Resin Bleed Research Articles

Epoxy Molding Compound Bleeding Reduction on Surface Mount Semiconductor Device

Epoxy Molding Compound (EMC) bleed consists of a transparent layer of resin which mainly could occur during molding injection at material packing stage. Resin bleeds on exposed pad of a Surface Mount Device (SMD) can significantly impact the solderability performances of the package, causing failure of the product. De-flashing process after molding is generally performed to eliminate the resin bleed. However, deflashing can be performed on post-plated leadframes without critical drawback, while it is not recommended on pre-plated leadframes due to the high risk of damaging the finishing surface. In this study, an alternative approach aimed to reduce or eliminate molding compound bleed by optimizing the inorganic part of the molding compound formulation is presented. Besides the quality improvement, the presented approach provides an economical advantage since, by tuning the properties of the filler inside the EMC, it is possible to eliminate the de-flashing process from the assembly flow of a package and consequently reduce the manufacturing cost.

A Thermally Enhanced Film Adhesive for Assembling High Power Density Electronic Devices

Assembly adhesives play a crucial role in packaging and thermal management design for microelectronics devices. The main function of the adhesives is attaching multiple components together to deliver designed functionalities and ensure integrity and reliability of the device. As requirements for performance and functionality increase, heat generated from these devices increases exponentially. For example, local heat flux can easily exceed 50 W/cm2 for IGBT power modules while it may reach several hundred W/cm2 at active areas for Gallium Nitride (GaN) power amplifiers. However, intrinsic thermal conductivity of polymer-based adhesives is around 0.2 W/mK, which increasingly becomes a thermal bottleneck in high power density electronics devices, especially those used in defense and aerospace systems. Effectively and safely dissipating the high heat and ensuring device reliability have become one of the critical tasks in the development of modern aerospace and defense systems. This paper presents an adhesive film that uses non-electrically conductive particles with high thermal conductivity as fillers in the adhesive to enhance thermal transport. The adhesive film offers an effective thermal conductivity around 0.7 W/mK, which is more than a 3X improvement over conventional assembly adhesives. In addition, the film adhesive has a long work life at ambient conditions with significantly improved processability and controllability, such as bondline thickness control and uniformity, no voids/air pockets, no or minimum resin bleed, etc. They can also be easily converted into preforms with complicated shapes/geometries with typical tolerances less than 10 mil to facilitate assembly processes and ensure a clean and waste free environment.

Deep Learning-Based Detection of Defects in Wafer Buffer Zone During Semiconductor Packaging Process

Effectively detecting defects in wafer buffer zones is crucial in semiconductor manufacturing processes. If defects are not detected in the middle of the semiconductor manufacturing process, the semiconductor eventually becomes a defective product after all processes are completed. To solve this problem, rule-based vision algorithms have been used to identify defects in the wafer buffer zone. After photographing the wafer buffer zone using a high-speed camera, defects are detected using the pixel values of the images. However, because of the resin bleed in the wafer, which is an epoxy compound, it is difficult to detect defects. Therefore, we introduced a deep learning method for semiconductor inspection and created a high-performance semiconductor inspection algorithm. The defects in the wafer buffer zone should be detected accurately and quickly. Furthermore, the approximate size of the defects must be extracted. We modified the Xception model to fit the wafer data characteristics considering both accuracy and speed. We proposed to extract the size of the defect using class activation mapping (CAM). We obtained an accuracy of 96.9% from the actual wafer dataset through this framework, and then managed to extract the size of the defect through CAM.

Effect of Resin Bleed Out on Compaction Behavior of the Fiber Tow Gap Region during Automated Fiber Placement Manufacturing.

Automated fiber placement is a state-of-the-art manufacturing method which allows for precise control over layup design. However, AFP results in irregular morphology due to fiber tow deposition induced features such as tow gaps and overlaps. Factors such as the squeeze flow and resin bleed out, combined with large non-linear deformation, lead to morphological variability. To understand these complex interacting phenomena, a coupled multiphysics finite element framework was developed to simulate the compaction behavior around fiber tow gap regions, which consists of coupled chemo-rheological and flow-compaction analysis. The compaction analysis incorporated a visco-hyperelastic constitutive model with anisotropic tensorial prepreg viscosity, which depends on the resin degree of cure and local fiber orientation and volume fraction. The proposed methodology was validated using the compaction of unidirectional tows and layup with a fiber tow gap. The proposed approach considered the effect of resin bleed out into the gap region, leading to the formation of a resin-rich pocket with a complex non-uniform morphology.

A multiphase flow approach to gas permeability of composite facesheets during Co-cure of honeycomb core sandwich structures

Co-curing composites facesheets to produce sandwich structures involves heating a prepreg-core assembly in a vacuum bag and compacting it by an external pressure until the prepreg cures and bonds to the core. During this process, vacuum pressure can cause atmospheric air in the honeycomb core to bubble through uncured prepreg facesheets and impact material quality, such as excess resin bleed and void growth in either the facesheet or the bondline. In this paper, we propose a multiphase flow model to predict core gas pressure throughout the process. Fully impregnated prepreg facesheets only become permeable to gas as liquid resin desaturates pores in the fabric weave. To solve for the core gas pressure, resin and gas transport equations are solved simultaneously. The permeability of each phase is a function of saturation–defined by the Brooks-Corey model to be a power law function of the capillary pressure. Deviating from a single phase (gas only) model, using dimensionless analysis we identify four material properties which control the outcome: (i) an intrinsic permeability, K i, which only depends on the fabric architecture and is invariant to all process variables (time, temperature, pressure), (ii) the ratio of fluid viscosities, μ w/ μ g, where resin viscosity varies strongly with temperature and degree of cure, (iii) a bubbling (capillary) pressure, P b, which defines the minimum pressure needed to penetrate the largest pore, and (iv) a pore size distribution parameter, λ, which defines the percentage of pores which may desaturate given the evolving capillary pressure. Changes in observed gas permeability due to time and temperature are explained by the evolving desaturation of facesheet resin and the known relationship to resin viscosity. By applying this model, a reduced set to material characterization is required and complex process cycles can quickly be interrogated. With our experiments we demonstrate that one single value for K i and one single value for λ can be used to predict the gas pressure decay (time dependent) at different temperatures, confirming that the intrinsic permeability is constant for the fiber network and changes in the relative permeability to air are due to the decrease in viscosity only. The limitations of this model are presented and discussed.

Resin bleed improvement on surface mount semiconductor device

Resin bleed is a transparent layer of epoxy compound which occurs during molding process but is difficult to be detected after the molding process. Resin bleed on the lead on the unit from the focused package, SOD123, can cause solderability failure at end customer. This failed unit from the customer will be considered as a customer complaint. Generally, the semiconductor company has to perform visual inspection after the plating process to detect resin bleed. Mold chase with excess hole, split cavity & stepped design ejector pin hole have been found to be the major root cause of resin bleed in this company. The modifications of the mold chase, changing of split cavity to solid cavity and re-design of the ejector pin proposed were derived after a detailed study & analysis conducted to arrive at these solutions. The solutions proposed have yield good results during the pilot run with zero (0) occurrence of resin bleed for 3 consecutive months.

Resin defect detection in appearance lumber using 2D NIR spectroscopy

Resinous defects incorporated into exterior, treated, appearance products, such as barge boards and weatherboards, can cause unsightly resin bleed and resin show-through, which is a discolouration of the paint layer above the resin feature. Manufacturers, therefore, go to great lengths to eliminate resinous wood via manual grading operations and also as part of automated grading systems. Images from resinous and non-resinous wood were acquired using a near-infrared (NIR) hyperspectral camera which produced spectral data for each pixel (0.94 × 1.00 mm). NIR data from these images in the range from 977 to 1565 nm were used to generate a partial least squares (PLS) model that was able to detect resin features reliably. Wavelength regions with peaks at 1180 and 1370 nm showed a distinct difference between non-resinous and resinous wood. Using a multi-step classification process, it was possible to filter out other wood features, such as sapstain or knots that might interfere with the correct identification of resinous areas. It was possible to average the spectral information of the PLS model to 18 wavelengths and the spatial information to 10 mm/data point without significant loss of resin-prediction ability. The optimised, reduced-information model predicted resin features in 24 out of 30 shooks correctly (80%). Solely shooks with small or narrow resin features or minor show-through were predicted incorrectly, i.e. they were typically predicted as rejects even though no resin show-through feature was apparent in the painted shook.

Bleeder Thickness Optimization for Controlling Resin Content in Thick Laminated Composites

Thick laminated composites are manufactured commonly by vacuum bagging of fiber-resin mix or prepregs on a suitable mould and, subsequently curing the lay-up at high temperature and pressure in an either autoclave or oven. At these pressures and temperatures, excess resin bleeds out of the lay-up during the initial stages of the curing. The amount of resin bleed is also a function of the bleeder parameters. Bleeder is a porous fibrous media that is laid around stacked lay-up to provide pathway for volatiles as well as absorb and hold the excess resin. Thicker or highly porous bleeders generally absorb higher amount of resin resulting in a resin starved laminate whereas very thin or denser bleeder leads to resin-rich areas within the laminate. It is thus important to select optimum bleeder parameters in order to achieve a desired resin volume fraction and its uniformity in a composite laminate upon curing. This paper details the simulation of the manufacturing of a thick laminated composite, where a significant amount resin is likely to flow out of a curing lay-up, leading to an optimization of bleeder parameters. A coupled, transient FE analysis is conducted that involves not only the heat transfer, resin flow and cure reaction kinetics simulation but also the simulation of the compaction of the wet laminate and the bleeder layers until the laminate is fully cured. Details of an experiment conducted to find compression characteristics of bleeder of varying thickness and the number of layers and related data that was used in the FE analysis are discussed in this paper. It is found that bleeder thickness significantly affects the amount of resin bleeding out from the curing laminate. As a result, the resin volume fraction of the laminate is affected. Case studies carried out to highlight the optimum bleeder thickness for a lay-up, and the method used to decide the thickness and the number of bleeder layers, are presented.

QFN 패키지의 Resin Bleed와 Melting 검출 알고리즘

There are many different types of surface defects on semiconductor Integrated Chips (IC's) caused by various factors during manufacturing process, such as Scratch, Flash, Resin bleed, and Melting. These defects must be detected and classified by an inspection system for productivity improvement and effective process control. Among defects, in particular, Resin bleed and Melting are the most difficult ones to classify accurately. The brightness value and the shape of Resin bleed and Melting defects are so similar that normally it is difficult to classify the Resin bleed and Melting. In this paper, we propose a segmenting method and a set of features for detecting and classifying the Resin bleed and Melting defects.

A linear finite element model to predict processing-induced distortion in FRP laminates

Abstract Manufacturing processes for laminated composites often produce parts whose dimensions do not match the mold from which they were made. This distortion is commonly referred to as ‘spring-in’. The amount of spring-in can depend on many factors including the manufacturing process (cure temperature, resin bleed, and applied pressure), the part (geometry, material, thickness, cure shrinkage, thermal expansion and layup sequence), and the tool (surface, thickness and thermal expansion). Much of the current work devoted to spring-in relies on extensive resin characterization. While this approach has been reasonably successful, it does little to assist the designer using material systems that have not been fully characterized (which is not always possible or feasible). This study considers the ability of a linear elastic finite element model to describe and quantify many of the factors contributing to spring-in. The aim of this study is to show that spring-in can be accurately predicted without a complete resin characterization. Numerical predictions based on relatively simple mechanical tests were observed to compare favorably with experimental measurements. Spring-in was dominated by thickness shrinkage, which contributed approximately 3/4 of the measured distortion. The mold stretching contribution diminished with thickness and was negligible for parts thicker than 2.5 mm (0.1 in.). While the material system at hand did not exhibit a fiber volume fraction gradient, its effects were included in the formulation of the model. For materials that have reported a gradient, it was found to account for approximately 10% of the part spring-in.

A surface energy approach for analyzing die attach adhesive resin bleed

Die attach adhesive resin bleed is a phenomenon that occurs when the resinous vehicle separates from the polymeric adhesive and migrates over the surface of the substrate beyond its intended boundary. A problem will occur if the material spreads over an area where wirebonding is to be performed. In this paper, a relatively new surface energy approach involving contact angle measurements (three-liquid probe method) was used to determine the mechanism of resin bleed. In addition, a surface profilometer was used to measure the roughness of the epoxy (fiber filled) substrate, and the effect of roughness of the substrate on resin bleed is also discussed. Preliminary results suggest that acid-base interactions play a significant role in resin bleed while no correlation was seen with roughness. The three-liquid probe method appears to hold promise for analysis of resin bleed behavior.

Resin bleed after painting from radiata pine boards treated with tributyltin naphthenate (light organic solvent preservative) or copper, chromium and arsenic compounds (water-borne preservative)

The degree of resin bleed from radiata pine boards was examined following preservation treatment, painting, and heating in an oven at 55°C for eight hours to simulate the effect of solar radiation. Two preservative types were compared: (a) tributyltin naphthenate solution (light organic solvent preservative or LOSP); and (b) copper chromium arsenic oxide solution (water-borne preservative or CCA). The CCA treatment, which involved a subsequent (90°C/60°C) kiln drying cycle, reduced resin bleed while LOSP treatment, after which kiln-drying is not appropriate, increased resin bleed. Spectroscopic analysis showed no evidence of reaction between resin components and CCA or LOSP treatment solutions. An acrylic paint coating system was more effective than an alkyd system in reducing resin bleed on LOSP-treated wood surfaces. Variability of resin bleed was associated with timber source and board sawing history. Both of these factors influenced LOSP uptake levels and undoubtedly contribute to the erratic occurrence of resin bleed in LOSP-treated boards. Control of the residual solvent level prior to exposure to solar radiation or heat may limit the potential for resin bleed after LOSP treatment. Wood of living trees contains the major resin acids (levopimaric, palustric and neoabietic) in approximately equal amounts but only traces of dehydroabietic acid (DHA). Heating of boards to 55°C followed by storage with air access for twelve months increased the proportion of DHA and its oxidation products and decreased the proportions of levopimaric, palustric and neoabietic acids.

Study on the different procedures for debleeding of plastic dip

Inevitable wear on encapsulation molds and small variances in the leadframes cause a small amount of the molding compound or resin to leak out. The leakage results in either excess flash around the body of the plastic package or a thin transparent film on the leadframe and leads, which subsequently affect the solder-dipping that protects the leadframe from corrosion. Thus, effective debleeding procedure must be performed before the solder-dipping process. In this study, nine different flow procedures are designed. The results of the study show that the resin bleed can be more effectively removed by a debleeding procedure carried out immediately after molding rather than post-cure. The wet blasting process used for the debleeding purpose is also found not suitable as it causes side effects to the package surface and leadframe.