Novel shear tools for viscoelastic characterization of packaging polymers

doi:10.1016/j.microrel.2006.09.032

Microelectronics Reliability

Volume 47, Issues 2–3, February–March 2007, Pages 240-247

https://doi.org/10.1016/j.microrel.2006.09.032 Get rights and content

Abstract

In order to calculate cure induced stresses, reliable material data is needed over a large modulus range. This paper concerns the derivation of correction equations needed for the interpretation of dynamic mechanical experiments on two different shear tools. For the first shear tool, consisting of an excitator plate sliding between two fixed parallel plates, the effect of the finite clamp stiffness on both the elastic and the viscoelastic modulus components are discussed. It turns out that this shear tool produces reliable results for modulus values in the range from 200 MPa and lower. The second shear tool consists of two parallel leaf springs between which the initially fluid sample is introduced. Clamping of the leaf spring ends and vertical displacement of the center then causes the sample to shear. Exact as well as approximate analytical solutions for the force–displacement relation are derived and are shown to compare well with finite element results. This second shear tool has a typical measuring range for modulus values of 20 MPa and higher.

Introduction

Polymer cracking, interface delamination and thermal fatigue are the major reliability concerns for microelectronic packages, caused mainly by mismatch in thermal expansion coefficients between the constituents. Packaging polymers are usually introduced to improve the thermo-mechanical reliability of the package and/or for environmental protection. However, the curing process of the applied polymers induces residual stress- and strain fields. To investigate the influence of the residual stress- and strain fields on the failure predictions and to optimize the product for reliability, cure-dependent viscoelastic constitutive relations for the applied polymers have recently been developed [1], [2], [3]. These describe the full temperature and conversion dependent viscoelastic behavior.

The experimental investigation of the shear relaxation modulus function, being one of the important model parameter functions, is complicated as at the beginning of cure the polymer material is in the liquid state and remains in a more or less liquid state during a first period of time, up to the so-called gelpoint and then gradually solidifies. This restricts the choice of possible sample geometries and measurement methods. Therefore, two different approaches are generally applied. One approach uses (solidified) molded samples of different degree of cure to be investigated thermal–mechanically. Disadvantage of such an approach is that it is rather time consuming. The second approach, which is not affected by this disadvantage, uses continuous monitoring of the thermal–mechanical properties from fluid to fully cured state, by applying the liquid resin in the (small) gap between parallel plates and then applying Dynamic Mechanical Analysis (DMA) during the curing process. So far, the second method was used with torsional shear deformation or with a double simple shear test set-up. However, the extremely large relative change of the shear relaxation modulus function during cure (from liquid to fully cured), cannot adequately be followed by commercially available DMA test facilities, for small-gap specimen. Therefore, in particular the measurements for larger modulus values are lacking accuracy. With application of a newly developed “Sandwich Beam” specimen, where the liquid resin is applied in the small gap between parallel leaf springs, the accuracy problem is completely solved. The interpretation of measurement data requires the theoretical solution of the viscoelastic behavior of the sandwich beam during DMA. For this an adequate approximation is proposed and verified through FEM viscoelastic simulations.

Section snippets

Simple shear tool

The initially liquid state of the resin requires a shear tool which prevents resin leakage. Initially, a simple shear tool was used for the characterization which consisted of two fixed parallel plates between which a third plate could move up and downwards. The two gaps between the fixed and the moving plates were filled with liquid resin. In this case surface tension prevents the sample from leaking. The middle plate was mounted to the actuator of a DMA facility, whereas the two outer plates

Sandwich beam set-up

The simple shear tool performs well for the lower shear modulus range. For higher modulus levels an alternative tool has to be developed which allows a shear stress to be reached at a much lower sample stiffness values. This cannot be realized by simply reducing the cross-sectional area since then inaccuracies in the sample shape will result in unacceptable errors in the modulus value. Also the ability to widen the gap is limited because of leakage problems. Therefore a completely different

Application to experimental data

In order to find out how the simple shear tool equation (Eq. (7)) and the sandwich beam equation (Eq. (36)) work out in practice we prepared an epoxy system consisting of stoichiometric amounts of novolac epoxy EPN 1180 (Ex Vantico) and bisphenol-A. As a catalyst 0.5 wt% TPP per 100 gram epoxy was used. These components were carefully mixed, degassed and applied in liquid form between the gaps of the SST and the SBT, respectively. The materials were cured to full extent inside the two shear

Conclusions

In this paper, correction equations for the effect of a finite clamp stiffness on the measured mechanical properties were derived. It was shown that the correction equations for the Simple Shear Tool could be used until the sample stiffness becomes comparable to that of the clamps or, more precisely, up to K_tot/K_clamp = 0.5. For higher sample stiffnesses, as occurring when a material solidifies, modulus values as measured with the Simple Shear Tool are no longer reliable since they become too

Acknowledgements

The authors acknowledge the European Community for their financial support (Mevipro project GRD1-2001-40296) and Vantico for the supply of the epoxy resin.

References (5)

Ernst LJ et al. Recent developments in thermal mechanical modeling of the curing process of filled thermoset polymers,...
Jansen KMB et al. Cure-temperature and time dependent constitutive modeling of moulding compounds. In: Proceedings of...

There are more references available in the full text version of this article.

Cited by (9)

Combining oscillatory shear rheometry and dynamic mechanical analysis to obtain wide-frequency master curves
2024, Polymer
In this paper, we present a novel approach combining widely available techniques, oscillatory shear rheometry and dynamic mechanical analysis to obtain wide-frequency range master curves. We carried out finite element analysis to investigate the influence of the specimen thickness and Poisson's ratio on the moduli measured with a shear sandwich clamp in solid-state. We demonstrated that the thickness/height ratio of the specimen greatly influences its shear stress state. Using polystyrene, which is a thermorheologically simple thermoplastic, we showed that continuous master curves can be generated from melt-state torsion and solid-state shear tests. Our results can facilitate the characterisation of polymers from viscous to elastic state and the development of test methods that use more accurate calculations.
Measuring and understanding cure-dependent viscoelastic properties of epoxy resin: A review
2022, Polymer Testing
Citation Excerpt :
The ISO standard recommends an analytical compliance correction method. The usability of shear DMA for the determination of cure-dependent properties was investigated by Van't Hof et al., who compared the measurement range of the described standard shear geometry according to ISO 6721-6 with a self-developed sandwich beam tool (SBT) [143]. After applying individual corrections for instrument compliance, they found that the standard shear geometry is capable of measuring in the rubbery regime but dramatically underestimates the glassy modulus.
The experimental determination of cure-dependent viscoelastic properties is of utmost importance for processing and use of epoxy (EP) resin-based materials and structures. Although this complex topic is frequently addressed in literature and in practice, no standards and few general guidelines for this task are available. Hence, a comprehensive review on rheological experiments and dynamic-mechanical analysis (DMA) including their preparation, conduction and evaluation was conducted. According to the results of our literature research, operators are well advised to rely on holistic test protocols that take care of the strong interaction between the curing process, the realistic device capabilities and the viscoelastic phenomena to be measured. We show that erroneous data or misinterpretations may occur if inappropriate measurement setups are employed. Methods for the prevention of both material- and device-related disturbances are suggested. Finally, remaining experimental challenges and future trends in the context of cure-dependent viscoelastic properties of EP are pointed out.
An investigation on curing rheokinetics of a nadic end-capped MDA-BTDA-ODA copolyimide resin by dynamic mechanical analysis
2019, Polymer Testing
Citation Excerpt :
In addition, the sensitivities of analytical techniques such as DSC (differential scanning calorimetry), FTIR (Fourier transform infrared spectroscopy), 13C-NMR (nuclear magnetic resonance) and SEC (size exclusion chromatography) are insufficient to provide information about the final period of the curing process [5–16]. The evaluation of the curing kinetics of thermoset resins by monitoring their viscoelastic behavior during the network formation process is a more challenging subject [17–22]. However, the kinetics of cross-link formation phenomenon can be examined by dynamic mechanical analysis which is distinguished from other analytical techniques by its sensitivity [23–26].
The curing behavior of a nadic end capped MDA-BTDA-ODA copolyimide resin was investigated using dynamic mechanical analysis in the shear mode. The studied copolyimide consists of 3,3′,4,4′-benzophenonetetracarboxylic acid dianhydride (BTDA) with a 50:50 M combination of 4,4′-diaminodiphenylmethane (MDA) and 3,4′-oxydianiline (ODA) monomers. The effects of curing temperature and time on the rheological properties of the resin were examined. At the beginning of the network formation process, an increase in the storage modulus, loss modulus and complex viscosity was observed. The gelation temperature, T_gel, was determined from the intersection points of the temperature-dependent storage and loss moduli curves. It was found that the rate of gelation of the resin was increased with increasing temperature. In terms of mechanical curing condition, the kinetics of isothermal network formation process at different isothermal curing temperatures were monitored by using Kamal-Sourour model. The model fitted well with experimental data. Therefore, mechanical conversion at different temperatures can be well predicted by the kinetics data.
Kinetic and thermo-viscoelastic characterisation of the epoxy adhesive in GLARE
2015, Composite Structures
Citation Excerpt :
The determination of thermo-viscoelastic response of fully cured composites is nowadays a standard procedure, e.g. [9,10]. The determination of the change in viscoelastic behaviour during cure is possible but much more challenging [11–15]. The need for consideration of the viscoelasticity during cure of epoxy FM-94 is also investigated in this paper.
Epoxy FM-94 has a temperature dependent and viscoelastic response which is used for metal bonding applications and composite laminates like GLARE. A Dynamic Mechanical Analysis is performed on the epoxy FM-94 adhesive. A master curve is produced using the time–temperature superposition principle and is modelled as a Prony series. In addition, the cure kinetics of the adhesive are measured using DSC (Differential Scanning Calorimetry) and fitted to a standard kinetic model. Both the kinetics and the thermo-viscoelastic models can be used in subsequent research to predict the cure induced and residual thermal stresses due to the composite processing steps.
Modeling and characterization of molding compound properties during cure
2009, Microelectronics Reliability
Citation Excerpt :
In that way the full range from fluid to high stiffness solid could be analyzed. Details can be found in [4]. The material injection process was done at 140 °C in a specially designed brass mold, which could be cooled down rapidly directly after injection to prevent cure during the sample preparation step.
During the encapsulation of electronic components stresses are generated due to curing effects and the difference in thermal shrinkage between molding compound and die. These residual stresses add up to the stresses generated during thermal cycling and mechanical loading and may eventually lead to product failure. In this paper we focus on three commercial molding compounds and analyze in detail the increase in elastic modulus and the change in viscoelastic behaviour during cure. This was done with a special shear tool which allows to measure mechanical properties with sufficient accuracy in the liquid as well as in the solid state. The cure dependent viscoelastic material behaviour was modeled using a cure dependent shift factor and rubber modulus.
The viscoelastic behaviour of the molding compounds is also shown not to be stable. During postcure the materials slowly continue to crosslink thereby systematically changing their viscoelastic behaviour. The material models presented here therefore only account for the initial curing stage and do not include postcure.
Prediction models for distortions and residual stresses in thermoset polymer laminates: An overview
2019, Journal of Manufacturing and Materials Processing

View all citing articles on Scopus

View full text