Abstract
Various aluminum foams were fabricated with a structure comparable to the Terrapene carolina (box turtle) shell hierarchy as a synthetic means of attaining the lightweight, yet impact-resistive, nature of the biological counterpart. Each foam was constructed from a single aluminum alloy but with different morphologies and foam densities. By borrowing from the sophistication of biological design, the aluminum foams were shown to exhibit robust mechanical performance. High strain rate experimentation, via split Hopkinson pressure bar, was utilized to reveal the strain rate sensitivity of the foams as well as a metric to compare impact performance. The structure-property relations, necessary for accurate material modeling, were also characterized by way of optical microscopy, scanning electron microscopy, energy dispersive X-ray spectroscopy, and nanoindentation tests. The robust varying mechanical performance was attributed to the biologically inspired materials design.
1 Introduction
Cellular solids, made up of an interconnected network of solid struts or plates that form the edges and faces of cells, have been used by man since antiquity [1]. Cellular materials are quite common in natural biological materials such as wood, cork, coral, honeycomb, tissue, and bone [1–3]. Biological structural materials are especially of interest to engineers and material scientists because of their hierarchical structures as well as mechanical properties superior to man-made counterpart materials [4]. The Terrapene carolina (box turtle) and Dasypus novemcinctus (nine-banded armadillo) shells, for instance, showed superior energy absorption ability and have a sandwich composite structure composed of a lamellar bony exterior and bony network of closed-cell foam interior that is very similar to the structure of the foams being investigated in this work [5–7]. The turtle shell carapace and armadillo front and rear shells behaved similarly to a typical metallic foam material with a stress plateau common among metallic foams loaded in compression [5–7].
Metallic foams have seen implementation in many areas because of their excellent properties valuable to many diverse applications. The large volume of air in foams makes them ideal for applications where acoustic and vibration damping are important. The ability to undergo large compressive deformations has led to applications where mechanical damping is necessary. Foams with controlled pore sizes have also been used as filters in high-temperature gas and fluid applications. Aluminum (Al) foams, especially, have occupied many researchers’ attention because they provide a unique combination of properties such as low density, high stiffness-to-weight ratio, strength and energy absorption, and recyclable and high performance-to-price ratio. Military research has recently given attention to these foams as such characteristics can provide protective solutions to the challenges facing threats and shed light on novel material/armor designs. These properties, coupled with the need for lightweight armor for soldiers and vehicles on the battlefield, have raised the interest in foam composite materials with robust blast characteristics [8]. Another current desirable application, lighter and more efficient yet safer consumer automobiles, has fueled resurgence in the interest of foams for automotive structural components with pronounced dynamic loading characteristics [9, 10]. Based on the vast interest in aluminum foams, an undertaking of novel enhancement of the aluminum foam morphology is shown to be a worthy cause. Also, with regard to mechanical properties, high strain rate experimentation is a key to identifying the robustness of Al foams and providing the necessary information for modeling and optimization of Al foam components.
The strain rate sensitivity of metallic foams has not received much experimental attention thus far. Indeed, most metals experience an increase in flow stress during dynamic loading [11]. Because metallic foams have rate-dependent parent materials, it follows that the foams themselves should exhibit rate-dependent behavior. Previous work by Mukai et al. [12] showed that the closed-cell Al foam with a relative density of approximately 0.1 exhibited a flow stress increase at a strain rate of 2500 s-1 when compared to quasi-static tests. Deshpande and Fleck [13] found an increase in the flow stress of approximately 30% for Al foams at strain rates up to 5000 s-1. Paul and Ramamurty [10] found similar results with strain rates ranging from 3.33×10-5 to 1.6×10-1 s-1. In another study, Mukai et al. [14] found that magnesium-based foam exhibited a substantial increase in plateau stress at strain rates above 800 s-1. In contrast, polymeric foams have been found to show very little strain rate sensitivity [15, 16].
Due to the superior mechanical response of the box turtle shell [5] and the ability to manufacture Al foams with comparative hierarchal structure, investigation into the benefits of these biologically inspired foams, with respect to mechanical response, is a priority. The aim of this study, then, is to investigate the strain rate sensitive mechanical response of different Al foam structures that resemble the structure of the box turtle shell, at high length scales, in an effort to develop biologically inspired structural systems by use of structure-property relations. The increased energy absorption of Al foams at higher strain rates coupled with observations from designs in nature and the structure of the foams being studied led to the present investigation that seeks to develop energy-absorbing components inspired by biological design. This work is also the first step on the road to developing biologically inspired armor systems by the use of high-fidelity physics-based modeling using corresponding experimentation for validation purposes.
2 Materials and methods
2.1 Foam selection and fabrication
To select the Al foams for development of the biologically inspired composite structure, observations of hierarchal structure of the box turtle shell were noted as the basis for its robust mechanical properties. Figure 1A shows that the turtle shell structure is composed of three main layers: two exterior bone layers and an interior foam layer. The marked difference in the bone layers and the foam layer is that the bone layers are highly dense, whereas the foam layer exhibits large pores or “cells”. Selection of foams, such as shown in Figure 1B, arises from controlled cell sizes that are comparable to the turtle shell.
Comparison of hierarchical structures between (A) biological structural material (turtle shell) and (B) synthetic counterpart material (metallic foam).
Using the structural layers as the metric for design of the Al foams, five different types of Al foams obtained from Cymat Technologies, Ltd. (Mississauga, ON, Canada) were selected for the development of the lightweight composite structure. Such foam materials included a natural small cell (NSC), an open one-side small cell (1SSC), a natural large cell (NLC), an open two-side large cell (2SLC), and a sandwich composite enclosed by thicker skins (3D). Morphology and structure comparison of different types of Al foams (e.g., NSC, 1SSC, NLC, 2SLC, 3D) are provided in Figure 2 and Table 1, respectively. The foams differed in their relative densities, cell sizes, and skin configurations. The parent materials for the foams were of the same compositions, only with differences due to the manufacturing methods.
Morphology of different types of aluminum foam materials obtained from a commercial source: (A) a natural small cell (NSC), (B) an open one-side small cell (1SSC), (C) a natural large cell (NLC), (D) an open two-side large cell (2SLC), and (E) a sandwich composite enclosed by thicker skins (3D).
Structure comparison of different types of aluminum foam materials.
| Foam designation | Relative density | Cell structure | Skin configuration |
|---|---|---|---|
| NSC | ∼0.27 | Closed cell | Both sides |
| 1SSC | ∼0.24 | Open cell | One side |
| NLC | ∼0.071 | Closed cell | Both sides |
| 2SLC | ∼0.092 | Open cell | None |
| 3D | ∼0.27 | Closed cell | Surrounding |
The fabrication process for producing such foam materials starts with a metal matrix composite (MMC), an Al alloy to which ceramic particles, in this case silicon carbides (SiC), have been added [17]. Once melted, the MMC is poured into a foaming box. Gas bubbles exiting the immersed rotating impellers in the air injection system are used to form the foam. After ordinary casting, the open-cell foam can be achieved. In order to manufacture closed-cell foam, the foam collects on the surface of the molten material that can be continuously drawn off to form a sheet. The cell size is controlled by the gas flow rate, impeller design, and impeller rotational speed. The rate and means by which the gas is introduced can be varied to produce foams with varying densities. The low-pressure casting process involves injection of foam into a mold. The pressure of injection is controlled so that it is sufficient to fill the mold precisely, but it is not so high that it collapses the unique cell structure of foam. This methodology enables making of foam core parts, thereby reducing the cost and weight of the parts substantially. In some applications, this methodology is able to replace solid cast parts with its foam parts. This process is also ideal for filling steel and aluminum tubes with foam. All of the foams except for the 3D structural foam are manufactured to produce “plates” of foam. The 3D structural foam is essentially a very low-pressure casting that allows for a variety of possible shapes.
2.2 Microstructure analyses
Elemental distribution of obtained samples was investigated under a SUPRA-40 field emission gun (FEG) scanning electron microscope (SEM) made by Zeiss SMT Ltd (Oberkochen, Germany). The SEM was also used to show micrographs of the geometrical structure of the samples. For each observation, the top and side surfaces of the samples were sectioned by using a diamond saw and prepared for microstructure observations. Sectioned specimens were cleaned by an ultrasonic cleaner and then mounted in epoxy. Mounted samples were then mechanically polished and slightly etched with 2% nitric acid.
Optical imaging was performed on a Zeiss optical microscope to quantify the present phases of the material. Image analysis was performed by using a computational code developed at the Center for Advanced Vehicular Systems at Mississippi State University. The code uses grayscale gradients to determine the significant properties such as particle size, nearest-neighbor distance, aspect ratio, and volume fraction. The experimental details of the image analysis procedures used can be found elsewhere [18]. The results of the image analyses of an undeformed specimen, a specimen tested at quasi-static strain rate, and a specimen tested at high strain rate help to establish the mechanisms responsible for failure.
2.3 Mechanical testing
Mechanical tests in this research included nano-indentation tests as well as quasi-static and high strain rate compression tests. Nano-indentation tests were conducted to compare micromechanical properties of various phases found in the samples. These tests were performed with a TriboIndenter®, from Hysitron Inc., (Minneapolis, MN) using a Berkovich-type indenter under load control. The loading cycle included a 20-s loading segment, a 2-s constant-load hold time, and a 20-s unloading segment. The loading and unloading rates of such tests were 450 μN/s. For this particular applied rate, the elastic modulus and hardness at the region of indenting point were determined from the unloading portion of the load-depth curve. The initial unloading elastic contact stiffness, S, is given by
where P is the applied load and h is the indenter displacement into the sample. Such stiffness is obtained from the slope of the initial portion of the unloading curve and is given by
where Er is the reduced modulus and A is the projected contact area of the indenter tip impression, which is a function of the indenter tip geometry as related to the depth of the indent. For a Berkovich tip, the projected contact area in relation to the indentation depth is given by
where hc is the contact depth. The reduced modulus is derived from the sum of the moduli of the indenter material and the sample and is given by
where E and v are the elastic modulus and Poisson’s ratio of the specimen and the indenter, respectively. The hardness has the normal definition and is defined by
where Pmax is the maximum indentation load and A is the resultant projected contact area at that load [19–21]. SEM images were taken and characterized for most of the indents made to obtain additional information regarding the deformation and fracture processes associated with indentation.
Quasi-static compression tests were performed on an Instron (Norwood, MA) 8869 electromechanical test apparatus with a maximum load capacity of 50 kN. All tests were carried out at room temperature with a strain rate of 10-3 s-1. The data from the load cell of the machine and crosshead displacement were used to calculate stress and strain. Specimens were square shaped with length and width of approximately 1 inch and the height equal to the thickness of the parent “plate”. For the large cell foams, a load cell of 10 kN was used, while a 100-N load cell was used for the small-cell foams.
High-rate compression tests were performed using a split Hopkinson pressure bar (SHPB). The SPHB is a common mechanism for the determination of the stress-strain response of materials at high loading rates. The SHPB generally consists of two long bars, called the incident and transmitted bars, used as transducers to record a stress wave propagating along the incident bar, through the specimen, and along the transmitted bar. A stress wave of length equal to twice the striker bar length and magnitude proportional to the striker bar velocity is initiated by propelling the striker bar with compressed gas. As the stress wave is purely elastic in the incident and transmitted bars, a theoretical solution exists for the propagation of the stress wave as well as the dispersion of the wave as it travels through the bars. Once the wave reaches the specimen after traveling the length of the incident bar, it reaches the specimen and a part of the wave is reflected back through the incident bar while the rest is transmitted through the specimen and into the transmitted bar. Strain gages applied to the incident and transmitted bars are used to capture the signals that are amplified and recorded using a computer based data acquisition system. The detailed description of the SHPB technique can be found elsewhere [22, 23]. Figure 3 shows a typical SHPB arrangement [24]. The SHPB used for this study consists of a 1.5-inch diameter Al7075 striker, incident, and output bars of 24, 60, and 48 inches in length, respectively.
A schematic of typical split Hopkinson pressure bar configuration in compression showing bar arrangement, specimen location, and strain gauge setup with an illustration of the bar velocities,
3 Results and discussion
3.1 Microstructure analysis
The chemical compositions of various Al foams used in this research were analyzed by a spectrometer and the results are depicted in Table 2. According to the manufacturer’s specification, the Al foams consist of a 4046 grade Al alloy matrix with SiC reinforcement phase. The actual compositions of various Al foams are chemically comparable and well satisfied the specification except Al and Si. This is attributed to the fact that the chemical analyses were carried out on the overall surface of the sample so that the SiC phase is forced to increase Si and to decrease Al contents. To clarify this, different phases in the samples were observed and analyzed by using an SEM and an energy-dispersive X-ray (EDX) spectroscopy technique, respectively. The parent materials for the foams were of the same compositions. All of the foams considered consisted of an aluminum-silicon-based alloy with silicon carbide reinforcing particles distributed throughout the matrix.
Nominal and actual chemical compositions (wt%) of various Al foam materials.
| Al | Si | Fe | Cu | Mn | Mg | Ti | Zn | Remark | |
|---|---|---|---|---|---|---|---|---|---|
| Specimen | |||||||||
| Matrix | >89 | 8–13 | 0–1.0 | 0–1.0 | 0–0.5 | 0.1–0.8 | 0–0.2 | 0–1.0 | 80–90% of total volume |
| Reinforcement phase | SiC 100% | 10–20% of total volume | |||||||
| NSC | 72.60 | 26.64 | 0.202 | 0.009 | 0.0091 | 0.321 | 0.092 | 0.008 | |
| 1SSC | 74.70 | 24.28 | 0.220 | 0.002 | 0.0160 | 0.322 | 0.090 | 0.042 | |
| NLC | 73.80 | 24.86 | 0.508 | 0.025 | 0.0140 | 0.349 | 0.090 | 0.001 | |
| 2SLC | 76.00 | 20.76 | 0.550 | 0.015 | 0.0130 | 0.391 | 0.083 | 0.001 | |
| 3D | 77.20 | 21.79 | 0.470 | 0.011 | 0.0049 | 0.369 | 0.097 | 0.001 | |
Figure 4 shows the microstructures of various Al foams obtained from the side surface of the samples. All Al foams showed typical eutectic microstructure of aluminum-silicon (Al-Si) alloy with reinforcement SiC particles. One can clearly observe that the microstructure reveals α-Al dendrites, areas with complex eutectics, and well-distributed reinforcement particles. From the chemical analysis and microstructure observation results, it is obvious that these Al foams are chemically comparable and consist of identical microstructure even though their overall shape and densities are qualitatively varied. The reinforcing particles are evident as large dark particles. A silicon-rich eutectic phase is also present as smaller, lighter, more elongated particles. EDX spectroscopy was performed to determine the chemical composition of the phases present in the foam and the results are presented in Table 3. The foams consist of a base Al-Si alloy of approximately 0.9 and 0.1 respective composition by weight. Silicon carbide particles are added for strengthening.
SEM micrographs of various Al foams obtained from the side surface of the samples; (A) a natural small cell (NSC), (B) an open one-side small cell (1SSC), (C) a natural large cell (NLC), (D) an open two-side large cell (2SLC), and (E) a sandwich composite enclosed by thicker skins (3D).
Phase percentages of aluminum, silicon, and carbon for the aluminum foam composite.
| Overall | Al α phase | Eutectic phase | SiC phase | |
|---|---|---|---|---|
| Aluminum | 0.694 | 0.947 | 0.825 | 0.008 |
| Silicon | 0.240 | 0.014 | 0.136 | 0.699 |
| Carbon | 0.053 | 0.026 | 0.028 | 0.292 |
Figure 5 provides a typical scanning electron micrograph of the Al foam showing dendritic structure with well-dispersed reinforcement particles and its EDX element mapping results. Al is surely detected from dendrite and eutectic regions, while Si is detected at particle and eutectic regions. Carbon (C) is detected at particle region, and this clearly reveals that the particles are SiC. Figure 6 shows the volume fractions of each phase performed on 3D sample by in-house image analysis software. SiC reinforcement particles are well distributed within the matrix and its volume fraction is about 11.2% that is well matched with the specification.
A typical SEM micrograph of the Al foam showing dendritic structure with well-dispersed reinforcement particles and its EDX element mapping results: (A) an SEM micrograph, and element mapping results of (B) Al, (C) Si, (D) C, (E) Mg, and (F) overall.
Volume fractions of each phase by image analysis software: (A) SiC particles 11.22%, (B) eutectic phase 22.90%, and (C) α-Al dendrite phase 65.88%.
3.2 Mechanical testing
3.2.1 Quasi-static compression testing
Compression tests were performed at a strain rate of 10-3 s-1 to characterize the quasi-static compressive behavior of the foams. The results are presented as the stress multiplied by the relative density vs. strain to show specific strength characteristics. Figure 7 shows the results for the quasi-static compression testing as stress-strain curves, and the specific energy absorption will be comparatively depicted later in this manuscript with that from high strain rate test results. The quasi-static mechanical response shows that the small-cell foams exhibited higher specific strength than the large cell foams. The initial yield of 3D and the NLC foams are followed by a significant plateau whereas the other foams do not. The initial elastic stiffness of the foams appears to vary as well. This shows that the different cell size and structure provides variation into specific strength, energy absorption, and stiffness.
Quasi-static (0.001 s-1) compressive stress-strain response with compressive stress multiplied by relative density (MPa×ρ/ρ0) plotted against strain. Note that the large-cell foams (i.e., 2SLC and NLC) are plotted against the secondary axis on the right side of the plot.
Images were captured to show the deformation process for the NSC and NLC foams with respect to the change in macroscopic structure. Figure 8 shows the cross section of the NSC and NLC at 0%, 30%, and 60% compressive strain. Notice that at 30% strain the NSC foam remains similar in structure, yet the NLC foam rapidly becomes denser. The images of the NLC foam show significant buckling of the cell walls that leads to densification during the deformation process. This could explain the difference in the specific strength and energy absorption characteristics.
Images captured from deformation of the (A) NSC and (B) NLC foams at 0%, 30%, and 60% compressive strains.
3.2.2 High strain rate compression testing
Figure 9 provides the dynamic compressive stress vs. strain responses for various Al foams, and the comparison of the specific energy absorption up to 20% strain is depicted in Figure 10. This strain level was chosen because significant densification had not occurred in any of the specimens. Significant findings from the tests include the yield stress obtained from the dynamic strain rate is higher than that from quasi-static strain rate. These strength increases with increasing strain rate for these Al foams could be attributed to materials properties as well as to viscous effects caused by gas flow through pores. The energy absorption of all foams increased significantly except for the NSC foam, which remained the same. This is attributed to the foam’s reduced hardening rate exhibited at the high strain rate. As the hardening rate decreases, the energy absorption decreases as well. The hardening rate decrease could be attributed to the embrittlement of the parent material as is typical of aluminum alloys at high strain rates. Although the global strain is low, the structure exhibits local regions of high strain that could lead to local fracturing of the cell walls at earlier strains than under the quasi-static condition. Once cell wall fracture occurs, these open cell walls do not resist deformation in the same manner as closed cell walls. Notice the stress levels for the foams between strains of 0.2 and 0.3 are lower than those for the quasi-static regime.
High strain rate (1000 s-1) compressive stress-strain response with compressive stress multiplied by relative density (MPa×ρ/ρ0) plotted against strain. Note that the large-cell foams (i.e., 2SLC and NLC) are plotted against the secondary axis on the right side of the plot.
Comparison of specific energy absorption between quasi-static and high strain rate tests on various Al foams.
3.2.3 Nanoindentation testing
Figure 11 shows the nano-indentation test results performed on the matrix and SiC particles of various Al foams. The elastic modulus and hardness of the matrix region and the particles incorporated in the various Al foams revealed comparable values. This implies that the fundamental micromechanical properties of tested Al foams were similar. However, the mechanical response previously shown reveals two orders of magnitude difference in the specific strength and energy absorption for the small- and large-cell foams. Therefore, the differences in mechanical response have little to do with the parent material characteristics but rather the hierarchal structure. The structure of these foams exhibit distinct differences in density, porosity, and morphology. These characteristics, then, are the major contributors to the varying mechanical response.
Nano-indentation hardness and elastic modulus results obtained from the matrix (A and B) and particle (C and D) regions incorporated in the various Al foams.
4 Summary and conclusions
Tested foams of Al MMCs of an Al alloy reinforced with SiC particles were fabricated for mimicking the T. carolina (box turtle) shell as a hierarchal biological inspired structure characteristic. Their microstructures consisted of the Al dendrites, Al-Si eutectic, and uniformly dispersed SiC reinforcements. Nano-indentation of the Al foam parent materials revealed comparable intrinsic material properties. However, macroscale observation and SEM micrographs showed significant variation in morphologies of the cell and densities. Elastic modulus, yield stress, and flow stress levels were strongly dependent on the morphologies and densities of the foam materials. Energy absorption abilities of tested Al foams were strongly reliant on physical bulk properties. Mechanical response of the Al foam exhibited strain rate sensitivity. High strain rate experimentation showed that some Al foams, such as the 3D, had much higher specific energy absorption during impact than other foams, such as the NLC foam. This is attributed to the fact that the 3D structure not only provided higher specific stiffness and strength but also possessed thicker cell walls compared to the NLC structure. Comparison research between the biological structural materials and the synthetic material counterpart could provide insight into novel safety systems design.
Acknowledgments
The authors would like to acknowledge the financial supports for this work from the Center for Advanced Vehicular Systems at Mississippi State University under grant CAVS Initiatives 190000-060803-021000 and the U.S. Department of Army (DOD) through grant TCN07173 07121203. They also thank Cymat Technologies, Ltd. for providing samples.
References
[1] Gibson LJ, Ashby MF. Cellular Solids: Structure and Properties, 2nd ed., Cambridge University Press: Cambridge, UK, 1997.Search in Google Scholar
[2] Niklas KJ. Plant Biomechanics: An Engineering Approach to Plant Form and Function, University of Chicago Press: Chicago, IL, 1992.Search in Google Scholar
[3] Cowin SC. Bone Mechanics, CRC Press, Inc.: Boca Raton, FL, 1989.Search in Google Scholar
[4] Meyers MA, Lin AYM, Seki Y, Chen P-Y, Kad BK, Bodde S. JOM 2006, 58, 35–41.10.1007/s11837-006-0138-1Search in Google Scholar
[5] Rhee H, Horstemeyer MF, Hwang Y, Lim H, El Kadiri H, Trim W. Mater. Sci. Eng., C 2009, 29, 2333–2339.10.1016/j.msec.2009.06.002Search in Google Scholar
[6] Rhee H, Horstemeyer MF, Ramsay A. Mater. Sci. Eng., C 2011, 31, 363–369.10.1016/j.msec.2010.10.012Search in Google Scholar
[7] Damiens R, Rhee H, Hwang Y, Park SJ, Hammi Y, Lim H, Horstemeyer MF. J. Mech. Behav. Biomed. Mater. 2012, 6, 106–112.Search in Google Scholar
[8] Gama BA, Bogetti TA, Fink BK, Yu C-J, Claar TD, Eifert HH, Gillespie, Jr JW. Compos. Struct. 2001, 52, 381–395.Search in Google Scholar
[9] Banhart J, Weaire D. Phys. Today 2002, 55, 37–42.10.1063/1.1506749Search in Google Scholar
[10] Paul A, Ramamurty U. Mater. Sci. Eng., A 2000, 281, 1–7.10.1016/S0921-5093(99)00750-9Search in Google Scholar
[11] Meyers, MA. Dynamic Behavior of Materials, John Wiley and Sons, Inc.: New York, NY, 1994.Search in Google Scholar
[12] Mukai T, Kanahashi H, Miyoshi T, Mabuchi M, Nieh TG, Higashi K. Scr. Mater. 1999, 40, 921–927.Search in Google Scholar
[13] Deshpande V, Fleck N. Int. J. Impact Eng. 2000, 24, 277–298.Search in Google Scholar
[14] Mukai T, Kanahashi H, Yamada Y, Shimojima K, Mabuchi M, Nieh TG, Higashi K. Scr. Mater. 1999, 41, 365–371.Search in Google Scholar
[15] Rinde JA, Hoge KG. J. Appl. Polym. Sci. 1971, 15, 1377–1395.Search in Google Scholar
[16] Thomas T, Mahfuz H, Carlsson LA, Kanny K, Jeelani S. Compos. Struct. 2002, 58, 505– 512.Search in Google Scholar
[17] Technical Manual for Cymat SmartMetalTM. Cymat Technologies Ltd.Search in Google Scholar
[18] Allison PG, Horstemeyer MF, Hammi Y, Brown HR, Tucker MT, Hwang Y-K. Mater. Sci. Eng., A 2011, 529, 335–344.10.1016/j.msea.2011.09.037Search in Google Scholar
[19] Oliver WC, Pharr GM. J. Mater. Res. 1992, 7, 1564–1583.Search in Google Scholar
[20] Lucas JP, Rhee H, Guo F, Subramanian KN. J. Electron. Mater. 2003, 32, 1375–1383.Search in Google Scholar
[21] TriboIndenter®User Manual. Hysitron Inc.Search in Google Scholar
[22] Gama B, Lopatnikov S, Gillespie J. Appl. Mech. Rev. 2004, 57, 223–250.Search in Google Scholar
[23] Kuhn H, Medlin D (eds.). Metals Handbook, 8, American Society for Metals: Materials Park, OH, 2000.Search in Google Scholar
[24] Gary G, DAVID Instruction Manual.Search in Google Scholar
©2015 by De Gruyter
This article is distributed under the terms of the Creative Commons Attribution Non-Commercial License, which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
