Walter de Gruyter . Berlin. New York

ISSN 0018-3830

iEditorial Guidelines

Please send original manuscripts (four copies) to:

Editorial Office HolzforschungWalter de Gruyter GmbH & Co. KGPostfach 30 34 21D-I0728 Berlin / F.R. of GermanyFax: ++ 3o-26005-325/Tel. ++ 30-26005-219e-mail: Holzforschung.editoria1@deGruyter.de

tables and figures. Short notes do not require a summary and will havepriority in publication.

introduction. This should define the problem and if possible the frameof existing knowledge. Please consider that people not working in yourparticular field are able to understand your intentions.

MateriaJs and Methods. Please be as precise as possible to enableother scientists to repeat your work.

Results. Only material pertinent to the subject must be included. Datamust not be repeated in figures and tables.

Discussion and Conclusion should interpret the results in view of theproblem as outlined in the introduction and of related observations byyourself or others. Implications for further studies or application maybe discussed. A conclusion should be added if results and discussion arecombined.

Acknowledgements may be used to credit support.

References. Relevant Literature must be cited in the text with theauthor's name and year of publication. In addition, the bibliographymust be listed at the end of the text in alphabetical order of the authors'names, together with the title of the paper and the full quotation of thebibliograpmcal reference, for example:

GeneralOnly unpublished original papers of experimental content, COlTe-sponding to the scientific requirement of HOLZFORSCHUNG, areaccepted for consideration. Papers submitted to symposia for publica-tion in a proceedings-volume are to be considered as pre-published.Therefore, it is not possible to accept these contributions unless theydiffer and are enriched by new findings and results. With the acceptanceof the manuscript for publication, the copyright is automaticallyassigned to the publishers and the author may not publish the samepaper elsewhere, not even in a foreign country.Only papers written in the English language will be accepted forpublication.To eliminate unnecessary, time-consuming correspondence, theauthors are requested to submit only manuscripts which are readyfor printing. This includes drafting and orgaoizing of the text inaccordance with the rules adopted by HOLZFORSCHUNG fortypesetting. AU manuscripts are subject to a thorough review byselected referees; the authors themselves are invited to recommendcompetent experts for reviewing.Before submitting a manuscript. the foUowing rules should be readcarefully. Manuscripts that differ from the specifications and do notcomply with our formal requirements wiU be returned to theauthors for correction before revjew. Please use a recent issue ofthis journal as a guideline in the preparation of your manuscript.For further information you can visit our bomepage www.deGruyter.de.

Specifications for Manuscripts

Submitted manuscripts must be both concise and precise, and consist ofno more than 8-10 single-sided typewritten sheets of 30 double-spacedlines. The length of the text should be in keeping with the resultsachieved. Contributions exceeding the standard length of 10 sheets (notincluding summary, tables, figures and references) by more than 50%may as a rule not be accepted for publication. It is therefore advisableto divide longer manuscripts into two or more continuing series. Papersof more than 15 pages may only be submitted after prior consent of theeditor.

Journals:Ralph J., S. Hatfield, S. Quideau, R.F. Helm, J.H. Grabber and H.-J.Jung. 1994. Pathway of p-coumaric acid incorlKJration into maize ligninas revealed by NMR. J. Am. Chem. Soc. 116, 9448-9456.

Books:

Panshin, A.J. and Co de Zeeuw. 1980. Textbook of Wood Technology.McGraw-Hill, New York. pp. 9448-9449.

Multi-author bOoks:

Lai, y:z. and K. V. Sarkanen. 1971. Isolation and SU1JctUral studies. In:Occurrence, FOmIation, Structure and FuDcUon. Eds. K. V. Sarkanen,C.H. Ludwig. Wiley Interscience, New York. pp. 150-163.

After the references full postal addresses of all authors should bementioned.

The numbers of Figures must be limited to the absolute minimum.Microphotographs must be high-gloss and rich in contrast. DrawiDpmust be suitable for reproduction (no photocopies) and reduction to awidth of 8.1 cm. Do not use block letters in the lettering of figures anddrawings. The legends for the figures must be concise and self-expla-natory so that they are intelligible even without reference to the text.The key to the symbols in the figures should be included in the legend,not form part of the figure itself. Each figure must be presented on asingle sheet using consecutive arabic numerals. The latter applies to theTables with concise and self-explanatory beadings. The IqendS of ellfigures should be listed on one separate sheet.

Dimensions and units. The metric system must be used. SI units arerecommended. Compound units are given with the proper exponentwithout a point, e.g. g °2 g-1 dw h-i.

Only those manuscripts which confOml to the above guidelines will beconsidered for publication. The Editor reserves the right to suggestabbreviations and text improvements.

GaUey proofs are forwarded to the authors in duplicate. The immediatereturn of one corrected proof to the Editor is requested. In view of thehigh costs, corrections must be limited to printing en'OI'S; amendmentsto or changes in the text which exceed 4 single corrections, or the cor-rection of a single sentence, will be cliarged.

20 reprints of each original paper will be forwarded to the author freeof charge. If further reprints are required, these can be ordered whenreturning the corrected proof to the Editor. The approximate price perreprint is given on the order form accompanying the galley proofs.

Diskettes

Please prepare also an electronic file. Please submit your text ondiskettes fonnatted for Windows '95. Indicate the operating system andthe programme name on the disk label.Order sections according to the fonnat of the journal.Use single tab stops to indent.Use single carriage returns between paragraphs.Save graphics and tables in separate files.Avoid using multiple space characters to fonnat or between sentences.Avoid the positioning of graphics and tables in the text.Avoid automatic numbering of pages and footnotes, manual pagebreakes.In short: the more simply you format your manuscript, the easier it willbe to reformat it for typesetting with a minimum of errors!

Composition of papen

Avoiding a lengthy introduction. papen must report of true scientificor technical progress, the major chapters being characterized byappropriate headings. Paragraphs of less importance should be markedfor setting in 'petit'.

The first page of the type-script should present: running title; full title;name(s) of the author(s); name and address of the laboratory where thework was carried out; Summary (approx. 112 sheet DIN A4) contain-ing project, procedure, results and conclusing (concise in abstract fonn)and Keywords.

For the benefit of the author and the reader it is advisable to publishclear-<:ut research results in the fonn of Short Notes (but no provisionalor intennediate reports), consisting of an extended abstl"a(:t of not morethan four double-spaced typewritten pages, in addition to the IIecessary

Subscription InfonnationHOLZFORSCHUNG is published bi-mondlly (6 issues per year).

Price of volume 55 (2001): DM 1.698.-. ~ 868.17; sFr 1,511.-;OS 12.395.-USA and Canada: US $ 960.-Single issue: DM 292.-; ~ 149.30; sFr 200.-; 6S 2.132.-USA and Canada: approx. US $ 160.-Orders for subscriptions may be placed widt any ~Iler or widt dte

publishers.

T.F. Shupe et al.: Hardwood Wettability 541

Holzfoncbung55 (2001) 541-548

An Investigation of Selected Factors that Influence HardwoodWettabilityBy Todd F. ShupeJ, Chung Y. Hse2 and Wan H. Wang)J School of Forestry, Wildlife, and Fisheries, Louisiana State University Agricultural Center, Baton Rouge, LA, USA2 USDA Forest Service, Southern Research Station. Pineville, LA, USA) Nanjing Forestry University, Department of Wood Science and Technology, Nanjing, Peoples Republic of China

SummaryWettability of sanded and non-sanded transverse and tangential sections of 22 southern hardwoodsspecies was judged by ~ment of contact angles using phenol fonnalddlyde rains. As expected,contact angle values on ttansverse sections were higher than those on tangential sections for both sand-ed and non-sanded surfaces. On sanded surfaces, hackberry had the highest mean contact angle (64.7°),and black oak had the \owest mean contact angle (SO. 1°). On DOlI-sanded surfaces, winged elm had thehighest mean contact angle (S9.1°), and sWeetgum had the lowest mean contact angle (4S.go). In addi-tion, 4 of the 22 species (southern red oak, sweetgum, white oak, and post oak) were selected to inves-tigate the effect of oven-dryin&, air-drying, and free-drying on wettability.The mean b'aDSvene contactwas 2.1 °-29.00 and S.l °-31.5° higher than ~al and tangential values. respectively. The contact anglepattern typically displayed for a given species and plane was generally oven-dry > air-dry> freeze-dry.The species pattern for DK)St methods and planes was: sweetguJD > white oak > post oak > soudlem redoak. White oak. and post oak pve similar contact angle values.

KeywonJ.r

Contact angle

DryingHardwoodPhenol fonnaldehyde resinWettability

Introduction Bonding properties may also vary widi regard to the plane(transverse, tangential, or radial) of wood. Due to the aniso-nupbic nature of wood, we know that it possesses uniquehygroscopic properties in its three fundamental directions:longitudinal, radial, and tangential. It is important to under-stand the bonding properties on these three planes for moreefficient utilization.

This project was initiated to 1) investigate die wettabili-ty of 22 southern hardwood species, 2) detennine the effectof wood plane on wettability, and 3) examine the effect ofthree drying methods on the wettability of four southernhardwoods.

Materials and Methods

The development of adhesive bonding technology has beenclosely related to surface quality research. Because woodadhesives are applied to the surface of wood, the propertiesof the wood surface are influential in detennining the wet-tability performance of an adhesive. Contact angle detenni-nation is a common method of evaluating the wettability ofwood surfaces. Contact angle is the adverse measure of wet-tability (Zisman 1964; Zisman 1976). It is thermodynami-cally detennined by the balance between adhesive forces,i.e., between liquid (adhesive) and wood (adherend) inter-faces, and cohesive forces within the liquid (Johnson andDettre 1993).

Numerous previous researchers have shown that thewettability of wood as determined through contact angleassessment is intimately associated with glue-line integrity(FTeeman 1959; Bodig 1962; Suchsland and Stevens 1968;Hse 1972; Scheikl and Dunky 1998). In North America,most previous studies on wood wettability have been con-ducted with southern yellow pine (Pinus sp.) or Douglas-fir(Pseudotsuga menziesil). Hameed and Roffael (1999) estab-lished that the wettability of sapwood from pine, Douglas-fir, and larch on cross, radial, and tangential sections withwater and various glues is better than that of heartwood.They found that in most cases, the tangential section of sap-wood and heartwood was less wettable than radial and crosssections. The literature is sparse with regard to wettabilitystudies of North American hardwoods.

Twenty-two hardwood species were selected for this study. Thespecies' common name. scientific name, pore distribution, and spe-cific gravity range are listed in Table 1. Ten uees with a diameterat breast height near 15.24 cm were selected for ~h species. Thesampling locations were broadly distributed throughout that por-tion of each species range occurring in the II-state area extendingfrom Vuginia to northern Florida and west to Arkansas and easternTexas. Only one tree of a particular species was cut at one location.

Our samples were unused samples from a previous study byChoong ef al. (1974). Therefore. the sample preparatlOO methodis similar. Disks that were 5.08 cm thick were removed at 1.8 mabove ground for eKh tree. Three rectangular-shaped samples werecut from e.:h disk using a fine-toothed handsaw. The ends of thesamples were either perpendicular to the grain (transverse) or tothe radial or tangential planes. The wood samples were sawn into

Holzforschung / Vol. 55 /2001/ No. 5,.. rnn,"';..ht ,)MI W.I A. n,.."..- . R-';" . N..., VnIfr

T.F. Shupe et aI.: Hardwood Wettability542

Overall Fiber radialMean (Darcy)2 diameter «Dm)3

Specificgravity range I

Scientific nameSpeciescommon name

Ring-porous

1.6880.712

16.7323.688

24.78145.53648.26159.0750.1104.873

39.00528.612

1.3562.767

42.9142.601

44.782

IS.2814;5112.3414.271S.3614.91IS.4i1S.0014.1914.7714.9815.4817.7S17.731S.211.1.42lS.4S-

0.70-0.860.71-0.91051-0.700.52-0.640.59-0.780.65-0.850.66-0.830.65-0.800.71-0.980.68-0.900.62-0.880.60-0.740.64-0.76051-0.710.63-0.820.62-0.770.64-0.85

Quercus marilandica Muenchh.Quercus alba L.

Celtis occidentalis L.Ulmus acericana L.Quercus nigra L.

Quercus velutina Lam.Quercus shumardii Buck!.

Quercus rubra L.Quercus stellata Wangenh.

Carya spp.Quercus falcata Michx.

Quercus iaurifolia Michx.Fraxinus americana L.

Fraxinus pennsylvanica Marsh.Quercus falcata vat. pagodaefolia Ell.

Ulmus alata Michx.Quercus coccinea Muenchh.

Blackjack oakWhite oak

HackberryAmerican elmWater oakBlack oakShumard oakNorthern red oakPost oakHickorySouthern red oakLaurel oakWhite ashGreen ashCherrybark oakWinged elmScarlet oak

Diffuse-porous

9.10314.54715.96813.3118.227

18.0222.8424.3026.9821.66

0.4~.600.46..{}.S70.36-{).5S0.3S-O.SS0.45'-0.67

Red mapleSweetgumYellow-poplarSweetbayBlack tupelo

Acer rubrum L.Liquidambar styracijlua L.Liriodendron tulipifera L.

Magnolia virginia L.Nyssa sylvatica Marsh.

1 Specific gravity determined from longitudinal permeability samples, based on oven-dry weight and dimensions (Cboong et al. 1974).2 Overall mean Darcy gas permeability values in the longitudinal direction at 0% moisture content (Cboong et al. 1974).3 Stemwood values from 15.24-cm diameter hardwood species ranging in age from 27-59 years (Manwiller no year given).

experimental variables were also detennined. All statistical analy-sis was conducted using SAS software (SAS 1989).

thin sections (0.3175 cm thick) from the end for contact anglemeasurement.

Contact angle determination was accomplished with a micro-scope equipped with a goniometer eyepiece. The microscope tube-was arranged horizontally. The specimen was placed on the stage,and a 0.06 ml droplet of phenol formaldehyde (PF) resin was appliedwith a pipette to the surface of the specimen. The contact angle wasmeasured by rotating the goniometer eyepiece so that the hairlinepassed through the point of contact between droplet and veneer andwas tangent to the droplet at that point. All measurements weremade 5 seconds after the resin had been dropped. For ring-porousspecies, all contact angles were determined randomly and regard-less of earlywood or latewoOd.

This study was conducted with a laboratory prepared orientedstrand board (OS8) core phenol formaldehyde (PF) resin that con-tained 44 percent solids, viscosity of 300 cps, and a mole ratioof 1.95:1:0.45 of formaldehyde to phenol to NaOH (sodiumhydroxide). Contact angle measurements were recorded on thetransverse and tangential sections of 22 species. Measurementswere conducted parallel to the grain direction on tangential sur-faces. For each specimen, one of the transverse and tangentialsurfaces was sanded with P320 extra fine sandpaper from 3M"'.The corresponding transverse and tangential surfaces on the samespecimens were left non-sanded. Therefore, each sample containedsanded and non-sanded transverse and tangential surfaces.

Analysis of variance was performed to determine the potentialsignificance of me main effects: species, surface preparation (sand-ed and not sanded), and me interaction effect. The Scheffe meanseparation test was employed to determine significant differencesbetween the different species. Correlation coefficients between

Results and Discussion

Species and sanding effect

The mean contact angles and Scheffe grouping for the 22species on sanded and non-sanded surfaces are presentedin Table 2 and Table 3, respectively. Table 4 summarizesthe results from the analysis of variance. As expected, therewere highly significant differences between the species(Table 4). For the sanded surfaces, white oak and water oakboth gave the highest mean contact angle on transversesurfaces (68.3°), and black oak yielded the lowest mean at51.2° (Table 3). Blackjack oak gave the highest mean con-tact angle for sanded tangential surfaces (62.3°), and blackoak again gave the lowest mean at 49.0°. On the non-sandedsurfaces, winged elm (68.6°) and cherrybark oak gave thehighest mean contact angles for transverse and tangentialplanes, respectively. Sweetgum (49.6°) and post oak (38.2°)gave the lowest mean contact angles on non-sanded trans-

verse and tangential planes, respectively (Table 4).It was expected that species would yield significantly

different contact angle values because of inherent differ-ences mostly attributable to differences in wood anatomyand chemistry. For example, in a study of spruce (Picea abies

Holzforschung / Vol. 55 / 2001/ No. 5

543T.F. Shupe et aI.: Hardwood Wettability

Karsten), pine (Pinus sylvestris L.), beech (Fagus sylvaticaL.), and poplar (Populus x euramericana Guinier), Scheikland Dunky (1998) found that peneb'ation was retardedwith increasing viscosity of the liquids and the smaller ceUdiameters in latewood in comparison to earlywood. Data

from previous studies on penneability and specific gravity(SG) (Choong et al. 1974) and fiber radial diameter (Man-willer no year given) were used to determine correlationbetween these properties and our contact angle values. Thesanded surfaces mean data was negatively correlated with

Table 1. Mean contact angle values and Schefff groupings (K 22 southern b8n1wood species on the ttansvene and radial (-=ea. S~mens were tested in the airdry condition, and the surface was sanded

RmaIXXWS

JJ2.3 (A)(4.4)

SS.5 (ABCDE)(5.2)

61.7 (AS)(4.3)

55.3 (CDE)(5.7)

58.5 (ABCDE)(3.0)

49.0 (F)(9.0)

53.6 (DPJ")(5.9)

57.9 (ABCDE)(6.6)

55.3 «:DE)(5.7)

61.0(ABC)(4.7)

56.3 (BCDE)(6.6)

60.9 (ABC)(5.9)

57.S (ABCDE)(6.1)

60.4 (ABC)(3.9)

S9.5(ABCD)(3.9)

59.0 (ASCDE)(6.2)

SS.8 (BCDE)(9.7)

62.Q1 ( ~(8.0>'

68.3 (A)(5.3)

67.6 (A)(4.7)

59.3 (CD)(8.4)

68.3 (A)(3.6)

51.2 (E)(8.9)

S6.5 (DE)(12.0)

63.5 (ABC)(4.4)

62.2(ABCD)(8.8)

63.1 (ABCD)(5.4)

62.3(ABCD)(9.6)

66.7 (AD)(4.9)

63.5 (ABC)(6.8)

62.3 (ABCD)(5.9)

W.3 (B(D)(3.9)

58.9 (CD)(6.4)

62.2 (ABCD)(3.9)

62.2..,G..1.BlM:kjack oak

9.8White oak

-0.1Ha:kbaTY

57.34.0American elm

63.49.8Water oak

2.2 50.1B lack. oak

2.9 55.1Shumard oak

60.7S.6Northern ~ oak

6.9Po-.O8t

2.1Hickory

6.0Soutbemn:doat

5.8Laurel oak

~.s6.0White ash

1.9 61.4~1Sb

59.90.8Oaryberkoak

59.0-0.1Wmgedelm

6.4 59.0Scarlet oak

Diffuse JMXOUI

53.4(EF)(7.3)

56.6 (ABCDE)(8.9)

S7.5 (ABCDE)(9.2)

S9.6 (ABCD)(4.5)

SB.O (ABCDE)(8.3)

-S6.76.6Red maple M.O (BCD)

(5.9)60.1 (BCD)

(8.3)65.0 (ABC)

(4.2)68.0 (A)

(4.8)62.0(ABCD)

(5.5)

62.4 (6.6)

3.5 5USweetgum

61.37.SYellow-poplar

63.88.4Sweetbay

4.0 60.0Block. tupelo

-MEAN 4.5 60.057.6 ('.4)

I Each ~ value ~sents 24 observations.2 Letters in parentheses represent Scheft'~ groupings. Species with similar letters are not statistically different at alpha = 0.05. S~es

comparisons were made within a particular surface (i.e., either transverse or tangential).3 Numbers in parentheses are coefficients of variation (%) = (standard deviation/mean) X 100.

Holzforscbung / Vol. 55 / 2001/ No.5

T.F. Shupe et al.: Hardwood WettabilityS44

(R = -0.52) but not at all related to the mean of the sandeddata (R = 0.00). Inferences regarding pore type are some-what limited because of the larger number of ring porousspecies than diffuse porous species included in the study.Furthermore, the diffuse porous species selected are all of

the Darcy permeability values (R = -0.33). The non-sanded

surfaces mean data was significantly correlated to SG(R = 0.47) and fiber radial diameter (R = -0.57).

The correlation analysis showed that pore type wasnegatively related to the mean of the non-sanded data

Table 3. Mean contact angle values and Scheffe groupings for 22 southern hardwood species on the transverse and radial faces. Speci.mens were tested in the airdry condition, and the surface was not sanded

Transverse (X) Tangential (T) X-T MeanSpecies

Ring porous

52.61 (DEFG)2

(9.0~59.6 (BCDE)

(6.8)58.5 (BCDEF)

(10.1)52.1 (EF(])

(7.6)63.0 (ABC)

(5.5)54.8 (CDEFG)

(11.8)61.3 (ABC)

(6.9)61.1 (ABCD)

(8.9)58.6 (BCDEF)

(9.4)62.0 (ABC)

(8.7)57.5 (BCDEFG)

(11.6)56.2 (BCDEFG)

(10.2)64.5 (AB)

(8.5)61.0 (ABCD)

(8.5)64.2 (AB)

(9.5)68.6 (A)

(5.3)57.6 (BCDEFG)

(8.3)

47.4 (ABCDE)(11.4)

51.5 (AB)(8.9)

48.0 (ABCD)(11.9)

43.0 (ABCDE)(13.9)

46.3 (ABCDE)(12.0)

44.7 (ABCDE)(10.9)

43.0 (ABCDE)(17.3)

44.7 (ABCDE)(11.2)

38.2 (E)(17.4)

49.8 (AB)(11.9)

49.8 (AB)(11.9)

43.4 (ABCDE)(14.8)

44.3 (ABCDE)(10.7)

47.2 (ABCDE)(11.3)

52.3 (A)(10.0)

49.5 (ABC)(5.1)

46.0 (ABCDE)(15.9)

J.2 50.0Blackjack oak

8.1 5$.6White oak

10.S 53.3Hackberry

9.t 47.6American elm

16.7 M.7Water oak

9.6 so.oBlack oak

18.3 S2.2Shumard oak

16.4 52.9Northern red oak

20.4 48.4Post oak

12.2Hickory

1",1 '3.1Soud1em red oak.

12.8Laurel oak

20.2White ash

13.8 54.1Green ash

.".9 58.3Cherrybark oak

19. 59.1Winged elm

SI.8.11..6Scarlet oak

Diffuse porous

12.1 S2.0S3.S (8.4) 45.6 (8.2)MEAN

1 Each mean value represents 24 observations.2 Letters in parentheses represent Scheff~ groupings. Species with-similar letters are not statistically different at alpha = 0.05. Species

comparisons were made within a particular surface (i.e., either transverse or tangential).3 Numbers in parentheses are coefficients of variation (%) = (standard deviation/mean) X 100.

HQlzfQ~hun21 Vol. 55 12001 1 No.5

54.5

Table 4. Summarized analysis of variance for the effect of 22 south-ern hanlwood species and surface preparauou technique (sandedand non-sanded) on contact angle. A separate analysis of variancewas perfonned for sanded and non-sanded data. The p-values weresimilar for all soun:es or variation for both analyses

df'2 p-value ----sov'

211

21

O.<MX>I"O.<MX>I..O.<MX>I..

SpeciesSurface preparation (SP)3Species x SP

r:~~,:.. DeDOtes significance at 8l1li- = 0.01.

1 Source of variation2 Dega=s of freedom3 Surface preparation was eidJer sanded or non-sanded '-", ~ ~.. I

~ ..r ,-~.." ~,~ k I

FIg. 1. Scanning electron micrograph of die ~ IUIf8Ce of.sanded yellow-poplar specimen.low density, and the pore type variable was significantly

conelated to SG (R = -0.80) (Table 5). Scheikl and Dunky(1998) found that the penetration behavior of liquids intowood surfaces depends on the different diameters of woodcells and the viscosity and molecule size of the penetratingliquids. However, pore type has been shown to be an impor-tant variable in wettability of wood.

The effect of surface preparation (i.e., sanding) washighly significant (Table 4). Contact angles on sanded spec-imens were greater than not sanded specimens by 3.90 and12.00 for transverse and tangential surfaces, respectively. Ingeneral, sanding of a wood decreases the true surface areaand decreases the roughness of the surface. It was expectedthat the smoother surface of sanded specimens would yieldlower contact angles than the non-sanded specimens. How-

ever, inexplicably the opposite occ~. The sanded surfaceswere visually detennined to be smoother than non-sandedsurfaces. This difference was continued from scanning elec-tron micrographs. The sanded yellow-poplar transverse sec-tion (Fig. 1) appears to be smoother than the non-sandedspecimen (Fig. 2). Similarly, the radial surface of a sandedsouthern red oak specimen (Fig. 3) is smoother than acorresponding non-sanded specimen (Fig. 4). It should benoced that surface roughness was not quantitatively meas-ured in this study.

The findings in this study are in agreement with those ofprevious studies that have shown surface roughness has aminimal impact on wettability (Gray 1961, 1962; Herczeg

~typel

Sanded

(X)2

Sanded(T~

Sandedmean.

Notsanded

(X)

Notsanded

(r)

sQ5 Dan:y6Notsandedmean

FibeTdi 7am.

Pore type 0.08'(0.72)9

-0.11(0.64)0.70

(0.00)

-0.00

(1.00)0.94

(0.00)0.900

(0.00)

-0.43

(0.04)

-0.02

(0.94)O.~

(0.73)0.03

(0.91)

-0.42

(0.05)

0.16

(0.49)0.47

(0.03)0.31

(0.15)0.33

(0.14)

-0.32(0.01)0.06

(0.78)0.29

(0.18)0.18

(0.43)0.86

(0.00)0.76

(0.00)

-0.80(0.00)-0.17(0.46)-O.OS(0.81)-0.13(0.51)0.40

(0.06)0.36

(0.10)0.47

(0.03)

-0.21(0.34)-0.27(0.22)~(0.11)-0.33(0.13)o.m

(0.84)o.~

(0.78)0.07(0.76)0.19

(0.40)

0.86(0.00)0.19(0.39)~.OO(1.00)0.12

(O.W)~.s3(0.01)~.37(0.09)-0..57(0.01)~.7S(0.00)~.)4(0..54)

Sanded (X) o.~(0.72)-0.11

(0.64)-0.00(1.00)-0.43(0.04)-0.42(O.QS)-0.52(0.01)-0.80(0.00)-0.21(0.34)0.86

(0.00)

Sanded (T 0.70(0.00)0.94

(0.00)-0.02(0.94)0.16

(0.49)0.06

(0.78)-0.17(0.46)-0.27(0.22)0.19

(0.39)

Sanded mean o.~(0.00)o.~

(0.73)0.47

(0.03)0.29

(0.18)-O.OS(0.81)

-0.35

(0.11)-0.00

(1.00)

Not sanded (X) 0.03(0.91)0.31(O.IS)0.18

(0.43)-0.13

(0.37)-0.33(0.13)0.12

(0.60)

Not sanded (T) 0.33(0.14)0.86(0.00)0.40

(0.06)o.os

(0.84)-0.53(0.01)

Not sanded mean 0.76(0.00)0.36(0.10)0.06

(0.78)-{).37(0.09)

so

Dan:y 0.19(0.40)-0.75

(0.00)

Fiber diam. -0.14(0.54)

I Species data entered as 1 = ring porous, 2 = diffuse pcxooua2 Trans~.) Tangential.4 Mean of b'lnsvax and tangential values.S Mean specific gravity (Choong et al. 1974).

0.47(0.03)0.07(0.76)-0.57(0.01)

6 Longitudinal penIIeability Dan:y values (0KIc.I8 eI aI. 1974:7 Fiber radial diameter (Manwiller no year given).. Pearson correlation coefficieut (R).9 Probability> IRI under Ho: Rho . 0, N . 22.

Holzforschung I Vol. SS I ~II No. S

T.F. Shupe et aI.: Hardwood Wettability546

1965). However, other studies have found decreasing wettingangles (improved wettability) with increasing roughness(Marian and Stumbo 1962a, b). It is acknowledged that sur-face roughness affects the contact angle measurementof wood; it is also apparent that other factors in addition tosurface roughness have a significant effect and must be con-

sidered when considering the relationship between surfaceproperties of the solid and a liquid. These other properties,including the surface tension and viscosity of the liquid,surface molecular packing, critical surface tension of thesolid, and the solid-liquid interaction all impact contact anglevalues.

Wood surface (transverse, radial, and tangential) effect

The mean contact angle on the transverse, radial, and tan-gential surfaces fOf southern red oak. sweetgum, white oak,and post oak is shown in Table 6. As expected, values on theradial and tangential surfaces were similar because of therelative similar anatomical composition of these longitudi-nal surfaces compared to the transverse surface. As expect-ed, contact angle values were higher on the transverse SUf-face than the radial and tangential surfaces. This findingcontradicts results from Gray (1962) who stated that thereis no consistent difference between wettability of some 15species of wood in different grain directions within the sta-tistical variation encountered with readings in the samedirection on the same specimen. The finding by Gray (1962)is peculiar given the inherent variability in wood surfacechemistry, permeability, and anatomical structure that existsbetween species and these properties' influence on contactangle determination and wettability.

Fig. 2. Scanning electron micrograph of the uansverse surface ofa non-sanded yellow-poplar specimen.

Fig. 3. Scanning electron micrograph of the radial surface of asanded southern red oak. specimen.

Drying method effect

Mean data for the four species included in the dryingmethod effect study are presented in Table 6 and a summa-rized analysis of variance is shown in Table 7. It was antic-ipated that oven-dry specimens would give higher contactangles because of the deactivation of the surface that occursdue to the oven-drying process (Gardner et oJ. 1996) andthe surface migration of extractives (Hse and Kuo 1988).The oven-dry specimens were dried at 105 °C for 24 hours.Oven-dry specimens yielded the highest mean contact angleof 63.3°, followed in decreasing order by air air-dry speci-mens (62.9°) and freeze-dry specimens (57.0°). Freeze-driedspecimens gave the lowest mean contact angle. The freez-ing process likely preserves, rather than degrades, the chem-ical properties of the wood surface. Therefore, surface deac-tivation is minimal and wetting is more favorable.

This component of the study also investigated contactangles on all three surfaces of each of the four species.Transverse values were consistently higher than radial andtangential values, which were nearly identical. This patternis largely attributable to the higher surface roughness andopen cell lumens on the transverse surface. Compared to thetransverse surface, the radial and tangential surfaces aremore anatomically similar and would likely have similarsurface roughness values.

It is acknowledged that there are other factors that in-fluence contact angle measurements. Chen (1970) reportedthat wood extractives can influence the contact angle. Jor-dan and Wellons (1977) found that extraction significantlyincreased the wetting of dipterocarp veneers. Kajita and Skaar(1992) attributed greater wettability of sapwood compared

Fig. 4. Scanning electron micrograph of the radial surface of anon-sanded southern red oak

HolzforschuR2/ Vol. SS /2001/ No.5

S47T.F. Shupe et aI.: Haniwood Wettability

Table 7. Summarized analysis of variance of the effect of species(southern red oak; sweetgum, white oak, and post oak), dryingmethod, and wood plane on contact angle detennination

transverse, tangential, and radial surfaces with regard to

three different drying methods.Contact angle values differed significantly between species

and between the 22 species, sanded and non-sanded sur-faces, and transverse and tangential planes. Contact angleson transverse planes were higher than those on tangentialplanes. Both transverse and tangential planes yielded highervalues on sanded surfaces as compared to non-sanded sur-faces. Contact angles were found to vary significantly ac-cording to drying method and wood surface. Contact anglevalues observed on transverse surfaces were higher thanthose observed on radial and tangential surfaces. Air-driedspecimens on average had the highest contact angles, andfreeze-dried specimens typically gave the lowest contactangles. Many of the differences in contact angle values arelargely attributed to surface roughness differences of thedifferent species and different wood surfaces.

df2 P-valueSOY!

0.1XX>1..0.1XX>1..0.1XX>1..0.1XX>1..0.1XX>1..0.0011..0.0019"

Species (S)Drying method (D)Plane (P)S*Ds*pD*PS*D*P

32266412

1 Source of variation.2 Degrees of freedom.** = Denotes significance at alpha = 0.01

to heartwood, to the higher extractive content of heartwood.Although extractives tend to dominate the wood surface, allthe chemical components comprising wood contribute to itssurface chemistry and thus affect surface activation (Gard-ner et aI. 1996). In addition, the surface tension and vis-cosity of the liquid, surface molecular packing, critical sur-face tension of the solid, and the solid-liquid interaction allimpact contact angle values. Surface roughness also affectscontact angle because it creates more than one metastablestate at the solid-liquid-vapor interface (Johnson and Dettre

1993).

AcknowledgementThis paper (No. 00-22-0216) is published with the approval of theDirector of the Louisiana Agricultural Experiment Station.

Conclusions

This study was initiated to determine the contact angle of22 southern hardwood species on sanded and non-sandedsurfaces of transverse and tangential surfaces. Moreover,four species were selected to determine contact angles on

ReferencesAdams, N.K. 1949. The Physics and Chemistry of Surfaces. Third

ed. Oxford University Press, London.Bodig, J. 1962. Wettability related to gluabilities of five Philippine

mahoganies. Forest Prod. J. 12(6),265-270.Chen, C.M. 1970. Effect of extractive removal on adhesion and

wettability of some tropical woods. Forest Prod. J. 20( I), 36-40.Choong, E.T., F.O. Tesoro and F.G. Manwiller. 1974. Permability

of twenty-two small diameter hardwoods growing on southern

pine sites. Wood Fiber 6(1),91-101.Freeman, H.A. 1959. Relation between physical and chemical prop-

erties of wood and adhesion. Forest Prod. J. 9(12),451-458.

Holzforschung / Vol. 55 / 200 1/ No.5

T.F. Shupe et al.: Hardwood Wettability548

face tension and the penetration of liquids into the wood sur-face. Holzforschung 52, 89-94.

Suschland, O. and R.R. Stevens. 1968. Gluability of southern pineveneer dried at high temperatures. Forest Prod. J. 18(1),38-42.

Wenzel, R.N. 1936. Resistance of solid surfaces to wetting bywater. Ind. Eng. Chern. 28, 988-994.

Zisman, W.A. 1964. Relation of the equilibrium contact angle toliquid and solid constitution. In: Contact Angle, Wettabilityand Adhesion. Advances in Chemistry Series No. 43. AmericanChemical Society. Washington, DC. pp. 1-51.

Zisrnan, W.A. 1976. 1nf1uence of constitution on adhesion. In:Handbook of Adhesives. 21M! ed. Van Nostrand Reinhold Co.,New York. pp. 33-71.

Received June 7th 2<XX>

Todd F. Shupel)Assistant ProfessorSchool of Forestry, Wtldlife, and FisheriesLouisiana State University Agricultural CenterBaton Rouge, LA 70803U.S.A.

Chung Y. Hse2)Principal Wood ScientistUSDA Forest ServiceSouthern Research StationPineville, LA 71360U.S.A.

Gardner, D.J., M.P. Wolcott, L. Wilson, Y. Huang, and M. Carpenter.1996. Our understanding of wood surface chemistry in 1995.In: Wood Adhesives. Eds. A. W. Christiansen and A.H. Conner.1995. Forest Products Society. Madison, WI. pp. 29-36.

Gray, V.R. 1961. Wetting, adhesion and penetraton of surface coat-ings on wood. J. Oil and Colour Chemists' Assn. 44,756-786.

Gray, V.R. 1962. The wettability of wood. Forest Prod. J. 12(9),452-461.

Hameed, M. and E. RoffaeL 1999. Uber die Benetzbarkeit vonSplint- und Kernholz der Kiefer, Douglasie und Urche. HolzRoh- Werkstoff 57,287-293.

Herczeg, A. 1965. Wettability of wood. Forest Prod. J. 15(11),499-505.

Hse, C. Y. 1972. Wettability of southern pine veneer by phenolformaldehyde wood adhesives. Forest Prod. J. 22(1), 51-56.

Hse, C. Y. and M.L. Kuo. 1988. Influence of extractives on woodgluing and finishing - a review. Forest Prod. J. 38(1), 52-56.

Johnson, R.E. and R.H. Dettre. 1993. Wetting of low-energy sur-faces. In: Wettability. Ed. J.C. Berg. Marcel Dekker Inc. NewYork. pp. 1-74.

Jordan, D.L. and J.D. Wellons. 1977. Wettability of dipterocarpveneers. Wood Sci. 10(1),22-27.

Kajita, H. and C. Skaar. 1992. Wettability of the surfaces of someAmerican softwoods species. Mokuzai Gakk. 38, 516-521.

Manwiller, F.G. Stemwood and branchwood fiber dimensions(transverse) in 6..inch-diameter hardwoods of 22 species grownon southern pine sites. Data in files of USDA Forest Serv.Southern Research Station. Study FS-SO-3201-1.42. Pineville,LA.

Marian, J.E. and D.A. Stumbo. 1962a. Adhesion in wood: Part I.Physical factors. Holzforschung 16(5), 134-148.

Marian, J.E. and D.A. Stumbo. 1962b. Adhesion in wood: Part ll.Pbysico-chemicaJ surface phenomena and the thermodynamicapproach to adhesion. Holzforschung 16(6), 168-180.

SAS Institute, Inc. 1989. SAS/STAT User's guide. Version 6, 4111ed., Vol. 2. Cary, N.C. 846 pp.

Scheikl, M. and M. Dunky. 1998. Measurement of dynamic andstatic contact angles on wood for the determination of its sur-

Wan H. WangAssociate ProfessorDepartment of Wood Science and TechnologyNanjing Forestry UniversityNanjingPeoples Republic of China

I) Author to whom correspondence should be directed.2) Addresses where the wod was perfomled.

Holzforschung / Vol. 5S /2001/ No. S