- Email: [email protected]
Nanoindentation behaviour of layered silicate reinforced unsaturated polyester nanocomposites
ARTICLE IN PRESS
POLYMER TESTING Polymer Testing 25 (2006) 846–852 www.elsevier.com/locate/polytest
Material Behaviour
Nanoindentation behaviour of layered silicate reinforced unsaturated polyester nanocomposites H.N. Dhakal, Z.Y. Zhang, M.O.W Richardson Department of Mechanical and Design Engineering, University of Portsmouth, Anglesea Road, Portsmouth PO1 3DJ, UK Received 18 February 2006; accepted 29 March 2006
Abstract The effect of various loading levels of nanoclay reinforcement on the nanomechanical properties of layered silicate nanoclay reinforced unsaturated polyester (UPE) nanocomposites were investigated by a nano-indentation test method. The clay was dispersed into a UPE matrix via blending using a mechanical stirrer. Structural studies were carried out using a wide angle X-ray diffraction (WAXD) method and correlated with the nanoindentation results. This shows that nanoindentation behaviour is strongly influenced by clay reinforcement and the degree of clay dispersion in the polymer matrix. The experimental results show that there is a strong correlation between nanomechanical properties and inter layer d-spacing of clay particles in the nanocomposite system. Incorporation of 1%, 3% and 5% by weight nanoclay into the polyester resin results in an improvement in hardness of 29%, 24% and 14%, respectively. The elastic modulus increased from 5393 MPa for unreinforced polyester to 6646 MPa (23% increase) with the introduction of 5% by weight nanoclay. r 2006 Elsevier Ltd. All rights reserved. Keywords: Nanoindentation; Nanocomposite; Intercalation; Hardness; Nanoclay; Reduced modulus
1. Introduction The use of clay/polymer nanocomposites technology has been demonstrated to be a useful way of producing materials with a wide range of engineering properties [1]. These improvements are achieved by the expansion of clay layers and the dispersion of the separated individual clay layers into the polymer [2]. Nanoindentation is a promising way of measuring the mechanical properties of materials at smaller length and load scales than allowed by other testing Corresponding author. Tel.: +44 23 9284 2396; fax: +44 23 9284 2351. E-mail address: [email protected] (H.N. Dhakal).
0142-9418/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.polymertesting.2006.03.017
methods, thus allowing individual constituents to be examined [3]. Nano and micromechanical testing involves the use of rigid indenters, typically with diamond or diamond-coated tips [4]. Nanoindentation is also known as depth sensing indentation and involves obtaining quantitative force versus displacement data and determining the elastic modulus, E, and hardness values, H of materials even beyond their elastic limit [5]. Knowledge of such mechanical properties at the nano level can be important for certain materials selection and design criteria and applications. In this study nanoindentation tests were carried out on nanoclay reinforced unsaturated polyester nanocomposite specimens manufactured by a hand lay up process. Varying clay loading levels were
ARTICLE IN PRESS H.N. Dhakal et al. / Polymer Testing 25 (2006) 846–852
847
used in an attempt to analyse the effect of reinforcement on nanomechanical properties. A nanoindentation test method was employed to determine load–displacement, reduced modulus, elastic modulus and elastic plastic depth for evaluating nanohardness. Results from these nanoindentation tests, in the form of hardness and elastic modulii, are quantified, discussed and also compared to E-glass fibre reinforced unsaturated polyester composites. 2. Experimental details Fig. 2. Symmetrical indentations (30 mm apart) (not to scale).
2.1. Apparatus All the nanoindentation tests in this study were performed using commercially available apparatus, namely the Nano TestTM (Micro Materials, UK). A Berkovich (three sided pyramidal) diamond indenter tip manufactured by Micro Materials was used throughout [6] and a schematic diagram of the nanotest system is shown in Fig. 1. Nine symmetrical indentations (in the form of a 3 3 matrix, 30 mm apart) as shown in Fig. 2 were made on each specimen. The coupons were cut from the composite laminates with approximate dimen-
sions of 18 mm 18 mm 3 mm. The specimens were mounted onto the nanoindentation fixture using a suitable adhesive. All tests were conducted at an approximate temperature of 27 1C. Typical experimental indentation parameters used for all measurements were as follows: Initial load: 0.1 mN. The maximum load for all indents:15 mN. Loading and unloading rate (strain rate): 2.00 mN/s. Dwell time or holding time at maximum load: 5 s. 2.2. Materials
Permanent magnet
Coil
Limit stop Frictionless pivot
Indenter Sample stub
Capacitor plates Balance weight
Optional Impulse assemby
Damping plate
Fig. 1. Schematic of the nanotest system.
Nanoindentation experiments were carried on four different nanocomposite systems using an organically modified layered silicate (referred to as LK-EP-C) as reinforcement. The matrix material was based on unsaturated polyester (UPE), Trade Name NORPOL 444-M888, mixed with a curing catalyst, methyl ethyl ketone peroxide (MEKP) at a concentration of 0.01% w/w of the matrix, supplied by Reichhold UK Ltd. This resin system was chosen for its low viscosity and low shrinkage. 2.3. Processing Four different types of nanocomposites sample (containing 0, 1%, 3% and 5% by weight layered silicate reinforcement, respectively) for nanoindentation behaviour characterisation were fabricated using a hand lay up method. The layered silicate nanoclay was first dried for 30 min at 100 1C to remove any moisture present. Then the varying concentrations of dried nanoclay were dispersed into the UPE matrix. The mixture
ARTICLE IN PRESS H.N. Dhakal et al. / Polymer Testing 25 (2006) 846–852
848
was stirred using a mechanical stirrer at a temperature of approximately 50 1C for 30 min at a speed of 2000 revolution per minute until no agglomerations were seen in the mixture. In order to eliminate the bubbles created during mixing, the mixture was degassed in a vacuum chamber at a pressure of 1000 mbar for about 60 min. The mixture of LKEP-C nanoclay and unsaturated polyester was kept at a temperature of 50 1C for about 15 h, and then a catalyst, MEKP, was added to the mixture of nanoclay and the matrix and stirred. Finally, the mixtures were poured into a pre-determined mould and cured at a temperature of 25 1C for 24 h, and post cured at 60 1C for approximately 2 h. 2.4. Wide angle X-ray diffraction (WAXD) WAXD analysis was performed to verify clay dispersion and gallery spacing using a Philips X-ray diffractometer with Cu Ka radiation generated at 20 mA and 40 kV. Diffraction spectra were obtained over a 2y range of 3.5101 to 151 in steps of 0.0201 with counting times of 1.250 s at each angular position. The d-spacing (basal distance between clay layers) values were calculated using Bragg’s Law, nl ¼ 2d sin y
(1)
where, n is an integer, d is inter layer d-spacing and l is wave length, with l ¼ 0:154 nm. 3. Results and discussion
d-spacing (basal distance between clay layers) values were calculated using Eq. (1). The X-ray diffraction patterns for all LK-EP-C nanoclay reinforced samples are different from each other. The interlayer distances of the samples were obtained from the peak position (d001-reflection) of WAXD traces. The d001-reflection for LK-EP-C was found at a 2y value of 4.881, which corresponds to an interlayer distance of 1.83 nm. The first peak at 2y value of 4.761 (1% w/w clay reinforced sample) illustrates the partial intercalated d-spacing of the clay at approximately 1.9 nm. For 3% LKEP-C samples, the peak at a 2y value of 4.561, has shifted towards a lower angle, indicating an intercalated d-spacing of 2.0 nm. The 5% LK-EP-C clay reinforced samples at a 2y value of 4.681, gave a d-spacing value of approximately 1.9 nm. 3.2. Nanoindentation behaviour 3.2.1. Maximum depth and hardness A schematic representation of one full typical loading–unloading cycle is presented in Fig. 4. The important quantities in this loading–unloading cycle are maximum load (Pmax), maximum depth (hmax), final depth after unloading (hf) and the slope of the upper portion of the unloading curve (S) known as the elastic contact stiffness. The hardness and elastic modulus can be derived using the method developed by Oliver and Pharr [7]. The fundamental relationships to calculate the hardness, H and modulus, E are
3.1. X-ray diffraction patterns H¼ The results of dispersibility in the UPE matrix obtained from the WAXD are shown in Fig. 3. The
Pmax , A
(2)
Pmax 350
1 % LKEPC loading 3 % LKEPC loading 5 % LKEPC loading LKEPC only
X-ray intensity
300 250 200
Loading
150 100
Unloading S
50 0 0
2.5
5
7.5
10
12.5
15
hf
hmax
Two theta degree Fig. 3. X-ray diffraction patterns for nanocomposite samples.
Fig. 4. Schematic representation of a typical loading–unloading curve.
ARTICLE IN PRESS 70.083 70.027 70.077 70.043 70.027 SD SD SD SD SD 3.12 3.16 3.12 3.25 3.13 70.397 0.097 70.115 70.233 70.160 SD SD SD SD SD 7.33 6.49 6.26 6.80 6.96 0.176 SD 70.003 0.198 SD 70.003 0.1940 SD 70.005 0.1928 SD 70.005 0.1818 SD 70.005 70.091 70.095 0.112 70.114 70.097 SD SD SD SD SD 5597 6425 6316 5864 5974 70.009 70.010 70.013 70.014 70.013 SD SD SD SD SD 301 387 372 343 330
For comparison. a
1676.41 SD 721.47 1514.93 SD 716.05 1538.28 SD 723.87 1602.14 SD 727.24 1617 SD 71617.94 Unreinforced UPE only UPE+1% w/w nanoclay UPE+3% w/w nanoclay UPE+5% w/w nanoclay UPE+E-glass 38 fibre % w/wa
Plastic depth (nm)
The average values of experimental data extracted from the loading–unloading curves from the test performed at a peak indentation load of 15 mN are presented in Table 1. Load–depth plots of indentations (made at a peak indentation load) on the four different samples fabricated with and without various clay loadings are presented in Figs. 5–8. A typical loading–unloading cycle data for an unreinforced polyester sample is presented in Fig. 5. The average maximum depth at peak load for this specimen is approximately 1676 nm. The loading–unloading behaviour of the 1% w/w clay loaded sample is shown in Fig. 6. The depth at peak load for this sample is approximately 1515 nm, which is about 11% lower than that of the unreinforced polyester sample. The indentation response of the 3% w/w clay loaded sample is shown in Fig. 7. The depth at maximum load for this sample is approximately 1538 nm, which is approximately 9% lower than that for the unreinforced polyester sample. Fig. 8 shows a typical load–unload curve for the 5% w/w nanoclay reinforced samples. The depth at maximum load for this sample is approxi-
Hardness (MPa)
(5)
Max depth (nm)
1 ðvs Þ2 . 1=E r 1 ðvi Þ2 =E i
Table 1 Average values from the nanoindentation test
Es ¼
Reduced Modulus (MPa)
where, Er is the reduced modulus, vs the Poisson’s ratio for the sample; (for polymer approximately 0.2), vi the Poisson’s ratio for the indenter (for diamond 0.07), E s the Elastic modulus for the sample and Ei the Elastic modulus for the indenter (1141 GPa are often used for a diamond). By rearranging Eq. (4), the equation for elastic modulus of the sample E s is
70.225 715.87 723.42 728.05 728.16
Elastic recovery
(4)
Specimens
1 ð1 vs Þ2 ð1 vi Þ2 ¼ þ , Er Es Ei
Plastic work (nJ)
where Er is the reduced elastic modulus which accounts for the fact that elastic displacement occurs in both indenter and sample and dp=dh ¼ S. The elastic modulus of the test material, Es is calculated from Er, reduced modulus, obtained from the test using
SD SD SD SD SD
where Pmax is the peak indentation load and A is the projected contact area at maximum load, and pffiffiffiffi p dP 1 pffiffiffiffiffi , Er ¼ (3) 2 dh A
849
1424.42 1264.14 1288.43 1343.32 1369.15
Elastic work (nJ)
H.N. Dhakal et al. / Polymer Testing 25 (2006) 846–852
ARTICLE IN PRESS H.N. Dhakal et al. / Polymer Testing 25 (2006) 846–852
18
18
16
16
14
14
12
12
Load (mN)
Load (mN)
850
10 8 6
6 4
2
2 0 0
500
1000 Depth (nm)
1500
2000
Fig. 5. Load versus indenter depth for the unreinforced polyester.
Load (mN)
8
4 0
18 16 14 12 10 8 6 4 2 0 0
500
1000 Depth (nm)
1500
2000
Fig. 6. Load versus indenter depth for 1% w/w UPE/LK-EP-C.
Load (mN)
10
18 16 14 12 10 8 6 4 2 0 0
500
1000
1500
2000
Depth (nm) Fig. 7. Load versus indenter depth for 3% w/w UPE/LK-EP-C.
mately 1602 nm, which is approximately 5% lower than the unreinforced polyester sample. As expected, the addition of 1% w/w clay into neat polyester has a great effect on the nanohardness. The nanohardness for the unreinforced polyester sample is 301 MPa, whereas for the 1% LK-EPC nanoclay reinforced sample it increases to
0
500
1000 Depth (nm)
1500
2000
Fig. 8. Load versus indenter depth for 5% w/w UPE/LK-EP-C.
387 MPa (29% increase). Similarly, the introduction of 3% w/w nanoclay increases the hardness from 301 to 372 MPa (an approximately 24% increase). The hardness value for the 5% LK-EP-C nanoclay reinforced specimen is 343 MPa, which is approximately 14% higher than for the unreinforced polyester sample. The increase in nanohardness can be correlated with the dispersion of clay and its d-spacing values. The 1% and 3% w/w clay loaded samples possess higher d-spacing values and exhibit better dispersion of clay compared to 5% w/w nanoclay loaded sample. Close observation of the load–unload curves (nanotest indentations for unreinforced polyester) indicates a uniform surface by virtue of the reproducibility of the nine indents at different locations. Similarly, for 1%, 3% and 5% w/w clay loaded samples, in all experiments, both loading and unloading curves appear to be continuous and consistent. The maximum indentation depths for these samples decrease and consequently the hardness values increase. In general, non-uniform penetration occurs as a result of either the onset of sudden plastic deformation or the formation of cracks [8]. However, for the all indentation curves in this present study there do not appear to be any discontinuities which would have indicated interfacial debonding, cracking, pop-in or pop-out during the test. The average load versus indentation depth for all four samples is shown in Fig. 9. The average load/ depth traces of clay reinforced nanocomposites show interesting behaviour. At a low level of clay loading (1% w/w), there is a decrease in penetration depth by approximately 11% compared to the UPE sample. However, at higher clay loading (5% w/w) the indentation depth of the nanocomposite
ARTICLE IN PRESS H.N. Dhakal et al. / Polymer Testing 25 (2006) 846–852
18
14 Load (mN)
Table 2 Calculated elastic modulus values
UPE only LKEPC 1 % LKEPC 3 % LKEPC 5 %
16
12
851
10 8 6
Specimens
Elastic modulus, Eindent (MPa)
Unreinforced UPE only 1% w/w UPE/LK-EP-C 3% w/w UPE/LK-EP-C 5% w/w UPE/LK-EP-C
5393 6192 (15% increase) 6074 (13% increase) 6646 (23% increase)
4 2 0 0
500
1000 Depth (nm)
1500
2000
Fig. 9. Average load versus indenter depth for four different samples.
increases to just 5% lower that that of the unreinforced sample. The hardness and the elastic modulus is dependant on contact depth [9]. For all studied samples, both hardness and elastic modulus decrease with increasing contact depth. The cause of this phenomenon is suspected to be related to how the microstructure of the nanocomposite changes with respect to the level of nanoclay loading and the degree of its dispersion [10]. This indicates that the degree of dispersion of the clay drops as the percentage of clay loading increases. The shape of the loading and unloading curves for all nanoclay reinforced samples are significantly different to the unreinforced UPE sample. As can be seen in Fig. 9, the indentation depth for all nanoclay reinforced samples is considerably lower than for the UPE sample. The extent of clay dispersion may have contributed to higher resistance to plastic deformation leading to a lower indentation depth. All the nanoclay loaded samples in this study have nanohardness values higher than that of a E-glass fibre (38% w/w) reinforced conventional unsaturated polyester composite which was tested for comparison purposes. 3.2.2. Reduced modulus and elastic modulus The reduced modulus measured for unreinforced polyester is 5597 MPa. For 1%, 3% and 5% w/w clay loaded samples, the reduced modulus is 6425 (approximately a 15% increase), 6316 (approximately a 13% increase) and 5864 (approximately a 5% increase) MPa, respectively. Interestingly, when the nanoclay concentration is increased, the reduced modulus seems to decrease. However, this is not the case for the elastic modulus calculated
using Eq. (5), which is presented in Table 2. As can be seen, as a result of clay reinforcement, the elastic modulus has increased. The elastic modulus for unreinforced polyester is 5393 MPa. For 1, 3 and 5% w/w clay loaded samples, the elastic modulus is 6192 MPa (approximately a 15% increase), 6074 MPa (approximately a 13% increase) and 6646 MPa (approximately a 23% increase), respectively. The trend of decreasing hardness and reduced modulus with the increase of clay loading level could be related to the uniformity of distribution of nanoclay within the host UPE matrix. On the other hand, the trend of increasing elastic modulus with increasing clay loading level could be related to the filler effect alone. Another interesting parameter to note from loading-unloading curves is that the reduced modulus from the nanoindentation test for all the samples is close to the calculated elastic modulus. The tip used in the indentation is diamond and the elastic modulus for diamond is very high, so the reduced modulus measured could effectively be the elastic modulus of the sample. A further investigation is needed to confirm the above assumptions.
4. Conclusions The effect of various loading levels of nanoclay reinforcement on the nanomechanical properties of layered silicate nanoclay reinforced unsaturated polyester nanocomposites have been investigated by a nanoindentation method. The results show that these systems exhibit significantly better mechanical properties than unreinforced UPE. The nanohardness of unreinforced polyester samples increase from 301 to 387 MPa (approximately 29%) with the introduction of 1% w/w of nanoclay and the elastic modulus increases from 5393 MPa for unreinforced polyester to 6646 MPa (23% increase) with the introduction of 5% w/w nanoclay. Homogeneous dispersion of nanoclay particles within the UPE matrix is crucial to achieving improved mechanical properties. The
ARTICLE IN PRESS 852
H.N. Dhakal et al. / Polymer Testing 25 (2006) 846–852
optimum layered silicate nanoclay concentration to give maximum improvement is 1% w/w for nanohardness and reduced modulus property and 5% w/w for elastic modulus (under current studied experimental conditions). References [1] C.M. Koo, H.T. Ham, S.O. Kim, K.H. Wang, I.J. Chung, D.C. Kim, W.C. Zin, Morphology evolution and anisotropic phase formation of the maleated polyethylene-layered silicate nanocomposites, Macromolecules 35 (2002) 5116–5122. [2] J.K. Pandey, K.R. Reddy, A.P. Kumar, R.P. Singh, An overview on the degradability of polymer nanocomposites, Polym. Degrad. Stab. 88 (2005) 234–250. [3] B.D. Beake, S. Chen, J.B. Hull, F. Gao, Nanoindentation behaviour of clay/poly(ethylene oxide) nanocomposites, J. Nanosci. Nanotechnol. 7 (2002) 73–79. [4] S.V. Hainsworth, H.W. Chandler, T.F. Page, Analysis of nanoindentation load–displacement loading curves, J. Mater. Res. 11 (1997) 1987.
[5] B. Wolf, Inference of mechanical properties from instrumented depth sensing indentation at tiny loads and indentation depth, Cryst. Res. Technol. 4 (2000) 377–399. [6] Micro Materials Nano Test User Manual Version 2.0, Micro Materials Ltd., Unit 3, The Byre, Wrexham Technology Park, Wrexham, LL13 7YP, United Kingdom 2002, pp. 70–71. [7] G.M. Pharr, W.C. Oliver, An improved technique for determining hardness and elastic modulus using load and displacement sensing indentation experiments, J. Mater. Res. 7 (1992) 1565–1566. [8] J.E. Bradby, J.S. Williams, M.V. Swain, Pop-in events induced by spherical indentation in compound semiconductors, J. Mater. Res. 19 (2004) 380–382. [9] C.M. Chan, G.Z. Cao, H. Fong, M. Sarikaya, Nanoindentation and adhesion of sol–gel-derived hard coatings on polyester, J. Mater. Res. 15 (2002) 150–151. [10] J.G. Wang, B.W. Choi, T.G. Nieh, C.T. Liu, Crystallization and nanoindentation behaviour of a bulk Zr–Al–Ti–Cu amorphous alloy, J. Mater. Res. 15 (2000) 802–803.
POLYMER TESTING Polymer Testing 25 (2006) 846–852 www.elsevier.com/locate/polytest
Material Behaviour
Nanoindentation behaviour of layered silicate reinforced unsaturated polyester nanocomposites H.N. Dhakal, Z.Y. Zhang, M.O.W Richardson Department of Mechanical and Design Engineering, University of Portsmouth, Anglesea Road, Portsmouth PO1 3DJ, UK Received 18 February 2006; accepted 29 March 2006
Abstract The effect of various loading levels of nanoclay reinforcement on the nanomechanical properties of layered silicate nanoclay reinforced unsaturated polyester (UPE) nanocomposites were investigated by a nano-indentation test method. The clay was dispersed into a UPE matrix via blending using a mechanical stirrer. Structural studies were carried out using a wide angle X-ray diffraction (WAXD) method and correlated with the nanoindentation results. This shows that nanoindentation behaviour is strongly influenced by clay reinforcement and the degree of clay dispersion in the polymer matrix. The experimental results show that there is a strong correlation between nanomechanical properties and inter layer d-spacing of clay particles in the nanocomposite system. Incorporation of 1%, 3% and 5% by weight nanoclay into the polyester resin results in an improvement in hardness of 29%, 24% and 14%, respectively. The elastic modulus increased from 5393 MPa for unreinforced polyester to 6646 MPa (23% increase) with the introduction of 5% by weight nanoclay. r 2006 Elsevier Ltd. All rights reserved. Keywords: Nanoindentation; Nanocomposite; Intercalation; Hardness; Nanoclay; Reduced modulus
1. Introduction The use of clay/polymer nanocomposites technology has been demonstrated to be a useful way of producing materials with a wide range of engineering properties [1]. These improvements are achieved by the expansion of clay layers and the dispersion of the separated individual clay layers into the polymer [2]. Nanoindentation is a promising way of measuring the mechanical properties of materials at smaller length and load scales than allowed by other testing Corresponding author. Tel.: +44 23 9284 2396; fax: +44 23 9284 2351. E-mail address: [email protected] (H.N. Dhakal).
0142-9418/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.polymertesting.2006.03.017
methods, thus allowing individual constituents to be examined [3]. Nano and micromechanical testing involves the use of rigid indenters, typically with diamond or diamond-coated tips [4]. Nanoindentation is also known as depth sensing indentation and involves obtaining quantitative force versus displacement data and determining the elastic modulus, E, and hardness values, H of materials even beyond their elastic limit [5]. Knowledge of such mechanical properties at the nano level can be important for certain materials selection and design criteria and applications. In this study nanoindentation tests were carried out on nanoclay reinforced unsaturated polyester nanocomposite specimens manufactured by a hand lay up process. Varying clay loading levels were
ARTICLE IN PRESS H.N. Dhakal et al. / Polymer Testing 25 (2006) 846–852
847
used in an attempt to analyse the effect of reinforcement on nanomechanical properties. A nanoindentation test method was employed to determine load–displacement, reduced modulus, elastic modulus and elastic plastic depth for evaluating nanohardness. Results from these nanoindentation tests, in the form of hardness and elastic modulii, are quantified, discussed and also compared to E-glass fibre reinforced unsaturated polyester composites. 2. Experimental details Fig. 2. Symmetrical indentations (30 mm apart) (not to scale).
2.1. Apparatus All the nanoindentation tests in this study were performed using commercially available apparatus, namely the Nano TestTM (Micro Materials, UK). A Berkovich (three sided pyramidal) diamond indenter tip manufactured by Micro Materials was used throughout [6] and a schematic diagram of the nanotest system is shown in Fig. 1. Nine symmetrical indentations (in the form of a 3 3 matrix, 30 mm apart) as shown in Fig. 2 were made on each specimen. The coupons were cut from the composite laminates with approximate dimen-
sions of 18 mm 18 mm 3 mm. The specimens were mounted onto the nanoindentation fixture using a suitable adhesive. All tests were conducted at an approximate temperature of 27 1C. Typical experimental indentation parameters used for all measurements were as follows: Initial load: 0.1 mN. The maximum load for all indents:15 mN. Loading and unloading rate (strain rate): 2.00 mN/s. Dwell time or holding time at maximum load: 5 s. 2.2. Materials
Permanent magnet
Coil
Limit stop Frictionless pivot
Indenter Sample stub
Capacitor plates Balance weight
Optional Impulse assemby
Damping plate
Fig. 1. Schematic of the nanotest system.
Nanoindentation experiments were carried on four different nanocomposite systems using an organically modified layered silicate (referred to as LK-EP-C) as reinforcement. The matrix material was based on unsaturated polyester (UPE), Trade Name NORPOL 444-M888, mixed with a curing catalyst, methyl ethyl ketone peroxide (MEKP) at a concentration of 0.01% w/w of the matrix, supplied by Reichhold UK Ltd. This resin system was chosen for its low viscosity and low shrinkage. 2.3. Processing Four different types of nanocomposites sample (containing 0, 1%, 3% and 5% by weight layered silicate reinforcement, respectively) for nanoindentation behaviour characterisation were fabricated using a hand lay up method. The layered silicate nanoclay was first dried for 30 min at 100 1C to remove any moisture present. Then the varying concentrations of dried nanoclay were dispersed into the UPE matrix. The mixture
ARTICLE IN PRESS H.N. Dhakal et al. / Polymer Testing 25 (2006) 846–852
848
was stirred using a mechanical stirrer at a temperature of approximately 50 1C for 30 min at a speed of 2000 revolution per minute until no agglomerations were seen in the mixture. In order to eliminate the bubbles created during mixing, the mixture was degassed in a vacuum chamber at a pressure of 1000 mbar for about 60 min. The mixture of LKEP-C nanoclay and unsaturated polyester was kept at a temperature of 50 1C for about 15 h, and then a catalyst, MEKP, was added to the mixture of nanoclay and the matrix and stirred. Finally, the mixtures were poured into a pre-determined mould and cured at a temperature of 25 1C for 24 h, and post cured at 60 1C for approximately 2 h. 2.4. Wide angle X-ray diffraction (WAXD) WAXD analysis was performed to verify clay dispersion and gallery spacing using a Philips X-ray diffractometer with Cu Ka radiation generated at 20 mA and 40 kV. Diffraction spectra were obtained over a 2y range of 3.5101 to 151 in steps of 0.0201 with counting times of 1.250 s at each angular position. The d-spacing (basal distance between clay layers) values were calculated using Bragg’s Law, nl ¼ 2d sin y
(1)
where, n is an integer, d is inter layer d-spacing and l is wave length, with l ¼ 0:154 nm. 3. Results and discussion
d-spacing (basal distance between clay layers) values were calculated using Eq. (1). The X-ray diffraction patterns for all LK-EP-C nanoclay reinforced samples are different from each other. The interlayer distances of the samples were obtained from the peak position (d001-reflection) of WAXD traces. The d001-reflection for LK-EP-C was found at a 2y value of 4.881, which corresponds to an interlayer distance of 1.83 nm. The first peak at 2y value of 4.761 (1% w/w clay reinforced sample) illustrates the partial intercalated d-spacing of the clay at approximately 1.9 nm. For 3% LKEP-C samples, the peak at a 2y value of 4.561, has shifted towards a lower angle, indicating an intercalated d-spacing of 2.0 nm. The 5% LK-EP-C clay reinforced samples at a 2y value of 4.681, gave a d-spacing value of approximately 1.9 nm. 3.2. Nanoindentation behaviour 3.2.1. Maximum depth and hardness A schematic representation of one full typical loading–unloading cycle is presented in Fig. 4. The important quantities in this loading–unloading cycle are maximum load (Pmax), maximum depth (hmax), final depth after unloading (hf) and the slope of the upper portion of the unloading curve (S) known as the elastic contact stiffness. The hardness and elastic modulus can be derived using the method developed by Oliver and Pharr [7]. The fundamental relationships to calculate the hardness, H and modulus, E are
3.1. X-ray diffraction patterns H¼ The results of dispersibility in the UPE matrix obtained from the WAXD are shown in Fig. 3. The
Pmax , A
(2)
Pmax 350
1 % LKEPC loading 3 % LKEPC loading 5 % LKEPC loading LKEPC only
X-ray intensity
300 250 200
Loading
150 100
Unloading S
50 0 0
2.5
5
7.5
10
12.5
15
hf
hmax
Two theta degree Fig. 3. X-ray diffraction patterns for nanocomposite samples.
Fig. 4. Schematic representation of a typical loading–unloading curve.
ARTICLE IN PRESS 70.083 70.027 70.077 70.043 70.027 SD SD SD SD SD 3.12 3.16 3.12 3.25 3.13 70.397 0.097 70.115 70.233 70.160 SD SD SD SD SD 7.33 6.49 6.26 6.80 6.96 0.176 SD 70.003 0.198 SD 70.003 0.1940 SD 70.005 0.1928 SD 70.005 0.1818 SD 70.005 70.091 70.095 0.112 70.114 70.097 SD SD SD SD SD 5597 6425 6316 5864 5974 70.009 70.010 70.013 70.014 70.013 SD SD SD SD SD 301 387 372 343 330
For comparison. a
1676.41 SD 721.47 1514.93 SD 716.05 1538.28 SD 723.87 1602.14 SD 727.24 1617 SD 71617.94 Unreinforced UPE only UPE+1% w/w nanoclay UPE+3% w/w nanoclay UPE+5% w/w nanoclay UPE+E-glass 38 fibre % w/wa
Plastic depth (nm)
The average values of experimental data extracted from the loading–unloading curves from the test performed at a peak indentation load of 15 mN are presented in Table 1. Load–depth plots of indentations (made at a peak indentation load) on the four different samples fabricated with and without various clay loadings are presented in Figs. 5–8. A typical loading–unloading cycle data for an unreinforced polyester sample is presented in Fig. 5. The average maximum depth at peak load for this specimen is approximately 1676 nm. The loading–unloading behaviour of the 1% w/w clay loaded sample is shown in Fig. 6. The depth at peak load for this sample is approximately 1515 nm, which is about 11% lower than that of the unreinforced polyester sample. The indentation response of the 3% w/w clay loaded sample is shown in Fig. 7. The depth at maximum load for this sample is approximately 1538 nm, which is approximately 9% lower than that for the unreinforced polyester sample. Fig. 8 shows a typical load–unload curve for the 5% w/w nanoclay reinforced samples. The depth at maximum load for this sample is approxi-
Hardness (MPa)
(5)
Max depth (nm)
1 ðvs Þ2 . 1=E r 1 ðvi Þ2 =E i
Table 1 Average values from the nanoindentation test
Es ¼
Reduced Modulus (MPa)
where, Er is the reduced modulus, vs the Poisson’s ratio for the sample; (for polymer approximately 0.2), vi the Poisson’s ratio for the indenter (for diamond 0.07), E s the Elastic modulus for the sample and Ei the Elastic modulus for the indenter (1141 GPa are often used for a diamond). By rearranging Eq. (4), the equation for elastic modulus of the sample E s is
70.225 715.87 723.42 728.05 728.16
Elastic recovery
(4)
Specimens
1 ð1 vs Þ2 ð1 vi Þ2 ¼ þ , Er Es Ei
Plastic work (nJ)
where Er is the reduced elastic modulus which accounts for the fact that elastic displacement occurs in both indenter and sample and dp=dh ¼ S. The elastic modulus of the test material, Es is calculated from Er, reduced modulus, obtained from the test using
SD SD SD SD SD
where Pmax is the peak indentation load and A is the projected contact area at maximum load, and pffiffiffiffi p dP 1 pffiffiffiffiffi , Er ¼ (3) 2 dh A
849
1424.42 1264.14 1288.43 1343.32 1369.15
Elastic work (nJ)
H.N. Dhakal et al. / Polymer Testing 25 (2006) 846–852
ARTICLE IN PRESS H.N. Dhakal et al. / Polymer Testing 25 (2006) 846–852
18
18
16
16
14
14
12
12
Load (mN)
Load (mN)
850
10 8 6
6 4
2
2 0 0
500
1000 Depth (nm)
1500
2000
Fig. 5. Load versus indenter depth for the unreinforced polyester.
Load (mN)
8
4 0
18 16 14 12 10 8 6 4 2 0 0
500
1000 Depth (nm)
1500
2000
Fig. 6. Load versus indenter depth for 1% w/w UPE/LK-EP-C.
Load (mN)
10
18 16 14 12 10 8 6 4 2 0 0
500
1000
1500
2000
Depth (nm) Fig. 7. Load versus indenter depth for 3% w/w UPE/LK-EP-C.
mately 1602 nm, which is approximately 5% lower than the unreinforced polyester sample. As expected, the addition of 1% w/w clay into neat polyester has a great effect on the nanohardness. The nanohardness for the unreinforced polyester sample is 301 MPa, whereas for the 1% LK-EPC nanoclay reinforced sample it increases to
0
500
1000 Depth (nm)
1500
2000
Fig. 8. Load versus indenter depth for 5% w/w UPE/LK-EP-C.
387 MPa (29% increase). Similarly, the introduction of 3% w/w nanoclay increases the hardness from 301 to 372 MPa (an approximately 24% increase). The hardness value for the 5% LK-EP-C nanoclay reinforced specimen is 343 MPa, which is approximately 14% higher than for the unreinforced polyester sample. The increase in nanohardness can be correlated with the dispersion of clay and its d-spacing values. The 1% and 3% w/w clay loaded samples possess higher d-spacing values and exhibit better dispersion of clay compared to 5% w/w nanoclay loaded sample. Close observation of the load–unload curves (nanotest indentations for unreinforced polyester) indicates a uniform surface by virtue of the reproducibility of the nine indents at different locations. Similarly, for 1%, 3% and 5% w/w clay loaded samples, in all experiments, both loading and unloading curves appear to be continuous and consistent. The maximum indentation depths for these samples decrease and consequently the hardness values increase. In general, non-uniform penetration occurs as a result of either the onset of sudden plastic deformation or the formation of cracks [8]. However, for the all indentation curves in this present study there do not appear to be any discontinuities which would have indicated interfacial debonding, cracking, pop-in or pop-out during the test. The average load versus indentation depth for all four samples is shown in Fig. 9. The average load/ depth traces of clay reinforced nanocomposites show interesting behaviour. At a low level of clay loading (1% w/w), there is a decrease in penetration depth by approximately 11% compared to the UPE sample. However, at higher clay loading (5% w/w) the indentation depth of the nanocomposite
ARTICLE IN PRESS H.N. Dhakal et al. / Polymer Testing 25 (2006) 846–852
18
14 Load (mN)
Table 2 Calculated elastic modulus values
UPE only LKEPC 1 % LKEPC 3 % LKEPC 5 %
16
12
851
10 8 6
Specimens
Elastic modulus, Eindent (MPa)
Unreinforced UPE only 1% w/w UPE/LK-EP-C 3% w/w UPE/LK-EP-C 5% w/w UPE/LK-EP-C
5393 6192 (15% increase) 6074 (13% increase) 6646 (23% increase)
4 2 0 0
500
1000 Depth (nm)
1500
2000
Fig. 9. Average load versus indenter depth for four different samples.
increases to just 5% lower that that of the unreinforced sample. The hardness and the elastic modulus is dependant on contact depth [9]. For all studied samples, both hardness and elastic modulus decrease with increasing contact depth. The cause of this phenomenon is suspected to be related to how the microstructure of the nanocomposite changes with respect to the level of nanoclay loading and the degree of its dispersion [10]. This indicates that the degree of dispersion of the clay drops as the percentage of clay loading increases. The shape of the loading and unloading curves for all nanoclay reinforced samples are significantly different to the unreinforced UPE sample. As can be seen in Fig. 9, the indentation depth for all nanoclay reinforced samples is considerably lower than for the UPE sample. The extent of clay dispersion may have contributed to higher resistance to plastic deformation leading to a lower indentation depth. All the nanoclay loaded samples in this study have nanohardness values higher than that of a E-glass fibre (38% w/w) reinforced conventional unsaturated polyester composite which was tested for comparison purposes. 3.2.2. Reduced modulus and elastic modulus The reduced modulus measured for unreinforced polyester is 5597 MPa. For 1%, 3% and 5% w/w clay loaded samples, the reduced modulus is 6425 (approximately a 15% increase), 6316 (approximately a 13% increase) and 5864 (approximately a 5% increase) MPa, respectively. Interestingly, when the nanoclay concentration is increased, the reduced modulus seems to decrease. However, this is not the case for the elastic modulus calculated
using Eq. (5), which is presented in Table 2. As can be seen, as a result of clay reinforcement, the elastic modulus has increased. The elastic modulus for unreinforced polyester is 5393 MPa. For 1, 3 and 5% w/w clay loaded samples, the elastic modulus is 6192 MPa (approximately a 15% increase), 6074 MPa (approximately a 13% increase) and 6646 MPa (approximately a 23% increase), respectively. The trend of decreasing hardness and reduced modulus with the increase of clay loading level could be related to the uniformity of distribution of nanoclay within the host UPE matrix. On the other hand, the trend of increasing elastic modulus with increasing clay loading level could be related to the filler effect alone. Another interesting parameter to note from loading-unloading curves is that the reduced modulus from the nanoindentation test for all the samples is close to the calculated elastic modulus. The tip used in the indentation is diamond and the elastic modulus for diamond is very high, so the reduced modulus measured could effectively be the elastic modulus of the sample. A further investigation is needed to confirm the above assumptions.
4. Conclusions The effect of various loading levels of nanoclay reinforcement on the nanomechanical properties of layered silicate nanoclay reinforced unsaturated polyester nanocomposites have been investigated by a nanoindentation method. The results show that these systems exhibit significantly better mechanical properties than unreinforced UPE. The nanohardness of unreinforced polyester samples increase from 301 to 387 MPa (approximately 29%) with the introduction of 1% w/w of nanoclay and the elastic modulus increases from 5393 MPa for unreinforced polyester to 6646 MPa (23% increase) with the introduction of 5% w/w nanoclay. Homogeneous dispersion of nanoclay particles within the UPE matrix is crucial to achieving improved mechanical properties. The
ARTICLE IN PRESS 852
H.N. Dhakal et al. / Polymer Testing 25 (2006) 846–852
optimum layered silicate nanoclay concentration to give maximum improvement is 1% w/w for nanohardness and reduced modulus property and 5% w/w for elastic modulus (under current studied experimental conditions). References [1] C.M. Koo, H.T. Ham, S.O. Kim, K.H. Wang, I.J. Chung, D.C. Kim, W.C. Zin, Morphology evolution and anisotropic phase formation of the maleated polyethylene-layered silicate nanocomposites, Macromolecules 35 (2002) 5116–5122. [2] J.K. Pandey, K.R. Reddy, A.P. Kumar, R.P. Singh, An overview on the degradability of polymer nanocomposites, Polym. Degrad. Stab. 88 (2005) 234–250. [3] B.D. Beake, S. Chen, J.B. Hull, F. Gao, Nanoindentation behaviour of clay/poly(ethylene oxide) nanocomposites, J. Nanosci. Nanotechnol. 7 (2002) 73–79. [4] S.V. Hainsworth, H.W. Chandler, T.F. Page, Analysis of nanoindentation load–displacement loading curves, J. Mater. Res. 11 (1997) 1987.
[5] B. Wolf, Inference of mechanical properties from instrumented depth sensing indentation at tiny loads and indentation depth, Cryst. Res. Technol. 4 (2000) 377–399. [6] Micro Materials Nano Test User Manual Version 2.0, Micro Materials Ltd., Unit 3, The Byre, Wrexham Technology Park, Wrexham, LL13 7YP, United Kingdom 2002, pp. 70–71. [7] G.M. Pharr, W.C. Oliver, An improved technique for determining hardness and elastic modulus using load and displacement sensing indentation experiments, J. Mater. Res. 7 (1992) 1565–1566. [8] J.E. Bradby, J.S. Williams, M.V. Swain, Pop-in events induced by spherical indentation in compound semiconductors, J. Mater. Res. 19 (2004) 380–382. [9] C.M. Chan, G.Z. Cao, H. Fong, M. Sarikaya, Nanoindentation and adhesion of sol–gel-derived hard coatings on polyester, J. Mater. Res. 15 (2002) 150–151. [10] J.G. Wang, B.W. Choi, T.G. Nieh, C.T. Liu, Crystallization and nanoindentation behaviour of a bulk Zr–Al–Ti–Cu amorphous alloy, J. Mater. Res. 15 (2000) 802–803.