Pe"anika 13(2), 159-163 (1990)

The Durability of Phenol-Resorcinol Formaldehyde Glue Joints onThree Malaysian Hardwood Species

MOHD. HAMAMI SAHRIFaculty ofForestlY

Universiti Pertanian Ntalaysia43400 UPM, SeJ'dang, SelnngoT Darnl EllSan, i\lfalnysia

Keywords: durability, delamination percentage, phenol-resorcinol formaldehyde, glne joinl.

ABSTRAK

Kekuatan dan ketahanan rekatan perekat fenol-1"esorsinol Jormaldehid ke alas tigo. spesies ka)'u heras lWalaysiatelah dinilni. Pelapisan spesies tunggal dan pelapisan spesies campuran di Iwsilkan menggunakan perekat diatas dan diuji pada mhu biM. Kekuatan Tekatan terhadap pekkangan di uji berdasarkan spesifikasi ASTMD 1101-59. Semua kombinasi peTekalan telah menghasilkan lahap peralllS pekkangan yang amat Tendah danmemenuhi keperluan seperti yangdipernntukan da/am spesifikasi Arrc. Peralus IJekkangan yangpaling linggiIeTha"il daripada kombinasi Kempas-Kempas dengan nilai 4. 7% dan yang palingTendah dmipada kombinasiSeJ"(l)'a-Sera)'a )'ang tidak menunjukkan sebarang pekkangan Tekatan_ Kombinasi spesies campuran padaumumnya memberikan peralus pekkangan )'ang kbik baik berbanding kombinasi spesies tunggaL

ABSTRACT

The integrity and durabilily ofphenol-TesoTcinolfOl7ooldehyde glue joints on thTee Malnysian haTdwood specieswere evaluated. Single and mixed-species laminations were produced using phenol-resorcinol fonnaldehydeadhesive al Toom lemperot"Te. The integrity ofthe glue bond to delamination was lested accoTding 10 the ASTi'vlD 1101-59 specificalion. All combinations, produced vI!/)' low delnmination percentage !£vels and met therequi,-"ment specified by AlTC. The highest delaminalion percentage was shown by Kempas-Kempas combinationwith only 4.7% and the lowest was Sera)la-Sera~~a combination with 0.0% delamination. Mixed-speciescombinations offered lower percentage ofde/aminalon that their singk-species counterparts.

INTRODUCTION

The suitability of glues and gluing techniquesfor the producti,:m of joint adequate forexterior service need to be evaluated regularly.The choice of wood species and woodadhesives is governed by many factors, but themost important one is the strength and integrityof the glue joint. For structural laminatedmembers such as laminated beams and deckingboards, the integrity of the glue joints shouldbe at least equal to that of the wood.

Few smdies have been conducted to checkthe strength and integrity of the glue joint oftropical hardwood using phenol-resorcinolformaldehyde and urea formaldeh)'de. Selbo(1967), found that more delamination occurred

in specimens bonded with urea formaldehydecompared to phenol-resorcinol bondedspecimens. ishihara et al. (1967) used Merantito compare the reliability and intergrity of theseglue bonds. He also found that phenol andphenol-resorcinol formaldeh)'de adhesiveoffered better delamination resistance anddelaminated onl)' 2% and 8% respectiveI)' incontrast to urea formaldehyde adhesive \\'ith51 % delamination. Wan Sardini (1979) agreedwith the previous researchers on theperformance of phenol-resorcinol formaldelyde adhesive on five Malaysian hardwoodspecies and on average all species alTeredencouraging results ,\~th only I%delamination.Bohlen (1972) indicated there was severe bond

MOHD. HAMMlI SAHRI

degradation in the specimens during thesecond cycle of delamination test resulting inthe failure of these glue lines meeting thestandard requirement.

The present study was conducted with onemain objective i.e. to evaluate the integrity ofphenol-resorcinol adhesive glue joints on threeMalaysian hardwood species, namely Seraya,Nemesu and Kempas.

MATERIALS AND METHODS

Three Malaysian hardwood species namelyNemesu (Slwreapauciflora), Kempas (Koompassiamalaccensis) and Seraya (Shorea curtisiz) wereused in this study. All selected species wereobtained from Mantega Forest Products (M)Sdn. Bhd., Pahang. The Sa""1 lumber weretaken from three different logs (one from eachspecies). Average sizes of the rough lumberwere 150 mm x 30 mm x 2 rn.

A phenol-resorcinol formaldehyde, PRY asynthetic thermo-setting adhesive was used inthis study. This is the most common adhesiveused for the production of laminated deckingboard in Malaysia. The sample specimens weremade by bonding wood strips with this adhesiveinto specimen sizes sufficient to conduct thetest in the laboratory.

Laboratory Sample FabricationAfter being air dried, the rough sawntimberwas resawn into shorter strips of 30 rom x 30mm x 350 mm using a multiple rip saw. Eachwood strip was planed just prior to edge-gluing.

Adhesive was mixed according to themanufacturers' specifications. For PRF, 10 partsof Norres 45 with 53% solid content was mixedwith 2 parts of hardener H-IO. The well mixedadhesive was then hand spread onto the pieceof wood with a metal spatula at a rate ofbetween 320 to 370 glm' followed by a closeassembly time of 15 to 30 minutes.

Seven different types of assemblies weremade for the glue joint integrity test. Threetypes consisted of single species combinationnamely Seraya-Seraya. Nemesu-Nemesu andKempas-Kempas; three types consisted of acombination of two species i.e. Seraya- Temesu,Seraya-Kempas and emesu-Kempas; and thelast type consisted of all species, Seraya- emesu-

KelIl pas. A total of 30 replicates were made foreach of these seven types.

The assemblies were pressed in a cold pressfor one hour with a panel pressure of 10 kglcm:.!. Each press contained 10 assemblies of thesame combination. The temperature and therelative humidity of the pressing are were 28°Cand 75% respectively. The assemblies werestored for a week at room temperature forproper conditioning.

lnteg>ity of Glue Joint (Delamination) TestThis test covers an accelerated means ofmeasm-ing the resistance to delamination ofstructural laminated wood members intendedfor exterior service and exposure.

Each test specimen consisted of a full crosssection of sample board and 75 mm in lengthalong the grain (Fig. 1). The test was performedin accordance to ASTM* 0 1101-59. Thefollowing sequence was followed in performingthe test.

i. The specimens were placed in a smallpressure vessel (3.0 m long x 380 mmdiameter). Sufficient water wasadmitted to submerge the specimens.

ii. 500 mm - 635 mm Hg of vacuum wasdrawn and held for 15 minutes. Thevacuum was then released and l.03 x10'KPa (kilopascal) of pressure wasapplied for a period of 2 hours.

111. The sequence was repeated tocomplete 2 cycles (a total of 4.5 hours).

iv. The specimens were then removedfrom the pressure vessel and dried fora period of 91.5 hours in a forced draftoven at a temperature of 27 - 29°C andrelative humidity of 25 - 30%. Duringdrying, the specimens were placed atleast 50 mm apart to allow for sufficientair movement.

v. The soaking and drying cycles werethen repeated at once to complete atotal period of 8 days.

* American $ode£)' for Tesling and Malerials

160 PERTANIKA VOL. 13 NO.2, 1990

THE DURABIUTYOF PHENOL·RESORCINOL FORMALDEHYDE GLUEJOh'ITS

\.<---- ~ZO"""" --<~

1<-7,---+7.-----1Fig. I: Fabrication oJ ~pecilJLem JUI" cifd.(lJJlinatioll lest

T

,/ /

,7

,

/

,

//

/

/

,

,

At the end of the final drying period, thedelamination if any, was measured by insertinga feeler gauge with a hlade thickness of 0.08mm to 0.10 mm into the delaminated gluelines.Measurements were made to the nearest 1.3mm from the end grain surfaces.

RESULTS AND DISCUSSION

The integrity of glue bonds to delaminationwas tested, on 210 different gluelines,representing seven combinations with 30replicates each. The delamination from the endgrain was measured and expressed aspercentage of total lengths of the glue line.The minimum, maximum and average lengthsof delamination in each combination are

presented in Table I and il1uSlrated in Figure2.

Single StJecies CombinationThe Seraya-Seraya combination produced thebest result as no apparent glueJine failures wereobserved. The Nemesu-Nemesu combinationproduced the second best result since 15 outof 30 specimens did not delaminate. Theaverage delamination length and percentagewere 2.8 mm and 3.7% respectively. Kempas·Kempas combination exhibited the highestdelamination percentage of 4.7% and all ofthe tested gluelines were delaminated to somedegree. All single species combinationdemonstrated delamination percentages ofless

TABLE 1Average values of delamination percentage of phenol-resorcinol glue joints

Combination No. of No. of glue- A\'erage length Percentagegluelines lines delamination of

tested delaminated (mm) delamination

No. % Max. Min. Ave. %

Single-speciesSeraya-Seraya 30 0 0Nemesu-Nemesu 30 15 50 6.5 1.3 2.8 3.7

Kempa~Kempas 30 30 100 7.8 1.3 3.6 4.7

Two speciesSeraya-Nemesu 30 72 3.3 3.9 1.3 2.3 3.0

Seraya-Kempas 30 11 36.7 3.6 1.3 2.0 26

Nemesu-Kempas 30 25 83.3 5.2 1.3 2.6 3.4

Three speciesSeraya-Nemesu-Kempas 30 17 56.7 6.5 1.3 2.4 3.1

PERTANIKA VOL. 13 NO.2, 1990 161

;\'IOHD. HAMA~H SA.HRI

~..

lvlixed-Species CombinalionsThe average values of delamination percentageof all gluejoints in each mixed-speciescombination were generally below 3.5%.Nemesu-Kempas cOlubination gave the highestdelamination percentage with 3.4%, followedby three species combination of Seraya-Nemesu­Kempas with 3.1 %, Seraya- emesu with 3.0%

CONCLUSION

The integrity test of the phenol-resorcinol gluejoin ts on three Malaysian hardwood speciesgave good results. All combinations producedvery low delamination percentage levels andmet the requirements specified by AlTC. Thehighest delamination percentage was shows byKempas-Kempas combination with only 4.7%.Excellent delamination resistance after soakingand drying cycles showed that the glue jointsproduced are adequate for exterior exposureservices.

Mixed-species combination offered lowerpercentage of delamination compared to thesingle-species counterparts, except for Seraya­Seraya combination which showed no apparentdelamination.

The results are considered preliminaryevidence and by no means complete indepicting the absolute integrity of the gluejoints. However, it provides information for theevaluation of adhesives for production ofproducts intended for exterior uses.

ACKNOWLEDGMENTS

and Seraya-Kempas with 2.6%. The mixed­species combination produced rather lowdelamination percentages and meet therequirement, as specified by AlTC. This alsoconfirmed that the gluejoints are likely to besuitable for exterior exposure.

Mixed-species combinations showed a lowerdelamination percentage as compared to thesingle species counterparts, except for SeraYd­Seraya combination. Kempas, being the most<len se species, under soaking and drying cyclesswelled and shrank in greater magnitude thanthe other two species. Combination of Kempasand Seraya showed a lower averagedelamination percentage. Seraya and Nemesuhas lower density value, thus lower coefficientof expansion. Combining two or more speciesof different wood densities, in this case wasslightly advantageous. Therefore. gluing variousspecies into a single board products. may resultin bener resistance to delamination andimprove other properties as well.

S·w.1(

~l

Fi~. 2.. hII'm/{" t1l'ltlll/lualul/l jJI'HPlllugf' oJ jJh,',W[-

Tesorci1101 glue joints

than 8% and met the requirements specifiedby AITC* for wet usc adhesives. Results fromthis study compared favourably with previollsworks. Moriya el at. (1970) using Keruing(Dipterocmpus sJJ/J) to compare the reliability andintegirty of phenol-resorcinol and phenolformaldehyde glues found the phenol­resorcinol glue join ts offered betterdelamination resistance than su"aight phenol.The phenol-resorcinol gluelines delaminatedby only 8-26% compared to 79-88% for thephenol formaldehyde glue. Hoshi (1977) alsofound that phenol-resorcinol glue jointsperformed better than straigh t phenol on 13hardwood species from Kalimantan, Indonesia.

Excellent resistance to delamination aftersoaking and drying cycles was shown by theglue joints especially with Seraya-Serayacombination. This shows that phenol-resorcinolformaldehyde can produce very durable gluejoints with these wood species.

AITC: American Institute of Timber ConstructionThe author wishes to thank the Dean, Facultyof Forestry, Universiti Pertanian Malaysia for

162 PERTANIKA VOL. 13 NO.2, 1990

THE DI;R-\BILITY OF PHENOL·RESORCINOL FOR,\IALDEHYDE GLUEjOlNTS

permission to usc all facilities required for thestudy. Our appreciation is extended to MentegaForest Products (Ml Sdn. Bhd. for providingthe wood samples and to 1 orsechem (M) Bhd.for the adhesive materials.

REFERENCES

A~[f.RICr\l'\ !!\STITUTE of TD.IBER CONTRUCTION

(AlTe). 1977. Voluntary Producl' Standard, P.S.56-73 for Strucmral Glued Laminated Timber.U.S. Dept. Of Commerce.

A:-'IERlr.r\;\l SOCIETY FOR TESTING A:"lD rVIATERIAI.

(ASTr....l). IHSO. Standard Test Method forlntegrity of GlueJoinLs in StnlCLUl'tll I.aminatedWaon Products for Exterior Use. D 1101-59.ASTM Standard, P;u't 22: 356-357.

BOHI.El", J.e. 1972. Shear Strength of HighTemperature Heat Lumber Laminated ,...·jthPhenol-resorcinol Glue. For, Prod. J. 22(12): 17~

24.

HOSHI, T. 1977. Cutting and Gluing Properties ofThinccn Specics of Kalimantan Woods. ; Gov.Bull. For. Exp. Sta. No. 200, 215-235. Tokyo.

MORJYA, K. M. SUGAJ.'10 and Y. CHInA. 1970. Gluingfaculties of Laminated Wood made of KeruingLumbcl' Grown in Malaya. BulL GOVl. For. Exp.Sea. No. 231, 45·53. Tokyo, Japan.

NISHIHARA, M.K MORIYA and M. SUGANO. 1967.Gluing Faculties of Laminated Wood made ofRed Meranti \Vood Grown in Sarav..ak Go\'. Bull.For. Exp. Sta. No. 190, 160-167. Tokyu.

SELBO, M.L. 1967. Long Term Effect ofPreseryatiollon Gluclincs in Laminated Beams. For Prod]17(5): 23-32.

W:\i'l SARnlN"1 ',VAN SAlJ.E.H. 1979. The Glue JointStrength of Laminated Timber Decking of FiveSelected Malaysian Timber Species. ;6 p.Unpuh!. B.Se. thesis.

(RRreived 19 iHarrh, 1989)

PERTANIKA VOL. 13 ?\o'D. 2, '9YO 163