Autoclaved beaten wood fibre-reinforced cement composites

Autoclaved beaten wood fibre-reinforced cement composites

Autoclaved beaten wood fibrereinforced cement composites R.S.P. COUTTS The need to beat (or refine) wood pulp fibres before using as a reinforcement...

509KB Sizes 2 Downloads 40 Views

Autoclaved beaten wood fibrereinforced cement composites R.S.P.

COUTTS

The need to beat (or refine) wood pulp fibres before using as a reinforcement in cement composites has been re-examined. In the case of Pinus radiata unbleached kraft pulp, it has been confirmed that beating the fibres does contribute to attaining optimum mechanical properties for composites fabricated from such modified fibres. Key words: composite materials; reinforced cement; toughness; cellulosic fibre; flexural strength; fibre/matrix interface; bond strength; beaten fibres

Previous work 1 on laboratory and pilot plant scales has shown that the production of an asbestos-free fibrereinforced cement sheet could be achieved using refined wood pulp fibres as the sole source of reinforcement. These findings have been confirmed by the fact that commercial products (Hardiflex II* and Hardiplank II*) with material properties acceptable to the Australian building industry are now on sale. 2 In the earlier work, general claims were made about the effect that refined (or beaten) fibres had on the properties of composites produced from Pinus radiata fibres. Current studies have shown that beating Phorraium tenax (New Zealand flax) fibres before inclusion in cement composites had little effect on composite properties 3. This finding has prompted a re-examination of the earlier study. The process of beating (or refining) wood fibres has three main effects: 1)

the fibres are shortened;

2)

external fibrillation occurs, causing partial or sometimes total removal of the primary wall and causing fibrils to form on the surface of the fibre; and

3)

internal fibrillation occurs, causing the fibre to become more conformable.

Fines are also generated. It was accepted in the preliminary study I that freeness, the parameter used to control refining or beating, was more a measure of the fines present in a given pulp than the condition of the fibres' surface. Freeness and wetness testers are widely employed by papermakers and have been used for a long time to indicate the degree of beating or refining a fibre has undergone. They measure the ease with which water drains away from the papermaking fibres while being formed into a wet mat on the drainage plate of the tester. In this * Hardiflex II and Hardiplank II are trade names for commercial products manufactured by James Hardie Industries Ltd of Australia. These products are made from Portland cement, silica, cellulose fibre and water, and are asbestos-free.

study, a Canadian Standard Freeness Tester (CSF) was used. 4 Before beating (or refining) a softwood kraft pulp will have a CSF value of about 750 ml whereas a well-beaten pulp is typically 150 to 200 ml. The advantage of freeness tests is that they can be carried out quickly and easily in a commercial process. Fines have little beneficial effect on the properties imparted to a composite prepared in the laboratory, b u t may greatly assist fabrication of a product in a continuous industrial process, by varying drainage rates on a Hatschek machine. 2

EXPERIMENTAL DETAILS Materials and samples The fibre used in this study was P. radiata kraft lap from Kinleith, New Zealand. The fibre was treated in the laboratory in a Valley beater with a bed-plate load of 5.5 kg and a stock concentration of 360 g oven-dried fibre in 23 1 water. The matrix was prepared from equal proportions of ordinary Portland cement and finely ground silica (Steetley brand 100,WQ). The beaten fibres were mixed in a slurry (using tap water) of approximately 20% solids with a fibre fraction ranging from 2 to t4% by mass. The mixture was stirred for 5 min and poured into a 125 × 125 m m evacuable casting box so that it could be distributed over the screen. An initial vacuum was drawn until the sheet appeared dry on the surface, it was then flattened carefully with a tamper. A vacuum of 60 kPa (gauge) was applied for 2 min. The sheet was then removed on the filter screen. The sheet and screen were stored between two steel plates and the procedure repeated until a stack of six sheets had been prepared. The stack of sheets was then pressed for 5 min at a pressure of 3.2 MPa. The load was applied slowly so as not to damage the sheets. The preparation was completed within 1 h from the working of the first slurry. After pressing, the screens were carefully removed from the sheets which were then stacked flat in a sealed plastic bag for 24 h. After removal from the bag, the sheets

0 0 1 0 - 4 3 6 1 / 8 4 / 0 2 0 1 3 9 - 0 5 $03.00 © 1984 Butterworth ~ Co (Publishers) Ltd. COMPOSITES. VOLUME 15. NO 2. APRIL 1984

139

were finally cured in an autoclave for 8 h at 0.86 MPa steam pressure.

toughness is strictly only valid for specimens of the same thickness.

Test methods

Water-absorption values and density measurements were obtained using the methods laid down in ASTM-C220-75.

Specimens were cut with a diamond saw to specified dimensions and stored under standard conditions prior to testing. These involved placing the samples (1) in a controlled atmosphere of 50 ___ 5% relative humidity (RH) and 22 + 2°C for 5 days; (2) heating in an oven at 100-105°C for 24 h then cooling in a desiccator (OD) and (3) soaking in water for 48 h (WET) with excess water being removed with a cloth prior to testing.

In all cases at least nine samples were tested for flexural strength, fracture toughness, density and water absorption. Standard deviations have been included in the Tables. RESULTS AND DISCUSSION In an earlier study of P. radiata fibre cement composites, ~ it was concluded that the flexural strength varied with the mass fraction of fibre and the degree of fibre refining. The fracture toughness of a sample tended to decrease with increased refining of the fibres.

Rectangular strips measuring approximately 125 x 40 mm (of thickness 5-8 mm) were used for flexural strength and fracture energy tests of samples containing beaten fibres. The flexural strength was measured in threepoint bending as:

When Phormium tenax fibre-reinforced cement composites of constant composition were studied? no obvious variation in flexural strength took place as the degree of beating of the pulp was increased, nor did the toughness decrease in a systematic manner with a change in freeness.

3Pl/2ban

where P is the maximum load recorded during the test, l is the specimen span, b is the specimen width and d is the specimen depth. A span of 100 mm and a deflection rate of 0.5 mm min -~ was used on an Instron testing machine (Model 1114). The results of the flexural tests were obtained using automatic data collecting and processing equipment. The fracture energy was calculated from the area under the load/ deflection curve. For the purpose of this paper, the fracture toughness is given by the fracture energy divided by the cross-sectional area of the specimen. The comparison of fracture energy or fracture Table 1.

The mechanical properties of beaten P. radiata fibre/ cement composites are reported in Table 1. Flexural strength vs fibre percentage (by mass) at different beatings is shown in Fig. 1 and indicates that at fibre content up to approximately 6% by mass, little variation in flexural strength is observed with the degree of beating.

Properties of fibre cement composites

Freeness (CSF)

Fibre content (% by mass)

727 727 708 727 708 708 708

2 4 6 8 10 12 14

15.6± 16.6 ± 19.9 ± 23.1 ± 21.3 ± 21.7 ± 17.6 ±

546 546 546 546 546 546 546

2 4 6 8 10 12 14

14.7 ± 1.2 1 7 . 7 ± 1.2 20.3 ± 2.7 22.6 ± 2.8 24.3±1.9 23.0 ± 2.9 23.0 ± 3.1

514 514 514 514 514 514 514

2 4 6 8 10 12 14

288 288 288 288 288 288 288 160 160 160 160 160 160 160

140

Flexural strength

Fracture toughness (kJ m-2)

Modulus of rupture (MPa) RH

OO

Water absorption

(%)

Density ( g c m -3)

RH

OD

WET

1.4 0.7 1.2 1.2 1.5 1.0 1.0

0.31 ± 0.03 0.57 ± 0.08 1.15±0.16 1.86 ± 0.34 1.92 ± 0 . 5 1 2.09 ± 0.33 2.18 ± 0 . 5 8

--

0.37 ± 0.05 0.88±0.15 1.81 ± 0.43 3.15±0.89 2.97±0.91 2.88 ± 0.97 2.59 ± 0.65

21.2 ± 0.6 25.5±1.1 31.9 ± 0.9 33.9±1.0 33.8±0.9 40.3 ± 0.7 42.0 ± 1.2

.65 ± 0.02 .49 ± 0.02 .37 ± 0.01 .31 ± 0 . 0 2 .30 ± 0.01 .19±0.01 •15 ± 0.02

--21.5 ± 1.9 -23.1 ± 2 . 0 -23.0 ± 1.5

11.9 ± 0.8 12.3 ± 0 . 8

0.22 ± 0.02 0.59 ± 0.06 0.93 ± 0 . 1 3 1.36 ± 0.29 1.99 ± 0.29 2.16±0.38 2.45 ± 0.70

--0.41 ± 0.05 -0.71 ± 0.11

0.28 ± 0.03 0.71 ± 0 , 0 9 -1.94 ± 0.63 -2.84 ± 0.54 --

22.3 23.1 26.0 27.8 31.6 32.4 34.1

0.5 1.0 1.0 0.7 1.6 1.0 1.0

.59 .54 .44 .39 .33 .30 .25

± ± ± ± ± ± ±

0.01 0.04 0.02 0.03 0.02 0.02 0.01

13.5 ± 1.4 16.6±t.5 18.5 ± 1.3 21.5 ± 1.7 24.1 ± 2 . 7 22.7 ± 1.3 21.4 ± 2.9

-20.9±2.7 -20.8 ± 1.6 -25.7 ± 3.6 --

9.6 ± 1.1

0.31 0.49 0.72 1.45 1.92 2.07 2.51

±0.03 ± 0.06 ± 0.07 ± 0.07 ± 0.33 ± 0.43 ±0.87

-0.29 ± 0.04 -0.54 ± 0.05

0.31 ± 0 . 0 3 -1.04 ± 0.26 -3.09 ± 0.46 3.65 ± 1.06

21.4±0.6 22.4 ± 0.4 25.5 ± 1.0 27.3 ± 1.1 29.6 ± 1.3 31.5 ± 0.6 34.3 ± 0.8

1.60 1.58 1.48 1.41 1.37 1.32 1.26

± ± ± ± ± ± ±

0.01 0.02 0.02 0.03 0.03 0.02 0.01

2 4 6 8 10 12 14

13.0 ± 15.4 ± 20.3 ± 20.0 ± -16.9 ± 19.1 ±

17.7 ± -19.8 ± -22.9 ± -19.1 ±

0.24 0.43 0.80 1.22 1.69 2.38 2.14

± ± ± ± ± ± ±

0.17±0.02 -0.38 ± 0.04

0.28±0.02 0.49 ± 0.06 -1.79 ± 0.58 -2.22 ± 0.30 --

21.2±0.5 23.0 ± 0.4 25.3 ± 1.0 28.1 ± 1.1 29.5 ± 0.8 33.5 ± 0.8 35.7 ± 1.2

1.65 1.55 1.49 1,41 1.39 1.31 1.26

± ± ± ± ± ± ±

0.02 0.02 0.02 0.02 0.02 0.01 0.02

2 4 6 8 10 12 14

14.6 ± 1.9 18.0 ± 1.3 20.9±1.7 19.2 ± 1.5 18.9+0.4 18.7 ± 1.6 17.1 ± 1 . 2

0.19 0.43 0.68 1.18 1.58 1.67 1.68

± 0.02 ± 0.8 ± 0.10 ±0.30 ±0.10 ±0.21 ± 0.20

0.5 1.1 0.7 1.3 0.8 0.7 0.6

1.62 1.52 1.47 1.40 1.33 1.30 1.25

± ± ± ± ± ± ±

0.02 0.01 0.02 0.02 0.01 0.01 0.01

1.7 1.6 1.2 2.3 2.3 1.6 1.9

1.0 1.2 1.9 1.6 1.1 1.1

WET

--------

1.5 1.7 2.4 4.8 2.9 0.5 7.6

2.4

± ± ± ± ± ± ±

15.4 ± 1.5 14.1 ± 1.1

13.5 ± 1.8 15.7 ± 0.8 11.6 ± 1.2 9.8 ± 2.0 10.0 ± 0.5

2.2 13.5±1.1 2.1 9.6 ± 1.3 1.1

--19.1 ± 2 . 3 -18.5±2.7 -17.7±0.9

11.4 ± 0.4 12.2 ± 1.4 9.7 ± 1.1

0.02 0.04 0.09 0.21 0.32 0.43 0.60

- -

-- -

- -

- -

- -

- -

0.99 ± 0.12

- -

0,94 ± 0.13 - -

- -

0.64 ± 0.08 -1.07 ± 0 . 1 0 - -

- -

0.35 ± 0.04 - -

0.77 ± 0.11 -0.92 ± 0.09

- -

- -

0.49 ± 0.05 -1.87 ± 0.40 -2.57 ± 0.55 --

± ± ± ± ± ± ±

2 1 . 3

±

23.8 25.8 29.6 32.2 32.4 35.1

± ± ± ± ± ±

COMPOSITES . APRIL 1 9 8 4

strength. The effect of soluble salts, used as pretreatments for wood fibres, had been noted in an earlier study 5 and may account for some of the variation in flexural strength of the samples. The flexural strength of composites containing between 2-12% unbeaten fibre reinforcement is very similar for both P. radiata and New Zealand flax fibres?

26 24 22 -

e / ~

R

"6 20 o_

18

il 160 CSF

i-/~/" u_

• o o •

12

I0

Fig. 1

I

I

2

4

288 514 546 700

CSF CSF CSF CSF

I

I

I

6 8 I0 Fibre content (% by moss)

I 12

14

Effect of fibre content on flexural strength at various freeness

values

Scanning electron micrographs of kraft P. radiata wood fibres and New Zealand flax fibres are observed at the same magnififcation in Fig. 2. It is noted that the P. radiata fibres are large and ribbon like and, although not obvious from the micrograph, have an aspect ratio (ratio of fibre length to fibre diameter) of approximately 75; whereas the smaller New Zealand flax fibres are more circular in cross-section and have an aspect ratio of approximately 190.3 Both fibres have an average fibre length of about 3.0 mm. It could be considered that the increase in flexural strength of the P. radiata fibre-reinforced composite with the degree of fibre beating is due to better composite formation; this being more noticeable at loadings > 6% by mass. The P. radiata fibres experience internal fibrillation during beating which makes them more collapsed or comformable than the unbeaten fibres. Conformable fibres produce less voidage (see Fig. 3) in the more highly loaded fibre composites during their preparation, and so may improve their mechanical properties. If, however, beating is excessive, the aspect ratio may be reduced below some critical value and so reduce strength and toughness. Due to its naturally more flexible nature (smaller fibre diameter) and greater aspect ratio, composites containing New Zealand flax do not show a similar effect in mechanical properties as the fibre is subjected to beating. Although beating reduces the aspect ratio of the flax fibres, it nevertheless must remain above the critical value and hence produce less variation in composite properties. The variation of flexural strength at various fibre loadings with moisture content is presented in Fig. 4. Fig. 5 shows an alternative representation, in which the flexural strength of the oven-dried and wet-tested samples are given as a percentage of the flexural strength of the same composite formulation, when tested at constant RH and temperature (see test 50 Unbeaten fibre /

/

f

composites

// Fig. 2 fibres

Scanning electron micrographs of P. radiata and Phorrnium tenax

Above 6% fibre content the properties of the fibrereinforced composites can be seen to improve as the fibres are beaten further (eg 514 and 546 ml CSF); however if beating of the fibres is excessive (eg 288 and 160 ml CSF), the strengths of the resulting composites are lower than those of the composites containing unbeaten fibres (700 ml CSF). For the range of beaten fibre-reinforced cement composites studied, optimum conditions resulted when the fibre freeness value was approximately 550 ml CSF and the sample contained approximately 10% fibres by mass. This agrees in general terms with the earlier findingsJ although the use of tap water in preference to water saturated with cement has resulted in higher values of flexural

COMPOSITES. APRIL 1984

45

-

/

/1

/

~

~mienfibre Ico posiles

> 40

35 0

I

I

I

I

I

I

I

2

4

6

8

I0

12

14

Fibre content (% by moss) Fig. 3

Effect of fibre content on void volume

141

_ I ~

50

Oven dried

~RH

tested

:~ 20

v

U-

00

I 2

I 4

I I I 6 8 I0 Fibre content (% by moss)

I 12

I 14

Fig. 4 Effect of fibre content on flexural strength following various preconditioning treatments

Fracture toughness Fig. 6 shows the fracture toughness of samples of various compositions (over the complete range of freeness values studied) when tested at different moisture contents. It can be observed that at lower fibre content (< 6% fibre by mass) there is a general area of overlap between the values of fracture toughness for samples tested wet and those tested at 50% RH. By contrast, oven-dried samples show a dramatically reduced fracture toughne.ss. This variation in toughness or shatter resistance has been attributed to the change in fracture mechanism s from one of predominantly fibre fracture in the oven-dried samples to a mixture of fibre pullout and fibre fracture in the samples at higher RH. A scanning electron microscopic examination of the fracture surfaces of this type of wood fibrereinforced composite has supported this hypothesis, 7 as has a theoretical study considering single sisal slivers embedded in cement- 8

180

160

140

120 RH valu -r I00 n~

o

80

6O

40-

20 0

I 2

from Fig. 5 that greater loads are needed to break the oven-dried fibre to matrix bonds than the same bond at higher RH. As the fibre content is increased (> 6%) there is an increase in voidage and a greater number of fibre to fibre bonds are formed. The increase in fibre to fibre bonds uses up more of the fibre surface area which reduces fibre to matrix bonding. The increase in voidage also reduces the interfacial area of contact between fibre and matrix and so diminishes the potential for a given fibre to be able to bond to the matrix. Thus at increasing fibre content there is a general overlap of flexural strength for oven-dried and RH-tested samples. The wet-tested samples continue to decrease in flexural strength due to the increasing role of fibre to fibre bonds (which involve hydrogen bonds and are destroyed when wet) in carrying an increased component of the load in the composite. ~

I 4

I I I 6 8 I0 Fibre content (% by moss)

I 12

I 14

Fig. 7 shows a plot of fracture toughness v s fibre freeness at various fibre percentages. In general terms, the toughness of a sample of constant fibre composition tends to decrease as the fibre is further beaten, although at fibre loadings > 10% an initial increase in toughness with beating is noted. It can also be seen 4.0

Fig. 5 Effect of fibre content on flexural strength of samples tested ovendry or wet, expressed as a percentage of the same composite tested at 50% RH and 22:1: 2"C

methods). The shaded regions representing oven-dried test values and wet-tested values also include the results obtained in the earlier work done with fibres beaten in a Valley beater. 1 Unfortunately in the earlier work, no single composite formulation had been tested under all three test conditions. The shaded regions include data for composites containing fibres from all beatings, as the percentage change of flexural strength at a given fibre weight is relatively constant for all freeness values in the freeness range 700-160 CSF. In Fig, 5 the flexural-strength results for low fibre contents show the greatest deviation between oven-dried and constant RH tests. At fibre contents above 6% little variation results. On the other hand, wet-tested samples tend to deviate further from the flexural strength value of RH-tested samples as the fibre content is increased. At low fibre content (< 6%), the number of bonds between fibres is small by comparison with the number of fibre to matrix bonds. It could be implied

142

-I-7

5.0

'E 2

2.0 0 0~

0

I.O

0 0

I 2

I 4

I I I 6 8 I0 Fibre content (% by mass)

I 12

I 14

Fig. 6 Effect of fibre content on fracture toughness following various preconditioning treatments

COMPOSITES. APRIL 1984

3.0

50

2.6 -

14% 0

0 0

4O

2.2

§ 3o

1.8

=o -~ 1.4 o

o ~

o

20

80

o

"~ I . O -

~'"""'--o

o

0.6 - o

6 °/0

101. I

o ~

0

o

--0---o

I

I

I

I

700

600

500

400

I

300 Freeness (CSF)

I

1.3

I

I

1.4 1.5 Density (g cm-3)

I

1.6

I

1.7

4% Fig. 8

0

0.2-

O

I I. 2

Relationship between water absorption and density

2°1.

I

I

200

I00

Fig. 7 Effect of freeness on fracture toughness at various values of fibre content

that the brittle matrix rapidly improves in toughness as the fibre content is increased (see Fig. 6).

strength of the composite materials reaches an optimum value when the fibre content is approximately 10% by weight and the fibre has a freeness value of approximately 550 CSF. Fracture toughness values for a composite tend to decrease (for a constant fibre content) when the fibre has been subjected to increased beating.

An explanation for the decrease in toughness with refining is dependent on the fact that one of the major effects of refining is to shorten the fibres. Hence the fibres offer less frictional resistance during fibre pullout from the matrix. ~.8.

ACKNOWLEDGEMENTS

Density and water absorption

REFERENCES

The variation of water absorption of samples at constant fibre content with density is not systematic, as the freeness of the fibre is changed by beating. The unbeaten fibre-reinforced composites are somewhat lower in density (apart from the 2% fibre content), this is in agreement with the voidage data shown in Fig. 3.

1

The density decreases and the water absorption increases (see Table 1) as the fibre content of the composite increases. The overall density of the composite reflects the changing proportions of the constituent fibres and matrix. However, the void volume of the composite also increases, but in a nonlinear fashion, as the fibre content increases and the fibres pack less efficiently. The change of density with change in water absorption was thought to be linear 9 but, as seen in Fig. 8, this is not the case when a sufficiently wide range of densities are studied.

The skillful assistance of P. Warden. R. Wicks and J. Wong is acknowledged with thanks. Thanks are also due to A. McKenzie for helpful discussions.

Coutts, R.S.P. and Ridikas, V. 'Refined wood fibre-cement products Appita 35 N o 5 ( M a r c h 1982) p p 395-400 2 Anon 'New - a wood fibre cement building board' CSIRO Industrial News No 146 (May 1982) 3 Coutts, R.S.P. Flax (Phormium tenax) fibres as a reinforcement in cement mortars' Int J Cement Composites and Lightweight Concrete 5 No 4 (November 1983) pp 257-262 4 Tappi Standard, 1~03m-50 (Tech Assoc Pulp and Paper

Industry, New York) 5 Campbell, M.D. and Coutts, R.S.P. 'Wood fibre reinforced cement composites' JMater Sci 15 (1980) pp 1962-1970 6 Coutts, R.S.P. and Kightly, P. 'Bonding in wood fibre-cement composites J Mater Sci (in press) 7 Coutts, R.S.P. and Kightly, P. 'Microstructure of autoclaved refined wood fibre cement mortars' JMater Sci 17 (1982) pp 1801-1806 8

Morrissey, F,E., Coutts, ILS.P. and Grossman, P.U.A.

(unpublished results. 1982) 9

Coutts, R.S.P. and Michell, A.J. 'Wood pulp fibre-cement composites' J Appl Polymer Sci (in press)

CONCLUSIONS

AUTHOR

The objective of this study was to re-examine earlier observations about the effect beating (or refining) softwood fibres had on the properties of cement composites into which the beaten fibres had been incorporated. It has been confirmed that the flexural

ILS.P. Coutts is a Principal Research Scientist in the Division of Chemical and Wood Technology, CSIRO. Inquiries should be addressed to: Dr R.S.P. Coutts, CSIRO, Division of Chemical and Wood Technology, Private Bag 10, Clayton, Victoria 3168, Australia.

C O M P O S I T E S . APRIL 1 9 8 4

143