Mechanical Characterization. Micromechanical Testing and Modelling

Thursday, May 3, 2001 8:30 AM in Room California

Start Invited? Item
8:30 AM E4/F1-1-1 Analysis of Depth-Sensing Indentation Tests with a Knoop Indenter
A.C. Fischer-Cripps (CSIRO, Australia); L Riester (Oak Ridge National Laboratories); T.J. Bell (CSIRO)
A method of analysing data obtained in a depth-sensing indentation test using a Knoop indenter which takes into account the elastic recovery along the direction of the short axis of the residual impression as the indenter is removed is presented. If elastic recovery is not accounted for, the elastic modulus and hardness are over-estimated by an amount that depends on the ratio of E/H of the specimen material. The new method of analysis expresses the elastic recovery of the short diagonal of the residual impression into an equivalent face angle for one side of the Knoop indenter. Conventional methods of analysis using this corrected angle provide results for modulus and hardness that are consistent with those obtained with other types of indenters.
8:50 AM E4/F1-1-2 Fracture Behavior of Nanocomposite Thin Films by Nanoindentation
A. Karimi, D. Bethmont (Swiss Federal Institute of Technology Lausanne (EPFL), Switzerland); E. Bergmann (Geneva School of Engineering, Switzerland)
Depth sensing nanoindentation measurement methods was used to study mechanical properties and fracture behavior of nanostructured thin films in terms of their microstructures. Two families of hard coatings in dry machining industry, i.e., diamond like carbon (DLC) and nanocomposite TiC@sub x@N@sub 1-x@ deposited onto steel and cemented carbide substrates were considered. Both sharp (pyramidal) and blunt (spherical) indenters were used to create interfacial and through thickness cracks in such films. DLC samples yield to cracking much easier than TiC@sub x@N@sub 1-x@ thin film. Based on the analysis of the energy release in cracking, the fracture toughness of the film was calculated. In addition, from the critical load for crack generation their rupture strength was estimated. The variation of hardness, elastic modulus, and fracture values of a number of DLC and TiC@sub x@N@sub 1-x@ films have been discussed with respect to the structural changes in the films which were investigated by electron and atomic force microscopy. Attempts are focused on modeling the nanoindentation measurements in order to subtract the substrate effects from the measured data for studying the role of film thickness on mechanical properties. The practical limits of the semi-infinite isotropic mechanical concepts for nanocomposite thin films were highlighted.
9:10 AM E4/F1-1-3 Mechanical Properties of Electroplated Multilayers
S.J. Bull, I. Arce-Garcia, S. Roy (University of Newcastle, United Kingdom)
The benefits of the multilayer (superlattice) structure in improving the hardness of vapour deposited coatings is well-known. As the multilayer period is reduced and the number of discrete interfaces increases so the hardness increases if there is a difference in shear modulus between the two layers. Large enhancements in elastic modulus are occasionally observed by these have been attributed to experimental artefacts. This has been clearly demonstrated for single crystal coatings on flat, smooth substrates and polycrystalline coatings on engineering substrates where the roughness is much higher. It has been recently shown that multilayer coatings can be produced by electrodeposition with similar layer periods. These coatings grow in a much more nodular fashion than is usually found for vapour deposited coatings but within individual nodules the layer structure is maintained in a similar fashion. In this study we have used nanoindentation to investigate the hardness and Young's modulus of electrodeposited Cu/Ni multilayers in order to determine if similar property enhancements are observed as the multilayer period is reduced. Enhanced hardening is observed for very low multilayer periods but this effect is small compared to what is usually observed for vapour-deposited coatings. This is due to the nodular microstructure - the effect is only apparent when the deformation associated with an indentation is confined to an individual nodule.
9:30 AM E4/F1-1-4 On Two Indentation Hardness Definitions
Zhiyong Li (School of Aeronautics and Astronautics); Y.-T. Cheng (General Motors R&D Center); S. Chandrasekar (Purdue University); HenryT. Yang (University of California at Santa Barbara)
With the development of indentation technology, the hardness measurement applies not only to metals, but also to many other materials, such as ceramics and polymers. There is, therefore, a need to reexamine the conclusions drawn by Stilwell and Tabor (Proc. Phys. Soc., Vol.78, P.169,1961), and to investigate the difference in hardness values between that under full load and that after complete unloading. In the present study, the difference between the contact radius under load and that after load removal is extensively studied for a wider range of materials using finite element analysis and dimensional analysis for conical indenters. Theoretically, both definitions can be expressed as combinations of mechanical properties of materials. When sinking-in occurs, there is a large difference between the hardness value calculated under full load and that after complete unloading. Fortunately, the Oliver-Pharr (J. Mater. Res. Vol.7, P.1564,1992) procedure can be used to estimate the hardness for the sinking-in cases. When piling-up occurs, the Oliver-Pharr method may cause error. In this study, the measured contact radius obtained from the residual profile, using the top points as the contact boundary, was found to give a good estimate of the real contact radius under load. Therefore, the conventional residual image measurement is suggested to obtain the hardness under full load. It is also interesting that the sinking-in under full load may sometimes change to piling-up after complete unloading, which may be defined as pseudo-piling-up. For the pseudo-piling-up cases studied, the top point can also give a good estimate of the contact radius. A new indentation methodology is proposed to better estimate the values of hardness and elastic modulus. Using NanoIndenter and Atomic Force Microscope (AFM), an indentation experiment was performed on steel to illustrate the difference between the Oliver-Pharr method and the AFM residual observation method.
9:50 AM E4/F1-1-5 Nanoindentation Testing of Carburised and Nitrided Gear Steels
S.J. Bull, A. Oila (University of Newcastle, United Kingdom); P.R. Chalker, P. Beahan (University of Liverpool, United Kingdom)
It is essential that the mechanical properties of gear steels are well known if accurate predictions of gear life and performance are to be made. For instance, to withstand the high contact stresses is a gear-tooth contact steel gears are usually heat treated or thermochemically treated to harden the surface. A knowledge of the variation of hardness with depth below the surface is necessary to ensure that the gears do not fail by plastic overload. The contact stresses also depend on the Young's Modulus of the material. Steel gears are often carburised, quenched and tempered which gives a tempered martensite structure with regions of retained austenite. These regions are sufficiently small (<10 micron) that only nanoindentation can be used to reliably measure their properties in correctly processed gears. However, the conventional analysis method to extract hardness and modulus from nanoindentation data due to Oliver and Pharr tends to greatly overestimate the hardness and modulus due to the effects of pile-up. Small precision gears can be nitrided and the carbonitrides which form in the nitrogen diffusion layer may have a higher hardness and modulus than the surrounding steel but the measurements are also affected by pile-up. In this study a combination of nanoindentation testing and AFM has been used to accurately determine the hardness and elastic modulus of nitrided and carburised steel. The results confirm that retained austenite is soft compared to the carburised steel as expected. A soft surface layer which may represent decarburisation of the gear steel during processing was also observed. The carbonitrides in nitrided steel are harder and stiffer than the surrounding steel but not to the extent previously reported. The white layer of iron nitrides present on the nitrided steel surface has a two layer structure with different mechanical properties in each layer.
10:10 AM E4/F1-1-6 An Improved Technique for Characterization of Mechanical Properties of Thin Films
S.P. Moylan, S. Chandrasekar (Purdue University)
Nano-indentation is a widely used technique for characterizing the Young's modulus and hardness of thin films and coatings that have been deposited onto substrates. It is known that nano-indentation results can be affected by the roughness of the surfaces being indented. Often, this observation appears to have not been given full consideration when studying the properties of thin films. Here, we show that by combining nano-indentation with taper-sectioning it is possible to minimize the role of surface roughness in the analysis of the film properties. The technique has been used to study the mechanical properties of hard and soft films on hard substrates. Application of the technique to study possible anisotropy in film properties is discussed.
10:30 AM E4/F1-1-7 Scratch Test of Graphite Surface
H. Zaidi, M.T. Nguyen, C. Berrier (Université de Poitiers, France)
The aim of this paper is to study the surface properties of graphite by scratch test. The scratch test was performed by diamond tip and also by steel ball with microscratch tester ST-3001 equipped with acoustic emission detector and optical microscopy. The progressive normal load is applied from 5N to 100N. The evolutions of friction force, derivative force and acoustic emission are recorded according to the applied normal load. The critical normal load Pc is determined for both contact (diamond tip/graphite) and for steel ball/graphite.@PARAGRAPH@There are numerous variables that can influence the critical normal load Pc. The principals are the contact geometry of tip: diamond tip or steel ball and the graphite humidity contents. Experimental results showed that the critical normal load Pc can be correlated to the friction coefficient which depends strongly on the tribocontact environment.@paragraph@The scrtch field in graphite will be modeled and the surface damage at critical normal load will be observed and discussed.
10:50 AM E4/F1-1-8 Nanotribology Sudies of Cr, Cr@sub 2@N and CrN Thin Films Using Constant and Ramped Loading Nanoscratch Techniques
G.H. Wei, T.W. Scharf, J.N. Zhou, A. Rar, J.A. Barnard (The University of Alabama)
The nanotribological behavior of dc magnetron sputtered Cr, Cr@sub 2@N, and CrN thin films has been studied using nanoscratch techniques. Constant and ramped load scratches were made using a Nano Indenter II system at various loads (1 mN, 2.5 mN and 5 mN) on 500nm-thick Cr, Cr@sub 2@N, and CrN thin films, respectively. Wear tracks were imaged by scanning electron and atomic force microscopy. The dependence of the displacement, residual wear depth, percent elastic recovery, and friction coefficient on load in constant load and ramped load tests is compared. Under the same (maximum) load, constant load tests exhibit higher displacements, residual depths and friction coefficients but lower percent elastic recoveries. The mechanism responsible for the difference in displacement between the ramped and constant load scratch tests has been analyzed. In addition, by measuring the cross-sectional profile of a constant load wear track by AFM, the hardness of the Cr, Cr@sub 2@N, and CrN thin films was calculated by sclerometry and compared with the result obtained from force-displacement curves using a Hysitron nanomechanical testing system. Adhesion tests are carried out on 30 nm-thick Cr, Cr@sub 2@N and CrN thin films and the critical loads are compared.
11:10 AM E4/F1-1-9 Characterization and Hardness Modelling of Alternate TiN/TiCN Multilayer Cathodic Arc PDV Coating on Tool Steel
E. Bemporad, F. Carassiti (Univ. of Rome - ROMA TRE, Italy); S. De Rossi, C. Pecchio (Istituto Scientifico Breda S.p.A., Italy)
The superficial hardness values of multilayered film wear-resistant coatings on soft substrate (such as tool steel) have to cope with the interaction among different layers and between the composite film and the substrate. The use of proper models take into account these interactions. Present TiCN coatings have in general complex structures, like multilayer or graded TiN / Ti(C@sub x@N@sub 1-x@) coatings to optimize film/substrate and the film/worked material interactions through a control of the internal stress state, the fatigue toughness, the hardness and the superficial composition.The paper presents the results of the investigations on a wear resistant coating made by alternate layers of TiN and Ti(C@sub x@N@sub 1-x@) 0,5µ each, deposited on a S600 tool steel by cathodic arc evaporation using a reactive gaseous mixture of methane and nitrogen. Microstructural characterization were carried out by ball crater method and Optical Microscopy, along with Scanning Electron Microscopy, Energy Dispersive Microanalysis and Image Analysis. Microhardness measurements were elaborated following the Chicot and Lesage volume law of mixture model (derived from Burnett and Rickerby), to evaluate the composite superficial hardness. In order to use this model, properties of each type of layers and of the substrate (Young modulus, hardness at infinitely small load and strain hardening coefficient) have to be used as input data. Properties values were obtained from "ad hoc" samples: single TiN 0,5µ layer film on substrate, single Ti(C@sub x@N@sub 1-x@) 0,5µ layer film on substrate and the uncoated substrate. Young modulus for the TiN is from literature, while Young modulus for Ti(C@sub x@N@sub 1-x@) is estimated by the use of the mixture rule. Thickness, composition profiles and microstructure of each film have been used to qualify data input. Experimental measurements on the composite superficial hardness and model prediction results are presented.
11:30 AM E4/F1-1-10 Elastic Modulus of Very Thin Diamond-Like Carbon Films
J.-W Chung (Korea Institute of Science and Technology (KIST)); K.-R. Lee (Korea Institute of Science and Technology (KIST), Korea); D.-H. Ko (Yonsei University, Korea); J.H. Han (Korea Research Institute of Standard Science); K.Y. Eun (Korea Institute of Science and Technology (KIST), Korea)
Elastic modulus of very thin diamond-like carbon (DLC) film could be measured by recently suggested free hang method. The DLC films were deposited on Si (100) wafer by r.f.-PACVD using C@sub 6@H@sub 6@ or CH@sub 4@ and filtered vacuum arc (FVA) of grphite. The free hang method could be applied to measure the elastic properties of the DLC films of various mechanical properties and of thickness less than 20nm. In contrast to other instrumental measurement method, the present method is not affected by the mechanical property of the substrate because the substrate was removed to obtain sinusoidal free hang of the DLC film. This advantage is more significant when measuring the mechanical property of very thin film deposited on a substrate of large difference in the mechanical properties. The measured biaxial elastic modulus were reasonable values as can be judged from the plain strain modulus of thick film measured by nanoindentation. While the biaxial elastic modulus of the film deposited by FVA is independent of the film thickness, the film deposited by r.f.-PACVD exhibited the decreased elastic modulus with decreasing film thickness when the film is thinner than 500nm. The reason for the different behavior will be discussed in terms of the difference in structural evolution during the initial stage of the film growth.
11:50 AM E4/F1-1-11 Microprobe-Type Measurement of Young's Modulus and the Poisson Coefficient by Means of Joint Depth Sensing Indentation and Acoustic Microscopy
C. Comte, J. von Stebut (INPL-Ecole des Mines, France)
Hyper frequency scanning acoustic microscopy (SAM) and nano-indentation are mechanical microprobe techniques (lateral resolution of less than 1µm) allowing the measurement of the near-surface elastic response of very thin layers. In the first case Young's modulus can be computed from the measurement of the Rayleigh wavelength if the density of the surface layer is assessed independently. In the second case the composite modulus E/(1-v@super 2@) can be assessed from the slope of the macro-elastic unloading plot. @paragraqph@ In the present paper we show that by joining both techniques on the same specimens, E and ? can be deconvoluted. @paragraph@ Effectiveness of the deconvolution method is first tested on two well-known materials. Results on fused silica (E=72.9GPa; v=0.17) as well as bulk aluminium (E=69.6GPa; ?=0.33) are in very good agreement with values commonly found in the literature.@paragraph@ The method is finally applied on AlCuFe alloys presenting similar compositions but with different phases (icosaedral, tetraedral and cubic). This study shows the importance of crystallographic structure on elastic behaviour, especially on Poisson coefficient which ranges from v=0.39 for the icosaedral phase down to v=0.12 for the tetragonal phase.