Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

High-Throughput Nanoindentation for Statistical and Spatial Property Determination

High-Throughput Nanoindentation for Statistical and Spatial Property Determination JOM, Vol. 70, No. 4, 2018 https://doi.org/10.1007/s11837-018-2752-0 2018 The Author(s). This article is an open access publication BEYOND INDENTATION HARDNESS AND MODULUS: ADVANCES IN NANOINDENTATION TECHNIQUES: PART II High-Throughput Nanoindentation for Statistical and Spatial Property Determination 1 2 1,3 ERIC D. HINTSALA, UDE HANGEN, and DOUGLAS D. STAUFFER 1.—Bruker Nano Surfaces, Eden Prairie, MN, USA. 2.—Bruker Nano Surfaces, Aachen, Germany. 3.—e-mail: [email protected] Standard nanoindentation tests are ‘‘high throughput’’ compared to nearly all other mechanical tests, such as tension or compression. However, the typical rates of tens of tests per hour can be significantly improved. These higher testing rates enable otherwise impractical studies requiring several thousands of indents, such as high-resolution property mapping and detailed statistical studies. However, care must be taken to avoid systematic errors in the mea- surement, including choosing of the indentation depth/spacing to avoid over- lap of plastic zones, pileup, and influence of neighboring microstructural features in the material being tested. Furthermore, since fast loading rates are required, the strain rate sensitivity must also be considered. A review of these effects is given, with the emphasis placed on making complimentary standard nanoindentation measurements to address these issues. Experi- mental applications of the technique, including mapping of welds, microstructures, and composites with varying length scales, along with studying the effect of surface roughness on nominally homogeneous speci- mens, will be presented. testing. This means that 10,000 indent maps can INTRODUCTION be completed in less than an hour—for perspective, Nanoindentation has been proven to be a power- this would generate a property map of ful tool for exploring mechanical behavior at small- 100 9 100 lm with 1-lm spacing. The speeds, res- length scales. This is due to the technique being olution, scan size, and sample preparation require- highly localized and only semi-destructive, while ments are comparable to a variety of SEM mapping simultaneously allowing extraction of a diverse set techniques, such as electron backscatter diffraction of properties including elastic, plastic, and fracture. (EBSD) and energy dispersive spectrometry (EDS). In addition, the sample preparation requirements The fact that they also give highly complementary are significantly less stringent than most other information means that correlated surveying of mechanical testing techniques, and the procedures microstructural features for their crystallographic, 1–5 are well established. However, standard nanoin- chemical and mechanical properties provides dentation testing requires several minutes per test researchers with a powerful tool. Besides this, for tasks such as locating suitable areas, the sample statistical indentation techniques allow users to approach, drift correction, and retraction of the tip. quickly determine parameters of significance, This makes certain applications, such as property screen materials, and identify more global trends 6,7 8 mapping by indentation grids or generation of from highly localized nanoindentation tests. In statistical data sets, extremely time consuming and, addition, statistical data sets help combat factors in some cases, too slow to be practical. Various producing data outliers, such as surface rough- 9–12 technologies have been developed in recent years ness, which are a hindrance to accurate nanoin- that have greatly accelerated nanoindentation test- dentation testing. Lastly, these techniques can also ing, with state-of-the-art speeds of up to 6 indents/ be applied to systems with environmental control, second representing at least two orders of magni- such as heating, controlled humidity, or even sub- tude improvement over standard quasi-static merged specimens, providing additional variables to 494 (Published online February 14, 2018) High-Throughput Nanoindentation for Statistical and Spatial Property Determination 495 Fig. 1. (a) Measured hardness of fused quartz after indenter tip area function calibration for three Berkovich indenters of varying tip radii. A dynamic indentation mode with a 0.2-nm displacement amplitude is capable of measuring values for low penetration depths. (b) Comparison of purely elastic and elastic–plastic indents in fused quartz, with inset schematics of the elastic zone (light) and the plastic zone (dark). During elastic indentation in the Hertzian regime the average contact stress is less than that required to initiate plasticity. The elastic–plastic indentation shows hysteresis in the load-depth curve. In this case a constant hardness is observed. explore. To date, hardness mapping has been uti- A second consideration involves the necessity of lized to explore spatial variations in a variety of high loading rates, which may induce strain rate 13 14 materials including cement pastes, concrete, sensitivity changes in the measured hardness. Both 15 6,7,15,16 tooth enamel, metal matrix composites, subjects will be covered in the following two sec- 17 6,18,19 intermetallics, metal alloys, and wood adhe- tions, ‘‘Indentation Spacing and Resolution’’ and sive bonds. ‘‘Strain Rate Sensitivity’’. However, properly conducting high-speed nanoin- dentation and interpreting its results requires one Indentation Spacing and Resolution to consider various factors, such as indentation When mapping surface properties, the in-plane spacing, strain rate effects, and indentation depth. spatial resolution is of primary concern; however, In addition, it can currently only be applied to defining this requires consideration of the full three- measuring hardness and elastic modulus because of dimensional shape of the indentation stress field or the restrictions on load function choice. Thus, high- the volume of material being tested. Since the stress speed indentation is not a replacement for standard field decays continuously as a function of distance indentation techniques. Rather, the approach that from the contact zone, boundaries can only be is advocated is that of a complementary technique defined by a specific stress or strain value. The for standard indentation, where the standard test- most important is the subdivision into a purely ing protocol allows one to assess indentation size elastic and an elastic–plastic zone with the bound- 21–23 24 effects, rate dependence, and spacing effects. ary set by the yield criterion of the material, as Here, we will review these key concepts first before illustrated in Fig. 1. Thus, the entire elastic zone presenting example application data emphasizing contributes to the modulus measurement, while property-mapping techniques. only the elastic–plastic zone contributes to the hardness, H, measurement. In terms of a more EXPERIMENTAL CONSIDERATIONS FOR practical definition for defining the indentation NANOINDENTATION MAPPING resolution, one can define an acceptable relative To produce high-quality property maps, consider- change of properties in proximity to a feature, ation of the stress field underneath the indenter tip such as a microstructural boundary or a previous is critical. Not only is there potential for the damage indent. zones from individual tests to overlap and invali- Some indenters have a geometry that can be date the results, but the ‘‘resolution’’ of the nanoin- described as self-similar, which is simply a tip shape dentation test is relevant when testing near with a constant ratio of the contact area to depth boundaries of features, such as grain and phase versus load. This property is maintained by common boundaries, weld zones, composite interfaces, and pyramidal indenters, including Berkovich and cube material gradients, for damage or composition. The corners, as well as conical tips, and implies that the indenter resolution needs to be carefully defined, as measured properties will not change as a function of the stress field occurs in three dimensions and indentation load. Notably, spherical tips are an consists of separately sized elastic and plastic zones. exception. Thus, for a self-similar indenter, all- 496 Hintsala, Hangen, and Stauffer Fig. 2. (a) Hardness and modulus versus neighbor-to-neighbor distance observed with indentations of 50-nm penetration depth on single crystal Al. (b) Hardness and modulus of an indent performed in Ti between two rigid TiN interfaces. important geometrical parameters can be expressed improves for a cube corner that has a steeper as a function of the contact radius, a. The contact contact radius to a depth ratio of 0.7 compared to radius is defined as the radius of a circle of 3.5 for a Berkovich. Since these tips are self-similar, estimated equivalent area to that of the actual the plastic zone size is proportional to the contact contact, thus allowing pyramidal probes to be radius and is reduced by approximately a factor of 5 described by the same parameter as spherical and as well. As previously discussed, the exact plastic conical. zone size is specific to a given tip-material-depth Regarding issues with indent-to-indent spacing, combination, so experimental evaluation is required there are several effects. If the second indent to determine this size precisely. This can be illus- overlaps with the residual impression or pileup trated by two case studies: (1) the influence of one from the previous, this clearly invalidates the semi- indentation on its neighbor and (2) the influence of infinite half-space assumption and the actual con- interfaces in the proximity of an indent. tact radius will deviate significantly from the To illustrate this effect, the indent spacing, d, was assumed value. Subtler is the overlap of plastic varied on an Al sample as tested with a Berkovich tip, zones, which extend further in the sub-surface of as shown in Fig. 2a. Interestingly, reducing d results the testing plane. The residual plastic zone could be in a corresponding decrease in hardness rather than considered cold-worked, thereby elevated hardness, an increase as would be expected for work hardening. but the exact interaction of plastically nucleated Additionally, no effect on modulus would be expected defects could also produce a softening effect by from plastic zone overlap. Therefore, the effect on providing dislocation sources. The radius of the modulus at the smaller values of d indicates the plastic zone, R , in relationship to the contact radius invalidation of the area function due to pileup. This pileup-affected zone begins at d = 750 nm, slightly is material specific because of differences in plas- less than the recommended distance of 5.6 times the ticity mechanisms. For metals, this ratio can range 27–29 contact radius, or 840 nm for a maximum displace- from 3a to 6a. For a sharp Berkovich tip with ment of 50 nm. Clearly, a trade-off occurs between 50-nm radius of curvature, reliable hardness mea- surements can be achieved at a depth of at least lateral resolution and hardness measurement accu- 15 nm as shown in Fig. 1 (note that the modulus racy, as using a sharper tip at smaller depths gives a was constant for the three tips over the entire depth reduced elastic–plastic zone. An additional tradeoff range). This corresponds to a maximum contact can be made by sacrificing the hardness measure- radius of 57 nm, thus requiring a 315-nm indent ment altogether and limiting testing to a purely spacing for a soft metal with R  6a. This situation elastic regime. The absence of a plastic zone and p High-Throughput Nanoindentation for Statistical and Spatial Property Determination 497 residual displacement and thus no pileup in a purely Strain Rate Sensitivity elastic indent allow for the contact radii to overlap One of the drawbacks of high-speed nanoinden- between individual indents. tation mapping is a loss of flexibility in the load The role of sample interfaces, important to high- function, where high loading rates are needed for resolution mapping of compositional and phase increased mapping speed. These high loading rates varied materials, is illustrated by mapping with a can influence the measured hardness, but this metal-ceramic cross-sectional sample. Here, indents depends again on the material type, tip shape, and are placed near a material interface in a sample, several other variables. The hardness from a specifically, a 750-nm-wide Ti layer sandwiched nanoindentation test is strain rate dependent and between two extremely hard (H  25 GPa) Ti-N fit by a power law relationship by a characterizing layers. The indenter was carefully placed in the @ ln H parameter, m  , where e _ is the strain rate. The @ ln e _ center of the Ti layer using in situ SPM imaging. A strain rate for indentation is defined proportionally profile of hardness and modulus as a function of as the displacement rate over the total displacement depth is shown in Fig. 2b, where correct hardness _ _ h=h or P=P correspondingly for loading rate over and reduced modulus values for Ti were only total load for materials that do not have depth measured at 15–30-nm depth. At larger depths, the indentation stress field increasingly interacts dependence to their response, and for self-similar with the TiN layers resulting in increasing modulus indenters. Since this relationship is fit by a power and hardness values. In some mapping scenarios, law, one can describe it as an order of magnitude the indentation grid is generated with a predeter- effect. As previously discussed, high-speed nanoin- mined spacing that will place indents at varying dentation techniques can run about two orders of distances from, or on top of, a phase boundary or magnitude faster than standard indentation tech- interface in the sample. This can produce measure- niques. Since typical strain rate sensitivity param- ment errors if plasticity mechanisms are affected by eter m values range from 0.001 to 0.1 in crystalline the presence of the boundary, such as providing materials, this corresponds to a hardness value shift defect sources, sinks, and barriers. These data between 0.4% and 37% compared to standard speed points can usually be filtered during analysis by indentation. However, one must look at the bigger examining the statistical distribution of measured picture, which is that strain rate sensitivity is properties and removing outliers. These boundary determined by the predominant deformation mech- effects can change the ideal indent spacing for anism and is strongly affected by variables that aid mapping. Therefore, examining the effect near or hinder operation of these mechanisms. These sample boundaries to determine the best spacing variables most notably include temperature, but value is recommended. also crystalline orientation and grain size. Special To summarize this section, the achievable limits cases, such as nanocrystalline or ultrafine grain of nanoindentation resolution depend strongly on materials, can possess high strain rate sensitivity the tip shape, material being tested, and, crucially, values because of the dominance of grain bound- 30,31 the deformation regime. Following Jakes et al., ary diffusion mechanisms or, in the case of glasses, unusual behavior due to shear transformation three dimensionless parameters can be defined that 33,34 zones. Some literature data are presented in control indent resolution: the contact area relative Table I, which shows how much of a hardness shift to the distance to a feature, A/d, and two material- would be expected by increasing the strain rate by dependent parameters, the ratio E/H and the Pois- son’s ratio. One can define a maximum modulus or two orders of magnitude as discussed above for some hardness change versus d using these parameters. of the more interesting scenarios. Therefore, the smallest achievable d values are Thus, the origins of strain rate sensitivity are found at the lowest indentation depths in the elastic complex and require considerations of many sub- regime. However, if hardness measurements are tleties. However, for several classes of materials the desired, testing in the elastic–plastic regime is effect is essentially marginal. The best approach is necessary and a balance between the accuracy of to directly measure the strain rate sensitivity for hardness and lateral resolution must be chosen. the materials of interest; these techniques have Table I. Hardness change expected for two orders of magnitude increase in loading rate Material m DH (%) Notes References UFG Al 0.03 to 0.1 14–37 From RT—250C 32 Single crystal Cr 0.08 to 0.003 37–1.4 From RT—300C 35 Ti alloy 0.005 to 0.04 2–18 Different grain orientations 36 Al–Li alloy  0.01 to 0.0035 4.6–1.6 Different aging recipes 37 Fused silica 0.0068 to 0.01 3.1–4.6 Room temperature 38 498 Hintsala, Hangen, and Stauffer recently been reviewed by Maier-Kiener and produced by laser and resistance techniques, map- Durst. As a final point, the role of indentation ping of phases, and grains in alloys, and evaluation depth should be acknowledged, as shallower indents of composites. In particular, dissimilar material are typical for indentation mapping and deeper welds produce complex microstructures and can indents for strain rate sensitivity measurements. be better engineered using nanoindentation data to Ideally, this should not affect results but as real tips establish statistically significant variables. are blunt, shallower indents are increasingly dom- The scale and resolution of high-speed indenta- inated by a spherical-like contact. tion can be demonstrated through a correlated EBSD and nanoindentation map of a 410 stainless Summary and Complimentary steel, which was laser clad onto a 4140 stainless Nanoindentation Validation steel substrate. The large-scale structure of the heat-affected zone from the laser cladding process is In the absence of sophisticated analysis and/or shown via a traditional stage automation method in modeling, one can simply advocate the approach of Fig. 3a. To investigate the transition from the exploring the effects of the parameter space on the cladding to the substrate in more detail, a fiduciary measured properties of interest whenever possible, marker was drawn around the interface using specifically, the combination of loading rate, inden- focused ion beam machining to facilitate testing of tation depth, and indentation spacing. Thus, a the same region by nanoindentation mapping and typical complimentary approach to validate a high- EBSD. In this case, a Berkovich tip was used with speed indentation map would include: 400 lN force and an indent spacing of 500 nm. The nanoindentation mapping shows little difference in 1. Measurement of depth sensitivity: Depth profil- modulus between the cladding and the substrate, ing is already frequently done to calibrate tip but a substantial change in hardness with  5 GPa area functions. A variety of methods can be average for the 4140 substrate and  8 GPa hard- used, including a varying depth indent arrays, ness average for the cladding. In the correlated partial unload load functions, or dynamic meth- 40,41 EBSD boundary map, it appears that the region of ods. highest hardness corresponds closely with regions of 2. Measurement of spacing sensitivity: With the a high density of high-angle grain boundaries, depth dependence established, the user can marked in blue. chose their desired depth for the indentation map. Next, the spacing effects at that desired High-Speed Nanoindentation at Elevated and depth can be studied with indent arrays. When Cryogenic Temperatures high spatial resolution is unnecessary or impractical because of the desire to map a larger High-temperature nanoindentation is a growing area by nanoindentation, conservatively large field of research for reactors, engines, turbines, and spacing could be used freely. more. One popular high-temperature material, a 3. Measurement of rate sensitivity: Either through SiC matrix and SiC fiber composite, is evaluated at grids with varying indent speed or characteriz- high temperature using 5-lm indent spacing, 7-mN ing the strain rate sensitivity coefficient. load, and a Berkovich tip. The difference in hard- ness between the fibers and the matrix is apparent, There could also be a desire to define the desired along with a region of low hardness along the indent spacing first, i.e., resolution of the map, then interface (Fig. 4). This is likely due to free volume what maximum depth can be used needs to be along the fiber/matrix interface, reducing the hard- determined, thereby reversing steps 1 and 2. ness through decreased material confinement. The distributions of hardness and modulus at 400C APPLICATIONS FOR NANOINDENTATION show a bimodal distribution, individually corre- MAPPING sponding to the fiber and matrix. As the tempera- In the following, several examples are presented ture is increased to 800C, the modulus values shift that highlight the capabilities of state-of-the-art slightly downward overall, as expected, but main- high-speed nanoindentation. The following exam- tains a bimodal distribution. The two measure- ples were all performed using a Hysitron TI-980 ments were taken from different regions of the TriboIndenter (Bruker Nano Surfaces, Minneapolis, sample, so the change in total counts for the two MN, USA) operating in Accelerating Property Map- phases is different. More interestingly, the hardness ping (XPM) mode. distribution is observed to shift towards a single peaked distribution at 800C. Correlated EBSD and Nanoindentation Cryogenic temperatures are of interest for mate- Mapping rials that are subjected to conditions such as outer space, arctic or winter environments, and part of The most obvious application for such technology cooling systems. A ubiquitous structural alloy, 1018 is mapping of small-scale material interfaces as steel, was studied from room temperature down to they cannot be easily evaluated at the bulk scale. 120C. A hardness map at 0C generated with 1- These include welds, especially microscale ones as High-Throughput Nanoindentation for Statistical and Spatial Property Determination 499 Fig. 3. (a) Stage automation map of the hardness distribution in a 410 laser cladding on 4140 substrate, with a black box denoting the location of the optical micrograph in (b) where a fiduciary marker was drawn around the boundary with FIB. This region was subjected to correlated EBSD, with the boundary map in (c) and the inverse pole figure in (d) and high-speed nanoindentation mapping with the hardness overlaid on the boundary map in (e). lm indent spacing, a peak load of 500 lN, and a High-speed nanoindentation techniques provide Berkovich tip clearly shows the two-phase ferrite an advantage when operating at extreme non- and pearlite microstructure, which is also reflected ambient temperatures through reduced contact by the hardness distribution as seen in Fig. 5. time, which reduces tip wear, and the relative effect Indents into the ferrite phase were done as part of of drift. Tip-sample thermal equilibrium is one of a decreasing temperature sweep, with the resulting the largest challenges to extreme temperature load displacement curves indicating a ductile-to- testing as it produces drift when they are brought brittle transition at  58C. Here, homogeneous into contact. The cryo and high-temperature stage dislocation plasticity at the above temperatures in this article utilizes a multi-element heating gave way to serrated flow, indicating dislocation microchamber that exposes the tip and sample to bursts. Since ferrite is BCC, the Peierls’ barrier is the same environment. However, it has been 46,47 relatively large compared to FCC metals and is thus shown that thermal stabilization in vacuum is reliant on thermal assistance for homogeneous more time consuming and difficult. Reducing the plasticity. This was also reflected in the hardness, typical time in contact from 20 s to 0.2 s reduces the which increased 44% over the tested temperature effect of drift on the measurement by two orders of range. This hardening for decreasing temperatures magnitude. This reduced time in contact also sig- also relates to the ductile to brittle temperature nificantly reduces tip wear, a major issue of high transition. This transition is more obvious when temperature testing. The tip sample contact can looking at the pop-in behavior of the indentation be modeled as a high-pressure diffusion couple in curves. The room temperature behavior is primar- thermodynamic software such as Thermal Calc. ily a smooth curve or with very short pop-ins to Simply reducing the time in contact has a dramatic approximate a predominately smooth curve. As impact on the number of indentations that can be temperatures decrease, the flow becomes more performed with a given tip/sample combination. stochastic with increasing pop-in size. There seems to be a transition between the  15C and  25C Generation and Utilization of Large Data Sets curves. This DBTT is lower than the 5C value In contrast to the heterogeneous samples tested typically reported for Charpy impact testing, which thus far, nanoindentation is often performed on is a much higher strain rate. samples that are relatively homogeneous, such as 500 Hintsala, Hangen, and Stauffer Fig. 4. The region of the SiC fiber-SiC matrix composite tested at 400C is shown by SPM in (a) and the corresponding hardness map is shown in (b). The property distributions at 400C and 800C are shown in (c) for the modulus and (d) for the hardness. foils, thin films on substrates, and the substrates compared to traditional Vickers microhardness 49–51 themselves. Even a layered sample with large dimen- testing. Here, 196 indent grids produced by sions in the sample plane can be considered locally high-speed nanoindentation with 5-mN load and a homogeneous. In these cases, statistics allow for Berkovich indenter are compared against a single generation of large data sets where the precise values Vickers indent in a line scan starting in the weld can be determined, with data histograms allowing for joint and moving progressively through the heat- the identification of statistical outliers. This can be affected zone, using an empirical relationship to compared to the number of tests run in a typical compare hardness. The grid of nanoindentations nanoindentation study, where n £ 10 in many cases. fit into approximately the same area as the single A simple experiment can be done looking at the microhardness test (Fig. 8). In this case, a variety of statistical distribution of hardness and modulus microstructures are encountered, from martensitic comparing vibratory polished (100) aluminum and rich regions near the weld joint progressively into aluminum polished with 600 grit paper. Arrays of bainite and finally ferrite. It can be observed that nine indents, 3 9 3 with 15-um spacing between although the hardness versus distance curves are indents, are placed in a larger 5 9 5 array (Fig. 6). comparable between the average value of the high- The total number of indents is now n = 225 for speed nanoindentation grids and microindentation, each sample, which is a factor of a 20 times larger the grids feature scatter bands due to the varying number of tests than for most nanoindention stud- microstructure. For instance, the spread is ies. The pileup corrected modulus and hardness increased near the weld joint, as the martensite histograms (Fig. 7) show both an increase in hard- clustered into islands around the grain boundaries, ness for the roughened/work-hardened sample and where as bainite and ferrite represent better dis- a corresponding increase in the spread of the data. persed microstructures. In these regions, a Combining both statistical analysis and mapping researcher could consider moving to more of a compared against high-speed nanoindentation for a mapping rather than statistical sampling type of railway weld joint in a railway steel can be experiment. High-Throughput Nanoindentation for Statistical and Spatial Property Determination 501 Fig. 5. The 1018 steel is evaluated at cryogenic temperatures, with a clear difference in hardness between the ferrite and pearlite phases in the (a) indentation map and corresponding hardness distribution (b). The inset in (b) is a gradient force SPM image of the microstructure. Example load–displacement curves for the temperature sweep in (c) and the resulting hardness in (d) show a DBT. Fig. 6. Load-displacement curves for the vibratory polished sample (a) and the sample roughened with 600 grit paper (b) and corresponding micrographs showing a 5 9 5 placement of 3 9 3 grids of indents in relationship to the surface topography. detailed information on small-scale regions of sig- CONCLUSIONS AND OUTLOOK nificant industrial importance that are not easily Overall, high-speed indentation techniques are tested on the bulk scale, such as welds, fine grain relatively underused given they possess many and phase structures, composites and interfaces, potential applications. Property mapping, especially and more. Statistical distributions can be generated when used in conjunction with correlated tech- simultaneously, which offer a variety of useful niques characterizing the structure, provide 502 Hintsala, Hangen, and Stauffer Fig. 7. Histograms for the vibratory polished modulus (a) and hardness (b) for both samples. The hardness comparison shows a much larger spread and increase in the value for the roughened sample in comparison to the vibratory polished (smooth) sample. Fig. 8. (a) Comparison between high-speed nanoindentation hardness grids and Vickers microhardness measurements as a function of distance from the weld boundary. SPM of a corresponding microhardness impression and nanohardness grid (total scan size 45 9 45 lm) is shown in (b), which is repeated for 50 points as shown in the stitched optical image in (c). information. However, these techniques do not behavior based on nanoscale measurements is very replace standard nanoindentation techniques, as attractive and may lead to future breakthroughs in the influence of the indentation size effect from materials design and fine-tuning of their perfor- depth, strain rate sensitivity, and spacing should be mance. As a rapid and highly localized mechanical studied in conjunction. measurement tool, high-speed nanoindentation Looking forward, there are many relatively unex- mapping should play an important role in this plored applications for these techniques. Though regard. there are a multitude of material microstructures to ACKNOWLEDGEMENTS potentially map and model, it is the authors’ opinion that sophisticated analysis of ‘‘big data’’ sets, The authors gratefully acknowledge assistance through techniques like machine learning, repre- from Richard Nay, Jared Risan, Robert Dietrich, sent one the biggest frontiers in materials science. Anqi Qiu, and Benjamin Stadnick with the speci- The theme of bridging length scales and producing men testing and Daniel Sorensen with sample cohesive understanding of bulk mechanical preparation. High-Throughput Nanoindentation for Statistical and Spatial Property Determination 503 23. M. R. Maughan, A. A. Leonard, D. D. Stauffer, D. F. Bahr, OPEN ACCESS Philos. Mag. (2017). https://doi.org/10.1080/14786435.2017. This article is distributed under the terms of the 24. B. N. Lucas, W. C. Oliver, G. M. Pharr, J.-L. Loubet, MRS Creative Commons Attribution 4.0 International Li- Online Proc. Libr. 436 (1996). cense (http://creativecommons.org/licenses/by/4.0/), 25. H. Hertz, Hertz’s Miscellaneous Papers (London, UK: which permits unrestricted use, distribution, and Macmillan, 1896). reproduction in any medium, provided you give 26. K.L. Johnson, Contact Mechanics (Cambridge, UK: Cam- appropriate credit to the original author(s) and the bridge University Press, 1985). 27. K. Durst, B. Backes, and M. Goken, Scripta Mater. 52, source, provide a link to the Creative Commons li- 1093–1097 (2005). cense, and indicate if changes were made. 28. ASTM Standard E384-17, Standard Test Method for Knoop and Vickers Hardness of Materials, (West Conshohocken, FUNDING PA: ASTM International, 2017) https://doi.org/10.1520/ e0384-17. Funding was provided by Bruker Nano Surfaces. 29. DIN ISO 6507-1:2018, Metallic materials—Vickers hard- ness test—Part 1: Test method (Geneva, Switzerland: REFERENCES International Organization for Standardization, 2018). 1. W.C. Oliver and G.M. Pharr, J. Mater. Res. 7, 1564–1583 30. J.E. Jakes, J. Mater. Res. 24, 1016–1031 (2009). (1992). 31. J.E. Jakes and D.S. Stone, Philos. Mag. 91, 1387–1399 2. A.C. Fischer-Cripps, Surf. Coat. Technol. 200, 4153–4165 (2011). (2006). 32. J.M. Wheeler, V. Maier, K. Durst, M. Go ¨ ken, and J. 3. A.A. Volinsky, J.B. Vella, and W.W. Gerberich, Thin Solid Michler, Mater. Sci. Eng., A 585, 108 (2013). Films 429, 201–210 (2003). 33. J. Lu, G. Ravichandran, and W.L. Johnson, Acta Mater. 51, 4. G.M. Pharr, D.S. Harding, and W.C. Oliver, Mechanical 3429–3443 (2003). Properties and Deformation Behavior of Materials Having 34. F.J. Dalla Torre, A. Dubach, M.E. Siegrist, and J.F. Lo ¨ ffler, Ultra-Fine Microstructures, ed. M. Nastasti, D.M. Parkin, Appl. Phys. Lett. 89, 091918 (2006). and H. Gleiter (Dordrecht, the Netherlands: Kluwer Aca- 35. V. Maier, A. Hohenwarter, R. Pippan, and D. Kiener, demic Publishers, 1993), pp. 449–461. Scripta Mater. 106, 42 (2015). 5. W.C. Oliver and G.M. Pharr, J. Mater. Res. 19, 3–20 (2011). 36. T.-S. Jun, D.E.J. Armstrong, and T.B. Britton, J. Alloys 6. N.X. Randall, M. Vandamme, and F.J. Ulm, J. Mater. Res. Compd. 672, 282 (2016). 24, 679–690 (2009). 37. H. Ovri and E.T. Lilleodden, Acta Mater. 89, 88 (2015). 7. G. Constantinides, K.S. Ravi Chandran, F.J. Ulm, and K.J. 38. R. Limbach, B.P. Rodrigues, L. Wondraczek, and J. Non- Van Vliet, Mater. Sci. Eng., A 430, 189–202 (2006). Cryst, Solids 404, 124 (2014). 8. L. Sorelli, G. Constantinides, F.J. Ulm, and F. Toutle- 39. V. Maier-Kiener and K. Durst, JOM 69, 2246–2255 (2017). monde, Cem. Concr. Res. 38, 1447–1456 (2008). 40. S.A.S. Asif, K.J. Wahl, and R.J. Colton, Rev. Sci. Instrum. 9. J.A. Greenwood and J.H. Tripp, Trans. ASME Ser. E J. 70, 2408 (1999). Appl. Mech. 34, 417–420 (1967). 41. X. Li and B. Bhushan, Mater. Charact. 48, 11–36 (2002). 10. W.G. Jiang, J.J. Su, and X.Q. Feng, Eng. Fract. Mech. 75, 42. L.E. Murr, J. Mater. Eng. Perform. 19, 1071–1089 (2010). 4965–4972 (2008). 43. B. Jones and C. Nachtsheim, J. Qual. Technol. 43, 1–15 11. M.S. Bobji and S.K. Biswas, J. Mater. Res. 14, 2259–2268 (2011). (1999). 44. O.L. Warren, S.A. Downs, and T.J. Wyrobek, Z. Metallkd. 12. A.C. Fischer-Cripps, Factors Affecting Nanoindentation 95, 287–296 (2004). Test Data (New York, NY: Springer, 2000), pp. 61–82. 45. U.S. Patent, US 20150185117 A1, Environmental condi- 13. M. Sebastiani, R. Moscatelli, F. Ridi, P. Baglioni, and F. tioning assembly for use in mechanical testing at micron or Carassiti, Mater. Des. 97, 372–380 (2016). nano-scales, Jul 2, 2015. 14. J.L. Cuy, A.B. Mann, K.J. Livi, M.F. Teaford, and T.P. 46. J.M. Wheeler, D.E.J. Armstrong, W. Heinz, and R. Schwaiger, Weihs, Arch. Oral Biol. 47, 281–291 (2002). Curr. Opin. Solid State Mater. Sci. 19, 354–366 (2015). 15. K. Morsi, V.V. Patel, K.S. Moon, and J.E. Garay, J. Mater. 47. S. Korte, R.J. Stearn, J.M. Wheeler, and W.J. Clegg, J. Sci. 43, 4050 (2008). Mater. Res. 27, 167–176 (2012). 16. H. Engqvist and U. Wiklund, Tribol. Lett. 8, 147–152 48. J.M. Wheeler and J. Michler, Rev. Sci. Instrum. 84, 101301 (2000). (2013). 17. H.-W. Chang, M.-X. Zhang, A. Atrens, and H. Huang, J. 49. R.R. Ambriz, D. Chicot, N. Benseddiq, G. Mesmacque, and Alloys Compd. 587, 527–532 (2014). S.D. De la Torre, Eur. J. Mech. A Solids 30, 307–315 18. C. Tromas, M. Arnoux, and X. Milhet, Scripta Mater. 66, (2011). 77–80 (2012). 50. K.H. Chung, W. Lee, J.H. Kim, C. Kim, S.H. Park, D. 19. T. Ogura, S. Hirosawa, and T. Sato, Sci. Technol. Adv. Kwon, and K. Chung, Int. J. Solids Struct. 46, 344–363 Mater. 5, 491–496 (2004). (2009). 20. J. Konnerth, A. Valla, and W. Gindl, Appl. Phys. A Mater. 51. G. Sun, F. Xu, G. Li, X. Huang, and Q. Li, Comput. Mater. Sci. Process. 88, 371–375 (2007). Sci. 85, 347–362 (2014). 21. W.W. Gerberich, N.I. Tymiak, J.C. Grunlan, M.F. Horste- 52. DIN ISO 14577-1:2015, Metallic materials—Instrumented meyer, and M.I. Baskes, J. Appl. Mech. 69, 433 (2002). Indentation Test for Hardness and Materials Parame- 22. W.D. Nix and H. Gao, J. Mech. Phys. Solids 46, 425–441 ters—Part 1: Test Method (Geneva, Switzerland: Interna- (1998). tional Organization for Standardization, 2015). http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png JOM Unpaywall

High-Throughput Nanoindentation for Statistical and Spatial Property Determination

Download PDF
10 pages

Loading next page...
 
/lp/unpaywall/high-throughput-nanoindentation-for-statistical-and-spatial-property-y1HjSwea5S

References (8)

Publisher
Unpaywall
ISSN
1047-4838
DOI
10.1007/s11837-018-2752-0
Publisher site
See Article on Publisher Site

Abstract

JOM, Vol. 70, No. 4, 2018 https://doi.org/10.1007/s11837-018-2752-0 2018 The Author(s). This article is an open access publication BEYOND INDENTATION HARDNESS AND MODULUS: ADVANCES IN NANOINDENTATION TECHNIQUES: PART II High-Throughput Nanoindentation for Statistical and Spatial Property Determination 1 2 1,3 ERIC D. HINTSALA, UDE HANGEN, and DOUGLAS D. STAUFFER 1.—Bruker Nano Surfaces, Eden Prairie, MN, USA. 2.—Bruker Nano Surfaces, Aachen, Germany. 3.—e-mail: [email protected] Standard nanoindentation tests are ‘‘high throughput’’ compared to nearly all other mechanical tests, such as tension or compression. However, the typical rates of tens of tests per hour can be significantly improved. These higher testing rates enable otherwise impractical studies requiring several thousands of indents, such as high-resolution property mapping and detailed statistical studies. However, care must be taken to avoid systematic errors in the mea- surement, including choosing of the indentation depth/spacing to avoid over- lap of plastic zones, pileup, and influence of neighboring microstructural features in the material being tested. Furthermore, since fast loading rates are required, the strain rate sensitivity must also be considered. A review of these effects is given, with the emphasis placed on making complimentary standard nanoindentation measurements to address these issues. Experi- mental applications of the technique, including mapping of welds, microstructures, and composites with varying length scales, along with studying the effect of surface roughness on nominally homogeneous speci- mens, will be presented. testing. This means that 10,000 indent maps can INTRODUCTION be completed in less than an hour—for perspective, Nanoindentation has been proven to be a power- this would generate a property map of ful tool for exploring mechanical behavior at small- 100 9 100 lm with 1-lm spacing. The speeds, res- length scales. This is due to the technique being olution, scan size, and sample preparation require- highly localized and only semi-destructive, while ments are comparable to a variety of SEM mapping simultaneously allowing extraction of a diverse set techniques, such as electron backscatter diffraction of properties including elastic, plastic, and fracture. (EBSD) and energy dispersive spectrometry (EDS). In addition, the sample preparation requirements The fact that they also give highly complementary are significantly less stringent than most other information means that correlated surveying of mechanical testing techniques, and the procedures microstructural features for their crystallographic, 1–5 are well established. However, standard nanoin- chemical and mechanical properties provides dentation testing requires several minutes per test researchers with a powerful tool. Besides this, for tasks such as locating suitable areas, the sample statistical indentation techniques allow users to approach, drift correction, and retraction of the tip. quickly determine parameters of significance, This makes certain applications, such as property screen materials, and identify more global trends 6,7 8 mapping by indentation grids or generation of from highly localized nanoindentation tests. In statistical data sets, extremely time consuming and, addition, statistical data sets help combat factors in some cases, too slow to be practical. Various producing data outliers, such as surface rough- 9–12 technologies have been developed in recent years ness, which are a hindrance to accurate nanoin- that have greatly accelerated nanoindentation test- dentation testing. Lastly, these techniques can also ing, with state-of-the-art speeds of up to 6 indents/ be applied to systems with environmental control, second representing at least two orders of magni- such as heating, controlled humidity, or even sub- tude improvement over standard quasi-static merged specimens, providing additional variables to 494 (Published online February 14, 2018) High-Throughput Nanoindentation for Statistical and Spatial Property Determination 495 Fig. 1. (a) Measured hardness of fused quartz after indenter tip area function calibration for three Berkovich indenters of varying tip radii. A dynamic indentation mode with a 0.2-nm displacement amplitude is capable of measuring values for low penetration depths. (b) Comparison of purely elastic and elastic–plastic indents in fused quartz, with inset schematics of the elastic zone (light) and the plastic zone (dark). During elastic indentation in the Hertzian regime the average contact stress is less than that required to initiate plasticity. The elastic–plastic indentation shows hysteresis in the load-depth curve. In this case a constant hardness is observed. explore. To date, hardness mapping has been uti- A second consideration involves the necessity of lized to explore spatial variations in a variety of high loading rates, which may induce strain rate 13 14 materials including cement pastes, concrete, sensitivity changes in the measured hardness. Both 15 6,7,15,16 tooth enamel, metal matrix composites, subjects will be covered in the following two sec- 17 6,18,19 intermetallics, metal alloys, and wood adhe- tions, ‘‘Indentation Spacing and Resolution’’ and sive bonds. ‘‘Strain Rate Sensitivity’’. However, properly conducting high-speed nanoin- dentation and interpreting its results requires one Indentation Spacing and Resolution to consider various factors, such as indentation When mapping surface properties, the in-plane spacing, strain rate effects, and indentation depth. spatial resolution is of primary concern; however, In addition, it can currently only be applied to defining this requires consideration of the full three- measuring hardness and elastic modulus because of dimensional shape of the indentation stress field or the restrictions on load function choice. Thus, high- the volume of material being tested. Since the stress speed indentation is not a replacement for standard field decays continuously as a function of distance indentation techniques. Rather, the approach that from the contact zone, boundaries can only be is advocated is that of a complementary technique defined by a specific stress or strain value. The for standard indentation, where the standard test- most important is the subdivision into a purely ing protocol allows one to assess indentation size elastic and an elastic–plastic zone with the bound- 21–23 24 effects, rate dependence, and spacing effects. ary set by the yield criterion of the material, as Here, we will review these key concepts first before illustrated in Fig. 1. Thus, the entire elastic zone presenting example application data emphasizing contributes to the modulus measurement, while property-mapping techniques. only the elastic–plastic zone contributes to the hardness, H, measurement. In terms of a more EXPERIMENTAL CONSIDERATIONS FOR practical definition for defining the indentation NANOINDENTATION MAPPING resolution, one can define an acceptable relative To produce high-quality property maps, consider- change of properties in proximity to a feature, ation of the stress field underneath the indenter tip such as a microstructural boundary or a previous is critical. Not only is there potential for the damage indent. zones from individual tests to overlap and invali- Some indenters have a geometry that can be date the results, but the ‘‘resolution’’ of the nanoin- described as self-similar, which is simply a tip shape dentation test is relevant when testing near with a constant ratio of the contact area to depth boundaries of features, such as grain and phase versus load. This property is maintained by common boundaries, weld zones, composite interfaces, and pyramidal indenters, including Berkovich and cube material gradients, for damage or composition. The corners, as well as conical tips, and implies that the indenter resolution needs to be carefully defined, as measured properties will not change as a function of the stress field occurs in three dimensions and indentation load. Notably, spherical tips are an consists of separately sized elastic and plastic zones. exception. Thus, for a self-similar indenter, all- 496 Hintsala, Hangen, and Stauffer Fig. 2. (a) Hardness and modulus versus neighbor-to-neighbor distance observed with indentations of 50-nm penetration depth on single crystal Al. (b) Hardness and modulus of an indent performed in Ti between two rigid TiN interfaces. important geometrical parameters can be expressed improves for a cube corner that has a steeper as a function of the contact radius, a. The contact contact radius to a depth ratio of 0.7 compared to radius is defined as the radius of a circle of 3.5 for a Berkovich. Since these tips are self-similar, estimated equivalent area to that of the actual the plastic zone size is proportional to the contact contact, thus allowing pyramidal probes to be radius and is reduced by approximately a factor of 5 described by the same parameter as spherical and as well. As previously discussed, the exact plastic conical. zone size is specific to a given tip-material-depth Regarding issues with indent-to-indent spacing, combination, so experimental evaluation is required there are several effects. If the second indent to determine this size precisely. This can be illus- overlaps with the residual impression or pileup trated by two case studies: (1) the influence of one from the previous, this clearly invalidates the semi- indentation on its neighbor and (2) the influence of infinite half-space assumption and the actual con- interfaces in the proximity of an indent. tact radius will deviate significantly from the To illustrate this effect, the indent spacing, d, was assumed value. Subtler is the overlap of plastic varied on an Al sample as tested with a Berkovich tip, zones, which extend further in the sub-surface of as shown in Fig. 2a. Interestingly, reducing d results the testing plane. The residual plastic zone could be in a corresponding decrease in hardness rather than considered cold-worked, thereby elevated hardness, an increase as would be expected for work hardening. but the exact interaction of plastically nucleated Additionally, no effect on modulus would be expected defects could also produce a softening effect by from plastic zone overlap. Therefore, the effect on providing dislocation sources. The radius of the modulus at the smaller values of d indicates the plastic zone, R , in relationship to the contact radius invalidation of the area function due to pileup. This pileup-affected zone begins at d = 750 nm, slightly is material specific because of differences in plas- less than the recommended distance of 5.6 times the ticity mechanisms. For metals, this ratio can range 27–29 contact radius, or 840 nm for a maximum displace- from 3a to 6a. For a sharp Berkovich tip with ment of 50 nm. Clearly, a trade-off occurs between 50-nm radius of curvature, reliable hardness mea- surements can be achieved at a depth of at least lateral resolution and hardness measurement accu- 15 nm as shown in Fig. 1 (note that the modulus racy, as using a sharper tip at smaller depths gives a was constant for the three tips over the entire depth reduced elastic–plastic zone. An additional tradeoff range). This corresponds to a maximum contact can be made by sacrificing the hardness measure- radius of 57 nm, thus requiring a 315-nm indent ment altogether and limiting testing to a purely spacing for a soft metal with R  6a. This situation elastic regime. The absence of a plastic zone and p High-Throughput Nanoindentation for Statistical and Spatial Property Determination 497 residual displacement and thus no pileup in a purely Strain Rate Sensitivity elastic indent allow for the contact radii to overlap One of the drawbacks of high-speed nanoinden- between individual indents. tation mapping is a loss of flexibility in the load The role of sample interfaces, important to high- function, where high loading rates are needed for resolution mapping of compositional and phase increased mapping speed. These high loading rates varied materials, is illustrated by mapping with a can influence the measured hardness, but this metal-ceramic cross-sectional sample. Here, indents depends again on the material type, tip shape, and are placed near a material interface in a sample, several other variables. The hardness from a specifically, a 750-nm-wide Ti layer sandwiched nanoindentation test is strain rate dependent and between two extremely hard (H  25 GPa) Ti-N fit by a power law relationship by a characterizing layers. The indenter was carefully placed in the @ ln H parameter, m  , where e _ is the strain rate. The @ ln e _ center of the Ti layer using in situ SPM imaging. A strain rate for indentation is defined proportionally profile of hardness and modulus as a function of as the displacement rate over the total displacement depth is shown in Fig. 2b, where correct hardness _ _ h=h or P=P correspondingly for loading rate over and reduced modulus values for Ti were only total load for materials that do not have depth measured at 15–30-nm depth. At larger depths, the indentation stress field increasingly interacts dependence to their response, and for self-similar with the TiN layers resulting in increasing modulus indenters. Since this relationship is fit by a power and hardness values. In some mapping scenarios, law, one can describe it as an order of magnitude the indentation grid is generated with a predeter- effect. As previously discussed, high-speed nanoin- mined spacing that will place indents at varying dentation techniques can run about two orders of distances from, or on top of, a phase boundary or magnitude faster than standard indentation tech- interface in the sample. This can produce measure- niques. Since typical strain rate sensitivity param- ment errors if plasticity mechanisms are affected by eter m values range from 0.001 to 0.1 in crystalline the presence of the boundary, such as providing materials, this corresponds to a hardness value shift defect sources, sinks, and barriers. These data between 0.4% and 37% compared to standard speed points can usually be filtered during analysis by indentation. However, one must look at the bigger examining the statistical distribution of measured picture, which is that strain rate sensitivity is properties and removing outliers. These boundary determined by the predominant deformation mech- effects can change the ideal indent spacing for anism and is strongly affected by variables that aid mapping. Therefore, examining the effect near or hinder operation of these mechanisms. These sample boundaries to determine the best spacing variables most notably include temperature, but value is recommended. also crystalline orientation and grain size. Special To summarize this section, the achievable limits cases, such as nanocrystalline or ultrafine grain of nanoindentation resolution depend strongly on materials, can possess high strain rate sensitivity the tip shape, material being tested, and, crucially, values because of the dominance of grain bound- 30,31 the deformation regime. Following Jakes et al., ary diffusion mechanisms or, in the case of glasses, unusual behavior due to shear transformation three dimensionless parameters can be defined that 33,34 zones. Some literature data are presented in control indent resolution: the contact area relative Table I, which shows how much of a hardness shift to the distance to a feature, A/d, and two material- would be expected by increasing the strain rate by dependent parameters, the ratio E/H and the Pois- son’s ratio. One can define a maximum modulus or two orders of magnitude as discussed above for some hardness change versus d using these parameters. of the more interesting scenarios. Therefore, the smallest achievable d values are Thus, the origins of strain rate sensitivity are found at the lowest indentation depths in the elastic complex and require considerations of many sub- regime. However, if hardness measurements are tleties. However, for several classes of materials the desired, testing in the elastic–plastic regime is effect is essentially marginal. The best approach is necessary and a balance between the accuracy of to directly measure the strain rate sensitivity for hardness and lateral resolution must be chosen. the materials of interest; these techniques have Table I. Hardness change expected for two orders of magnitude increase in loading rate Material m DH (%) Notes References UFG Al 0.03 to 0.1 14–37 From RT—250C 32 Single crystal Cr 0.08 to 0.003 37–1.4 From RT—300C 35 Ti alloy 0.005 to 0.04 2–18 Different grain orientations 36 Al–Li alloy  0.01 to 0.0035 4.6–1.6 Different aging recipes 37 Fused silica 0.0068 to 0.01 3.1–4.6 Room temperature 38 498 Hintsala, Hangen, and Stauffer recently been reviewed by Maier-Kiener and produced by laser and resistance techniques, map- Durst. As a final point, the role of indentation ping of phases, and grains in alloys, and evaluation depth should be acknowledged, as shallower indents of composites. In particular, dissimilar material are typical for indentation mapping and deeper welds produce complex microstructures and can indents for strain rate sensitivity measurements. be better engineered using nanoindentation data to Ideally, this should not affect results but as real tips establish statistically significant variables. are blunt, shallower indents are increasingly dom- The scale and resolution of high-speed indenta- inated by a spherical-like contact. tion can be demonstrated through a correlated EBSD and nanoindentation map of a 410 stainless Summary and Complimentary steel, which was laser clad onto a 4140 stainless Nanoindentation Validation steel substrate. The large-scale structure of the heat-affected zone from the laser cladding process is In the absence of sophisticated analysis and/or shown via a traditional stage automation method in modeling, one can simply advocate the approach of Fig. 3a. To investigate the transition from the exploring the effects of the parameter space on the cladding to the substrate in more detail, a fiduciary measured properties of interest whenever possible, marker was drawn around the interface using specifically, the combination of loading rate, inden- focused ion beam machining to facilitate testing of tation depth, and indentation spacing. Thus, a the same region by nanoindentation mapping and typical complimentary approach to validate a high- EBSD. In this case, a Berkovich tip was used with speed indentation map would include: 400 lN force and an indent spacing of 500 nm. The nanoindentation mapping shows little difference in 1. Measurement of depth sensitivity: Depth profil- modulus between the cladding and the substrate, ing is already frequently done to calibrate tip but a substantial change in hardness with  5 GPa area functions. A variety of methods can be average for the 4140 substrate and  8 GPa hard- used, including a varying depth indent arrays, ness average for the cladding. In the correlated partial unload load functions, or dynamic meth- 40,41 EBSD boundary map, it appears that the region of ods. highest hardness corresponds closely with regions of 2. Measurement of spacing sensitivity: With the a high density of high-angle grain boundaries, depth dependence established, the user can marked in blue. chose their desired depth for the indentation map. Next, the spacing effects at that desired High-Speed Nanoindentation at Elevated and depth can be studied with indent arrays. When Cryogenic Temperatures high spatial resolution is unnecessary or impractical because of the desire to map a larger High-temperature nanoindentation is a growing area by nanoindentation, conservatively large field of research for reactors, engines, turbines, and spacing could be used freely. more. One popular high-temperature material, a 3. Measurement of rate sensitivity: Either through SiC matrix and SiC fiber composite, is evaluated at grids with varying indent speed or characteriz- high temperature using 5-lm indent spacing, 7-mN ing the strain rate sensitivity coefficient. load, and a Berkovich tip. The difference in hard- ness between the fibers and the matrix is apparent, There could also be a desire to define the desired along with a region of low hardness along the indent spacing first, i.e., resolution of the map, then interface (Fig. 4). This is likely due to free volume what maximum depth can be used needs to be along the fiber/matrix interface, reducing the hard- determined, thereby reversing steps 1 and 2. ness through decreased material confinement. The distributions of hardness and modulus at 400C APPLICATIONS FOR NANOINDENTATION show a bimodal distribution, individually corre- MAPPING sponding to the fiber and matrix. As the tempera- In the following, several examples are presented ture is increased to 800C, the modulus values shift that highlight the capabilities of state-of-the-art slightly downward overall, as expected, but main- high-speed nanoindentation. The following exam- tains a bimodal distribution. The two measure- ples were all performed using a Hysitron TI-980 ments were taken from different regions of the TriboIndenter (Bruker Nano Surfaces, Minneapolis, sample, so the change in total counts for the two MN, USA) operating in Accelerating Property Map- phases is different. More interestingly, the hardness ping (XPM) mode. distribution is observed to shift towards a single peaked distribution at 800C. Correlated EBSD and Nanoindentation Cryogenic temperatures are of interest for mate- Mapping rials that are subjected to conditions such as outer space, arctic or winter environments, and part of The most obvious application for such technology cooling systems. A ubiquitous structural alloy, 1018 is mapping of small-scale material interfaces as steel, was studied from room temperature down to they cannot be easily evaluated at the bulk scale. 120C. A hardness map at 0C generated with 1- These include welds, especially microscale ones as High-Throughput Nanoindentation for Statistical and Spatial Property Determination 499 Fig. 3. (a) Stage automation map of the hardness distribution in a 410 laser cladding on 4140 substrate, with a black box denoting the location of the optical micrograph in (b) where a fiduciary marker was drawn around the boundary with FIB. This region was subjected to correlated EBSD, with the boundary map in (c) and the inverse pole figure in (d) and high-speed nanoindentation mapping with the hardness overlaid on the boundary map in (e). lm indent spacing, a peak load of 500 lN, and a High-speed nanoindentation techniques provide Berkovich tip clearly shows the two-phase ferrite an advantage when operating at extreme non- and pearlite microstructure, which is also reflected ambient temperatures through reduced contact by the hardness distribution as seen in Fig. 5. time, which reduces tip wear, and the relative effect Indents into the ferrite phase were done as part of of drift. Tip-sample thermal equilibrium is one of a decreasing temperature sweep, with the resulting the largest challenges to extreme temperature load displacement curves indicating a ductile-to- testing as it produces drift when they are brought brittle transition at  58C. Here, homogeneous into contact. The cryo and high-temperature stage dislocation plasticity at the above temperatures in this article utilizes a multi-element heating gave way to serrated flow, indicating dislocation microchamber that exposes the tip and sample to bursts. Since ferrite is BCC, the Peierls’ barrier is the same environment. However, it has been 46,47 relatively large compared to FCC metals and is thus shown that thermal stabilization in vacuum is reliant on thermal assistance for homogeneous more time consuming and difficult. Reducing the plasticity. This was also reflected in the hardness, typical time in contact from 20 s to 0.2 s reduces the which increased 44% over the tested temperature effect of drift on the measurement by two orders of range. This hardening for decreasing temperatures magnitude. This reduced time in contact also sig- also relates to the ductile to brittle temperature nificantly reduces tip wear, a major issue of high transition. This transition is more obvious when temperature testing. The tip sample contact can looking at the pop-in behavior of the indentation be modeled as a high-pressure diffusion couple in curves. The room temperature behavior is primar- thermodynamic software such as Thermal Calc. ily a smooth curve or with very short pop-ins to Simply reducing the time in contact has a dramatic approximate a predominately smooth curve. As impact on the number of indentations that can be temperatures decrease, the flow becomes more performed with a given tip/sample combination. stochastic with increasing pop-in size. There seems to be a transition between the  15C and  25C Generation and Utilization of Large Data Sets curves. This DBTT is lower than the 5C value In contrast to the heterogeneous samples tested typically reported for Charpy impact testing, which thus far, nanoindentation is often performed on is a much higher strain rate. samples that are relatively homogeneous, such as 500 Hintsala, Hangen, and Stauffer Fig. 4. The region of the SiC fiber-SiC matrix composite tested at 400C is shown by SPM in (a) and the corresponding hardness map is shown in (b). The property distributions at 400C and 800C are shown in (c) for the modulus and (d) for the hardness. foils, thin films on substrates, and the substrates compared to traditional Vickers microhardness 49–51 themselves. Even a layered sample with large dimen- testing. Here, 196 indent grids produced by sions in the sample plane can be considered locally high-speed nanoindentation with 5-mN load and a homogeneous. In these cases, statistics allow for Berkovich indenter are compared against a single generation of large data sets where the precise values Vickers indent in a line scan starting in the weld can be determined, with data histograms allowing for joint and moving progressively through the heat- the identification of statistical outliers. This can be affected zone, using an empirical relationship to compared to the number of tests run in a typical compare hardness. The grid of nanoindentations nanoindentation study, where n £ 10 in many cases. fit into approximately the same area as the single A simple experiment can be done looking at the microhardness test (Fig. 8). In this case, a variety of statistical distribution of hardness and modulus microstructures are encountered, from martensitic comparing vibratory polished (100) aluminum and rich regions near the weld joint progressively into aluminum polished with 600 grit paper. Arrays of bainite and finally ferrite. It can be observed that nine indents, 3 9 3 with 15-um spacing between although the hardness versus distance curves are indents, are placed in a larger 5 9 5 array (Fig. 6). comparable between the average value of the high- The total number of indents is now n = 225 for speed nanoindentation grids and microindentation, each sample, which is a factor of a 20 times larger the grids feature scatter bands due to the varying number of tests than for most nanoindention stud- microstructure. For instance, the spread is ies. The pileup corrected modulus and hardness increased near the weld joint, as the martensite histograms (Fig. 7) show both an increase in hard- clustered into islands around the grain boundaries, ness for the roughened/work-hardened sample and where as bainite and ferrite represent better dis- a corresponding increase in the spread of the data. persed microstructures. In these regions, a Combining both statistical analysis and mapping researcher could consider moving to more of a compared against high-speed nanoindentation for a mapping rather than statistical sampling type of railway weld joint in a railway steel can be experiment. High-Throughput Nanoindentation for Statistical and Spatial Property Determination 501 Fig. 5. The 1018 steel is evaluated at cryogenic temperatures, with a clear difference in hardness between the ferrite and pearlite phases in the (a) indentation map and corresponding hardness distribution (b). The inset in (b) is a gradient force SPM image of the microstructure. Example load–displacement curves for the temperature sweep in (c) and the resulting hardness in (d) show a DBT. Fig. 6. Load-displacement curves for the vibratory polished sample (a) and the sample roughened with 600 grit paper (b) and corresponding micrographs showing a 5 9 5 placement of 3 9 3 grids of indents in relationship to the surface topography. detailed information on small-scale regions of sig- CONCLUSIONS AND OUTLOOK nificant industrial importance that are not easily Overall, high-speed indentation techniques are tested on the bulk scale, such as welds, fine grain relatively underused given they possess many and phase structures, composites and interfaces, potential applications. Property mapping, especially and more. Statistical distributions can be generated when used in conjunction with correlated tech- simultaneously, which offer a variety of useful niques characterizing the structure, provide 502 Hintsala, Hangen, and Stauffer Fig. 7. Histograms for the vibratory polished modulus (a) and hardness (b) for both samples. The hardness comparison shows a much larger spread and increase in the value for the roughened sample in comparison to the vibratory polished (smooth) sample. Fig. 8. (a) Comparison between high-speed nanoindentation hardness grids and Vickers microhardness measurements as a function of distance from the weld boundary. SPM of a corresponding microhardness impression and nanohardness grid (total scan size 45 9 45 lm) is shown in (b), which is repeated for 50 points as shown in the stitched optical image in (c). information. However, these techniques do not behavior based on nanoscale measurements is very replace standard nanoindentation techniques, as attractive and may lead to future breakthroughs in the influence of the indentation size effect from materials design and fine-tuning of their perfor- depth, strain rate sensitivity, and spacing should be mance. As a rapid and highly localized mechanical studied in conjunction. measurement tool, high-speed nanoindentation Looking forward, there are many relatively unex- mapping should play an important role in this plored applications for these techniques. Though regard. there are a multitude of material microstructures to ACKNOWLEDGEMENTS potentially map and model, it is the authors’ opinion that sophisticated analysis of ‘‘big data’’ sets, The authors gratefully acknowledge assistance through techniques like machine learning, repre- from Richard Nay, Jared Risan, Robert Dietrich, sent one the biggest frontiers in materials science. Anqi Qiu, and Benjamin Stadnick with the speci- The theme of bridging length scales and producing men testing and Daniel Sorensen with sample cohesive understanding of bulk mechanical preparation. High-Throughput Nanoindentation for Statistical and Spatial Property Determination 503 23. M. R. Maughan, A. A. Leonard, D. D. Stauffer, D. F. Bahr, OPEN ACCESS Philos. Mag. (2017). https://doi.org/10.1080/14786435.2017. This article is distributed under the terms of the 24. B. N. Lucas, W. C. Oliver, G. M. Pharr, J.-L. Loubet, MRS Creative Commons Attribution 4.0 International Li- Online Proc. Libr. 436 (1996). cense (http://creativecommons.org/licenses/by/4.0/), 25. H. Hertz, Hertz’s Miscellaneous Papers (London, UK: which permits unrestricted use, distribution, and Macmillan, 1896). reproduction in any medium, provided you give 26. K.L. Johnson, Contact Mechanics (Cambridge, UK: Cam- appropriate credit to the original author(s) and the bridge University Press, 1985). 27. K. Durst, B. Backes, and M. Goken, Scripta Mater. 52, source, provide a link to the Creative Commons li- 1093–1097 (2005). cense, and indicate if changes were made. 28. ASTM Standard E384-17, Standard Test Method for Knoop and Vickers Hardness of Materials, (West Conshohocken, FUNDING PA: ASTM International, 2017) https://doi.org/10.1520/ e0384-17. Funding was provided by Bruker Nano Surfaces. 29. DIN ISO 6507-1:2018, Metallic materials—Vickers hard- ness test—Part 1: Test method (Geneva, Switzerland: REFERENCES International Organization for Standardization, 2018). 1. W.C. Oliver and G.M. Pharr, J. Mater. Res. 7, 1564–1583 30. J.E. Jakes, J. Mater. Res. 24, 1016–1031 (2009). (1992). 31. J.E. Jakes and D.S. Stone, Philos. Mag. 91, 1387–1399 2. A.C. Fischer-Cripps, Surf. Coat. Technol. 200, 4153–4165 (2011). (2006). 32. J.M. Wheeler, V. Maier, K. Durst, M. Go ¨ ken, and J. 3. A.A. Volinsky, J.B. Vella, and W.W. Gerberich, Thin Solid Michler, Mater. Sci. Eng., A 585, 108 (2013). Films 429, 201–210 (2003). 33. J. Lu, G. Ravichandran, and W.L. Johnson, Acta Mater. 51, 4. G.M. Pharr, D.S. Harding, and W.C. Oliver, Mechanical 3429–3443 (2003). Properties and Deformation Behavior of Materials Having 34. F.J. Dalla Torre, A. Dubach, M.E. Siegrist, and J.F. Lo ¨ ffler, Ultra-Fine Microstructures, ed. M. Nastasti, D.M. Parkin, Appl. Phys. Lett. 89, 091918 (2006). and H. Gleiter (Dordrecht, the Netherlands: Kluwer Aca- 35. V. Maier, A. Hohenwarter, R. Pippan, and D. Kiener, demic Publishers, 1993), pp. 449–461. Scripta Mater. 106, 42 (2015). 5. W.C. Oliver and G.M. Pharr, J. Mater. Res. 19, 3–20 (2011). 36. T.-S. Jun, D.E.J. Armstrong, and T.B. Britton, J. Alloys 6. N.X. Randall, M. Vandamme, and F.J. Ulm, J. Mater. Res. Compd. 672, 282 (2016). 24, 679–690 (2009). 37. H. Ovri and E.T. Lilleodden, Acta Mater. 89, 88 (2015). 7. G. Constantinides, K.S. Ravi Chandran, F.J. Ulm, and K.J. 38. R. Limbach, B.P. Rodrigues, L. Wondraczek, and J. Non- Van Vliet, Mater. Sci. Eng., A 430, 189–202 (2006). Cryst, Solids 404, 124 (2014). 8. L. Sorelli, G. Constantinides, F.J. Ulm, and F. Toutle- 39. V. Maier-Kiener and K. Durst, JOM 69, 2246–2255 (2017). monde, Cem. Concr. Res. 38, 1447–1456 (2008). 40. S.A.S. Asif, K.J. Wahl, and R.J. Colton, Rev. Sci. Instrum. 9. J.A. Greenwood and J.H. Tripp, Trans. ASME Ser. E J. 70, 2408 (1999). Appl. Mech. 34, 417–420 (1967). 41. X. Li and B. Bhushan, Mater. Charact. 48, 11–36 (2002). 10. W.G. Jiang, J.J. Su, and X.Q. Feng, Eng. Fract. Mech. 75, 42. L.E. Murr, J. Mater. Eng. Perform. 19, 1071–1089 (2010). 4965–4972 (2008). 43. B. Jones and C. Nachtsheim, J. Qual. Technol. 43, 1–15 11. M.S. Bobji and S.K. Biswas, J. Mater. Res. 14, 2259–2268 (2011). (1999). 44. O.L. Warren, S.A. Downs, and T.J. Wyrobek, Z. Metallkd. 12. A.C. Fischer-Cripps, Factors Affecting Nanoindentation 95, 287–296 (2004). Test Data (New York, NY: Springer, 2000), pp. 61–82. 45. U.S. Patent, US 20150185117 A1, Environmental condi- 13. M. Sebastiani, R. Moscatelli, F. Ridi, P. Baglioni, and F. tioning assembly for use in mechanical testing at micron or Carassiti, Mater. Des. 97, 372–380 (2016). nano-scales, Jul 2, 2015. 14. J.L. Cuy, A.B. Mann, K.J. Livi, M.F. Teaford, and T.P. 46. J.M. Wheeler, D.E.J. Armstrong, W. Heinz, and R. Schwaiger, Weihs, Arch. Oral Biol. 47, 281–291 (2002). Curr. Opin. Solid State Mater. Sci. 19, 354–366 (2015). 15. K. Morsi, V.V. Patel, K.S. Moon, and J.E. Garay, J. Mater. 47. S. Korte, R.J. Stearn, J.M. Wheeler, and W.J. Clegg, J. Sci. 43, 4050 (2008). Mater. Res. 27, 167–176 (2012). 16. H. Engqvist and U. Wiklund, Tribol. Lett. 8, 147–152 48. J.M. Wheeler and J. Michler, Rev. Sci. Instrum. 84, 101301 (2000). (2013). 17. H.-W. Chang, M.-X. Zhang, A. Atrens, and H. Huang, J. 49. R.R. Ambriz, D. Chicot, N. Benseddiq, G. Mesmacque, and Alloys Compd. 587, 527–532 (2014). S.D. De la Torre, Eur. J. Mech. A Solids 30, 307–315 18. C. Tromas, M. Arnoux, and X. Milhet, Scripta Mater. 66, (2011). 77–80 (2012). 50. K.H. Chung, W. Lee, J.H. Kim, C. Kim, S.H. Park, D. 19. T. Ogura, S. Hirosawa, and T. Sato, Sci. Technol. Adv. Kwon, and K. Chung, Int. J. Solids Struct. 46, 344–363 Mater. 5, 491–496 (2004). (2009). 20. J. Konnerth, A. Valla, and W. Gindl, Appl. Phys. A Mater. 51. G. Sun, F. Xu, G. Li, X. Huang, and Q. Li, Comput. Mater. Sci. Process. 88, 371–375 (2007). Sci. 85, 347–362 (2014). 21. W.W. Gerberich, N.I. Tymiak, J.C. Grunlan, M.F. Horste- 52. DIN ISO 14577-1:2015, Metallic materials—Instrumented meyer, and M.I. Baskes, J. Appl. Mech. 69, 433 (2002). Indentation Test for Hardness and Materials Parame- 22. W.D. Nix and H. Gao, J. Mech. Phys. Solids 46, 425–441 ters—Part 1: Test Method (Geneva, Switzerland: Interna- (1998). tional Organization for Standardization, 2015).

Journal

JOM – Unpaywall

Published: Feb 14, 2018

There are no references for this article.