Fracture behaviour of the basalt fibre reinforced composites with polysiloxane-derived matrix.
Fibre reinforced composites find their utilization in the rapidly broadening field of applications. Some composite types can be considered as structural materials for usage at elevated temperatures. Ideally they should be lightweight, chemically and thermally stable, possessing good mechanical properties and cheap. Actually, no real material reveals these properties altogether.
Recently, attention has been devoted to continuous basalt fibres (CBF) whose primary advantage (according to http://www.basaltex.com/) consists in their low cost, good resistance to acids and solvents, and good thermal stability (under very low stress up to 1250 [degrees]C, under common load only to 500 [degrees]C). They were investigated as reinforcement in composites with concrete (Sim et al., 2005) or polymer-epoxy (Park et al., 1999), polypropylene (Matko et al., 2003), (Czigany, 2005), (Bashtannik et al., 1997) or phenol-formaldehyde resin (Ozturk, 2005), (Artemenko, 2003) matrix. Fibre--matrix interface properties were studied in various basalt fibre--polymer matrix systems in, e.g., (Park et al., 1999) or (Matko et al., 2003).
Good thermal stability of CBF allows even utilizing their reinforcing function in fibrous composites manufactured by means of an additional heat treatment, e.g., pyrolysis (up to 750 [degrees]C in nitrogen) of a polymer matrix composite, which yields a ceramic matrix composite. For this purpose, selected polysiloxane resins are suitable as matrix precursors (Cerny et al., 2005).
Mechanical properties of CBF at elevated temperatures have been investigated in (Cerny et al., 2007). Here, a significant loss of tensile modulus above 400 [degrees]C was established. When loaded in tension by 10 MPa and heated, an unlimited elongation of the CBF tow commenced at 580-640 [degrees]C (depending on the fibre type). This fibre behaviour can play a certain role in forming the microstructure of the pyrolysed CBF-reinforced composites, because the latter are subjected to varying temperature and stress fields during manufacture. In this way, mechanical properties of the pyrolysed composites can be affected by parameters of the manufacturing process.
In this paper selected mechanical properties of unidirectional composites with polysiloxane-derived matrix and CBF reinforcement are presented. A special attention is devoted to the impact of long term exposition in hot air (aimed as a simulation of the anticipated operating conditions) to the composite failure under flexural load.
Two types of commercially available continuous basalt fibre tows were utilized for composite manufacturing in laboratory: Basaltex and Kamennyj Vek. Their basic properties are summarized in Table 1. These fibres were exploited either in their original state or after removing the lubrication by extraction in toluene (because of the a-priori unknown behaviour of the lubricant after heat treatment). Two types of polysiloxane resins supplied by the Lucebni zavody Kolin (Czech Republic) were employed as matrix precursors: polymethylsiloxane resin M130 and polymethylphenylsiloxane resin L901. Their behaviour when thermally cured and subsequently pyrolysed has been investigated recently (Cerny et al., 2005). By combining two precursor types and two fibre types 4 composite batches were obtained (Table 2). Each batch broke down to 2 subtypes: with and without fibre lubrication.
Unidirectional composites were made by the wet-winding (prepreg) route described in (Cerny et al., 2005). Eight prepreg layers were stacked in a heated mould, cured gradually for 6 h at 160-225 [degrees]C and post cured at 250 [degrees]C for 4 h under uniaxial pressure of 0.6 MPa. The received polymer-matrix composites (specimen size approximately 45 x 4 x 1.5 [mm.sup.3]) were further pyrolysed in nitrogen atmosphere to 650 or 750 [degrees]C. During pyrolysis the polymer matrix was gradually transformed into an inorganic matrix (similar to silicon oxycarbide glass), which was accompanied by release of volatiles, mass loss, and void formation. In order to assess their thermal stability some specimens were--prior to measuring their properties--exposed to hot air (4 h at 650 or 750 [degrees]C).
An electrodynamic resonant frequency tester Erudite (CNS Electronics Ltd., London, UK) was used for measurement of resonant frequencies of composite specimens up to 100 kHz. The tensile modulus E was then determined from the basic longitudinal resonant frequency of a beam with free ends, and the in-plane shear modulus G was obtained from flexural vibrations by a resonant frequency method (Cerny and Glogar, 1998). For static measurements of the tensile modulus in a four-point flexural arrangement (thickness to span ratio (1.5-2.0) / 40) a testing machine Inspekt (made by Hegewald-Peschke, Germany) with a 5 kN load cell was used. The flexural strength was determined with the same apparatus in a three-point arrangement.
In contrary to other specimens, those exposed additionally to hot air broke during the flexural test in a brittle manner enabling thus to study their fracture surfaces by scanning electron microscopy (SEM) using the JEOL JSM-5410 apparatus. For this purpose, the fractured specimens were provided with an Au-Pd coating in order to increase their surface conductivity.
In Fig. 1 and Fig. 2 the results for E and G moduli from resonant frequency measurements are plotted. Removing the fibre lubrication deteriorated the elastic properties of the investigated materials which is most notable for the KV composites pyrolysed to 750 [degrees]C (Fig. 2). Minute interlaminar cracks emerging during pyrolysis are probably responsible for the shear modulus decline.
[FIGURE 1 OMITTED]
[FIGURE 2 OMITTED]
[FIGURE 3 OMITTED]
[FIGURE 4 OMITTED]
The static measurements yielded for the tensile modulus and flexural strength the data plotted in (Fig. 3) and (Fig. 4). During the flexural strength tests the specimens did not break in a brittle manner and they kept integrity even after failure (at large deformations).
The flexural strength is more sensitive to the treatment temperature (650 or 750 [degrees]C, Fig. 4) than the modulus (Fig. 3, Fig. 1). While slightly higher values of modulus (especially in case of lubricated fibres) were detected with the specimens treated to 750 [degrees]C, a distinct fall of strengths occurred for all specimens treated to this temperature instead of 650 [degrees]C. It follows that the composites should rather be pyrolysed to lower temperatures which, on the other hand, implies lowering their threshold operational temperature.
EXPOSITION TO HOT AIR
Exposition to hot air ("oxidation", 4 h at 650 or 750 [degrees]C) deteriorated the flexural strength of the studied composites to 60-120 MPa (Fig. 5).
[FIGURE 5 OMITTED]
In all cases the fracture was brittle and the specimens broke cleanly to (at least) two parts. The fracture surfaces were investigated by scanning electron microscopy in a secondary electron mode emphasizing the surface topography.
[FIGURE 6 OMITTED]
[FIGURE 7 OMITTED]
[FIGURE 8 OMITTED]
ANALYSIS OF FRACTURE SURFACES
Total of 16 fracture surfaces were examined and many photographs ware taken, mostly at tilt 20[degrees] (perpendicular view or tilt 40[degrees] were used sporadically).
Only in two cases (the B2 composites with lubricated fibres oxidized to 650 and 750 [degrees]C) the specimens failed partly by delamination (Fig. 6). As both delaminated specimens originated from the same batch the reason for delamination could be sought in a defect lay-out of the involved prepreg. Indeed, in Fig. 8 a considerable level of fibre non-parallelism in the delamination plane can be seen. Non-ideally arranged fibres do not allow regular filling of inter-fibre gaps by the matrix and may cause an easier separation of the lamina.
Most of the investigated fracture surfaces, however, split the broken specimen into 2 parts and reveal small to moderate vertical articulation where ridges separate mutually inclined plateaus with no or small numbers of protruding fibres, Fig. 9. Fibre pullout (the basic toughening mechanism for fibrous composites) is present only occasionally, namely in the B2 composite with lubrication (Fig. 10) as well as without lubrication (Fig. 11, Fig. 12).
[FIGURE 9 OMITTED]
Much more often the propagating crack cuts the fibres before than they can break farther from the crack plane (Fig. 13). It explains the brittle fracture and the mostly catastrophic pattern of the load-displacement characteristic of the flexural test (Fig. 7). As a consequence, the specimen behaves more or less like a monolith and does not fully utilize the potential of reinforcing fibres.
The reason for the absence of pull-out can be sought in a too strong bonding of fibres to matrix. Even some coalescence between adjacent fibres is not excluded at 750 [degrees]C, especially in case of insufficient resin penetration into the tow (Fig. 14, Fig. 15). These unfavorable fracture features of the oxidized specimens should be remedied by some appropriate surface treatment of the fibres.
[FIGURE 10 OMITTED]
[FIGURE 11 OMITTED]
[FIGURE 12 OMITTED]
[FIGURE 13 OMITTED]
[FIGURE 14 OMITTED]
[FIGURE 15 OMITTED]
One feature of the fracture is typical for the KV fibres, namely the proneness to break in very complex modes. Especially in the composites with the M130-derived matrix the propagating crack cuts the fibre cross-section in different directions and in a non-planar way (Fig. 16, Fig. 18, Fig. 20 and Fig. 22). This feature was not observed with the Basaltex fibres. Fig. 17 reveals the extent of a monolithic behaviour: the fracture plane cuts the fibres in an oblique direction, not distinguishing between fibres and matrix.
Further photographs present more cases of cracks parallel to or slightly inclined to the fibre axis (Fig. 19, Fig. 21).
A very specific appearance of the inner part of fractured specimens was observed in the composites made from the KV fibres (both the lubricated and de-lubricated ones) and M130 resin. In those oxidized at 750 [degrees]C a large amount of open caverns exists impinging upon both matrix and fibres (Fig. 23, Fig. 24, Fig. 25 and Fig. 26). Possibility of an inherent origin of the caverns is very improbable because the composites in question are made from different prepregs and their manufacture processes have no point of contact. Therefore, there should exist some common reason for cavern occurrence, which is independent of possible manufacture flaws. As a working hypothesis an enhanced permeability of these materials to gas (perhaps a system of open, communicating pores) can be proposed.
[FIGURE 16 OMITTED]
[FIGURE 17 OMITTED]
[FIGURE 18 OMITTED]
[FIGURE 19 OMITTED]
[FIGURE 20 OMITTED]
[FIGURE 21 OMITTED]
[FIGURE 22 OMITTED]
[FIGURE 23 OMITTED]
[FIGURE 24 OMITTED]
[FIGURE 25 OMITTED]
[FIGURE 26 OMITTED]
In many cases the fracture discloses the appearance of individual fibres, which varies broadly from total bareness (Fig. 27) through partial coverage by minute flakes (Fig. 28) to massive coverage with matrix blocks (Fig. 29).
[FIGURE 27 OMITTED]
[FIGURE 28 OMITTED]
[FIGURE 29 OMITTED]
The bare fibres are found mostly (but not exclusively) in the composites made with de-lubricated fibres (Fig. 30, Fig. 31, Fig. 32).
The fibre surface is sometimes undulated (Fig. 33, Fig. 34) but it is difficult to decide on the nature of the undulated surface. The case depicted in Fig. 35 may result from some reaction with the matrix.
A very specific appearance of the fibre surface can be found in the composites reinforced with lubricated KV fibres in combination with the L901-derived matrix (Fig. 36 and Fig. 37). The high surface roughness gives evidence on a strong interaction with the matrix.
It was interesting to find out whether the appearance of fibre surfaces is due solely to their thermal treatment or whether it results from the interaction with matrix in the thermally treated composites. For this purpose an additional investigation of the fibres alone, which were treated in the same way as those embedded in the composite specimens, was undertaken. In all cases the fibre surfaces were smooth and clean (Fig. 38, Fig. 39, Fig. 40) suggesting thus that the presence of matrix is a key factor in formation of the surfaces like those in Fig. 35.
[FIGURE 30 OMITTED]
[FIGURE 31 OMITTED]
[FIGURE 32 OMITTED]
[FIGURE 33 OMITTED]
[FIGURE 34 OMITTED]
[FIGURE 35 OMITTED]
DISCUSSION AND CONCLUSIONS
It is desirable to analyze the fracture behaviour from the causal viewpoint, i.e., to identify the influence of: a) the choice of raw materials (i.e., fibre and matrix type) and b) the processing temperature on the resulting appearance of the fracture surface.
[FIGURE 36 OMITTED]
[FIGURE 37 OMITTED]
[FIGURE 38 OMITTED]
[FIGURE 39 OMITTED]
[FIGURE 40 OMITTED]
Attempts were made to trace the influence of the treatment temperature and/or the surface state of fibres (existence/non-existence of lubrication) on the fracture behaviour but the results are not convincing, as the investigated fracture surfaces are qualitatively similar.
Very few conclusions can be drawn from the acquired results:
* Some features are common to a vast majority of the investigated materials, e.g., a very low occurrence of fibre pull-out and fibre coalescence. These are chiefly responsible for the observed brittle fracture.
* Other features are restrained to certain composite types:
1. undulation of the disclosed fibre surface exists only in the composites pyrolysed and treated in air at 750 [degrees]C (Fig. 33, Fig. 34, Fig. 35). This treatment temperature and its duration are to be blamed.
2. breaking of individual fibres in a very complicated mode (including cracking parallel to the fibre axis) occurred only with the KV fibres, predominantly in combination with the M130 matrix.
3. failure by delamination took place only with the B130 composites treated to 650 and 750 [degrees]C, which both stem from the same prepreg stacking. It suggests that flaws in the lay-out of the prepreg can be a common reason for the delamination behaviour.
We have also attempted to relate the measured mechanical properties (strength, moduli) to the level of occurrence of voids in the fracture surface (the feature allowing a rough quantification from the SEM images). The hypothesis was a simple one: the larger is the void fraction the lower should be the modulus or strength. It was but partially proved for the composites with the M130-derived matrix (B130, KV130) pyrolysed and oxidized at 650 [degrees]C, which generally possess less voids than those with the L901-derived one. Within the composites treated to 750 [degrees]C, those with the M130-derived matrix also possess no or a small amount of voids (in contrary to the composites with the L901-derived matrix) but no dependence of the mechanical parameters on the void content was established. The mechanical properties are obviously governed by other factors (fibre volume fraction, interface strength and properties, etc.).
To conclude, the investigated composites (pyrolysed polysiloxane matrix with basalt fibre reinforcement) worsened their properties after treatment in air at elevated temperature (650-750 [degrees]C). The originally non-catastrophic flexural failure (Fig. 6) changed to a brittle fracture, which was accompanied with a flexural strength decrease (Fig. 5). The coalescence of fibres (Fig. 14, Fig. 15) and their strong interaction with the matrix (Fig. 29, Fig. 35) are probably the main reasons for the onset of brittleness. Unless these drawbacks are resolved (perhaps, by proper surface treatment of basalt fibres aiming at optimization of the fibre/matrix interaction and a better soaking of the fibre tows) the service temperature of the composites in air will probably not exceed approximately 500 [degrees]C.
This study was supported by the Czech Science Foundation within the project No. 106/05/0817 and by the Project No. 423 of the Institute of Rock Structure and Mechanics, ASCR (institute research plan AVOZ30460519). Remarks by Z. Chlup are gratefully acknowledged.
Artemenko, S.E.: 2003, Polymer composite materials made from carbon, basalt, and glass fibres. Structure and properties, Fibre Chem. (35), 226-229.
Bashtannik, P.I., Ovcharenko, V.G. and Boot, Y.A.: 1997, Effect of combined extrusion parameters on mechanical properties of basalt fiber-reinforced plastics based on polypropylene, Mech. Compos. Mater. (33), 600-603.
Czigany, T.: 2005, Basalt fiber reinforced hybrid polymer composites, Mater. Sci. Forum (473-474), 59-66.
Cerny, M., Glogar, P., Sucharda, Z. and Machovic V.: 2005, Properties and performance of polysiloxane-derived ceramic matrix in heat resistant composites reinforced with R-glass or fine ceramic fibres, Ceram.-Silik. (49), 145-152.
Cerny, M., Glogar, P., Golias, V., Hruska, J., Jakes, P., Sucharda, Z. and Vavrova, I.: 2007, Comparison of mechanical properties and structural changes of continuous basalt and glass fibres at elevated temperatures, Ceram.-Silik, (accepted for publication).
Cerny, M. and Glogar, P.: 1998, Flexural vibrations of a Timoshenko beam: a numerical solution to the frequency equations and its application for assessment of shear moduli of a carbon-carbon composite, J. Mat. Sci Lett. (17), 425-427.
Matko, S., Anna P., Marosi, G., Szep, A., Keszei, S., Czigany, T. and Posloskei, K.: 2003, Use of reactive surfactants in basalt fiber reinforced polypropylene composites, Macromol. Symp. 202, 255-267.
Ozturk, S.: 2005, The effect of fibre content on the mechanical properties of hemp and basalt fibre reinforced phenol formaldehyde composites, J. Mater. Sci. (40), 4585-4592.
Park, J.M., Shin, W.G., and Yoon, D.J.: 1999, A study of interfacial aspects of epoxy-based composites reinforced with dual basalt and SiC fibres by means of the fragmentation and acoustic emission techniques, Compos. Sci. Technol. (59), 355-370.
Sim, J., Park, C. and Moon, D.Y.: 2005, Characteristics of basalt fiber as a strengthening material for concrete structures, Compos. Pt. B Eng. 36, 504-512.
Petr GLOGAR (1) *, Martin CERNY (1) and Zdenek TOLDE (2)
(1) Institute of Rock Structure and Mechanics, Academy of Sciences, Czech Republic
(2) Faculty of Mechanical Engineering, Czech Technical University in Prague, Czech Republic
* Corresponding author's e-mail: firstname.lastname@example.org
(Received December 2006, accepted March 2007)
Table 1 Basic properties of the used basalt fibre tows. Producer Average filament Average number cross-section of filaments ([micro][m.sup.2]) in a tow B Basaltex 131.3 [+ or -] 61.5 1095 [+ or -] 8 [http://www. basaltex. com/] KV Kamennyj 129.6 [+ or -] 29.3 974 [+ or -] 5 Vek KV [http://www. basfiber.com/en/ roving.shtml ] Total tow Linear density cross-section ([10.sup.-6] ([mm.sup.2]) kg.[m.sup.-1]) B 0.143 374 KV 0.126 344 Table 2 Labelling of the investigated composite types. Fibre Resin Composite Fibre Resin Composite Resin Basaltex L901 B1 Kamennyj Vek L901 KV1 M130 B2 M130 KV2
|Printer friendly Cite/link Email Feedback|
|Author:||Glogar, Petr; Cerny, Martin; Tolde, Zdenek|
|Publication:||Acta Geodynamica et Geromaterialia|
|Date:||Apr 1, 2007|
|Previous Article:||Interactive seismic viewer "OP"--short review.|
|Next Article:||Co-pyrolysis of coal/waste polymers mixtures.|