Materials Science Forum
Vols. 537-538
Vols. 537-538
Materials Science Forum
Vols. 534-536
Vols. 534-536
Materials Science Forum
Vols. 532-533
Vols. 532-533
Materials Science Forum
Vols. 530-531
Vols. 530-531
Materials Science Forum
Vols. 527-529
Vols. 527-529
Materials Science Forum
Vol. 526
Vol. 526
Materials Science Forum
Vols. 524-525
Vols. 524-525
Materials Science Forum
Vols. 522-523
Vols. 522-523
Materials Science Forum
Vols. 519-521
Vols. 519-521
Materials Science Forum
Vol. 518
Vol. 518
Materials Science Forum
Vol. 517
Vol. 517
Materials Science Forum
Vols. 514-516
Vols. 514-516
Materials Science Forum
Vol. 513
Vol. 513
Materials Science Forum Vols. 524-525
Paper Title Page
Abstract: In this paper we will discuss the impact of residual stresses on the reliability of microelectronic
components and the materials used therein. The following issues will be particularly emphasized:
First, the tendency toward delamination and subsequent cracking along interfaces, such as between
silicon dies, organic substrates, glues, and underfill material; second, the fatigue of electrolytically
deposited copper vias within the substrate and FR4 board material; third, the accumulation of
irreversibly accumulated plastic (creep) strain in lead containing as well as leadfree solders; the microstructural
change observed during thermo-mechanical use within the bulk as well as at the interface of
solder interconnects. We will present state-of-the-art numerical techniques that allow to quantify the
development of stresses and strains within the aforementioned materials, mostly by finite element
analysis, as well as the coupling between local stresses and diffusion processes, which is theoretically
based on phase field models. Further emphasis is put on proper knowledge and determination of the
inherent material parameters and how theoretical predictions can be linked to and validated by experimental
observations and facts.
1
Abstract: A rigorous strategy for (X-ray) diffraction stress measurements at fixed
penetration/information depths is described. Thereby errors caused by lack of penetration-depth
control in traditional (X-ray) diffraction (sin2ψ) measurements are annulled. The ranges of
accessible penetration/information depths and experimental aspects are briefly discussed. The power
of the method is illustrated by the analysis of an only small stress gradient in a sputter-deposited
nickel layer.
13
Abstract: Stress gradients have been investigated employing a measurement strategy for diffraction
measurements at constant penetration/information depths. Two examples have been considered: (i)
sputter-deposited copper thin films on silicon wafers and (ii) γ’-Fe4N1-x layers on α-Fe substrates
obtained by gaseous nitriding. In the Cu thin films rather low tensile stresses, increasing in
magnitude with increasing penetration/information depth have been found. An evaluation of the
measured lattice strains has been performed on the basis of the f(ψ) method, where the X-ray elastic
constants (XEC’s) have been calculated as weighed averages of the corresponding Voigt and Reuss
XEC’s and the weighing parameter has been taken as a fitting parameter. This evaluation reveals
that the grain interaction changes with increasing penetration/information depth from near-Reuss
type towards Neerfeld-Hill type. In the γ’-Fe4N1-x layers stress gradients occur due to surface
relaxation near the surface and deeper in the layer due to a nitrogen concentration gradient which is
built up during nitriding. First measurements in a laboratory diffractometer show the effect of
surface relaxation on the stress-depth profile near the surface. As no single-crystal elastic constants
are available for γ’-Fe4N1-x, the mechanical elastic constants have been employed in diffraction
stress analysis. The results indicated that single-crystal elastic anisotropy occurs. From the measured
data also a concentration – depth profile has been deduced.
19
Abstract: A new algorithm is proposed to determine the through-thickness residual stress gradient
by X-ray Diffraction measurements on progressively thinned components. The procedure is based
on a chemical or electrochemical attack of the component surface, which is then measured at each
thinning stage. The simple algorithm provided for by a specific norm has been revised to take into
account the X-ray absorption effects and the conditions of mechanical equilibrium of the
component. The new procedure is illustrated for a typical case of study concerning a shot-peened
metal component.
25
Abstract: The detection of near surface residual stress gradients in real space requires high depth
resolution for any orientation of the diffraction vector with respect to the sample co-ordinate system.
In order to meet this demand, the slits are no longer being fixed in the laboratory co-ordinate system
as in strain scanning experiments but directly coupled with the sample. Hence, the gauge volume
orientation within the sample remains constant and allows performing depth-resolved sin2ψ
measurements in real space. The method’s accuracy is determined by the gauge volume definition,
which is investigated in detail. Apart from the evaluation of the σ(τ) versus σ(z) relation, which is of
fundamental interest in X-ray residual stress gradient analysis, the method will be shown to have a
unique applicability in rather delicate sample geometries such as multilayer systems.
37
Abstract: In this work, the effect of steel grade on the fatigue resistance of deep-rolled
crankshafts is analysed. In the first part of this paper, the mechanisms leading to the increase
of the fatigue resistance brought by the deep rolling treatment, is presented. This reinforcement
is mainly linked to crack arrest due both to a decrease of the in-depth stress concentration
factor and to remaining compressive residual stresses induced by the deep rolling. In a second
part, an analytical model of residual stresses generation by deep-rolling and fatigue is
presented. In this model the low cyclic fatigue behaviour of the steel is taken into account, and
the residual stress stability with bending fatigue cycling can be predicted. After a presentation
of the experimental validation on two different microstructures (baintic and ferrito- perlitic),
this model is used for analysing the main parameters of the deep-rolling process and fatigue
resistance.
45
Abstract: In the work presented here the residual stress states of sintered iron (ASC 100.29) were
studied after mechanical surface treatments. This included the investigation of the stability of compressive
residual stresses of deep rolled sintered iron at different angles relative to the rolling
direction at quasistatic and cyclic bending loading. An increase of the compressive residual stress in
the transversal deep rolling direction at fatigue loading was found and will be discussed in this
presentation.
51
Abstract: Different classes of metallic materials (aluminum alloys, steels, titanium alloys) were
mechanically surface treated by deep rolling and laser shock peening and isothermally fatigued at
elevated temperature under stress control. The fatigue tests were interrupted after different numbers
of cycles for several stress amplitudes and residual stresses and FWHM-values were measured by
X-ray diffraction methods at the surface and as a function of depth. The results summarize the
response of the surface treatment induced residual stress profiles to thermomechanical loading
conditions in the High Cycle Fatigue (HCF)- as well as in the Low Cycle Fatigue (LCF) regime.
The effects of stress amplitude, plastic strain amplitude, temperature and frequency are addressed in
detail and discussed. The results indicate that residual stress relaxation during high temperature
fatigue can be predicted for sufficiently simplified loading conditions and that thermal and
mechanical effects can be separated from each other. A plastic strain based approach appears to be
most suitable to describe residual stress relaxation. Frequency effects were found to be not very
pronounced in the frequency range investigated.
57
Abstract: Deformation by rolling induces in general a strong crystallographic texture, hence an
important material anisotropy. This is a reason why the cross-rolling is sometimes applied in order
to symmetrize the crystallographic texture. Such an operation modifies also residual stresses. The
goal of this study was to characterize residual stress and texture changes during simple and crossrolling
in polycrystalline copper and ferritic steel. The obtained results show that important
modification of the first order residual stresses occurs during cross-rolling, while the level of the
second order ones is approximately constant. Experimental results were analysed using an elastoplastic
deformation model.
63