NANOCOATINGS

Fatigue
&
Residual Stress Distribution
in
Nanostructured Films and Coatings
by
Synchrotron Radiation


 
 

Rutgers, The State University of New Jersey
College of Engineering
Department of Ceramics and Materials Engineering
98 Brett Road, Piscataway, New Jersey 08855-8058
 
 
Diffraction Figure


Energy Dispersive X-ray Diffraction Profiling of Residual Stress Distribution Across Nanostructured Coating/Substrate Interfaces Using Synchrotron Radiation

Research Team
Principal Investigator: Dr. Thomas Tsakalakos, Department of Ceramics and Materials Engineering
Co-Principal Investigator:Dr.  Mark Croft, Department of Physics
Research Investigator:Dr.  Igor Zakharenchenko
Research Student: Haquan Ge
 
 
 

February 6 2000

A. K. Vasudevan
Scientific Officer
Materials Division, Code-332
Office of Naval Research
800 N. Quincy Street
Arlington, VA 22217-5660
 
 

NATO PAPER

XRD Strain and Stress determination in Nanostructured Films and Coatings
 
Thomas Tsakalakos, Department of Ceramic and Materials Engineering
Mark Croft, Department of Physics
Rutgers University
Piscataway, NJ 08854, USA
 

 
1. Abstract
The residual stress determination of nanostructured films and coatings is reviewed. A method of measuring the depth dependence of the stress profile using energy dispersive
x-ray diffraction with a high-energy "white" beam synchrotron radiation source is presented. The profiling is accomplished with the aid of a highly collimated incident and scattered x-ray beams and with micro positioning of the sample-interface. The depth of the profiling is on the order of mm and the resolution of the profiling is on the order of a few microns. The technique allows the three dimensional profiling of the stress distribution in these materials.
 
2. Introduction
X-ray diffraction has of course been the central technique for nearly a century for determining interatomic distances in solids. Applications of this technique to profiling residual strains, as a function of depth from a surface, have also been developed on a more limited basis. Such conventional residual surface strains analysis has utilized the "sin 2? " diffraction method [1], where the varying penetrating power of the x-rays with varying incident angle is exploited. This technique is limited both by the rather broad convolution of scattering depths via the exponential penetration function, and by the limitation of the penetration length to a few microns.
At the nanoscale level Tsakalakos has worked extensively on stress distribution around interfaces of similar or dissimilar multilayered materials using advanced x-ray methods [2,3]. Earlier work [4] on fatigue, corrosion-fatigue and stress-corrosion-cracking demonstrated the ability to measure excess dislocation densities near the surface by x-ray rocking curve analysis. This method, however, failed to quantify the residual stress as a function of distance because it relied on successive removal of surface layers, which resulted in redistribution of the stresses in the remaining material.
In this paper we review some aspects of the energy dispersive diffraction technique as a nondestructive method from which the three dimensional state of stress can be deduced. We also report some new results on WC/Co nanostructured coatings prepared by plasma spraying coating methods.
The expansion of the availability of high-energy synchrotron "white beam" sources (20-100 keV) has led to the expansion of the use of energy dispersive x-ray diffraction. This technique (see Fig. 1. for a schematic) has several unique advantages: these x-rays penetrate deeply; the data collection geometry is extremely stable with stationary incident/diffracted-beam paths; this stability and the high intensity of the synchrotron beam allow extremely high incident/diffracted beam collimation thereby defining very small highly position-specific diffraction volumes; the data collection is very efficient/rapid allowing good signal-to-noise diffraction spectra to be taken quickly on small volumes; and finally, recent advances in precise sample micro positioning allow precision scanning of the diffraction volume laterally and as a function of depth through the sample.
 
It is of great technological importance to be able to predict the failure of materials subjected to static stressing in aggressive environments or dynamic or repeated loading in mechanical and structural applications. This ability is often hindered when inhomogeneous distributions of residual, interfacial or other type of elastic stresses due to the processing or manufacturing conditions, are present. Thus, for deformation processes which are known to be highly sensitive to preexisting large elastic stresses, the evaluation of these inhomogeneous stresses with respect to depth from the surface or interface is equally essential to the comprehensive characterization of deformation induced damages. Because the EDXRD method is nondestructive, it represents an ideal tool with which to elucidate structural changes, and the degradation in mechanical performance such as fatigue failure and fracture.
 
3. Energy dispersive technique

Here we report some preliminary EDXRD measurements on small diffraction volumes in a WC/Co coating on a high-strength-steel base micro-positioned to separately probe the base and coating. These measurements were carried out on Beam Line X17C at the National Synchrotron Light Source at Brookhaven Lab with the indispensable aid of Dr. J. Z. Hu. The enhanced high energy white beam on this line is supplied by a superconducting wiggler insertion device. A schematic of the typical geometry for these preliminary measurements is shown in Fig. 1.
The standard columniation slits on this beam line allow aperturing of the incident beam down to 10 microns. The diffracted beam columniation depends on the variable detector distance from the fixed scattering volume. The existing micro- positioning stage on the line was also utilized. The sample used consisted of a 100 micron WC/Co coating plasma sprayed onto a 2-mm thick steel base.
In Fig. 2, the diffraction pattern for a small volume in the steel is shown along with its bcc indexing. This pattern was collected over just a couple of minutes in the geometry shown in the schematic Fig. 1. This spectrum emphasizes the utility of this hard radiation for collection of high quality diffraction data on small volumes even when buried beneath a high-Z coating.
In Fig. 3, EDXRD pattern for the WC/Co coating is displayed along with the WC indexing. The weal W2C impurity lines labeled are typical for such materials. The W-K and W-L fluorescence lines are also noted.
Figures 4 and 5 demonstrate the applicability of the method in terms of strain resolution, which is about 10-4 and the penetration depth, which reaches 3 mm depth with a diffracting volume size of about 5  mm [1]. The measurements were performed on controlled samples of Al2O3 cylindrical fibers embedded in CP Titanium matrix composite. The solid lines represent the theoretically obtained stress distributions from the interface showing the excellent agreement between theory and experiment.
 
 4. Discussion and Conclusions
In this paper we have reviewed a new technique for measuring residual stresses in nanostructured films and coatings using energy dispersive x-ray diffraction. The scattering volume was scanned through this interface of substrate/coating via micro positioning of the sample. The samples were placed into a geometry that allowed x-ray access in various orientations. Lateral profiling of the strains, approaching the edges, were also possible to be performed to probe for strain relief effects at such surfaces. These experiments were carried out on beam lines at the National Synchrotron Light Source at Brookhaven Lab.
A key factor for the accurate measurement of the residual and interfacial stress distribution is the ability to penetrate the coating and the substrate sufficiently to allow for long range stress relaxation from the surface and interface of the nanostructured coatings.
The energy dependence of the absorption length (1/e decay length) for x-rays in WC/Co is shown in Fig.6. Note that the density of commercial WC/Co of 15 g/cm3 was used. The effective penetration lengths in the 30 -150 keV range are sufficiently large to allow the proposed diffraction profiling. Indeed, as shown in Fig. 6, the penetration depth for WC/Co is in the range of a few mm for the working energy conditions.
The detected volume is determined by the intersection of the collimated beam and the diffracted beam.  Incident beam columniations of down to 5  have been reported as possible both at NSLS [6] and CHESS [7].  Additional incident and diffracted beam aperturing will be used.
The diffraction experiments proposed utilize a constant incident and diffracted beam positions. Thus positioning of the diffraction volume is determined by the translational positioning of the sample. Sample micro positioning is typically available on beamlines (e.g. 2.5  mm step size reported for X23A3 Beamline at NSLS).  For additional required accuracy Newport Corporation produces a wide range precision positioners (e.g. 0.1 mm repeatable positioning with 400 mm travel).
In summary, for nanostructured films and coatings with thickness in the range of 100 mm we can measure the stress distributions with spatial resolution down to 1 mm due to new developments in beam aperturing and micro-position devices. We can measure strains down to 10 -4 with accuracy of about 2 to 5 % (as determined by the peak position described above). Our original control experiments of WC/Co coatings demonstrated the feasibility of the method and revealed some of the expected working conditions for the stress determination. For films of larger thickness, a better resolution can be obtained by increasing the diffracted volume.
In the future, it is hoped that this approach to the problem of the residual stress distribution can eventually transcend the laboratory research stage and be used directly as a diagnostic tool for technological applications. For example, an in-situ fatigue stage development in conjunction with the EDXRD technique could provide invaluable information on the determination of prefracture damage, crack initiation and the remaining lifetime of the materials.
Acknowledgments
The Authors gratefully acknowledge the support of the Office of Naval Research under contract ONR N000149910424- 99PRO4801-00. These measurements were carried out on Beam Line X17C at the National Synchrotron Light Source at Brookhaven Lab with the indispensable aid of Dr. J. Z. Hu.
References
1. Kuntz, T., Wadley. H. and Black D. (1993) Residual strain gradient determination in metal matrix composites by Synchrotron X-ray Energy Dispersive Diffraction Metallurgical Transactions A, 24A, 1117.
2. Lee, J-W., Mayo, W.E. and Tsakalakos, T. (1992) Elastic and plastic contributions to x-ray line broadening of InGaAsP/InP heterostructures, Journal of Electronic Materials, 21, 867-875.
3. Chassapis, C.S. and Tsakalakos, T.(1997) Multidimensional optimization of a stochastic model for X-ray diffraction from superlattices, Computer Physics Communications 99,163-179.
4. Panborn R.N., Yazici, R., Tsakalakos, T., Weissmann, S. and Kramer, I.R. (1980) Determination of prefacture damage in fatigued and stress-corroded materials by x-ray double crystal diffractrometry, National Bureau of Standards Special Publication 567, Proceedings of Symposium on Accuracy in Powder Diffraction held at NBS, Gaiathersburg, MD, 433-49.
5. see numerous references in Advances in X-ray analysis: Volume 39. Proceedings of the Forty-Fourth Annual Conference on Applications of X-Ray Analysis (Col. Springs Col., 1995), Gilfrich, J., Goldsmith, C. Huang, T., Jenkins, R., Predecki, P. K. and Smith, D. (eds.) (Plenum Press, NY, 1997).
6. see Microbeam Diffraction at NSLS (July 97 NSLS  Newsletter).
7. see CHESS B1 Beamline Facilities Description.
 
 
 
 
 
 
 
 

BACK TO HOME PAGE