Materials and Design 32 (2011) 1704–1709

Contents lists available at ScienceDirect

Materials and Design

journal homepage: www.elsevier .com/locate /matdes

Technical Report

Abrasive wear behaviour of hard powders filled glass fabric–epoxyhybrid composites

N. Mohan, S. Natarajan ⇑, S.P. KumareshBabuDepartment of Metallurgical and Materials Engineering, National Institute of Technology, Tiruchirappalli 620 015, India

a r t i c l e i n f o a b s t r a c t

Article history:Received 12 May 2010Accepted 28 August 2010Available online 29 September 2010

0261-3069/$ - see front matter � 2010 Elsevier Ltd. Adoi:10.1016/j.matdes.2010.08.050

⇑ Corresponding author. Tel.: +91 431 2503000; faxE-mail address: drsnatarajan.nitt@gmail.com (S. N

The effect of incorporation of tungsten carbide (WC) and tantalum niobium carbide (Ta/NbC) powders onthree-body abrasive wear behaviour in glass fabric–epoxy (G–E) composites was investigated and find-ings are analysed. A vacuum assisted resin transfer moulding (VARTM) technique was employed to obtaina series of G–E composites containing different fillers (WC and WC + Ta/NbC). Dry sand rubber wheelabrasion test was carried out at 200 rpm speed. The effect of different loads (22 and 32 N) and abradingdistances (from 135 to 540 m) on the performance of the wear resistance were measured. The wear vol-ume loss of the composites was found increasing with the increase in abrading distances and under thesame conditions the specific wear rate decreases. The hard powders filled G–E composite systems exhibitlower wear volume loss and lower specific wear rate as compared to unfilled G–E composite system. Thefeatures of worn surfaces of the specimen were evaluated at higher and lower abrading distances at loadof 32 N were using scanning electron microscope (SEM) and results indicate more severe damage tomatrix and glass fiber in unfilled composite system as compared to hard powder filled composites.

� 2010 Elsevier Ltd. All rights reserved.

1. Introduction

Polymers and their composites are extensively used in tribolog-ical sectors because of light weight, excellent strength to weightratios, resistance to corrosion, self lubricating properties, bettercoefficient of friction and better wear resistance [1]. Thermosetepoxy resins are extensively studied as matrix materials for com-posite structures as well as adhesives for aerospace, aviation andhydrospace applications because they exhibit low shrinkage, high-er mechanical properties, easy fabrication, excellent chemical andmoisture resistance, good wettability and good electrical charac-teristics. These materials are widely used for a variety of engineer-ing applications in automotive, marine and aerospace. Theimportance of tribological properties and wear behaviour of poly-meric composites were studied in detailed by Kishore et al. [2].The improvement of the tribological properties of a polymer withthe incorporation of fibers/fillers is well known and it showed bothpositive and negative results on the tribological properties of apolymer [1,3]. Three-body abrasive wear is caused by interactionsof hard asperities (hard debris or foreign particles trapped betweenthe polymer and mating surface) on one surface move across asofter surface under load, it penetrate and remove material fromthe softer surface and also leave grooves on the softer surfaces thatmay further increase or decrease the wear rate by several orders

ll rights reserved.

: +91 431 2500133.atarajan).

[4]. Three-body abrasive wear is often has considerable practicalimportance, for example in coal handling equipments in powerplants, gear pumps handling industrial fluids and agricultural ma-chine components, sleeve bearing and bushes operating in abrasiveenvironment, lower sleeve bearing in vertical sewage pumps, vehi-cle spring bushes, marine stern tube bearings, chain wear strips,rope sleeve bearings, etc. [5,6]. Suresha and Chandramohan [7]showed that silicon carbide loaded glass fabric–vinyl ester com-posite gives better abrasive wear performance as compared tographite filled composite system. Crivelli Visconti et al. [8] studieda 6 vol.% of silica (SiO2) and tungsten carbide (WC) filler filled glassfabric reinforced with epoxy matrix composites and they foundthat tungsten carbide filled G–E composite shows excellent wearresistance. Commonly used filler materials in fiber reinforced poly-mer composites are graphite, molybdenum disulfide, tungsten car-bide and silicon carbide. Graphite (Gr) and molybdenum disulfide(MoS2) possess self-lubrication properties and they are widelyused in bearing liner applications. Silicon carbide (SiC) has beenextensively used in abrasive machining processes such as grinding,honing, water-jet cutting and sand blasting due to its high hard-ness. The SiC and WC in epoxy imparts good abrasion resistanceand strength [7–9,12]. Keeping the above aspects in view, it is clearthat there is a lot of scope for the study of abrasive wear behaviourof polymer matrix composites. The present study was focus on thepreparation of glass fabric reinforced-epoxy composites reinforcedwith tungsten carbide (WC) and tantalum niobium carbide(Ta/NbC) as fillers in powder form and to investigate the influenceof these hard powders on three-body abrasive wear behaviour.

Table 1Formulations of composite specimen with measured physical properties.

Specimen code Glassfiber(wt.%)

Matrix(wt.%)

Filler(wt.%)

Density(g/cc)

HardnessShore-D

Glass fabric–epoxycomposite (G–E)

60 40 – 1.78 85

Tungsten carbide andtantalum niobiumcarbide filler filled glassfiber–epoxy composite(WC + Ta/NbC – G–E)

60 34 4 1.82 86

Tantalum niobium carbidefiller filled glass fiber–epoxy composite(Ta/NbC – G–E)

60 34 4 1.80 87

N. Mohan et al. / Materials and Design 32 (2011) 1704–1709 1705

2. Experimental

2.1. Materials

A bidirectional E-Glass woven fabric 360 g/m2 was procuredfrom M/s. Reva Composites, Bangalore, India. The glass fabric ofdimension 18 lm diameter was used as reinforcement is shownin Fig. 1. Bifunctional epoxy resin (LY 5052) and room temperaturecuring cyclo aliphatic amine (HY 5052) (system) were obtainedfrom M/s. HAM, India. The resin is a clear liquid with viscosity at25 �C, 1000–1500 mPa and specific gravity 1.17 g/cc. The hardeneris a liquid and its viscosity is 40–60 mPa and specific gravity0.94 g/cc. The commercially available tantalum niobium carbidein the ratio of 60:40 (Ta/NbC) and tungsten carbide (WC) powder(density of Ta/NbC = 10.8 g/cc and WC = 15.6 g/cc, size = 50–55 lm) was used as filler. The powders were procured from M/s.Kennametal, Bangalore, India.

2.2. Fabrication of hybrid composite specimen

The composite fabrication consist of three steps: (a) mixing ofthe epoxy resin and filler using a mechanical stirrer, (b) mixingof the curing agent with the filled epoxy resin, and (c) fabrica-tion of composites. In the first step, a known quantity of fillerwas mixed with epoxy resin using a high speed mechanical stir-rer to ensure the proper dispersion of filler in the epoxy resin. Inthe second step, the hardener was mixed into the filled epoxyresin using a mechanical stirrer. The ratio of epoxy resin to hard-ener was 100:38 on a weight basis. In the last step, the epoxyresin was manually smeared onto the glass fabric and the resul-

Fig. 1. SEM image of dry glass fabric of diameter 18 lm, (a) 250� and (b) 500�before fabrication.

tant composites were fabricated using the VARTM process as de-scribed elsewhere [10].

The composites were cured at room temperature under a pres-sure of 14 psi for 24 h and it is post cured up to 3 h at 100 �C. The glassfiber:matrix (epoxy):filler ratio was 60:36:4. The unfilled glassepoxy composites were designated as G–E, Ta/NbC filled G–E com-posites as Ta/NbC – G–E and composite containing both WC andTa/NbC in a weight ratio of 2:2 was designated as WC–Ta/NbC –G–E. The details of the composites are provided in Table 1. The lam-inate of dimensions 300 mm � 300 mm � 2.6 ± 0.2 mm was fabri-cated and the specimens for the required dimensions were cutusing a diamond tipped cutter. Density of the composites specimenswas determined using a high precision digital electronic weighingbalance of 0.1 mg accuracy by using Archimedes principle.

2.3. Techniques

The modified dry sand/rubber wheel abrasion test set up as perASTM-G65 standard was used to conduct the three-body abrasivewear experiments (Fig. 2). The surface of the specimens wascleaned with a soft paper soaked in acetone before the test. Thespecimen weight was recorded using a digital electronic balance(0.1 mg accuracy) before it was mounted in the specimen holder.The difference between initial and final weight of the specimenwas a measure of wear loss. A minimum of three trials were con-ducted to ensure repeatability of test data. The silica sand of angu-lar shape (53–75 lm) was used as abrasives in this study (Fig. 3).The abrasive was fed at the contacting face between the rotatingrubber wheel and the test specimen. The tests were conducted ata rotational speed of 200 rpm. The rate of feeding the abrasivewas 250 ± 10 g/min. The experiments were carried out for loadsof 22 N and 32 N with a constant rubbing velocity of 2.33 m/s. Fur-ther, the abrading distances were varied in steps of 135 m from135 to 540 m. The wear behaviour was measured by the loss inweight, which was then converted into wear volume using themeasured density data. The specific wear rate (KS) was calculatedfrom the equation;

KS ¼DV

L� dm3=N m ð1Þ

where DV is the volume loss in m3, L is the load in Newton, and d isthe abrading distance in meters.

3. Results and discussion

3.1. Wear volume loss and specific wear rate

The graphical plot of wear volume loss versus abrading distanceof filled and unfilled composite systems are shown in Fig. 4a and b.

Fig. 2. A digital photo of dry sand rubber wheel abrasive wear machine.Fig. 4. Wear volume loss as function of abrading distance of G–E, WC + Ta/NbC – G–E and Ta/NbC – G–E composites at (a) 22 N and (b) 32 N loads.

Fig. 3. SEM image of the silica sand abrasives used for testing.

Fig. 5. Glass epoxy composites (a) abraded and (b) un-abraded surfaces.

1706 N. Mohan et al. / Materials and Design 32 (2011) 1704–1709

Abrasive wear of composites is strongly influenced by the fillerincorporation and operating parameters [6,7,11–17]. From the

Fig. 4, it was obvious that the abrasive wear volume loss increasedwith the increase in abrading distance. When the abrasive particlesgot entrapped between the rubber wheel and the specimen, due tohigh stress of the hard abrasive particles, the ploughing actiontakes place on the specimen surface leading to removal of matrixmaterial. The continuous removal of matrix material on the surfaceof the specimen produces groove with wear scars (Fig. 5). Thisgroove further increases in-depth as the load increases. After the

N. Mohan et al. / Materials and Design 32 (2011) 1704–1709 1707

removal of the matrix layer, the cutting action takes place on theglass fibers. The wear volume loss of 0.104 � 103 mm3 was ob-served for Ta/NbC – G–E composite and where for unfilled it was0.252 � 103 mm3 at 32 N load.

Abrasive wear loss increases with increase in the applied load.This is due to energy barrier created at the junction of the surface.At lower loads, the energy generated by abrasive particles is notsufficient to break the surface energy barrier and at higher loads,particles gain energy from the high speed rubber wheel, hence highwear volume loss was observed. In case of filler filled G–E compos-ite, wear loss increased with the increase in applied load. Energybarrier created at the surface is greater due to the reinforcementof hard filler particles into the matrix. Hence at even higher loads,energy generated by abrasive particles is not sufficient. As a resultabrasive particles cannot get penetrate deeper into the matrixmaterial. The gained energy from the high speed rubber wheel issufficient for microcutting of large amount of glass fibers[18–21]. Hence, similar observations are found in the presentinvestigation. The mixture of WC and Ta/NbC-filled composite sys-tem showed higher abrasion resistance up to 12% and Ta/NbC-filled composite system up to 20% as compared to unfilled G–Ecomposite. The wear resistance offered by different composites ofthe present research investigation shows the following sequence:Ta/NbC – G–E > WC + Ta/NbC – G–E > G–E. At this juncture, theauthors are pleased to quote the research work of Bijwe et al.[12,22–27]. Which confirm that addition of 4% of fillers such as

Fig. 6. Specific wear rate as a function of abrading distance of G–E, WC + Ta/NbC –G–E and Ta/NbC – G–E composites at (a) 22 N and (b) 32 N loads.

Fig. 7. SEM images of worn surface of G–E composites at 32 N load: (a) 135 m and(b) 540 m abrading distances.

Ta/NbC made in their present study also has brought in consider-able improvement in abrasion resistance.

The graphical plot of specific wear rate (Ks) as a function ofabrading distance of various composite systems is shown inFig. 6a and b. The wear data revealed that, the specific wear rate(Fig. 6) decreases with increasing the abrading distance and ap-plied load. The lower value of specific wear rate was noticed forfilled G–E composites as compared to the unfilled G–E composite.The highest specific wear rate of 3.68 � 10�11 m3/N m for un filledG–E composite and lowest specific wear rate of 1.25 � 10�11

m3/N m for Ta/NbC filled G–E composite system was noticed. Thespecific wear rate for Ta/NbC filled G–E composite varies between2.8 � 10�11 m3/N m and 1.25 � 10�11 m3/N m and for hybrid com-posite (WC + Ta/NbC – G–E) varies between 3.14 � 10–11 m3/N mand 1.33 � 10�11 m3/N m. As compared to G–E composite andWC + Ta/NbC filled G–E composite system a Ta/NbC-filledcomposite system shows lower wear rate at all condition. This isdue to Ta/NbC powders in epoxy matrix act as good interfacialadhesion with G–E composites and improves the wear resistance.

1708 N. Mohan et al. / Materials and Design 32 (2011) 1704–1709

From Fig. 6 it was noticed that at lower abrading distances(135–270 m) a drastic reduction in specific wear rate and with fur-ther increase in abrasion distance from 270 m to 540 m, a linearreduction of specific wear rate was observed for all the compositespecimen. This can be attributed to the fact that at lower abradingdistance, low modulus polymer matrix (soft component) was ex-posed to abrasion, which is less hard as compared to silica sandand hence high specific wear rate resulting in matrix damageand less exposure of fiber as shown in SEM images (Figs. 7a, 8aand 9a).

At higher abrading distance, high modulus glass fabric was ex-posed to abrasion, which shows lower wear rate. This is because ofhigher hardness provides better resistance against abrasion and inturn, abrasive particles have to work more to create failure in thefibers (i.e., much higher amount of energy is required to facilitatefiber failure). Similar observations were noticed by Satheeshet al. [28] and Unal et al. [29] in their study of abrasive wear behav-iour of composite materials. Thus the rate at which the materialwas removed with respect to the abrading distance decreases.Hence the glass fiber resists the further abrasion resulting in fiberfracture, fiber damage and fiber cutting.

Fig. 8. SEM images of worn surface of WC + Ta/NbC – G–E-composites at 32 N load:(a) 135 m and (b) 540 m abrading distances.

3.2. Worn surface morphology

The surface morphology of the composites was examined usinga scanning electron microscope (SEM) (S3000, V-1, HITACHI). TheSEM features of the surfaces of unfilled and hard powder filledG–E composite are shown in Figs. 7–9 (white big arrow indicatesthe direction of abrasion).

Fig. 7 indicates unfilled G–E composite depicts that the matrixand fiber damage were due to the cutting action by abrasive parti-cles. The Fig. 7a shows matrix fracture, debris formation and part-ing up off of top and bottom layer marked as (i) and (ii). At higherabrading distance (540 m) in Fig. 7b fiber fracture and pulveriza-tion were predominately seen and this resulted in more severedamage happen to fiber and matrix due to heavy cutting action.

A parting up of matrix appears to take place in three layers asshown in the Fig. 7b. As compared to top and middle layer a moredamage was observed at bottom layer. The presence of character-istic wear debris as small white particles in the middle layer maybe attributed to the ploughing and cutting action of the abrasiveparticles.

Fig. 9. SEM images of worn surface of Ta/NbC – G–E composites at 32 N load: (a)135 m and (b) 540 m abrading distances.

N. Mohan et al. / Materials and Design 32 (2011) 1704–1709 1709

Fig. 8a and b shows the worn out surface features of hard pow-der (mixture of WC + Ta/NbC) filled glass–epoxy composites at135 m and 540 m abrading distances at 32 N load. The amount ofobserved fiber fracture is relatively low compared to the unfilledG–E composite. At a lower abrading distances (135 m), fromFig. 8a it is observed that there is less fiber and matrix damageon the surface. A mixture of hard powder in the matrix strongly re-sist the penetration of abrasive particles which was strongly envel-oped by fiber.

At higher abrading distance (Fig. 8b) there is an evidence ofexposure of longitudinal fibers in the direction perpendicular theabrasion with no damage which indicates that the mixture oftwo fillers in G–E composite have poor interfacial adhesion.

Fig. 9a and b shows the worn out surface feature of Ta/NbC filledG–E composites at 135 m and 540 m abrading distances at 32 Nload. At lower abrading distances in Fig. 9a, Ta/NbC particles con-taining the epoxy matrix present on the surface of the compositeis inferred to act as an effective barrier to prevent the damageand exposure of the fiber. At higher abrading distance (Fig. 9b) agood interfacial bonding between filler and matrix exist that resiststhe damage of fiber and protects it in position.

A significant difference between the SEM images of unfilled andfilled G–E composite could be observed. Both at high and lowabrading distances, the Ta/NbC filled G–E composite (Fig. 9) exhib-ited relatively less damage of matrix and fiber compared to theunfilled and WC + Ta/NbC filled G–E composite. The SEM imagesof the G–E composite containing a mixture of WC and Ta/NbC(Fig. 8) revealed a better and improved abrasion resistance com-pared to unfilled G–E composites. The observed fragmentation ofmatrix and fiber was relatively low as compared to unfilled G–Ecomposite. At the 135 m abrading distance (Fig. 9a), only looseningof the matrix could be observed without the exposure of any fibers.At the 540 m abrading distance (Fig. 9a), a little detachment ofmatrix and single longitudinal fiber can be seen. The improvementin the abrasion resistance was observed in Ta/NbC filled G–E com-posite compared to WC + Ta/NbC filled and unfilled G–E composite.This is due to Ta/NbC in G–E composite that exhibits a good adhe-sion between the fiber matrix interface.

4. Conclusions

From abrasive wear studies of G–E, WC + Ta/NbC – G–E, and Ta/NbC – G–E composites, the following conclusions are drawn:

� Abrasive wear volume loss increases with the increase in abrad-ing distance/applied load for all the composites. However, theTa/NbC filled G–E composite showed better abrasion resistance.� Abrasive wear rate ratio is found to be 3:5 for filled and unfilled

composites respectively.� SEM studies of worn out surfaces support the damage to the

matrix, exposure of fibers, crushed and fragmented fibers anddebonding of matrix and filler in WC + Ta/NbC – G–E compositesystem.

Acknowledgements

The authors wish to express their gratitude for the provision ofthe excellent experimental facilities, established under TEQIPin the Department of Metallurgical and Materials Engineering,National Institute of Technology, Tiruchirappalli. One of theauthors is grateful to the management of Dr. Ambedkar Institute

of Technology, Bangalore and AICTE, New Delhi for their help andvaluable support during the course of this work.

References

[1] Pihtili Hasim, Tosun Nihat. Investigation of the wear behaviour of a glass–fiberreinforced composite and plain polyester resin. Compos Sci Technol2002;62:367–70.

[2] Kishore, Sampathkumaran P, Seetharamu S, Vynatheya S, Murali A, Kumar RK.SEM observations of the effects of velocity and load on the sliding wearcharacteristics of glass fabric–epoxy composites with different fillers. Wear2000;237:20–7.

[3] El-Tayeb NSM, Yousif BF. Evaluation of glass fibre reinforced polyestercomposite for multi-pass abrasive wear applications. Wear2007;262:1140–51.

[4] Zum Gahr KH. Wear by hard particles. Tribol Int 1998;31:87–596.[5] Stachowiak Gwidon W. Wear-materials, mechanisms and practice. Tribology

of polymeric solids and their composites. USA: Johan Wile; 2005. p. 223–67.[6] Harsha AP, Tewari US. Two-body and three-body abrasive wear behaviour of

polyaryletherketone composites. J Polym Test 2003;22:403–18.[7] Suresha B, Chandramohan G. Three-body abrasive wear behaviour of

particulate-filled glass–vinyl ester composites. J Mater Proc Technol2008;200:306–11.

[8] Crivelli Visconti I, Langella A, Durante M. The wear behaviour of compositematerials with epoxy matrix filled with hard powder. J Appl Compos Mater2001;8:179–89.

[9] Suresha B, Chandramohan G, Renukappa MN, Siddaramaiah. Mechanical andtribological properties of glass–epoxy composites with and without graphite. JAppl Polym Sci 2007;103:2472–80.

[10] Chisholm N, Mahfuzi H, Rangari V, Rodgers R, Jeelani S. Synthesis andmechanical characterization of carbon/epoxy composites reinforced with SiCnano particles. NSTI Nanotechnol 2004;3:302–7.

[11] Mohan N, Natarajan S, KumareshBabu SP, Siddaramaiah. Investigation on two-body abrasive wear behavior of silicon carbide filled glass fabric–epoxycomposites. J Min Mater Charact Eng 2010;9:231–46.

[12] Bijwe J, Jhon Rajesh J, Jeyakumar A, Gosh A, Tewari US. Influence of solidlubricants and fiber reinforcement on wear behavior of polyethersulphone.Tribol Int 2000;33:697–706.

[13] Chand N, Pandey A. High stress abrasive wear study on fly-ash filledpolypropylene/polystryrene blends. J Met Mater Process 2000;12:91–6.

[14] Mohan N, Natarajan S, KumareshBabu SP. Studies on two-body abrasive wearbehavior of graphite filled glass fabric epoxy composites. Int J Mater Sci2010;5:974–1836.

[15] Budinski KG. Resistance of particle abrasion of selected plastics. Wear1997;203–204(1):302–9.

[16] Bijwe J, Logani CM, Tewari US. Influence of fillers and fiber reinforcement onabrasive wear resistance of some polymeric composites. Wear1990;138:77–92.

[17] Tewari US, Bijwe J, Mathura JN, Sharma Indu. Studies on abrasive wear ofcarbon fiber (short) reinforced polyamide composites. Tribol Int1992;29:53–60.

[18] Chand N, Naik A, Neogi S. Three-body abrasive wear of short glass fibrepolyester composite. Wear 2000;242:38–46.

[19] Basavarajappa S, Joshi Ajith G, Arun KV, Praveen Kumar A, Prasanna Kumar M.Three-body abrasive wear behaviour of polymer matrix composites filled withSiC Particles. J Polym Plast Technol Eng 2010;49:8–12.

[20] Thorp JM. Abrasive wear of some commercial polymers. Tribol Int1982;15:59–68.

[21] Harsha AP, Tewari US. Tribo performance of polyaryletherketone composites.Polym Test 2002;21:697–709.

[22] Bijwe J, Indumathi J, Jhon Rajesh J, Fahim M. Friction and wear behavior ofpolyetherimide composites in various wear modes. Wear 2001;249:715–26.

[23] Mohan N, Natarajan S, Kumaresh Babu SP, Siddaramaiah. Investigation onsliding wear behaviour and mechanical properties of jatropha oil cake-filledglass–epoxy composites. J Am Oil Chem Soc, in press.

[24] Tanaka K, Satoshi K. Effect of various fillers on the friction and wear of PTFE-based composites. Wear 1982;2:221–34.

[25] Bahadur S, Fu Q, Gong D. The effect of reinforcement and the synergismbetween CuS and carbon fiber on the wear of nylon. Wear 1994;178:123–30.

[26] Chand N, Gautam KKS. Influence of load on abrasion fly-ash glass fiberreinforced composites. J Mater Sci Lett 1994;13:230–3.

[27] Srivastava VK, Pathak JP. Friction and wear properties of bushing bearing ofgraphite filled short glass fiber composite in dry sliding. Wear1996;197:45–150.

[28] Satheesh Kumar MN, Yaakob Z, Mohan N, KumareshBabu SP, Siddaramaiah.Mechanical and abrasive wear studies on biobased jatropha oil cakeincorporated glass–epoxy composites. J Am Oil Chem Soc 2010;87:929–36.

[29] Unal H, Sen U, Mimaroglu A. Abrasive wear behaviour of polymeric materials.Mater Des 2005;26:705–10.