Experimental Mechanics

High Temperature Nanoindentation Response of RTM6 Epoxy Resin at Different Strain

Rates
--Manuscript Draft--

Manuscript Number: EXME-D-14-00038

Full Title: High Temperature Nanoindentation Response of RTM6 Epoxy Resin at Different Strain
Rates

Article Type: Research paper

Keywords: high temperature nanoindentation; epoxy resins; drift effects; influence of strain rate;
Nano-DMA

Corresponding Author: Patricia María Frontini, Ph.D.

Universidad Nacional de Mar del Plata
Mar del Plata, ARGENTINA

Corresponding Author Secondary

Information:

Corresponding Author's Institution: Universidad Nacional de Mar del Plata

Corresponding Author's Secondary

Institution:

First Author: Patricia María Frontini, Ph.D.

First Author Secondary Information:

Order of Authors: Patricia María Frontini, Ph.D.

Saeid Lotfian

Miguel A. Monclús, Ph.D.

Jon M. Molina-Aldareguia, Ph.D.

Order of Authors Secondary Information:

Abstract: Instrumented indentation techniques allow obtaining hardness and Young's modulus
from load-displacement curves at the very near-surface of materials. Routinely,
indentation experiments are conducted at room temperature; however, in some cases,
it is necessary to derive local mechanical properties at high temperatures.
Characterization of glassy polymers by indentation has always been challenging
because these materials exhibit time-dependent mechanical properties. Their visco-
elastic properties are characterized by their storage and loss moduli, whereas the
visco-plastic response of the material can be associated to its hardness. Since these
properties are time-dependent, the thermal stability of the measuring technique is
critical to achieve reliable results. The most important source of erro r in indentation is
thermal drift and it becomes increasingly important as the measuring temperature
increases. We show that conventional methods applied for drift correction in
nanoindentation of inorganic materials are not applicable to glassy polymers and we
propose an alternative method. To this aim, we carried out nanoindentation tests at
varying temperatures and strain rates on the commercial aerospace grade epoxy resin
RTM6. Using the proposed method, it was possible to determine the viscoelastic-
plastic properties of RTM6 between RT and 200 ºC.

Additional Information:

Question Response

Please suggest up to three (3) possible Prof. Y. Charles Lu, Email: chlu@engr.uky.edu
reviewers's names and email addresses Tony Maxwell, E-mail: tony.maxwell@npl.co.uk
for your manuscript. If you do not want to FELICIA STAN, email: felicia.stan@ugal.ro
suggest reviewers, you need to enter the
text: "Reviewers are not being
suggested." Because this is a required

Powered by Editorial Manager® and ProduXion Manager® from Aries Systems Corporation
field text box, typed text must be entered
into it in order to build and submit your
manuscript. Thank you.

Powered by Editorial Manager® and ProduXion Manager® from Aries Systems Corporation
Cover Letter

Hareesh V. Tippur
Main editor
Experimental Mechanics

Mar del Plata 10th February, 2014

Dear Prof. Tippur,

Please find enclosed the manuscript and figures of the paper entitled: High
Temperature Nanoindentation Response of RTM6 Epoxy Resin at Different
Strain Rates by Patricia Frontini, Saeid Lotfian, Miguel A. Monclús and Jon M.
Molina-Aldareguia which is submitted for publication in Experimental
Mechanics.

Regards,

PATRICIA FRONTINI
pmfronti@fi.mdp.edu.ar;pmfronti@gmail.com
Instituto de Investigaciones en Ciencia y Tecnología de Materiales
(INTEMA)
FACULTAD DE INGENIERIA (<http://www.fi.mdp.edu.ar/>)
UNIVERSIDAD Nacional de Mar del Plata
J. B. Justo 4302 Mar del Plata ARG. (B7608FDQ)
Tel 54-223-481-6600 . int. 204.
- FAX 54-223-481-0046
Manuscript
Click here to download Manuscript: rtmnano_mm_jma_v5-only text-2_MM2_JM (1).doc

Title :
1
2
3 High Temperature Nanoindentation Response of RTM6 Epoxy Resin
4
5
6
at Different Strain Rates
7
8
9
10
11
12 Authors:
13
14 Patricia Frontini1, Saeid Lotfian, Miguel A. Monclús and Jon M. Molina-Aldareguia
15
16
17
18
19 The affiliation(s) and address(es) of the author(s):
20
21
22 IMDEA Materials, C/ Eric Kandel, 2, Tecnogetafe, 28906 Getafe, Madrid – Spain
23
24
25 Corresponding author:
26
27 PATRICIA FRONTINI
28 pmfronti@fi.mdp.edu.ar;pmfronti@gmail.com
29
30 Instituto de Investigaciones en Ciencia y Tecnología de Materiales (INTEMA)
31
32 FACULTAD DE INGENIERIA (<http://www.fi.mdp.edu.ar/>)
33 UNIVERSIDAD Nacional de Mar del Plata
34
35 J. B. Justo 4302 Mar del Plata ARG. (B7608FDQ)
36
37 Tel 54-223-481-6600 . int. 204.
38 - FAX 54-223-481-0046
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57 1
On leave from INTEMA- UNMdP-CONICET, Av. J. B. Justo 4302, B7608FDQ, Mar del
58 Plata, Argentina. e-mail: pmfronti@fi.mdp.edu.ar
59
60
61
62
63
1
64
65
 Abstract
1
2
3
4
5 Instrumented indentation techniques allow obtaining hardness and Young’s modulus from load-
6
7 displacement curves at the very near-surface of materials. Routinely, indentation experiments
8 are conducted at room temperature; however, in some cases, it is necessary to derive local
9
10 mechanical properties at high temperatures. Characterization of glassy polymers by indentation
11
12 has always been challenging because these materials exhibit time-dependent mechanical
13 properties. Their visco-elastic properties are characterized by their storage and loss moduli,
14
15 whereas the visco-plastic response of the material can be associated to its hardness. Since these
16
17 properties are time-dependent, the thermal stability of the measuring technique is critical to
18 achieve reliable results. The most important source of error in indentation is thermal drift and it
19
20 becomes increasingly important as the measuring temperature increases. We show that
21
22 conventional methods applied for drift correction in nanoindentation of inorganic materials are
23 not applicable to glassy polymers and we propose an alternative method. To this aim, we carried
24
25 out nanoindentation tests at varying temperatures and strain rates on the commercial aerospace
26
27 grade epoxy resin RTM6. Using the proposed method, it was possible to determine the
28 viscoelastic-plastic properties of RTM6 between RT and 200 ºC.
29
30
31
32
33
Keywords: high temperature nanoindentation; epoxy resins; drift effects; influence of strain
34
35 rate; Nano-DMA .
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
2
64
65
 Introduction
1
2 The so-called Nanoindentation or Instrumented indentation technique can be used to measure
3
4 mechanical properties, such as Young’s modulus and hardness, at the micro and nano-scale [1].
5
6 Its high spatial resolution and depth sensitivity allow measuring the variations in the local
7 mechanical properties of nano and microcomposites, which constitutes the basis for multiscale
8
9 optimization of mechanical properties of many advanced materials [2].
10
11 Instrumented indentation techniques have been generally developed for room
12 temperature. Nevertheless, in the last few years, there has been an increasing demand for
13
14 conducting nanoindentation tests under relevant service temperatures. In order to get reliable
15
16 measurements, high-temperature nanoindentation requires high thermal stability, since thermal
17 drift can have a large effect on the measured displacements, which are usually in the nanometer
18
19 range. Driven by this, many improvements have been made to both instrumentation and
20
21 methods by the leading manufacturers, and with some of the current commercial models, it is
22
now possible to derive local mechanical properties at service temperature.
23
24 Most of the work up to date on high temperature nanoindentation has focused on
25
26 inorganic materials. However, this capability might be very valuable for testing epoxy resins
27
used in carbon fiber reinforced plastics (CFRP) for aerospace applications, that work in very
28
29 harsh environments (temperature changes, radiation, moisture ….) that promote hygrothermal
30
31 aging. Hygrothermal aging typically refers to the deterioration of the performance of a material
32
due to prolonged exposure to moisture and variable temperature conditions [3-6]. The ability to
33
34 measure the local mechanical properties by nanoindentation would allow determining the
35
36 hygrothermal aging suffered by the resin in real CFRP subjected to service conditions [7].
37
38
To this end, nanoindentation tests should allow extracting material parameters relevant
39 for the calibration of constitutive models that describe the mechanical response of polymers.
40
41 Contrary to most inorganic materials, the deformation of polymeric materials is inherently time
42
43 dependent. Their visco-elastic properties are characterized by their storage and loss moduli,
44 whereas the visco-plastic response of the material can be associated to its strain-rate dependent
45
46 hardness [8-11]. This implies performing tests at a wide range of temperatures and strain rates.
47
48 However, the nanoindentation of rigid polymeric materials at high temperature is still an
49 emerging discipline [12,13]. There is very scarce literature regarding the characterization of
50
51 glassy polymers at varying temperatures and strain rates via nanoindentation, counting mainly
52
53 on a few reports covered by the leading indentation equipment manufacturers [14-17].
54 Moreover, generally none have addressed a multiscale characterization approach in order to
55
56 predict macroscopic behaviour from property determination at local scales [18-20].
57
58 The objective of this paper is to investigate the potential of the nanoindentation
59 technique to determine the mechanical response of polymers in the glassy regime as a function
60
61
62
63
3
64
65
 of temperature and strain rate. The paper shows that conventional drift correction methods do
1
not work well on these materials due to their time-dependent properties. Instead, an alternative
2
3 method that combines dynamic indentation with high temperature testing to correct for spurious
4
5 drift effects is presented and applied to the characterization of RTM6, a commercial aerospace
6
grade epoxy resin.
7
8
9 Background
10
11
12 Characterization of polymers by indentation has always been challenging because these
13 materials exhibit strong temperature and time-dependent mechanical properties. Viscoelasticity
14
15 of polymers manifests itself even at room temperature
16
17 During an indentation test under varying loading conditions, polymers exhibit a marked
18 creep behavior, which acts during the whole loading history. It manifests itself mainly during
19
20 the dwell phase and may be observed also at the unloading portion of the load-displacement (P-
21
22 h) curve. Hardness and modulus are affected by creep because the slope of the P-h curve
23 computed from the upper part of the unloading curve, following the standard Oliver and Pharr
24
25 (O&P) method [1], contains not only elastic but also viscoelastic-viscoplastic contributions. If
26
27 these phenomena appear, in the worst case, the indentation depth can increase even when the
28 indenter is unloaded and a characteristic nose appears at the beginning of the unloading curve
29
30 rendering an abnormally high (and even negative) contact stiffness, resulting in erroneous
31
32 calculation of Young’s modulus and/or hardness.
33 Creep influence in polymer indentation is usually taken into account and corrected by
34
35 holding the indenter at maximum load for a long time resulting in a negligible residual creep
36
37 rate compared with the unloading rate [21]. This technique allows short-lived creepy materials
38 to accommodate to the load. However, this strategy cannot be used aiming to determine the
39
40 strain-rate-dependent properties of materials using self similar-indenters, since the strain-rate
41
42 dependency will be erased during the dwell period.
43
An alternative method that can be used in time-dependent polymers is based on
44
45 superimposing a small oscillating ac force on the primary dc load on the indenter during
46
47 indentation and measuring the phase and relative amplitude of the indenter-sample contact
48
response by a lock-in amplifier, which can then be used to determine the viscoelastic properties
49
50 of the material - storage modulus (E´), loss modulus (E´´) and tan  (E´´/E´) [21-28]. This
51
52 method is referred to as dynamic indentation and has been given different names depending on
53
54 the instrument manufacturer (Continuous Stiffness Measurement (CSM) (Agilent), Dynamic
55 compliance Testing (Micro Materials) or Nano-DMA (Hysitron)). Extraction of accurate
56
57 material properties with this method relies on an appropriate model for the dynamic response of
58
59 the indentation system and on the application of a suitable constitutive equation. The sample is
60 usually represented by a parallel spring and dashpot arrangement (Kevin-Voigt cell), as this
61
62
63
4
64
65
 reduces the complexity of any mathematical analysis involved to deconvolute the instrument
1
response function from the measured data. Other more complicated models (i.e. adding
2
3 elements to the simple Kevin-Voigt model), can be required to successfully obtain loss/viscosity
4
5 parameters.
6
Since polymer properties are time-dependent, the use of experimental techniques with
7
8 very high thermal stability is required so that the results are not discredited by the inherent error
9
10 caused by thermal drift of the measuring device. Drift rates might vary in a relatively short time;
11
12
therefore they must be monitored and corrected for each test. Unfortunately, conventional
13 methods that are typically used for correcting drift effects rely on measuring drift rates either at
14
15 the initial contact or at the end of the unloading process, by maintaining a small load for a dwell
16
17
period and recording the displacement. On inorganic materials, these conventional methods
18 work well because any displacement recorded during this hold period can only be attributed to
19
20 spurious drift effects. In this paper, we will show that they are not applicable in glassy polymers
21
22 because substantial creep effects might take place during these hold periods, giving rise to the
23 incorrect determination of drift rates. This paper presents an alternative approach to correct for
24
25 drift in glassy polymers based on dynamic nanoindentation, refer to as reference drift correction
26
27 method. The method is successfully applied to determine the visco-elastic and visco-plastic
28 parameters of RTM6, an aeronautical grade epoxy resin.
29
30
31
32
33 Experimental Procedure
34
35
36 Material
37
38
Sample tests were conducted on a Hexcel® HexFlow® RTM 6 Epoxy System for Resin
39
40 Transfer Moulding Monocomponent System (Tg = 190° C), which was cured according to the
41
42 corresponding manufacturers' instructions.
43
44
Surface preparation was performed by mechanical grinding using silicon carbide papers
45 of progressively finer grade (between 320-2400) with water lubrication and final polishing
46
47 using 1, 0.3 and 0.02 μm alumina particle size aqueous suspensions on velvet cloths (NAP
48
49 Struers). The surface thus obtained had a roughness (Ra) of  1 nm according to AFM
50 measurements (see Figure 1), which provides a smooth surface for nanoindentation tests.
51
52 Samples with a thickness of 500 m were used in order to avoid temperature gradients across
53
54 the sample thickness during high temperature nanoindentation. This is very important in glassy
55
polymers due to the large temperature dependency of Young modulus and their low thermal
56
57 conductivity. Large differences between the top and bottom temperature might lead to a
58
59 rigid/compliant system leading to serious compliance contributions in the measurement from the
60
61
62
63
5
64
65
 bottom part of the sample, especially close to Tg. The samples were mounted on metal discs,
1
using a high temperature adhesive (MINCO, Forta Fix, AUTOSTIC FC 6 250 ml). Surface
2
3 temperature was monitored by placing a thermocouple onto the sample surface.
4
5 Despite using thin specimens, a distinct gap between the temperature set point of the
6
heating platform and the measured surface temperature was found, which varied between 10 and
7
8 50 ºC, depending on testing temperature. The maximum testing temperature was limited to
9
10 200 ºC, at which point the color of the sample changed dramatically due to thermal degradation
11
12
(see Figure 2).
13
14
15
16 Dynamic Mechanical Spectroscopy (DMA)
17
18 The dynamic mechanical behavior of the RTM6 resin was studied with a Q800 DMA analyzer.
19
20 Prismatic samples with nominal dimension of 26 x 10.0 x 1.6 mm were milled from the plates
21
22 for oscillatory tests using a single-cantilever clamp. The tests were carried out under controlled
23 strain in the linear domain of viscoelasticity of the material, applying a heating rate of 3 ◦C/min
24
25 over a temperature range of 23 to 300 ºC. Oscillatory tests were performed with a span of 50
26
27 mm at a frequency of 1 Hz and a strain amplitude of 30 µm.
28
29 NanoIndentation Testing
30
31
32 Nanoindentation testing was carried out on a Hysitron TI950 Triboindenter equipped with the
33
nanoDMA III option (http://www.hysitron.com/products/ti-series/ti-950-triboindenter). The
34
35 sample was mechanically clamped on a heating stage that allows heating at temperatures up to
36
37 400 ºC. The nanoDMA option allows the application of an oscillatory load superimposed to the
38
main quasi-static dc load, for the continuous measurement of hardness, storage and loss moduli,
39
40 and tan , as a function of depth, frequency and time. In this method, described in detail by
41
42 Oliver and Pharr [29], the kinematic model of the system, based on a damped harmonic
43
44 oscillator, is used to relate the stiffness (S) and the damping (D) of the contact, to the phase and
45 amplitude response. The reduced storage and loss modulus (Er’ and Er’’) and hardness (H) are
46
47 then derived as a function of indentation depth (h) through the equations of the contact:
48
49
50
S  D 
51 Er'  (1  s 2 ) (1) Er''  (1  s 2 ) (2)
52 2 A h 2 A h
53
54
F
55 H (3)
56 A(h)
57
58
59
60
61
62
63
6
64
65
 where the contact area A(h) is obtained from the tip area function, which was calibrated
1
2
beforehand from indentations on fused silica;  is the oscillation frequency, and  is the
3 Poisson’s ratio of the test material.
4
5 As elastic displacements occur both in the specimen (with modulus of elasticity Esample
6
7 and Poisson’s ratio sample) and in the indenter, the elastic modulus of the sample is calculated
8 from the reduced modulus, Er using
9
10
11
12  1 1  indenter 2 
13 Esample  (1  s )  
2
 (4)
14  Er Eindenter 
15
16
17
18
19
20
21 Assuming Ediamond = 1141 GPa, diamond = 0.07 and epoxy = 0.35
22
23
24
25
26 The oscillation frequency was set to 200 Hz with variable force amplitude in order to attain
27 oscillation amplitudes of  1-2 nm. The maximum DC force was varied between 5.5 and 6 mN,
28
29
depending on testing conditions, in order to reach maximum indentation depths of  1 m.
30 Indentation tests were carried out at prescribed indentation strain rates of 0.25, 0.05 and 0.01 s-1,
31 at temperatures between 23 and 200ºC, using a three-side pyramidal (Berkovich) diamond
32  
33 indenter. The indentation strain rate ( ) is given by Eq. (4), where h is indenter displacement
34
35 velocity, and h is depth[30].:
36
37 
 h
38   (4)
39 h
40
41
42 For all indentation tests, an initial contact load of 2 N was maintained constant for a
43
44 period of 60 s and the displacement monitored over time. This segment is conventionally used
45 in indentation to compute the drift rate prior to the test from the last 30 s of the hold segment.
46
47 However, as will be shown in the results section, this approach leads to an incorrect
48
49 determination of the drift rate, which was finally determined using the reference method
50 proposed below.
51
52
53
54 Results and Discussion
55
56 Dynamic Mechanical Spectroscopy
57
58
59 Figure 3 shows the temperature dependence of the storage modulus E’ and the loss factor tan
60
E’’/E’). The drops in the E’ evolution and the peaks of the tan value characterize the
61
62
63
7
64
65
 physical transitions of the polymer. A small but distinct drop of E’ can be noticed between 50
1
and 100 ºC, which is related to a sub-Tg transition. The temperature position of the maximum of
2
3 tan δ, corresponding to this relaxational process occurs approximately at 95 ºC (Tω~95 ºC). The
4
5 main temperature damping of the RTM6, the so-called -transition, occurs at Tg ~225 ºC. It is
6
7 associated to the glass transition involving long distance molecular motions. This relaxation is
8 characterized by a very sharp drop of the E’ storage modulus (about two decades). Moreover,
9
10 the tan  evolution shows that above 200 ºC the viscous response of the polymer becomes
11
12 significant (Fig. 3). Two different molecular relaxations have been already reported by other
13 authors in similar epoxy systems [31-33]. According to Ochi and co-workers [34] the secondary
14
15 viscoelastic transition, can be associated with motions of the p-phenylene groups. Other authors
16
17 [35-37] consider this additional relaxation as a probe of the structural or molecular
18 arrangements within the network resulting from moisture absorption. In spite of all these results,
19
20 the ω-relaxation assessment in terms of molecular mobility remains uncertain in literature.
21
22
23 Nanoindentation Tests using conventional drift correction method
24
25 In this section, we present results of conventional dynamic indentation tests. By conventional,
26
27 we refer to the tests in which drift correction was carried out by measuring the drift rate during
28
29 the initial contact at the beginning of the test, and correcting the data assuming a constant drift
30 rate during the whole test. Figure 4 plots the storage modulus computed as a function of
31
32 indentation depth at three different indentation strain rates (0.01, 0.05, 0.25 s-1) and three
33
34 different temperatures (25, 90, 165 ºC). The results show a large spread and a lack of
35 reproducibility in the measurements, specially a low strain rates and high temperatures. The
36
37 room temperature results show a plateau in storage modulus at indentation depths larger than
38
39 ≈ 400 nm, with a value of 3.8 GPa, independent of indentation strain rate. The size effect
40 encountered at shallow indentation depths for polymers is still under debate and has been
41
42 attributed to different reasons, including sample preparation. However recent investigations
43
44 performed on glassy polymers have revealed that the polymer surface stiffening mechanism is
45 related to the creation of a mechanically unique interfacial region between the probe and the
46
47 polymer surface [38]. The results are consistent with the expectation that storage modulus
48
49 should be independent of indentation strain rate [39]. This suggests that conventional drift
50 correction does a fair job at room temperature, presumably because drift values determined were
51
52 relatively small (typically < 0.1 nm/s).
53
54 For the lower strain rates, due to longer loading times, thermal drift effects become
55 increasingly important and the data becomes more unreliable. In addition and unlike at room
56
57 temperature, as testing temperature increased (Fig. 4b and 4c), the storage modulus did not
58
59 saturate with indentation depth and the results were very scattered, especially at low indentation
60 strain rates, giving rise to spurious storage moduli’s dependency with indentation strain rate.
61
62
63
8
64
65
 The drift rates determined at initial contact at 90ºC and 165ºC varied between -0.81 and +0.1
1
nm/s, much larger than at room temperature and much larger than drift values obtained in the
2
3 same machine when testing metals and ceramics at similar conditions. Errors in drift estimations
4
5 arise from the viscoelastic-viscoplastic behavior of the polymers themselves, which make
6
traditional drift determination methods unpractical, as the polymer reacts by deforming over
7
8 time (even at the small applied initial contact load of 2 N), especially at higher temperatures.
9
10 Incorrect drift determination and correction affects more the data obtained at low indentation
11
12 strain rates, for which the test lasted up to 250 s, thus explaining their larger scatter.
13 The same spurious effects of the wrong drift estimation can be found when the hardness
14
15 is plotted as in Figure 5, which illustrates the hardness profiles obtained at different
16
17 temperatures as a function of strain rate. The hardness was fairly constant at depths greater than
18 ~400 nm, but again results showed a large scatter, especially at low indentation strain rates. In
19
20 these circumstances, the variation in hardness with indentation strain rate is far from the
21
22 expected behaviour of the RTM6 epoxy´s strain rate sensitivity [40].
23
24
25 Nanoindentation Tests using the reference method for drift correction
26
27
28 In this section, we present an alternative analysis carried out in the data presented above to
29 avoid the errors associated with conventional drift correction methods in polymeric materials.
30
31 The analysis relies on mainly two hypotheses: (1) we assume that drift effects are negligible at
32
33 the largest indentation strain rate of 0.25 s-1. And (2), we assume that storage modulus should
34 remain practically unchanged at each temperature independently of the strain rate imposed.
35
36 The first hypothesis is reasonable as the total indentation time for a 1000 nm deep
37
38 indent at the prescribed strain rate of 0.25 s-1 is 17 s; therefore, it is safe to assume that, at the
39 machine drift rates of < 0.1 nm/s, thermal drift effects will not be significant.
40
41 Regarding the second hypothesis, it is accepted that the storage modulus should be
42
43 independent of the penetration rate and should only be a function of the frequency and
44 temperature applied [39].
45
46
47
48 Comparison of computed Storage modulus with DMA spectroscopy
49
50
51 To weight the validity of the first hypothesis, Figure 5 plots the storage modulus determined
52
53 from indentation tests carried out at a strain rate of 0.25 s-1, as a function of temperature. No
54 drift correction was applied to these data and a Poisson ratio of 0.35 was assumed for RTM6 at
55
56 all temperatures (see section 3.3). These results may be compared with tensile (Young´s)
57
58 modulus values and DMA results.
59
60
61
62
63
9
64
65
 Quantitatively speaking, the average modulus value of 3.5 – 4.0 GPa obtained by
1
nanoindentation at room temperature is larger than the 2.9 GPa reported for the Young modulus
2
3 of RTM6 resin [41]. This finding is consistent with the general trend reported for indentation
4
5 measurements in polymers [22]. This is also true when comparing the indentation storage
6
modulus with the storage modulus E’ obtained from DMA spectroscopy (2.3 GPa from Figure 3
7
8 a). In this case, an additional factor affecting the mismatch between the absolute values of the
9
10 elastic moduli determined from both methods could be the fact that, in DMA, data were
11
12
obtained at a much lower superimposed frequency (1 Hz vs. 40 Hz) and in the linear viscoelasic
13 regime.
14
15 Nevertheless, and from a qualitatively viewpoint, the trends observed in storage modulus
16
17
with temperature for DMA (Figure 3) and for nanoindentation (Figure 6) are remarkably
18 similar, corroborating the validity of the first hypothesis. Worth mentioning is the fact that the
19
20 nanoindentation data in Figure 6 is able to capture the smooth drop of the storage modulus E’
21
22 between 50 and 120 ºC associated with the Tω transition reported in Figure 3 a. Given the
23 profound differences in test method and sample geometry, the agreement with DMA is
24
25 promising.
26
27
28
29 Application of the reference method for drift correction for the lower strain rates
30
31
32 The reference method proposed relies on the second hypothesis mentioned at the beginning of
33 section 4.3, i.e., that storage modulus should remain constant with strain rate at each
34
35 temperature [39]. Under such an assumption, the indentation curves corresponding to the two
36
37 lower indentation strain rates were corrected for drift, not by the drift rate monitored by the
38 instrument at the beginning of the test, but by assuming a drift that matches the elastic modulus
39
40 at each temperature to the elastic modulus determined by the indentations performed at the
41
42 fastest strain rate of 0.25 s-1. Figure 7 and 8 plot the storage modulus and hardness for the three
43
strain rates at three different temperatures (25, 90 and 165 ºC) after the application of the
44
45 reference drift correction to all indentation curves. The applied correction leads to reproducible
46
47 and repeatable results that give rise to a clear increasing trend of hardness with strain rate at all
48
temperatures tested.
49
50
51
52
53 Analysis of corrected hardness data: Eyring's Plot and influence of T transformation
54
55 The stress-strain response of most amorphous polymers is known to be dependent on
56
57 temperature and strain rate. Accordingly, the load-displacement curves obtained after reference
58
59 drift correction (Figure 9) show the expected tendency with strain rate and temperature relative
60 to the mechanical response of glassy polymers providing additional support to the validity of the
61
62
63
10
64
65
 proposed correction. At a fixed testing temperature, load- displacement curves show a clear
1
influence of the strain rate. A higher loading rate yielded a higher load for the same
2
3 displacement into the surface.
4
5 The curves also exhibit creep at maximum indentation loads during the hold period of 5
6
s, as well as a pseudo-elastic behavior on unloading irrespectively of the original loading rate.
7
8 From the viewpoint of potential applications, a simple correlation of indentation
9
10 hardness to macroscopic mechanical parameters is of growing interest in the polymer field since
11
12
this may offer interesting possibilities with respect to quality control of load-bearing polymer
13 products [42]. For metals with almost fully plastic deformation behavior, an excellent
14
15 correlation between microhardness and yield stress already exists as it was shown by Tabor
16
17
[43]. In the case of polymers a straightforward correlation has not being yet postulated since the
18 ratio of hardness to yield stress highly depends on the deformation behavior itself and loading
19
20 mode [42]. However in the case of shear-controlled deformation behavior, and this is the case
21
22 of thermoset resins, such correlations are possible and comparable with Tabor’s relationship for
23 metals [42].
24
25 In most cases, the yield stress obeys thermally activated processes following the well
26
27 known Eyring's model [6], and this has been corroborated in epoxy resins [7]. Eyring's model
28 proposes a linear relationship between yield stress divided by the absolute temperature and the
29
30 logarithmic strain rate, displaying the same slope for each particular temperature. In an attempt
31
32 to prove whether the applied correction is robust or not, all the corrected hardness data were
33 collected and plotted according to Eyring’s model as shown in Figure 10. The plot shows that
34
35 the corrected indentation hardness over temperature (H/T) data fall into straight lines, following
36
37 the same stress-activated Eyring flow model than the yield stress data for epoxy resins and
38 confirming the good correlation between yield stress and hardness for epoxy resins over a wide
39
40 range of temperatures and strain rates.
41
42 An abnormal behavior was displayed by experiments performed at temperatures within
43 the T transition region, where the data scatter was also larger. Data points obtained at 70 ºC
44
45 practically overlap with the data determined at 90 ºC (see Figure 10). It has been stated [32, 33]
46
47 that thermal ageing treatment may affect the ω-relaxation process in the temperature range
48
where it appears. Indeed this effect may explain the apparent abnormal behavior found in this
49
50 regime, if we consider that the sample was being thermally aged during the experiment itself.
51
52 This is because yielding is not attributable to a unique relaxation transition. Annealing close to
53
secondary relaxation transitions usually leads to an increase of the yield stress [44].
54
55
56
57
58
59
60
61
62
63
11
64
65
 1
Conclusions
2
3
4
5 In the present study, the effect of strain rate on the hardness and storage modulus of an RTM6
6
aeronautical epoxy resin has been experimentally explored by dynamic nanoindentation at high
7
8 temperatures. Tests were performed between RT and 200 ºC at three different indentation strain
9
10 rates (0.01, 0.05, 0.25 s-1) . The storage modulus determined by nanoindentation at 0.25 s- 1 were
11
12
comparable to the results obtained by DMA spectroscopy within the temperature range scanned.
13 The measurements appear to be sensitive enough to detect subtle T transitions at temperatures
14
15 below Tg. It was made clear that conventional drift correction routines hindered the real
16
17 response of material, especially at lower indentation strain rates. To overcome this problem, the
18 load-displacement curves were corrected for drift by applying a novel reference correction
19
20 method that relies on two hypothesis: (1) machine drift rates are sufficiently low to render drift
21
22 correction unnecessary for the fastest strain rate (0.25 s -1) and (2) storage modulus should only
23 be a function of temperature and dynamic frequency, and independent of indentation strain rate.
24
25 The robustness of the proposed method was confronted by plotting the corrected hardness data
26
27 in an Eying plot. The expected linear behavior was confirmed, supporting the validity of the
28 method. Future work is in progress in order to derive the multiaxial viscoelastic-viscoplastic
29
30 constitutive equations representing global mechanical behavior, by using the reference drift
31
32 corrected load-indent displacement plots.
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
12
64
65
 1
2 Acknowledgements
3
4
5 The authors want to acknowledge Xavier Morelle, PhD sudent at Université catholique de
6
7 Louvain - Institute of Mechanics, Materials and Civil Engineering (iMMC) for providing epoxy
8 samples and Dr. Juan Pedro Fernández from IMDEA materials for performing standard DMA
9
10 measurements. P. M. Frontini was recipient of a sabbatical stay fellowship from Ministerio de
11
12 Educación y Ciencia (Spain) .
13
14
15
16
17 References
18
19
20
21 1. Oliver WC, Pharr GM (1992) An Improved Technique for Determining Hardness
22
23 and Elastic-Modulus Using Load and Displacement Sensing Indentation
24 Experiments. J. Mater. Res. 7(6): 1564-1583.
25
26 2. Kim JH, Ko JH, Bae B-S (2007) Dispersion of silica nano-particles in sol-gel
27
28 hybrid resins for fabrication of multi-scale hybrid nanocomposite. Journal of Sol-
29 Gel Science and Technology 41(3): 249-2551.
30
31 3. http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/20090023547_2009022929.pdf.
32
33 4. Lin YC, Chen X (2005) Investigation of the effect of hygrothermal conditions on
34 epoxy system by fractography and computer simulation. Materials letters 59(29):
35
36 3831–3836.
37
38 5. Couture A, Laliberte J, Li C (2013) Mode I Fracture Toughness of Aerospace
39 Polymer Composites Exposed to Fresh and Salt Water. Chemical and Materials
40
41 Engineering 1(1): 8-17.
42
43 6. Miller SG, Roberts GD, Bail JL, Kohlman LW, Binienda WK (2012) Effects of
44 hygrothermal cycling on the chemical, thermal, and mechanical properties of
45
46 862/W epoxy resin. High Performance Polymers 24(6): 470-477.
47
48 7. LLorca J, González C, Molina-Aldareguía JM, Segurado J, Seltzer R, Sket F,
49 Rodríguez M, Sádaba Muñoz R, Canal LP (2011) Multiscale modeling of
50
51 composite materials: a roadmap towards virtual testing. Advanced Materials 23:
52
53 5130-5147.
54
8. Richeton J, Ahzi S, Vecchio K S , Jiang FC, Makradi A (2007) Modeling and
55
56 validation of the large deformation inelastic response of amorphous polymers over a
57
58 wide range of temperatures and strain rates . International Journal of Solids and
59
Structures 44: 7938–7954.
60
61
62
63
13
64
65
 9. Anand L, Ames NM (2006) On modeling the micro-indentation response of an
1
amorphous polymer International Journal of Plasticity 22: 1123–1170.
2
3 10. Jordan JL, Foley JR Siviour CR (2008) Mechanical properties of Epon 826/DEA
4
5 epoxy Mech Time-Depend Mater 12: 249–272.
6
11. Zhang CY, Zhang YW, Zeng KY and Shen L (2006) Characterization of
7
8 mechanical properties of polymers by nanoindentation tests. Philosophical
9
10 Magazine 86(28): 4487–4506.
11
12
12. Gray A, Orchid D, Beake BD (2009) Nanoindentation of advanced polymers under
13 non-ambient conditions: creep modelling and tan delta. J Nanosci Nanotechnol
14
15 9(7):4514-9.
16
17
13. Lu YC, Jones DC , Tandon GP , Putthanarat S, Schoeppner GA (2010) High
18 Temperature Nanoindentation of PMR-15 Polyimide. Experimental Mechanics
19
20 50:491–499.
21
22 14. http://www.hysitron.com/LinkClick.aspx?fileticket=ByGFq9XtNDg%3d&tabid=33
23 0.
24
25 15. http://www.nanovea.com/Application%20Notes/temperature-anoindentation.pdf
26
27 16. http://www.micromaterials.co.uk/wp-content/uploads/2013/01/mm-polymeric.pdf
28 17. http://cp.literature.agilent.com/litweb/pdf/5990-5761EN.pdf (2010)
29
30 Nanoindentation, Scratch, and Elevated Temperature Testing of Cellulose and
31
32 PMMA Films . Agilent Technologies Application Note, 2010;
33 18. Seltzer R, Kim J, Y. Mai YW (2011) Elevated temperature nanoindentation
34
35 behaviour of polyamide 6. Polymer International 60: 1753-1761.
36
37 19. Kranenburg JM, Tweedie CA, van Vliet KJ, Schubert US (2009) Challenges and
38 Progress in High-Throughput Screening of Polymer Mechanical Properties by
39
40 Indentation . Adv. Mater. 21: 3551–3561.
41
42 20. Rodríguez M, Molina-Aldareguía JM, González C, LLorca J (2012) Determination
43 of the mechanical properties of amorphous materials through instrumented
44
45 nanoindentation. Acta Materialia 60: 3953-3964
46
47 21. Mark R. VanLandingham (2003) Review of Instrumented Indentation. [J. Res. Natl.
48 Inst. Stand. Technol. 108: 249-265 .
49
50 22. Beyaoui M, Mazeran PE, Arvieu MF, Igerelle M, Guigon M (2009) Analysis of
51
52 nanoindentation curves in the case of bulk amorphous polymers. Int. J. Mat. Res.
53 100: 1-7 .
54
55 23. Hay J , Herbert E (2010) Measuring the Complex Modulus of Polymers by
56
57 Instrumented Indentation Testing. Experimental Techniques. doi: 10.1111/j.1747-
58 1567.2010.00618.x
59
60
61
62
63
14
64
65
 24. VanLandingham MR, Villarrubia JS , Guthrie WF and Meyers GF (2001)
1
Nanoindentation Of Polymers: An Overview. Macromol. Symp. 167: 15-43.
2
3 25. Odegard GM , Gates TS, Herring HM, Chakravartula A, Komvopoulos k (2005)
4
5 Characterization of Viscoelastic Properties of Polymeric Materials Through
6
Nanoindentation. Experimental Mechanics 45(2): 130–136.
7
8 26. Chakravartula A, Komvopoulos K, (2006) Viscoelastic properties of polymer
9
10 surfaces investigated by nanoscale dynamic mechanical analysis. Applied Physics
11
12
Letters 88(13): 131901-131903.
13 27. Kranenburg JM, Tweedie CA, van Vliet KJ, Schubert US (2009) Challenges and
14
15 Progress in High-Throughput Screening of Polymer Mechanical Properties by
16
17
Indentation .Adv. Mater. 21: 3551–3561.
18 28. Hayes SA, Goruppa AA, Jones FR (2004) Dynamic nanoindentation as a tool for
19
20 the examination of polymeric materials. Journal of Materials Research 19(11):
21
22 3298-3306 .
23 29. Oliver WC, Pharr GM (2004) Measurement of hardness and elastic modulus by
24
25 instrumented indentation: Advances in understanding and refinements to
26
27 methodology. J. Mater. Res. 19(1): 1-20.
28 30. Fisher-Cripps C (2004) Nanoindentation, 2nd ed. Springer-Verlag, New York,
29
30 pp146.
31
32 31. Terekhina S, Fouvrya S, Salviaa M, Bulanov I (2010) An indirect method based on
33 fretting tests to characterize the elastic properties of materials: Application to an
34
35 epoxy resin RTM6 under variable temperature conditions. Wear 269: 632–637.
36
37 32. Schroeder JA, Madsen PA, Foistier RT (1987) Structure/property relationships for a
38 series of crosslinked aromatic/aliphatic epoxy mixtures Polymer 28 : 929-940.
39
40 33. Colombini D, Martinez-Vega JJ, Merle G (2002) Influence of hydrothermal ageing
41
42 and thermal treatments on the viscoelastic behavior of DGEBA-MCDEA epoxy
43 resin. Polymer Bulletin 48: 75-82
44
45 34. Ochi M, Yoshizumi M, Shimbo M (1987) Mechanical and dielectric relaxations of
46
47 epoxide resins containing the spiro-ring structure. II. Effect of the introduction of
48 methoxy branches on low-temperature relaxations of epoxide resins. J. of Pol. Sci.
49
50 B: Pol Phys. 25 (9): 1817–1827.
51
52 35. Mikols WJ, Seferis JC, Apicella A, Nicolais L (1982) Evaluation of Structural
53 Changes in Epoxy Systems by Moisture Sorption-Desorption and Dynamic
54
55 Mechanical Studies. Polymer Composites 3( 3 ):118-124.
56
57 36. Maxwell IA, Pethrick RA (1983) Dielectric studies of water in epoxy resins.
58 Journal of Applied Polymer Science 28(7) : 2363–2379 .
59
60
61
62
63
15
64
65
 37. Wang JY , Ploehn HJ (1995) Dynamic Mechanical Analysis of the Effect of Water
1
on Glass Bead-Epoxy Composites. Journal of Applied Polymer Science 59: 345-
2
3 357 .
4
5 38. Tweedie CA, Constantinides G, Lehman KE, Brill DJ, Blackman GS, Van Vliet, KJ
6
7 (2007) Enhanced Stiffness of Amorphous Polymer Surfaces under Confinement of
8 Localized Contact Loads. Adv. Mater. 19: 2540–2546.
9
10 39. Herbert EG, Oliver WC, Pharr GM (2008) “Nanoindentation and the dynamic
11
12 characterization of viscoelastic solids. J. Phys. D: Appl. Phys. 41: 1–9.
13 40. Morelle X, Lani F, André S, Maxime M; Bailly C, Pardoen T (2012) The Elasto-
14
15 Viscoplasticity and Fracture Behaviour of RTM6 Epoxy resin. Proccedings of
16
17 SAMPE Europe Student Conference , Paris, France, du 23/03/2012 au 25/03/2012).
18 41. http://www.hexcel.com/Resources/DataSheets/RTM-Data-
19
20 Sheets/RTM6_global.pdf.
21
22 42. Tabor D (1951) The Hardness of Metals. Clarendon Press, Oxford.
23
43. Koch T, Seidler S (2009) Correlations Between Indentation Hardness and Yield
24
25 Stress in Thermoplastic Polymers. Strain 45( 1): 26–33.
26
27 44. Odegard GM, Bandyopadhyay A (2011) Journal of Polymer Science PartB:
28
Polymer Physics 49(24) : 1695-1716
29
30
31
32 Captions to Figures
33
34
35
36
37 Fig. 1: Surface before polishing (a). Surface after polishing (b).
38
39 Fig.2: Change in color of the RTM6 sample after heating to 200 ºC.
40 Fig. 3 : Dynamic mechanical spectra of RTM6 epoxy resin: (a) Storage modulus vs.
41
42 Temperature showing relaxation and relaxation processes at Tg  225C and Tw  95 ºC
43
44 respectively; (b) Storage modulus (log) , Loss Modulus (log) and Tan  (log) vs. Temperature .
45
46
Fig. 4: Storage Modulus as a function of strain rate against indent displacement for three
47
48 different temperatures. Values were corrected by the conventional drift rate determination used
49
50 in conventional indenters. Error bars on the symbols represent one standard deviation.
51
52 Fig. 5 : Hardness as a function of strain rate against indent displacement for three different
53
54 temperatures Values were corrected by the conventional drift rate determination used in
55
56 conventional indenters. Error bars on the symbols represent one standard deviation.
57
58
59
60
61
62
63
16
64
65
 Fig. 6: Storage modulus as a function of temperature at 0.25 s-1 indentation strain rate. Plotted
1
values were taken at 900 nm indentation depth. Data correspond to various runs, but error bars
2
3 were omitted for clarity.
4
5
6 Fig. 7: Storage Modulus as a function of strain rate against indent displacement after reference
7 drift correction for three different temperatures. Error bars on the symbols represent one
8
9 standard deviation.
10
11
12 Fig. 8: Hardness as a function of strain rate against indent displacement after reference drift
13 correction for three different temperatures. The error bars correspond to one standard deviation.
14
15
16 Fig. 9: Inferred Load - indent displacement plots after reference drift correction for three
17
18
different temperatures.
19
20 Fig. 10: Eyring plot constructed from corrected indentation hardness divided by absolute
21
22 temperature (H/T) against logarithm strain rate. Data correspond to various runs, but error bars
23 were omitted for clarity.
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
17
64
65
Figure 1 a
Click here to download Figure: figure 1a-.pdf

a
Figure 1 b
Click here to download Figure: figure 1b.pdf

.
Figure 2
Click here to download Figure: Fig2.pdf

Degradated sample
Figure 3 a
Click here to download Figure: figure 3a.pdf

2500
a
Tω

E', Storage Modulus (MPa) 2000

1500

1000 Tg

500

0
0 50 100 150 200 250

Temperature (°C)
Figure 3 b
Click here to download Figure: figure 3b.pdf

1000 1000
b
100 100
E', E'' (MPa)

10 10
Storage Modulus

Tan δ
Tan Delta
Loss Modulus
1 1

0,1 0,1

0,01 0,01
0 50 100 150 200 250
Temperature (°C)
Figure 4-25C
Click here to download Figure: Fig 4-25C.pdf

6,0 6,0

5,5 5,5

5,0 5,0

4,5 4,5
Storage Modulus (Gpa)

4,0 4,0

3,5 3,5

3,0 3,0

2,5 2,5
Strain Rate
2,0 2,0
0,25
1,5 0,05 T = 25 º C 1,5

1,0 0,01 1,0

0,5 0,5

0,0 0,0
0 250 500 750 1000
Indent. Displacement (nm)
Figure 4-90C
Click here to download Figure: Fig 4-90C.pdf

6,0 6,0

5,5 5,5

5,0 5,0

4,5 4,5
Storage Modulus (GPa)

4,0 4,0

3,5 3,5

3,0 3,0

2,5 2,5

2,0 Strain Rate 2,0

0.01 T = 90°C
1,5 1,5
0.05
1,0 0.25 1,0

0,5 0,5

0,0 0,0
0 250 500 750 1000
Indent. Displacement (nm)
Figure 4-165C
Click here to download Figure: Fig 4-165C.pdf

6.0 6.0

5.5 5.5

5.0 5.0

4.5 4.5
Storage Modulus (Gpa)

4.0 4.0

3.5 3.5

3.0 3.0

2.5 2.5

2.0 2.0
Strain Rate
1.5 0,01 1.5

1.0
0,05 T = 165 º C 1.0
0,25
0.5 0.5

0.0 0.0
0 250 500 750 1000

Indent. Displacement (nm)

Figure 5-25C
Click here to download Figure: Fig5-25C.pdf

0,45 0,45

0,40 0,40

0,35 0,35

0,30 0,30
Hardness (Gpa)

0,25 0,25

0,20 0,20

0,15 Strain rate 0,15

0,01
0,10
T = 25 º C 0,10
0,05
0,25
0,05 0,05

0,00 0,00
0 250 500 750 1000
Indent. Displacement (nm)
Figure 5-90C
Click here to download Figure: Fig5-90C.pdf

0,45 0,45

0,40 0,40

0,35 0,35

0,30 0,30
Hardness (GPa)

0,25 0,25

0,20 0,20

0,15 0,15
Strain Rate
0,10
0.01 T = 90 º C 0,10
0.05
0,05 0.25 0,05

0,00 0,00
0 250 500 750 1000
Indent. Displacement (nm)
Figure 5-165C
Click here to download Figure: Fi5-165C.pdf

0,45 0,45

0,40 0,40

0,35 0,35

0,30 0,30
Hardness (Gpa)

0,25 0,25

0,20 0,20

0,15 0,15
Strain Rate
0,10 0,01 0,10
0,05 T = 165 º C
0,05 0,25 0,05

0,00 0,00
0 250 500 750 1000
Indent. Displacement (nm)
Figure 6
Click here to download Figure: Fig.6.pdf

6 6

5 5
Storage Modulus (Gpa)

Tω
4 4

3 3

2 2

1 1

0 0
0 20 40 60 80 100 120 140 160 180 200 220
Temperature (°C)
Figure 7 .25C
Click here to download Figure: Fig.7 25C.pdf

6,0 6,0

5,5 5,5

5,0 5,0

4,5 4,5
Storage Modulus (GPa)

4,0 4,0

3,5 3,5

3,0 3,0

2,5 2,5

2,0 2,0
Strain Rate
1,5 0.01 T = 25 º C 1,5
0.05
1,0 1,0
0.25
0,5 0,5

0,0 0,0
0 250 500 750 1000
Indent. Displacement (nm)
Figure 7 .90C
Click here to download Figure: Fig.7 90C.pdf

6,0 6,0

5,5 5,5

5,0 5,0

4,5 4,5
Storage Modulus (GPa)
4,0 4,0

3,5 3,5

3,0 3,0

2,5 2,5

2,0 Strain Rate 2,0

1,5 0.01 1,5

0.05 T = 90 º C
1,0 1,0
0.25
0,5 0,5

0,0 0,0
0 250 500 750 1000
Indent. Displacement (nm)
Figure 7 .165C
Click here to download Figure: Fig.7 165C.pdf

6,0 6,0

5,5 5,5

5,0 5,0

4,5 4,5
Storage Modulus (Gpa)

4,0 4,0

3,5 3,5

3,0 3,0

2,5 2,5

2,0 2,0
Strain Rate
1,5 1,5
0,01
T = 165 º C
1,0 0,05 1,0
0,25
0,5 0,5

0,0 0,0
0 250 500 750 1000
Indent. Displacement (nm)
Figure8-25C
Click here to download Figure: Fig.8-25C.pdf

0,45 0,45

0,40 0,40

0,35 0,35

0,30 0,30
Hardness (GPa)

0,25 0,25

0,20 0,20

0,15 Strain Rate 0,15

0.01
0,10 T = 25 º C 0,10
0.05
0.25
0,05 0,05

0,00 0,00
0 250 500 750 1000
Indent. Displacement (nm)
Figure8-90C
Click here to download Figure: Fig.8-90C.pdf

0,45 0,45

0,40 0,40

0,35 0,35

0,30 0,30
Hardness (GPa)

0,25 0,25

0,20 0,20

0,15 0,15
Strain Rate
0,10 0.01 0,10
T = 90 º C
0.05
0,05 0.25 0,05

0,00 0,00
0 250 500 750 1000
Indent. Displacement (nm)
Figure8-165C
Click here to download Figure: Fig.8-165C.pdf

0,45 0,45

0,40 0,40

0,35 0,35
Hardness (Gpa)

0,30 0,30

0,25 0,25

0,20 0,20

0,15 0,15

0,10
Strain Rate 0,10
0,01
0,05 T = 165 º C
0,05 0,05
0,25
0,00 0,00
0 250 500 750 1000
Indent. Displacement (nm)

Fig. 8: Hardness as a function of strain rate against indent displacement after reference
drift correction for three different temperatures. The error bars correspond to one
standard deviation.
Figure 9-25C
Click here to download Figure: Fig.9-25C.pdf

6000 6000

5000 Strain rate 5000

0.01
0.05
4000 0.25 4000
Load (μN)

3000 3000

2000 2000

1000 1000
T = 25 º C

0 0
0 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300
Indent. Displacement (nm)
Figure 9-90C
Click here to download Figure: Fig.9-90C.pdf

6000 6000

Strain Rate
5000 0.01 5000
0.05
4000 0.25 4000
Load (μN)

3000 3000

2000 2000

1000 1000
T = 90°C

0 0
0 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300
Indent. Displacement (nm)
Figure 9-165C
Click here to download Figure: Fig.9-165C.pdf

6000 6000

5000 Strain Rate 5000

0.01
0.05
4000 0.25 4000
Load (μN)

3000 3000

2000 2000

1000 1000
T = 165 º C

0 0
0 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300
Indent. Displacement (nm)
Figure 10
Click here to download Figure: Fig.10.pdf

-3 -3
1,1x10 1,1x10
-3 -3
1,0x10 1,0x10
-4 -4
9,0x10 9,0x10
-4 -4
8,0x10 8,0x10
-4 -4
7,0x10 7,0x10
H/T (Gpa/°K)

-4 -4
6,0x10 6,0x10
-4 -4
5,0x10 5,0x10
-4 -4
4,0x10 4,0x10
-4 -4
3,0x10 3,0x10
-4 T (°C) -4
2,0x10 2,0x10
25 90 165
-4
1,0x10 50 115 200 1,0x10
-4

70 140
0,0 0,0
0,01 0,1
Strain Rate (1/S)