Improved Fracture Toughened Epoxy Matrix System Reinforced with Recycled Milled Carbon Fibre

Abstract

A stereology study of fracture surfaces of 1 to 10 wt. % short milled carbon fibre (SMCF) reinforced epoxy was carried out. Single edge notch bending (SENB) sharp crack samples over a fracture length of 10 mm, covering over 90 % fast crack growth, show that fibre – to – fibre distance significantly decreases with increased SMCF content. For 1% SMCF, the mean free path between fibres is 600 µm, whereas for 10 % SMCF, the mean free path is 110 µm, based on over 200 SEM images for each composition. In SENB sharp crack specimens tested at 2.8 mm/min, the slow crack growth region (length nearly zero for neat epoxy and maximum 1.14 mm for 10% SMCF-epoxy composite) shows intensive debonding and pull-out mechanisms. Fracture toughness (KIC) increased from 0.78 MPam1/2 for neat epoxy to 2.71 MPam1/2 for 10 weight % SMCF/epoxy composites thanks to the combined effect of fibre stereology and the debonding/pull-out mechanisms. Notably, the flexural modulus of the 10 wt. % SMCF reinforced epoxy was 58 % higher than that of the neat epoxy.

Keywords: Fracture mechanics; Electron microscope; Polymers; Particulate filler; Reinforcement

Abbreviations

K_IC: Fracture Toughness; SENB: Single Edge Notch Bending; ESMCF XX: Epoxy Short Milled Carbon Fibre Series; CV: Coefficient of Variation; Ds: Inter-particle/Nearest Particle Distance; CFRP: Carbon Fibre Reinforced Plastic

Introduction

In recent years, epoxy is finding advantageous uses as a composite matrix in infrastructure applications [1-3]. But many times, the epoxy matrix fails to yield the required toughness (resistance to crack initiation and propagation) in the structure and hence many trials are undertaken to increase the fracture resistance of the epoxy. This has been attempted by adding second phase materials such as elastomers [4-6] and/or a rigid phase like alumina, silica or glass beads [7], which enhances fracture properties by a range of margins, however at a substantially increased material and processing cost and occasionally with reduction in modulus and or strength. A recent publication [8] using 3 weight % carbon nanotubes (CNT) in epoxy showed 17% improvement in fracture toughness (K_IC) of the epoxy and 48% improvement in K_IC of the composite laminate using 40% volume fraction CNT.

The present study aims to increase fracture property of the epoxy using cheap and commercially available recycled short milled carbon fibre (SMCF) without applying any surfactant or involving any additional mixing process- thereby keeping the fabrication cost of the modified epoxy matrix very low and at the same time obtaining a much tougher and stiffer modified epoxy.

The authors’ earlier work [9] confirmed that the reinforcement SMCF a) does not interact chemically with epoxy and b) does not affect the curing time at room temperature.

Normally when epoxies cure, they undergo shrinkage of typically 3-7% which can impose a mechanical/physical compressive/ squeezing force upon the filler. Subsequently during mode I testing, if the fillers undergo debonding from the epoxy interface and then separate through pull-out mechanism, the processes will be valuable energy absorbing sources [6].

Uniform dispersion and distribution of the reinforcement in the matrix are very important as well. If the distance between the fillers is small, and they are un-clustered and well distributed, then the fillers can act as more sites for arresting crack/micro-crack growth - implying that for cracks to re-start the system will have to go through multiple initiation stages (high energy absorbing) and subsequent propagation modes thereby enhancing the fracture properties of the modified epoxy [10]. Figure 1 provides a schematic diagram of the possible distribution features of fibre in matrix where the best and most even fibres scatter over the whole area (Figure 1D), providing optimum properties for crack arrest and reduction of subsequent initiation/propagation. This paper examines the uniformity of SMCF distribution in the epoxy matrix which is necessary for using the SMCF/epoxy system to generate future high quality continuous CFRP [11] systems. The inter-particle distance (Ds) of SMCF will naturally affect considerably the enhancement of the properties after fibre addition [12].

Materials and Methods

Materials

Commercial SMCF (MF100) supplied by ELG Carbon Fibre Ltd, had average diameter 7.5 µm, length 100 to 300 µm, and density 1.8 gm/cc as per information provided by the supplier. And a general purpose low shrinkage laminating epoxy system (EL-M) consisting of DGEBA and cyclo-aliphatic polyamine curing agent (hardener) was obtained from Barnes Pvt. Ltd, Sydney, Australia.

Sample fabrication

SMCF fibres were mixed with epoxy resin and stirred mechanically followed by 45 min ultrasonication after which hardener was added and stirred. The mixture was then allowed to cure at room temperature for 9 days in a silicon mould. The mould was casted in a way to produce Single Edge Notch Bending (SENB) samples having dimensions as per ASTM Standard D5045 [13] with an average notch length of 2.54 mm, as confirmed by a traveling microscope and a Vickers micro-hardness tester. Representative moulded samples are shown in Figure 2. The notch tip radius of the SENB samples was 0.25 mm as per the ASTM standard. Once the SENB samples were tested for fracture toughness, the separated parts were subsequently used as 3-point bending samples for measuring modulus of elasticity and flexural strength. The advantage of this procedure is that the same castings provide various properties, whilst maintaining the same material characteristics and dimensional requirements of ASTM D790 [14].

In the results described below, epoxy/SMCF composites are designated as ESMCF 00 (neat epoxy), ESMCF 01, ESMCF 02, ESMCF 03, ESMCF 05 and ESMCF 10 representing the wt. % of SMCF added viz. 1, 2, 3, 5, and 10 %. The volume fractions were observed as 0, 0.16, 1.37, 2.29, 2.75 and 6.41% respectively as shown in Table 1.

SEM studies

A Hitachi tabletop TM 3000 scanning electron microscope (SEM) was used to study the fracture surfaces of the samples at different magnifications after the samples were gold coated using a Leica EM SCD050 coater. The gold coating was done at 60 mA current for 45 sec resulting in a15 nm thick coating.

Stereology

The distribution of the SMCF in the resin matrix is determined in terms of coefficient of variation (CV). For each composition3 samples were used, and each sample generated over 100(~60 in X direction and over 90 in Y direction) SEM images at 600X magnification (approximate 268 X 200 µm) taken at various places on the sample as shown schematically in Figure 3. From these observations the standard deviation and mean of the fibre population are calculated. The CV is the ratio of the standard deviation and the mean as determined by this procedure and is related to the variation in distribution. Lower CV values indicate better dispersion and usually CV close to 10% is considered as adequately uniform dispersion [15,16].

The same micrographs are used to determine inter-particle distance (Ds) using following equation 1 [12,17]:

math 1

where d_p is filler diameter (7.5 µm) and V_p is volume fraction which is taken as the ratio of the area covered by the number of fibresto the area covered by the total structure [18].

3-Point Bending Tests for flexural modulus and strength

Flexural strength and modulus of elasticity of the fabricated samples were determined in an INSTRON 5982 instrument using, as mentioned earlier, a 3-point bend test meeting ASTM D790 specification. The span length (S) was selected as 60 mm in order to maintain the span to depth ratio close to 16, to minimize the development of shear stresses in the sample [14]. The following equations were used to calculate flexural strength (s_f) and modulus (E_f)

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

References

1. Pintaude G. Introduction of the Ratio of the Hardness to the Reduced Elastic Modulus for Abrasion. 2013.
2. Oliver WC, Pharr GM. An improved technique for determining hardness and elastic modulus using load and displacement sensing indentation experiments. J Mater Res. 1992; 7: 1564-1583.
3. Oliver WC, Pharr GM. Measurement of hardness and elastic modulus by instrumented indentation: Advances in understanding and refinements to methodology. J Mater Res. 2004; 19: 3-20.
4. Saha DR, Mada MR, Datta A, Bandyopadhyay S, Chakravorty D. Nanoindentation measurements on nanostructured silver grown within a gel derived silica glass by electrodeposition. J Applied Physics. 2014. 115: 214308.
5. Hajra P, Saha DR, Mada MR, Dutta S, Brahma P, Bandyopadhyay S, et al. High creep strain rates observed in nanocrystalline a-Fe2O3 particles by nanoindentation measurement. Mater Sci Engi A. 2014; 605: 1-7.
6. Beake B. Modelling indentation creep of polymers: a phenomenological approach. J Physics D: Applied Physics. 2006; 39: 4478-4485.
7. Fischer-Cripps AC. A simple phenomenological approach to nanoindentation creep. Mater Sci Engineering A. 2004; 385: 74-82.
8. Ngan AHW, Tang B. Viscoelastic effects during unloading in depth-sensing indentation. J Mater Res. 2011; 17: 2604-2610.
9. Marshall DB, Lawn BR, Evans AG. Elastic/Plastic Indentation Damage in Ceramics: The Lateral Crack System. J American Ceramic Society. 1982; 65: 561-566.
10. Chen J. Indentation-based methods to assess fracture toughness for thin coatings. J Physics D: Applied Physics. 2012; 45: 203001.
11. Fischer-Cripps AC. Nanoindentation Testing. 2011: 21-37.
12. Fischer-Cripps AC. Time-dependent Nanoindentation. 2011: 125-145.
13. Ma Z, Long S, Pan Y, Zhou Y. Creep behavior and its influence on the mechanics of electrodeposited nickel films. J Mater Sci Technol. 2009; 25: 90.
14. Mandal S, Kose S, Frank A, Elmustafa AA. A numerical study on pile-up in nanoindentation creep. Int J Surface Sci Engineering. 2008; 2: 41.
15. Kucharski S, Jarzabek D. Depth Dependence of Nanoindentation Pile-Up Patterns in Copper Single Crystals. Metallurgical and Materials Transactions A. 2014; 45: 4997-5008.
16. Cheng YT, Cheng CM. Effects of 'sinking in' and 'piling up' on estimating the contact area under load in indentation. Philosophical Magazine Letters. 2010; 78: 115-120.
17. Fu K, Chang Y, Tang Y, Zheng B. Effect of loading rate on the creep behaviour of epoxy resin insulators by nanoindentation. J Mate Sci: Mater Electron. 2014; 25: 3552-3558.
18. Fang T-H, Chang W-J. Nanoindentation characteristics on polycarbonate polymer film. Microelectronics Journal. 2004; 35: 595-599.
19. Mencík J. Determination of mechanical properties by instrumented indentation. Meccanica. 2006; 42: 19-29.
20. Joslin DL, Oliver WC. A new method for analyzing data from continuous depth-sensing microindentation tests. J Mate Res. 2011; 5: 123-126.
21. Bhushan B, Li X. Nanomechanical characterisation of solid surfaces and thin films. International Materials Reviews. 2003; 48: 125-164.
22. Doerner MF, Nix WD. A method for interpreting the data from depth-sensing indentation instruments. J Mate Res. 1986; 1: 601-609.
23. Yang S. Analysis of nanoindentation creep for polymeric materials. J Applied Physics. 2004; 95: 3655.
24. Atlas of Stress-strain Curves. 2nd edn. ASM International. 2002.