COMPARING MECHANICAL PROPERTIES OF ...

[AWA Journal, Vol. 18 ([), 1997: 77 -88

COMPARING MECHANICAL PROPERTIES OF SECONDARY WALL

AND CELL COR.~ER MIDDLE LAMELLA IN SPRUCE WOOD

by

Rupert Wimmer J & Barry N. Lucas -

SUMMARY

Mechanical characterizations of the 52 layers and the cell comer middle lamella in the axial direction were investigated in spruce wood . A me­ chanical properties microprobe capable of measuring hardness and Young's modulus on a spatially resolved basis similar to that of an elec­ tron beam microprobe was used. Hardness of the cell comer middle lamel­ la was found to be almost as high as that of the secondary wall. but the Young's modulus of the cell corner middle lamella was 50% less than that of the 52' The 52 showed constant hardness over its range of Young's modulus, but the cell comer middle lamella exhibited a strong correlation (R2 = 0.55) between hardness and the Young 's modulus. Further inves­ tigations are needed to dire ctly combine chemical and micromechanical properties and also to investigate the mechanical effects of the high varia­ bility of cell corner middle lamella chemistry.

Key words: Spruce, Picea, hardness, Young 's modulus, secondary wall, middle lamella, mechanical property. INTRODUCTION

High resolution mechanical property measurements of cell wall material have been done previously. Such studies have included micro-tensile tests of microtomed (usu­ ally 80 urn thick) wood sections (e.g., Wellwood et al. 1965; Kennedy & Ifju 1962; Grozdits & Ifju 1969) , tensile tests on individual fibers obtained through chemical maceration (e.g., Leopold & Macintosh 1961; Schniewind et al. 1965), and a micro­ compression test done perpendicular to the grain (Wilson 1964). Because the 52 layer is the thickest cell wall layer it is been suggested that it controls the strength of the entire fiber (Wardrop & Dadswell 195J; Abe et al. 1991). An analogy can be drawn between the structure of the 52 layer and that of a unidirectional fiber-reinforced com­ posite material, in which the cellulose fibrils represent the fiber reinforcement, and the mostly amorphous hemicellulose and lignin represent the composite matrix. However, the mechanical behavior of tissues cannot be explained solely by the material proper­ ties of secondary cell walls and other factors need to be considered, one being the physical properties of the middle lamella. I) Univer sitat fur Bodenkultur Wien, Gregor Mendel-Strasse 33. 1180 Vienna. Austria. 2) Nano Instruments. Inc., P.O. Box 14211, Knoxville, Tennessee 37014. U.S.A.

78

IAWA Journal, Vol. 18 (1), 1997

In unlignified woody tissues, the primary walls of adjoining cells are separated by a pectic layer, called the middle lamella, that essentially acts as a cementing agent. Lignin first is deposited in the primary wall at the cell comer and then in the cell comer middle lamella, tangential compound middle lamella, radial compound middle lamella, and secondary wall (Donaldson 1991). The impregnation of lignin in the cell wall and the middle lamella can alter the material properties of these structures . When lignification occurs, the distinctions between the primary cell walls of adjoining cells and the in­ tervening middle lamella are lost, and the three layers are then referred to as the com­ pound middle lamella. The volume fraction of the compound middle lamella in conif­ erous wood is c. 10-12% of the woody tissue volume (Fergus et a1. 1969). To measure the mechanical properties of the 52 and the compound middle lamella, depth-sensing indentation was applied as a testing method in this study. According to Brinell, the impression of a steel-ball on a smooth metallic surface, not too close to the edges, delivers clear, reproducible results . At the beginning of the century, Janka (1906) proposed and developed a modified Brinell-hardness test for wood. The static force required to completely embed a steel hemisphere 0.444 inch (11.5 rom) in diameter, corresponding to a projected hemispherical surface of 1 ern", into the wood was meas­ ured. Because the steel hemisphere covers a wide range of cellular structures, the meas­ ured hardness is approximately proportional to the density of the wood or cell mass per unit volume . The distinguishing feature of the indentations used for this investigation is their very small size. Using a mechanical properties microprobe allows one to evaluate the mechanical response of a sample with submicron spatial resolution (Oliver 1986; Oliver & Pharr 1992; Page et a1. 1992) and this method has recently been used on woody tissues (Wimmer et al. 1997). The objectives of the current study were to 1) simultane­ ously quantify the patterns of hardness and Young's modulus of the cell comer middle lamella (CCML) and the secondary wall (52)' and 2) look for differences and trends in these properties across tree rings of spruce. MATERIAL AND METHODS

From a 80-year-old red spruce tree (Picea rubens Sarg.) a 5 mm increment core was taken and a sequence of five tree rings prepared so that cross-sectional areas of about 1 mm- were obtained . Samples were prepared as usually done for a transmission elec ­ tron microscopy study. To minimize leaching effects as well as possible changes in chemical composition, specimens were air dried at room temperature for several days and then oven dried at 70 °C for about 20 minutes before embedding in resin according to the method of Spurr (1969) . The compact structure of the woody cell wall allows no penetration of the embedding medium and thus a phase boundary exists between the cell wall and the polymeride (Jayme & Fengel 1961; Fengel 1967). Under vacuum. the resin filled the lumina and then the resin was cured in the usual way. A surface on the embedded specimen was sectioned using a Reichert-Ultramicrotome with a diamond knife . Mechanical tests were done of pure cell wall material and of the cell corner middle lamella . The rnicrotomed samples provided sufficient surface quality for the

Wimmer & Lucas - Mechanical properties of cell wall and middle lamella

79

A 1000 500

o

o

o

0

Fig. 1. Three-dimensional reconstructed atomic-force-microscope (AFM) images: a) remaining plastic deformation of a single 52 indentation (indentation depth = 80 run): b) remaining plastic deformation of a cell corner middle lamella indentation close to [he tracheid celt wall.

indent ation tes ts. Th e spec ime ns were glue d on aluminum stu bs using a laser tech­ niqu e for perpendicul ar alignm ent. Moun ted specime ns were fixed in the stage of the Na no Ind en ter IJ® (Nano Instruments, Inc., Kn oxvill e , TN). Th e operation pri nci ples of the co mp uter -controlled Nano Indenter Il ®are described in deta il elsewhe re (Oliver & Ph arr 1992 ; Will ems et a1. 1993; Wimmer et a1. 1997 ).

80

IAWA Journal, Vol. 18 (1) ,1997

The Nano Indenter II®was enclosed in a heavy wood en cabinet, the major purpo se of which is to ensure thermal stability during an experiment. The indenter itself was sus­ pended on a pneumatic antivibration table to isolate it fro m building vib rations. The apparatus was located in a room stabilized at 21 °C and relativ e humidity was around 60 % throughout the experiment. With the Nano Indenter ll ®, loads on the order of a few I-lN can be applied to a pyramidal diamond indenter, with a resolution better than 50 nN, and the resulting depth displacement can be measured to 0.16 nm. A load­ controll ed instrument mode is used in examining hardness at shallow depths. The sys­ tem controls the load as well as the loading rate continuously during the loading and unload ing segments of the indentati on procedure. With knowledge of the indenter geo­ metry, the contact area can be determined and the nanohardness value for the applied load can be calculated (Oliver 1986 ; Wimmer et a1. 1997) . Unlike conventional hard­ ness testers, it is not necessary to optically determine the area of an indentation in order to calculate the hardness (Fig. 1a). Because the radial and tangential oriented compound middle lamella were too nar­ row « 0.5 um), the indent ation tests were done solely on cell com er compound middle lamell a (CCML) which provided surfaces of approximately 2 x 2 um (Fig. 1b). Two sets of measurements were made . In the first set 5 tree ring samples were prepared and subsequently mea sured with the Nano Indenter II®. Each indentation was check ed carefully using refle cted light microscopy and for more detailed analy sis an atomic force microscope in contact mod e (Park Scientifi c lnstruments'P) was used. In total 267 single measurements were obtained. All these mea surements were done solely in latew ood becau se their tracheid wall s were wide enough for locatin g the indentation s. The second set of measurements characterized properties along a radial file of rracheids , In total , properties of 12 co nsecutive trach eid s, and the adjacent CCML, were meas­ ured over a distan ce of 250 urn, Measnrernents started with the last latewood tracheid and proceeded toward the earlywood. RESULTS

Dataset J Fignres 2a and b are box-and-whisker plots showin g Young 's modulus and hard­ ness of radial and tangential S 2-walls, CCML and also Spurr's resin as a control. In this data set (267 measurements), the average Young 's modulus of tangenti ally and radially oriented trach eid walls was 16 GPa and the hardne ss was 0.30 GPa. Differ­ ences between radial and tangential cell wall s were not significant. Both hardness and Young's modulus of the CCML were lower than the S2, but to a diffe rent degree. The S 2 layer displayed an elasti c modulus nearly twice that of CCML but the S2 was only 10% harder than CCML. Dataset 2 The second data set comprised measurements mad e across a 250 J..IIl1 transect along a radi al file of trach eids in a single tree ring (Table 1). The Young's modulus of S 2 walls in this second dataset was nearl y three time s as high as the CCML, and the hard­

Wimmer & Lucas - Mechanical properties of cell wall and middle lamella

81

a

o

o 0

o

o

0

o

CCML

co

Resin

0

10

5

20

15

25

30

Youngs' modulus (GPa)

b Sj-tang

00 0

5 2- rad

0

o

- -1 em

0 0

CCM L

(ID O

0

Re sin

0.0

0.1

0.2

0.3

0.4

0.5

Hardness (GPa) Fig. 2. Box-and-Whisker Plot of Young's modulus (a) and hardness (b) for the anatomical com­ ponents. Number of measurements: Sz-tang (85). Sz-rad (62), CCML (97). Resin (23) .

ness of the S 2 wa lls was approxim ately 20 % more than the CC M L. Coefficien ts of variatio n wer e co nside rably high er for CCML than for 5 2 walls . Figures 3a and b dep ict the mec hanical trend s for CCML and 5 2 layers for the me asure d di stance fro m latewood tow ard s ear lywood .

IAWA Journ al, Vol. 18 ( 1), 1997

82

30

1 - - - - - - - - - - - - - --;::::::=======:1

a

25

20

15

10

5

latewood

o

earlywood

50

100 150 Distance (urn)

200

250

300

0.50 T:'""------A-------::::::======~

b

-...- 5 2 layer -0--- CCML

OA5 OAO '2 c,

0.35

v: ~

0.30

o '-'

"§

:I:

0.25

0.20 0.15

early wood

latewood

o

50

100 150 Distance (urn)

200

250

300

Fig. 3. Young's modulus (a) and hardness (b) of radial 52 layer and cell comer middle lamella along a 250 urn distance in a tree ri ng.

Three out of 76 CCML hardness values were above those of the S 2 walls, which could be due to scatter. The hardness of the very last latewood tracheids in the growth ring is low, but hardness increases significantly in the subsequent 1-2 tracheids. In contrast, CCML values do not show this difference. The lowest values for Young's

Wimmer & Lucas -

Mechanical properties of cell wall and middle lamella

83

Table I. Comparison of longitudinal hardness and Young' s modulus (GPa) of secondary wall layers (S 2 layer) and cell corner middle lamellas (CCML) . Hardness

S2 layer

Young' s modulus

CCML

5 2 layer

CCML

mean

0.335

0.267

19.70

6.89

SD

0.030

0.058

3.14

2.26

CV O/O m in

9 0.226

max

0.383

counts

50

21

IS

21 0.125

10.34

4.14

0.475

27.29

11.69

SO

76

76

modu lns and hardness of both the S2 and the CCML occ urred at the dis tance of 220 urn. Figure 4 shows a scatter plot presenting hard ness values plotted aga ins t their cor­ responding Young's modu lus for CCML and S z. Diffe rent clusters are observed for the S2 and CCML. Hardness and Young's modul us of CCML are high ly correlated to each another (R2 = 0.55, p < 0.00 1), but no such correlations occur for the 5 2 (R2 = 0.09 , n.s.) .

0.50

•

0 0

0.45

.

0 0.40

S'

~

S '" ." '"

'"~

iA~:

0.35 0.30

'0

::c

0.2 5

:£J

0.20

•

0

•

··+X~· •• • •• •

=

S2 lay er CCML

.'

•

0

0 0 0

0.15 0.10 0

5

10

15

20

25

30

Young's modu lus (GPa)

Fig . 4. Hardness-Young's modu lus scatterploLshowing relation ships for cell comer midd le la­ mella and 52'

84

IAWA Journal, Vol. 18 (I) , 1997

DISCUSSION

Because the mechanical properties microprobe allowed measurements on a spatially resolved basis that was smaller than the width of single cell walls, the effects of cell wall proportions and wood density on mechanical properties were not being directly measured as in other wood property studies (e.g. Wellwood et al. 1965). In this study. the mechanical measurements of S2 walls and the CCML reflect the ultrastructural and chemical variability present in wood . The S2 layer and CCML differ chemically and structurally, thus knowledge of the factors controlling the biochemical and bio­ mechanical processes occurring during cell wall formation are important. Cell comers of sprucewood are highlylignified(e.g., Downes et al.199l; Wimmer & McLaugWin 1996), as is the CCML (Fergus et al. 1969). Because no data were found for direct mechanical CCML measurements, results that refer to the mechanical behavior of isolated lignin are discussed. Cousins et al. (1975) obtained Young's modulus of di­ oxane lignin from a continuous indentation test. The lignin was chemically dissolved using dioxane and precipitated to a thick syrup in a thin stream and dried . The dry lig­ nin powder resulted in a rod and appeared to be hard and brittle. The Young's modulus of these lignin samples was 3.3 GPa with a coefficient of variation of 12%. Young's modulus for CCML was twice the dioxane lignin elasticity. A few points might be worthy of discussion. First, isolating lignin using dioxane introduces the problem that all methods of isolation have. namely. these methods either fundamentally or at least partially change the native lignin structure (Fengel & Wegener 1989). Further, during the lignin isolation process extractives and other chemical components are removed or, at least, have changed their molecular structure. Cousins (1976) has compared different lignins and found periodate lignin to be the one that undergoes the fewest changes in either physical or chemical structure during isolation. This type of lignin had the high­ est moduli but was also most sensitive to moisture changes. The present work was done at rue equilibrium moisture of 10.9% which gave mechanical values 10% under those expected at 0% moisture (see Wimmer et al. 1997 for further discussion) . The periodate lignin (Cousins 1976) would give 4 GPa at 10.9% equilibrium moisture content. Meier (1961) analyzed spruce and pine fibers at different stages of maturation and found that the middle lamella/primary wall of cambial cells was rich in pectic acids, arabinose and galactose. Therefore, it is likely that a change of these chemical compo­ nents could alter the mechanical properties of the CCML. Pectin plays a important role during the lignification of wood cells, involving the cation Ca 2+ . Studies have demon­ strated that the cambium has an especially high Ca 2 + concentration (Wardrop 1976). Pectin is a good chelator of Ca 2+, building strong 'egg in the box' structures that can be formed between polygalacturonic acids and Ca 2 + (Grant et al. 1973). Westermark et al. (1986) found Ca 2+ to be high in unlignified woody tissues at high lignin concen­ trations indicating that pectin may act as a selective binder for Ca 2 + in the camhial cells. Pectin degrades or will be removed prior to the lignification apparently releasing Ca 2+ ions that are used in lignification. The released and positively charged Ca 2 + ions bind to negatively charged groups of lignin and this could affect the mechanical prop­ erties of the CCML.

Winuner & Lucas -

Mechanical properties of cell wall and middle lamella

85

From cell wall fracture investigations it is known that fractures occur predominant­ ly between the middle lamella and 5 I layers or between the 5 I and S21ayers (Donaldson 1995). The rapid change from lignin-rich to lignin-poor lamellae as observed within the 51 layer (Maurer & Fenge11991) favors cell wall fracture. Mark (1967) demon­ strated that wood fracture is rarely initiated within the compound middle lamella. It is more likely that the S I layer is the first to undergo mechanical failure, usually because of shearing . Mark (1967) showed that the shear stresses in the 5 I and S2 layer are typically opposite in direction, as might be anticipated from the difference in their microfibrillar orientation angles. This agrees with finding high CCML hardness which is an indicator for high mechanical strength. The mechanical measurements along a 250 11m radial transect showed the CCML to have a higher variability than the 52' The high variability of the CCML could be ex­ plained by the high variability of its chemistry. In a recent study, Tirumalai et a1. (1996) used Raman microprobe spectroscopy to study the concentration of ligno­ cellulosics in the CCML of black spruce. They found that the relative concentration of lignin and cellulose varies considerably, partially because of lignin deficient regions . Similar results were found in birch (Daniel et a1. 1991). 5 2 layers have a significantly higher Young's modulns than the CCML, but are not particularly harder than the CCML. Maximum Young's modulus did not exceed 27 GPa and the mean is less than theoretically calculated values for S2, which vary be­ tween 28 GPa for earlywood and 35 GPa for latewood (Cave 1968). The final 1 or 2 latewood tracheids in a tree ring are usually very narrow radially. The mechanical properties, particularly hardness, of these final tracheids were significantly lower than those of other tracheids. A change in cell wall chemistry likely explains this finding. Fukazawa and lmagawa (1981) showed in a UV-microscopic study with Abies that cells in the terminal zone of the latewood have a distinctive high lignin content. Wilson and Wellwood (1965) and also Wu and Wilson (1967) found that in conifers early wood lignin content was consistently 2 to 3% higher than Iatewood lignin content. Con­ versely, the per cent cellulose content is higher in latewood (De Zeenw 1965; Fengel 1969) and there are longer cellulose chains ,better packing and higher crystallinity (Lee 1961). These characteristics explain the mechanically stronger latewood (Wimmer et al. 1997) with the exception of the very final (1-2) latewood tracheids which have a different chemistry. A discussion of the trends in Figure 4 leads to the question of positional dependen­ cies of the indentations within CCML and 5 2' Measurements in CCML, if done very close to 52, could be affected by the 5 2 mechanical properties and vice versa. How­ ever, if this were the case, the Young's modulus should be affected first due to the longer range of the elastic strain fields during an indentation. Therefore, the 52 Young's modulus should be affected much more by position within S2 than is hardness, the latter remaining nearly constant. As far as the strong correlation between hardness and Young's modulus of CCML is concerned, it is possible that there is not a positional effect within the CCML, but rather that the high chemical variability due to lignin deficient regions is the main cause for this phenomenon. Due to the variability in chem­ istry and, consequently, packing density, hardness and Young's modulus are affected

86

IAWA Journal, Vol. 18 (I), 1997

in the same way resulting in a strong correlation. On the other hand, in the Sz the amount of cellulose is more or less constant over the measured distance of 250 urn and therefore packing density has not affected hardness which stays equal along a large range of Young's moduli . But the orientation of the cellulose in the Sz (microfibrillar angle) might have been altered to some extent influencing the Young's modulus con­ siderably as found by Cowdrey and Preston (1966). CONCLUSIONS

For a better understanding of the mechanical behavior of wood it is essential to have basic information on the property values of single wood components such as the Sz and the CCML. The mechanical properties microprobe used in this investigation was capable of measuring hardness and Young's modulus on a spatially resolved basis similar to that of an electron beam microprobe. The CCML was found to be mechani­ cally strong with hardness values almost as high as the secondary walls. As a hypoth­ esis, in the lignin enriched middle lamella the positively charged Ca 2+ ions could bind to the negatively charged groups at the lignin, providing additional strength to the CCML. Although we have measured mechanical properties from Sz and CCML solely, it has to be reemphasized that in solid wood the individual cells are not isolated and they rarely operate in isolation from neighboring cell walls. Due to the high variability of the chemistry in the CCML as well as cellulose orientation in the S2' further investiga­ tions are needed that link chemical and micromechanical properties at a high spatial resolution . ACKNOWLEDGMENTS The authors gratefully acknowtedge the assistance of Drs . W. C. Oliver and T. Y. Tsui for their collaborative role . Helpful ideas and suggestions were provided by Drs, B. Gardner and B. Hinter­ stois ser, Thanks are due to Prof. E.A. Wheeler for inten sive editing, and to two anonymous reviewers. Research was conducted while the senior author was a Visiting Scientist at the Oak Ridge National Laboratory. Oak Ridge, TN 37831 partly with joint funding by the Austrian Science Foundation (Schrodinger Seholarship 1799-BIO) and by the U.S . Environmental Protection Agency under Interagency agreement with the U.S. Department of Energy.

REFERENCES Abe, H., J. Ohtani & K. Fukazawa. 1991 . FE-SEM observations on the microfibrillar orientation in the secondary wall of tracheids. IAWA Bull . n. s. 12: 431-438. Cave, J.D. 1968. The anisotropic elasticity of the plant cell wall. Wood Sci. Techno!. 2: 268­ 278. Cousins, W. J. 1976. Elastic modulus of lignin as related to moisture content. Wood Sci. Technol, 10: 9-17. Cousins, W.J., R.w. Armstrong & W.H. Robinson. 1975 . Young 's modulus of lignin from a continuous indentation test. J . Mat. Sci. 10: 1655-1658.

Wimmer & Lucas -

Mechanical properties of cell wall and middle lamella

87

Cowdrey, D.R. & F.R.S. Preston. 1966. Elasticity and microfibrillar angle in the wood of Sitka spruce. Proc. Roy. Soc . B 166: 245-272. Daniel, G., T. Nils son & B. Pettersson . 1991. Poorly and non-lignified regions in the middle lamella cell comers of birch (Betula vermcosa) and other wood species . IAWA Bull. n. s. 12: 70-83. De Zeeuw, C. 1965 . Variability in wood. In: w.A. Cote Jr. (ed .), Cellular ultrastructure of woody plants: 457-471. Syracuse Uni versity Pre ss. Donaldson, L.A. 1991. Seasonal changes in lignin distribution during tracheid development in Pinus radiata D. Don. Wood Sci. Techuol. 25: 15-24. Donaldson, L.A. 1995. Cell wall fraction properties in relation to lignin distribution and cell dimensions among three genetic groups of radiata pine. Wood Sci . Technol, 29: 51-63 . Downes, G. M., J.Y. Ward & N.D. Turvey. 1991. Lignin distribution across trach eid cell walls of poorly lignified wood from deformed copper deficient Pinus radiata (D. Don). Wood Sci. Techno!. 25: 7-14. Fengel, D. 1967. Ultramicrotomy, its application in wood research. Wood Sci . Technol. I : 191­ 204. Fengel, D. 1969. The ultrastructnre of cellulose from wood . Pan I: Wood as the basic material for the isolation of cellulose. Wood Sci . Technol, 3: 203-217. Fengel, D. & G. Wegener. 1989. Wood - chemistry, ultrastructure, reactions. Walter De Gruyter, Berlin, New York Fergus, B. J., A.R. Procter, J.A.N. Scott & D .A . Goring. 1969. The distribution of lignin in sprucewood as determined by ultraviolet microscopy. Wood Sci . Techno!. 3: 117-138. Fukazawa, K. & H. Imagawa. 1981. Qnantitati ve analysis of lignin using an UV microscopic image analyzer. Variation within one growth increment. Wood Sci. Techno!. 15: 45 -55. Grant, G.T.. E.R. Morris, D.A. Rees, P.J.c. Smith & D. Thorn. 1973. Biological interactions between polysaccharides and divalent cations. The egg-box model. pEBS Letters 21: 195­ 198. Grozdits, G .A. & G. Ifju. ]969. Development of tensile strength and related properties in differentiating coniferous xylem. Wood Sci. 1: 137-147. Janka, G. 1906. Die Harte der Hol zer. Cb!. Ges . Forsrwes. 32 : 193-202. Jayme, G. & D. Fenge!. 196]. Beobachtungen an Ultradunnschnitten von Fichtenholz. Holz Roh- u. Werkstoff 19: 50-55 . Kennedy, R.W . & G . Ifju . 1962. Applications of microtensile testing to thin wood sections. Tappi45:725-733. Lee, Ci L, 1961. Crystallinity of wood cellulose fibers studied by X-ray methods. For. Prod. J. II: 108-112. Leopold, B. & D. C. McIntosh . 1961 . Chemical composition and physical properties of wood fibres. III. Tensile strength of individual fibers from alkali extracted loblolly pine holocellu­ lose. Tappi 44: 235-240. Mark, R. E. 1967 . Cell wall mechanic s of tracheids, Yale University Press, New Haven, London. Maurer, A. & D. Fengel. 1991. Electron microscope representation of structural details in soft­ wood cell walls by very thin ultramicrotome sections. Holz Roh - n. Werkstoff 49: 53­ 56 . Meier, H. 1961. The distribution of polysaccharides in wood fibers . J. Polyrn. Sci. 51: 11-18. Oliver, W.C. 1986. Progress in the development of a mechanical properties microprobe. MRS Bull. 11 (5): 15-19. Oliver, W.C. & G .M. Pharr. 1992. An improved technique for determining hardness and elastic modulus using load and displacement sensing indentation experiments. J. Mater. Res. 7: 1564-1583.

88

IAWA Journal , Vo!. 18 (1),1997

Page, T.F., WC . Oliver & Ci J. McHargue . 1992. The deformation behavior of ceramic crystals subjected to very low load (nano ) indentations. J. Mater. Res. 7: 450-473. Schniewind, A.P., G. Ifju & D.L. Brink. 1965. Effect of drying on the flexural rigidity of single fibers. In: Consolidation of the paper web. Cambridge Symposium 1965. Technical section of the British paper and board makers association: 538-543 . London . Spurr, A. R. 1969 A low-viscosity epoxy resin embedding medium for electron microscope. J. Ultrastruct. Res. 26: 31-43. Tirurnalai, V.C., U. P. Agarwal & J.R. Obst. ] 996. Heterogeneity of lignin concentration in cell comer middle lamella of white birch and black spruce . Wood Sci. Technol, 30: 99-104. Wardrop, A.B. 1976. Lignification of the plant cell wall. Appl. Polym. Symp . 28: 1041-1063. Wardrop, A.B. & RE . Dadswell. 1951. Helical thickenings and micellar orientation in the sec­ ondary wall of conifer tracheid s, Nature 168: 610. Wellwood, R.W., G. Ifju & J.W Wilson. 1965. Intra-increment physical propertie s of certain western Canadian coniferous species. In: W A. Cote J r (ed.), Cellular ultrastructure of woody plants: 539-549. Syracuse University Press. Westerrnark, u., R-L. Hardell & T. Iversen . 1986. The content of protein and pectin in the lig­ nified middle lamella/primary wall from spruce fibers. Holzforschung 40: 65-68. Willems, G., J.P. Celis, P. Lambrechts , M. Braern & G. Vanherle. 1993. Hardness and Young's modulus determined by nanoindentation technique of filler particle s of dental restorative materials compared with human enamel. J. Biomed . Mat. Res. 27: 747-755 . Wilson, J.W. 1964. Wood characteristic s. Ill. Intra-increment physical and chemical propertie s. Pulp Paper Res. Inst. Can. Res. Note No. 45. Wilson, J.W & R.W Wellwood. 1965. Intra-increment chemical properties of certain Western Canadian coniferous species. In: W.A. Cote Jr (ed.), Cellular ultrastructure of woody plants: 551-559. Syracuse University Press. Wimmer, R., B.N. Lucas, T.Y. Tsui & W.e. Oliver. 1997. Longitudinal hardness and Young's modulus of spruce tracheid secondary walls using nanoindentation technique. Wood Sci. Techno!. 31 (in press). Wimmer, R. & S. B. McLaughlin. 1996. Possible relationships between chemistry and mechanical properties in the microstructure of red spruce xylem . In: I.S. Dean, D.M. Meko & T.W Swetnam (eds.), Tree rings , environment and humanity. Proceedings of the International Conference, Tucson , Arizon a, 17-21 May 1994: 659-668. Radiocarbon . Wu, Y. & J.W Wilson . 1967. Lignification within coniferous growth zones. Pulp Pap. Mag . Can. 68: Tl59-T171.