromophores derived from knots in the wood.
3.3 Wood extractives
Low-molecular wood components soluble in organic solvents or water.
3.4 Sound knot
A sound knot is a knot intergrown to more than VA with the surrounding wood. (Nordic Timber -Grading rules for pine and spruce sawn timber).
4 Test panels
4.1 Wood
The raw material for the test panels shall be battens of pine (Pinus silvestris) free from visible cracks, blue stain and rot damage. The battens shall be dried to moisture content class 18 in accordance to INSTA 141. The drying temperature shall not exceed 70 °C during any part of the drying schedule. After drying, the battens shall be stored in an atmosphere in accordance with EN 23 270 [23 ±2°C och (50 ± 5) % relative humidity] for one (1) month.
From these dried battens test panels are prepared in the following dimension 230 x 70 x 10 mm with at least one sound knot with a diameter at least as large as the measuring aperture of the apparatus for colour measurement. The test panels shall be cut so no part of the test panel contains material deriving closer from the surface of the battens that 10 mm. If test panels are not used within 24 hours they shall be stored below -18 °C.
For each coating system under test select five (5) test panels on a random basis from the available supply.
4.2 Preparation and coating of panels
Apply the coating system using the method specified by the manufacturer with the product or products under test. Unless otherwise specified, dry the coated test panels vertically in air in an atmosphere according to EN 23 270 [23 ± 2 °C och (50 ± 5) % relative humidity].
4.3 Preparation of coated panels
A representative sample of the product to be tested (or of each product in the case of a multicoat system) shall be taken as specified in EN ISO 15528.
5 Apparatus
5.1 Apparatus for accelerated ageing.
Accelerated ageing of the coated test panels shall be performed in an apparatus in accordance with ASTM G26-92, Standard Practice for Operating Light-Exposure Apparatus (Xenon-Arc Type) With and Without Water for Exposure of Nonmetallic Materials. Suitable apparatus is Atlas Weather-o-Meter.
5.1 Apparatus for accelerated ageing. (Alternative)
Accelerated ageing of the coated test panels shall be performed in an apparatus in accordance with ISO 11507 Paint and varnishes - Exposure of coatings to artificial weathering - Exposure to
fluorescent UV and water. Suitable apparatus is QUV (Q-Panel Company).
5.2 Apparatus for colour measurements
The colour shall be measured in CIELAB colour coordinates with an apparatus with illuminant D65 as specified in ISO 7724/2-1997.
6 Procedure
6.1 Exposure (Atlas Weather-o-Meter)
The coated test panels are mounted in the exposure cabinet with the coated surface towards the Xenon-Arc lamp.
The coated test panels are exposed according to the exposure test method A specified in ASTM G 26-92 6.2. (a cycle of 102 minutes of light followed by a cycle of 18 minutes of light and water spray). The coated test panels are exposed continuously according to the test method specified for 72 hours. After 72 hours of exposure, remove the coated test panels from the exposure cabinet, blot lightly to remove any water after which the colour difference is measured as specified in 7.1.
6.1 Exposure (QUV, Q-Panel Company) (Alternative)
The coated test panels are mounted in the exposure cabinet with the coated surface towards the fluorescent lamps.
The coated test panels are exposed according to ISO 11507 but without condensation or water spray. The coated test panels are exposed continously for 72 hours.
After 72 hours of exposure, remove the coated test panels from the exposure and measure the colour difference as specified in 7.1.
7 Measurements
7.1 Measurement of colour difference
The colour difference between the coated surface over the knot and the coated surface beside the knot is measured with an instrument specified in 5.2. Measure the colour in CIELAB colour
coordinates with an apparatus with illuminant D65 as specified in ISO 7724/2-1997.
7.2 Calculation of AE value
Calculation of AE value is done for every single test panel with the method described in ISO 7724/3-1997.
7.3 Statistical evaluation
The "AE value" for a coating system under test is reported as the arithmetric mean value for all separate measurements and the standard deviation of the measurements.
8 Test report
The test report shall contain at least the following information:
a) the type and identification of the coating or coating system under test; b) a reference to this part of EN 927 (EN 927-7);
c) the items of supplementary information referred to in the "Scope and Field of application" in this standard;
d) type and manufacturer of apparatus for accelerated ageing according to 5.1 and type and manufacturer of apparatus for colour measurements specified in 5.2.
e) The result of the test (arithmetic mean value and standard deviation according to 7.3); f) any deviation, by agreement or otherwise, from the test method specified;
g) the date of the test.
Annex (informative)
A 1 . General
Wood material consists of both high-molecular components (cellulose, hemicellulose and lignin) and low-molecular components. The low-molecular components can have lipophilic a s well a s hydrophilic character. These low-molecular compounds are often refered as wood extractives. The material class "wood extractives" include the following classes of compounds: Fats and fatty acids, steryl esters and sterols, terpenoids and waxes (long-chain alcohols and their acid esters and sugars).
During outdoor exposure of wood containing wood extractives, these substances can be mobilised and transported through the coating. Although colouriess, these low-molecular extractives can undergo chemical reactions introducing chromophoric groups in the molecule thus causing discoloration of the coating.
A 2. Wood material
The rather complicated procedure of preparing test panels is done to ensure that the resins in the knots has not cross-linked, or hardend during panel preparation or storage.
A 3. Precision
The amount of discoloration of coatings due to resin exudation from knots is greatly dependent of the type of knot, the amount and type of wood extractives in the knot and the degree of polymerisation of wood extractives during storage of test samples.
In the test procedure five replicates are used. Due to large deviation in the properties of the knots a relative large standard deviation are to be expected. This will, on the other hand, reflect the actual situation when using wood material containing knots in outdoor use.
Bilaga 2.
Suttie, E. och Ekstedt, J.
"Evaluation of a method to determine
discoloration of paints on wood due to
tannin staining from knots.
Evaluation of a method to determine discoloration of paints on
wood due to tannin staining from knots
Ed Suttie* & Jan Ekstedt**
•Centre for Timber Technology & Construction, BRE. Garston, Watford, WD25 9XX, UK **Trätek, Drotting Kristinas väg 67, Stockholm, Sweden
Abstract
The natural exposure and water absorption tests carried out within European Standard EN927 use clear, knot free wood in order to minimise variability in the results. However, knots are common in practice in many coated timber components, and it is widely
acknowledged that the performance of paints can be compromised over knots or timber that possess a high tannin content. As part of the European Standard for evaluating the
performance of coatings on wood a validated method is required for evaluating the discoloration due to tannin staining and resin exudation from knots. CEN/TC139 Working Group 2 has been charged with identifying a reliable method of testing for tannin stain resistance and producing a standard; identified as EN927-7 "Resistance to tannin staining". This paper presents results from an evaluation of a Trätek standard procedure by two laboratories. Results of discoloration are presented for panels containing knots that have been coated with one of two test coatings. Reflectance spectrometry and digital imaging based methods of evaluating discoloration are used and an interpretation of the
comparability of the techniques is provided. The suitability of the method for promoting to WG2 as a standard is presented on the basis of these results.
Background
Tannins and other chromophoric compounds present in wood are naturally occurring and in species such as European softwoods can be found in association with knots. Such
compounds are mobilised by moisture and heat and can migrate to the surface of wood coatings. At the surface the compounds are degraded by the action of UV radiation into coloured breakdown products that appear as a yellow or brown stain on the surface of the wood coating. Blocking, or stain resistant, wood primers are often used as a guard against tannin staining. There Is a need to evaluate the relative ability of paints and primers at preventing tannin staining.
Method
The method used for the preparation of coated panels is based on the method of Trätek (2000) which should be referred to for a more detailed description of the method. However, the weathering protocol followed prEN927-6 using exposure in a QUV weathering device (fluorescent tubes) and in a laboratory window for a 'natural' weathering exposure behind glass.
Scots pine boards (50 x 120 x 3000 mm) were dried to a moisture content of about 16% at temperatures below 70*'C at the sawmill. The boards were then stored in a refrigerator at (-18"C). Before preparation of the test panels (20 x 70 x 230 mm) the boards were conditioned for 1 month at 20''C and 65% RH. The test panels were prepared to contain at least one sound knot with a diameter of at least 20 mm. The panels were then split to produce two matched knot panels, one for testing at each laboratory. At all times between preparation and use the panels were stored in the refrigerator. Each knot was individually identified and mapped.
Five replicate panels were prepared for each of the two test paints for each of the three exposures for each of the laboratories. The two waterborne test paints were supplied by Becker Acroma and included one of known poorer performance at resistance to tannin staining than the other.
Paint A: (2 component high molecular weight epoxy primer with a one component acrylic top coat)
Paint B: (1 component acrylic primer with a one component acrylic top coat) The following panel preparation protocol was followed by both laboratories:
Day 1 Remove samples from refrigerator to controlled conditions (20°C and 65% RH) Day 3 Application of primer
Day 4 Application of top coat, leave to dry and harden
Day 8 Measurement of unexposed panels (day 0), expose panels The three exposure methods were as follows:
1. QUV with spray: Using the exposure cycle in proposed prENV 927-6 (doc. CEN/TC139 WG2 N305) but omitting the first COND step (Step 1)
Step Function Temperature Duration Remark
3 UV 60°C 2,5 hr UVA-340 Lamp output: 0,77 4 SPRAY 0,5 hr 6-7 Ipm, UV lamp off
2. QUV without spray: Continuous UV radiation, UVA-340 Lamp output: 0,77.
3. Window exposure: The samples are exposed approximately 45° to the incident solar radiation in a window facing south-west. These recreates a 'natural' weathering exposure behind glass.
Colour over the knot area and over an area of clear timber was measured after 0 , 1 , 2, 3, 6 and 9 days exposure. In addition colour was recorded for the samples exposed in the window after 28 and 119 days.
Colour measurements were recorded at both institutes following an in-house method. At BRE a Macbeth colour eye reflectance spectrometer using a 25 mm diameter measuring window centred on the knot was used. At Trätek an 8 mm window was used on a similar instrument (Topcon RD-100 Colour difference meter) centred on the knot and the darkest spot as well as a digital imaging technique. Control measurements were recorded of the coatings over clear timber for each replicate. Discoloration was calculated as the colour difference between the coating over the knot and the coating over clear timber.
Results & Discussion
The colour change results over the coated knot due to tannin staining are presented as photographic images (Figure 1) and also as colour co-ordinate plots. Overall the colour change due to tannin staining and resin exudation were highest for paint B in all QUV exposures. For some panels exposed without spray high levels of resin exudation were recorded - particularly for panels coated with paint B. Paint B was known to perform poorly in comparison to paint A for resistance to tannin staining and resin exudation.
The panels are matched but there are a number of parameters in the experimental procedure that might influence the repeatability and reproducibility of the method. These include the
preparation of the paints (initial mixing), the application of paint to the panels (target
applications were used), the QUV weathering devices, the window exposures (Sweden and UK), the operator and assessment technique. In addition knot size is not uniform and the potential for tannin staining of each knot may vary even though the panels are carefully prepared.
The data presented provide an evaluation of the differences between the coatings, the exposures and the laboratories (BRE and Trätek).
Overall performance of coatings in QUV weattiering:
Figure 2 presents the overall discoloration with exposure as colour change (AE) calculated from the colour change of the coating on clear timber subtracted from the colour change over the knot.
The tannin staining and resin exudation is less on panels coated with paint A than paint B. Paint A is better at resistance to tannin staining than paint B. The colour change increases with time as more tannin staining occurs with a notable increase in yellowing (Ab*). Ab* (yellowness) is the principal component of the overall colour change. The Trätek exposure produced a greater overall colour change than the BRE exposure.
An indication of variability in the data is presented in Tables 1 and 2 which present the range of AE values after 6 days exposure for each individual panels and the overall mean and the standard deviation data. The standard deviations of the colour change data is high
(approximately AE = 4) and therefore the 95% confidence limits of the data will also be high. There is no significant difference between exposure (spray or no spray) for paint A (BRE), paint A (Trätek) and paint B (Trätek). There is a significant difference in exposure (spray or no spray) for paint B (BRE). The spray option appears to reduce the colour change during this exposure as though the stain, or resin, were being washed from the surface.
Examples of the colour change over the exposure period for individual replicate panels are presented for paint A in Figures 3 and 4. These plots show the similar nature of all the colour change curves and the different levels of colour change recorded for the individual panels. A significant influence on these is the knot itself as the ranked order of discoloration for
individual matched panels was broadly similar for both laboratories. The repeatability of the method is strongly influenced by the knot present in the panel which varied in size from 20 -35 mm in diameter between replicate panels.
There is no significant difference between panels exposed in the QUV weathering device with spray (exposure method 1) and no spray (exposure method 2).
Overall performance of coatings befiind glass:
The data from the longer exposures of samples as a 'natural' weathering behind glass are presented in Figure 5. In both laboratory exposures paint B discoloured more than paint A. At the end of the exposure period (119 days = approximately 4 months) the colour changes for both BRE and Trätek samples were equivalent at AE = 1.0 for paint A and AE = 3.5 to 4.0 for paint B. Paint B shows visible discoloration of sufficient intensity to constitute a rejection of the coated wood product whilst paint A to the naked eye still looks unstained.
The similar results obtained from these exposures suggest that many of the parameters, including the preparation and application of the paints, the different window exposures and the assessment technique, are not significant influences on the reproducibility of the method. The significant difference in results from the two laboratories for the QUV weathered samples suggest that the influence of the QUV weathering device is an important parameter in this test method. Differences in operating temperatures and irradiation densities appears to have a significant effect on the reproducibility between laboratories.
Conclusions
Although the performance of paint A was better at resisting tannin staining than paint B, it still performed poorly in the QUV exposures. The operating temperature of the QUV weathering device was considered to be too high at 60°C and future work should concentrate on
lowering the operating temperatures. It is noted that the ASTM (2001) method uses a condensation cabinet at an operating temperature of 39°C with samples evaluated after 16 hours exposure (for cedar panels).
The performance of coatings at resisting tannin stain when exposed in a window behind glass was similar for both coatings irrespective of the laboratory that conducted the test method. This suggests that the difference between coating mixing, application and assessment techniques is not of significance to the reproducibility of the tannin staining method. However differences were noted when samples were exposed in the QUV
weathering device. The difference in the operation and output of individual QUV weathering devices is important, particulariy the operating temperature and irradiation density. This is consistent with findings of round robin tests of artificially weathering coatings.
This points to the necessity of including intemal comparison products into any future method. A coating that performs well and one that performs pooriy should be included and the validity of the test might be confirmed by the achievement of a threshold value for colour change for the paint that performs badly.
Good progress has been made towards evaluating a rapid method of assessing resistance to tannin staining. Refinements of the test protocol described include running the QUV
weathering device at a temperature below 60°C to minimise resin exudation. Results of AE, AL*, Aa*, Ab* and experimental refinements to the Trätek method will be presented to CEN/TC139 WG2 to provide the momentum for a method agreement, wider validation and progress towards a standard.
Recommendations
Future studies should concentrate on a reduced operating temperature for the QUV
weathering machine and including a wider range of test coatings of unknown performance. The tannin staining test method should be based on:
• Lower operating temperature of the QUV weathering device (40°C) • Exposure period of 24 hours to induce significant discoloration in samples • QUV exposure without spray
• Intemal comparison products to validate test
Aclcnowiedgements
The authors thank Mr Tommy Sebring and Becker Acroma for their technical assistance and the supply of coatings.
Mr Ed Suttie would like to thank the Construction Industry Directorate of the UK
Government's Department of Trade and Industry (DTI) for funding BRE's contribution to this work. Mr Jan Ekstedt would like to thank the Association for Swedish Wood Products Research (Träforsk) and the Swedish Wood Association (Svenskt Trä) for funding Tratek's contribution to this work.
References
ASTM (2001) Standard test method for evaluation of tannin stain resistance of coatings D6686-01
CEN/TC139/WG2 document N305 Draft text for prEN927-6 Paints and varnishes. Coating materials and coating systems for exterior wood. Artificial weathering.
Trätek (2000) Determination of discoloration of paints on wood due to resin exudation from knots. Trätek rapport 0006009.
Table 1 Range of colour change (AE) values across the five replicates for each laboratory at
6 days exposure in the QUV weathering device
AE A/spray A/no spray B/spray B/no spray BRE 4 - 1 2 3 . 5 - 1 1 6 - 1 9 1 2 - 2 7 Trätek 1 0 - 2 0 8 - 2 0 1 6 - 2 4 1 9 - 2 8
Table 2 Colour change (AE) mean values and standard deviations
Tratek exposure time / hours A A/QUV-Spray sd B B/QUV-Spray sd A A/QUV-No spray
sd
B B/QUV-No spraysd
A window B window BRE A A/QUV-Spraysd
B B/QUV-Spraysd
A A/QUV-No spraysd
B B/QUV-No spraysd
A window B window 0 24 48 72 138 2856 0.1 6.4 10.1 12.2 14.7 0.07 3.45 4.35 4.16 3.90 0.1 14.7 17.2 18.2 19.8 0.08 4.52 4.19 4.98 3.66 0.0 7.2 10.5 12.2 14.0 0.04 3.80 4.04 4.26 4.75 0.1 13.9 18.2 20.5 22.7 0.07 143 1.90 2.40 3.^4 0.1 0.0 0.1 0.2 0.1 1.26 0.2 0.2 0.4 0.5 0.6 3.58exposure time / hours
0 24 48 72 138 2856 0.12 3.02 6.55 7.01 7.35 0.05 1.75 3.21 3.40 3.6^ 0.04 9.28 8.93 9.04 7.90 0.02 4.74 5.54 5.40 5.14 0.12 2.94 4.15 4.61 5.49 0.08 1.44 2.40 2.47 2.97 0.05 12.91 13.52 14.72 17.09 0.01 5.47 4.35 5.08 5.20 0.13 0.18 0.17 0.33 0.02 0.95 0.07 0.26 0.57 0.67 0.89 3.95 23
A B A B A
B A B A B
Figure 1 a) Panels after 3 days exposure at BRE in QUV with spray option. The test
paint is identified as A or B.
B B
B B B
b) Panels after 3 days exposure at BRE in QUV without spray. The test paint is identified as A or B.
Trätek LLi 0) D - .A/QUV-Spray . . B/QUV-Spray 60 80 100 120 140 160 Time (hr) .A/QUV-No spray ,B/QUV-No spray BRE UJ Q O 20 40 60 80 100 120 140 Time (hrs) - -A/QUV-Spray - - B/QUV-Spray 160 • A/QUV-No spray B/QUV-No spray
Figure 2. Overall colour change (AE) of paints A and B plotted against exposure time in the QUV weathering device.
Trätek Paint A QUV spray 20.0
3 10.0
Time (hr)
BRE Paint A QUV spray
S 10 Panel 1 Panel 3 Panel 5 Panel 7 Panel 9 20 40 60 80 100 120 140 160 Time (hr)
Figure 3. Colour change of individual replicate panels exposed in the QUV weathering
device with spray option.
Trätek Paint A QUV No spray
50 100 Time (hr)
BRE Paint A QUV no spray
0 20 40 60 80 100 120 140 160 Time (hr) - • — P a n e l 13 B - F^nel 15 Panel 17 Panel 19 • Panel 21
Figure 4. Colour change of individual replicate panels exposed in the QUV weathering device with no spray.
Trätek behind glass exposure
1500 Time (hr)
2500 3000
BRE behind glass exposure
500 1000 1500 2000 2500 3000 Time (hr)
Figure 5. Overall colour change (AE) of paints A and B plotted against exposure time in the laboratory window for a 'natural' weathering exposure behind glass.
Detta digitala dokument skapades med anslag från
Stiftelsen Nils och Dorthi
Troédssons forskningsfond
Trätek
INSTITUTET FOR TRATEKNISK FORSKNING
Box 5609. 114 86 STOCKHOLM Besöksadress: Drottning Kristinas väg 67 Telefon: 08-762 18 00
Telefax: 08-762 18 01
Vidéum Science Park, 351 96 VÄXJÖ Besöksadress: Luckligs plats 1 Telefon: 0470-59 97 00 Telefax: 0470-59 97 01 Skeria 2. 931 77 SKELLEFTEÅ Besöksadress: Laboratorgränd \ Telefon: 0910-28 56 00 Telefax: 0910-28 56 01
