Ultrathin wafer level chip size package

The ShellCase wafer-level packaging process uses commercial semiconductor wafer processing equipment. Dies are packaged and encapsulated into separate enclosures while still in wafer form. This wafer level chip size package (WLCSP) process encases the die in a solid die-size glass shell. The glass encapsulation prevents the silicon from being exposed and ensures excellent mechanical and environmental protection. A proprietary compliant polymer layer under the bumps provides on board reliability. Bumps are placed on the individual contact pads, are reflowed, and wafer singulation yields finished packaged devices. This WLCSP fully complies with Joint Electron Device Engineering Council (JEDEC) and surface mount technology (SMT) standards. Such chip scale packages (CSP's) measure 300-700 μm in thickness, a crucial factor for use in various size sensitive electronic products     

Critical issues of wafer level chip scale package (WLCSP) withemphasis on cost analysis and solder joint reliability

Some of the critical issues of wafer level chip scale package (WLCSP) are mentioned and discussed in this investigation. Emphasis is placed on the cost analysis of WLCSP through the, important parameters such as wafer-level redistribution, wafer-bumping, and wafer-level underfilling. Useful and simple equations in terms of these parameters are also provided. Furthermore, the effects of microvia build-up layer on the solder joint reliability of WLCSP on printed circuit board (PCB) through the creep responses such as the deformation, hysteresis loops, and stress and strain are presented. Only solder-bumped with pad-redistribution WLCSPs are considered in this study     

圆片级封装的研究进展

圆片级封装(wafer level package,WLP)因其在形状因数、电性能、低成本等方面的优势,近年来发展迅速。概述了WLP技术近几年的主要发展。首先回顾标准WLP结构,并从焊球结构等方面对其进行了可靠性分析。其次介绍了扩散式WLP工艺以及它的典型应用,并说明了扩散式WLP存在的一些可靠性问题。最后总结了WLP技术结合硅通孔技术(TSV)在三维叠层封装中的应用。     

Effects of build-up printed circuit board thickness on the solderjoint reliability of a wafer level chip scale package (WLCSP)

The creep analyses of solder-bumped wafer level chip scale package (WLCSP) on build-up printed circuit board (PCB) with microvias subjected to thermal cyclic loading are presented. The emphasis of this study is placed on the effects of the thickness of the PCB with a microvia build-up layer on the solder joint reliability of the WLCSP assembly. The 62Sn-2Ag-36Pb solder joints are assumed to follow the Garofalo-Arrhenius creep constitutive law. The shear stress and creep shear strain hysteresis loops, shear stress range, creep shear strain range, and creep strain energy density range at different locations in the corner solder joint are presented for a better understanding of the thermal-mechanical behaviors of the solder-bumped WLCSP on build-up PCB with microvias. It is found that, due to the large coefficient of thermal expansion of the build-up resin, the effects of thickness of the PCB with microvia build-up layer become much more significant than that without the microvia build-up layer     

Ultra CSPTM-a wafer level package

There has been a significant amount of work over the past five years on chip scale packaging. The majority of this work has been an extension of conventional integrated circuit (IC) packaging technology utilizing either wire bonders or tape automated bonding (TAB)-type packaging technology. Handling discrete devices during the IC packaging for these type of chip scale packages (CSPs) has resulted in a relatively high cost for these packages. This paper reports a true wafer level packaging (WLP) technology called the Ultra CSP^TM. One advantage of this WLP concept is that it uses standard IC processing technology for the majority of the package manufacturing. This makes the Ultra CSP ideal for both insertion at the end of the wafer fab as well as the facilitation of wafer level test and burn-in options. This is especially true for dynamic random access memory (DRAM) wafers. Wafer level burn-in and wafer level processing can be used for DRAM and other devices as a way to both reduce cost and improve cycle time. Thermal cycling results for Ultra CSPs with a variety of package sizes and input/output (I/O) counts are presented. These test vehicles, assembled to FR-4 boards without underfill, cover a range of footprints typical of flash memory, DRAM and other devices. The electrical and thermal performance characteristics of the Ultra CSP package technology are discussed     

用于圆片级封装的金凸点研制

介绍了电镀法进行圆片级封装中金凸点制作的工艺流程,并对影响凸点成型的主要工艺因素进行了研究.凸点下金属化层(UBM,under bump metallization)溅射、厚胶光刻和厚金电镀是其中的工艺难点,通过大量的实验研究,确定了TiW/Au的UBM体系,得到了优化的厚胶光刻工艺.同时,研制了用于圆片级封装金凸点制作的垂直喷镀设备,选用不同的电镀液体系和光刻胶体系,对电镀参数进行了控制和研究.对制作的金凸点与国外同类产品的基本特性进行了对比,表明其已经达到可应用水平.     

Computational parameter study of chip scale package array cooling

This paper describes the results of a computational investigation into the thermal management of chip scale package arrays. The parameters considered include power dissipation, cooling air inlet velocity, module spacing, and circuit board conductivity. The geometry used throughout the study was an array of five modules placed on board with forced air cooling along the axis of the array both above and below the circuit board. Each module was the same size and dissipated the same amount of power. Free convection was included with gravity aligned normal to the plane of the circuit board. The effects of thermal radiation were neglected and the flow was considered to be laminar. Three dimensional solutions were generated using the commercial computational fluid dynamics code FLOTHERM. Results are presented in the form of thermal resistances for each package in the array. A number of interesting results were found. For the case of low conductivity circuit boards, the resistance for the first package in the array was a function of inlet velocity only. However, this was not the case when power planes were present and energy was conducted more effectively along the board. For low inlet velocities, when there are strong natural convection effects, there was a temperature overshoot where the highest temperature does not occur in the last package of the array. Finally, when the effects of natural convection were small, the thermal resistance was relatively insensitive to the power dissipation     

Microwave frequency interconnection line model of a wafer level package

In this paper, we introduce the microwave transmission characteristics of interconnection lines on a wafer level package (WLP) and also propose a precise microwave-frequency model of the WLP interconnections. The slow wave factor (SWF) and attenuation constant are measured and discussed. High-frequency measurement is described, based on two-port S-parameter measurements, using an on-wafer microwave probe with a frequency range of up to 5 GHz. The extracted model is represented in the form of distributed lumped circuit model elements and can be easily merged into SPICE simulations. From the extracted model, it was found that line capacitance and inductance per unit length are 0.110 pF/mm and 0.286 nH/mm, respectively. We have successfully applied the extracted model to the design and analysis of a Rambus memory module for time domain simulation and signal integrity simulation. From the simulation, it was found that the WLP has better high-frequency performance, because of its low package inductance, compared with the /spl mu/BGA package, but longer propagation delay, because of the relatively high package capacitance.     

Anand model and FEM analysis of SnAgCuZn lead-free solder joints in wafer level chip scale packaging devices

In this paper, the constitutive response of SnAgCuZn solder was studied under uniaxial tension. Anand model was used to represent the inelastic deformation behavior of the lead-free solders. The material parameters of the constitutive relations for SnAgCuZn solders were determined from separated constitutive relations and experimental results. In addition, the constitutive model was used to analyze the thermally induced inelastic deformation of the solder joints in wafer-level chip scale packages mounted on printed circuit boards based on finite element analysis. And it is found that the addition of Zn can enhance the fatigue life of SnAgCu solder joint.     

Solder joint reliability of a polymer reinforced wafer level package

Wafer level packages (WLPs) have demonstrated a very clear cost-advantage vs traditional wire-bond technologies, especially for small components that have a high number of dice and I/O per wafer. Ultra CSP^® is a WLP developed by the Kulicke & Soffa Flip Chip Division (formally Flip Chip Technologies). Typical products utilizing the Ultra chip scale package (CSP) have 5×5 or less area arrays at 0.5 mm pitch. This relatively small array has been limited by the inherent solder joint reliability of WLPs. A much larger subset of higher I/O IC’s could benefit from WLPs provided that standard reliability requirements are achieved without the use of underfill.A new polymer reinforcement technology, “Polymer Collar WLP™”, has been developed by K&S Flip Chip Division. Polymer Collar WLP utilizes a polymer reinforcement structure surrounding the solder joint and it has demonstrated more than 50% increase in solder joint life in thermal cycling tests. The most attractive feature of the Polymer Collar WLP process is its simplicity. A simple replacement of the standard solder flux with Polymer Collar material during the solder attach process is all that is required. This simplicity makes Polymer Collar the most cost-effective solution for adding a polymer reinforcement structure to the solder joint. Other methods in use today require additional complex and costly manufacturing steps.This Polymer Collar WLP is expected to widen the WLP market to include larger arrays where the Ultra CSP did not have suitable solder joint reliability.     

芯片尺寸封装焊点的可靠性分析与测试

利用ANSYS有限元分析软件,将芯片尺寸封装(CSP)组件简化为了二维模型,并模拟了CSP组件在热循环加栽条件下的应力应变分布;通过模拟发现了组件的结构失效危险点,然后对危险点处的焊点热疲劳寿命进行了预测;最后进行了CSP焊点可靠性测试.结果表明,用薄芯片可提高焊点可靠性.当芯片厚度从0.625 mm减小到0.500 ...     

Effect of parameter variations at chip and wafer level on clockskews

In this paper, a methodology is proposed to determine clock skews and the performance of clock architectures considering parameter variations in an early stage of technology development. With this methodology, it is possible to separate process-induced clock skew from other effects like imperfect loading. Parameter variations are seen as one of the most important effects influencing chip performance in future. By comparing a 0.45- and a 0.25-μm technology, it is shown that in the future, process variations will increase clock skew. The clock skews are determined by measuring the relevant device and metal line parameters as a function of position over chip and wafer. In the past, parameters like IDS, Vth, and resistances could be measured very precisely, although it was difficult to measure low capacitances of single metal lines in the range of femto farad. Thus a new measurement method is used to determine interconnect capacitances extremely precisely. Based on these measurement data, a netlist of a defined clock tree is created by a C-program, and the clock signal delay is simulated. From the delay simulation, we calculate the clock skew for each chip dependent on the parameter variations. Experimental results are separated into a basic random fluctuation part and processing-related contributions on the chip and wafer levels. In addition, the effect of temperature gradients on each chip to the clock skew is simulated. The methodology presented is not restricted to just one clocktree but allows investigation of all kinds of clock distribution circuits. The method has clear advantages with respect to chip area against clocktree realizations on a testchip. No direct and costly measurement of signal delays by voltage contrast methods is required, since all parameters are determined by measurement on the device level     

Finite element analysis for solder ball failures in chip scale package

The failure mechanism of solder ball connect in chip scale package (CSP) utilizing wire-bonded ball grid array was elucidated using finite element analysis in this study. The macro-micro-coupling technique was used in the current model. There exist two factors which contribute to solder ball cracking: shear stress due to thermal expansion mismatch between the package and the PCB and warpage of the package itself. This study revealed that shear stress due to the thermal expansion mismatch prevailed over warpage of the package in causing the solder ball cracking in the present type of CSP.     

Microwave frequency crosstalk model of redistribution line patterns on wafer level package

As the operating frequency of systems increases above the gigahertz frequency range, the electrical performance of a package becomes more critical. Wafer level package (WLP) is a promising solution for future high-speed packaging needs. Because the length of the interconnection lines on the WLP is limited to die size, the WLP has a minimum number of electrical parasitic elements. Because the crosstalk generates significant unwanted noise in nearby lines, causing problems of skew, delay, logic faults, and radiated emission, the crosstalk phenomena is drawing more attention than ever among the electrical characteristics of the WLP. Consequently, the modeling of the crosstalk parameters of the WLP is very important when used in high-speed systems. In this paper, we first report the crosstalk model parameters of the WLP, especially for the redistribution layer. These can be easily embedded into SPICE circuit simulation. The model is represented by the distributed lumped circuit elements, such as the mutual capacitance and the mutual inductance. The crosstalk model was extracted from two-step on-wafer S-parameter measurements and was fitted to the measurements made at up to 5 GHz.     

Board level reliability assessment of chip scale packages

A large program had been initiated to study the board level reliability of various types of chip scale package (CSP). The results on six different packages are reported here, which cover flex interposer CSP, rigid interposer CSP, wafer level assembly CSP, and lead frame CSP. The packages were assembled on FR4 PCBs of two different thicknesses. Temperature cycling tests from −40°C to +125°C with 15 min dwell time at the extremes were conducted to failure for all the package types. The failure criteria were established based on the pattern of electrical resistance change. The cycles to failure were analyzed using Weibull distribution function for each type of package. Selected packages were tested in the temperature/humidity chamber under 85°C/85%RH for 1000 h. Some assembled packages were tested in vibration condition as well. In all these tests, the electrical resistance of each package under testing was monitored continuously. Test samples were also cross-sectioned and analyzed under a Scanning Electronic Microscope (SEM). Different failure mechanisms were identified for various packages. It was noted that some packages failed at the solder joints while others failed inside the package, which was packaging design and process related.     

Failure mechanisms of lead-free chip scale package interconnections under fast mechanical loading

The reliability of chip scale package (CSP) components against mechanical shocks has been studied by employing statistical, fractographic, and microstructural research methods. The components having high tin (Sn0.2Ag0.4Cu) solder bumps were reflow soldered with the Sn3.8Ag0.7Cu (wt.%) solder paste on Ni(P)|Au- and organic solderability preservative (OSP)-coated multilayer printed wiring boards (PWBs), and the assemblies were subjected to the standard drop test procedure. The statistically significant difference in the reliability performance was observed: the components soldered on Cu|OSP were more reliable than those soldered on Ni(P)|Au. Solder interconnections on the Cu|OSP boards failed at the component side, where cracks propagated through the (Cu,Ni)₆Sn₅ reaction layer, whereas interconnections on the Ni(P)|Au boards failed at the PWB side exhibiting the brittle fracture known also as “black pad.” In the first failure mode, which is not normally observed in thermally cycled assemblies, cracks propagate along the intermetallic layers due to the strong strain-rate hardening of the solder interconnections in drop tests. Owing to strain-rate hardening, the stresses in the solder interconnections increase very rapidly in the corner regions of the interconnections above the fracture strength of the ternary (Cu,Ni)₆Sn₅ phase leading to intermetallic fracture. In addition, because of strain-rate hardening, the recrystallization of the as-soldered microstructure is hindered, and therefore the network of grain boundaries is not available in the bulk solder for cracks to propagate, as occurs during thermal cycling. In the black pad failure mode, cracks nucleate and propagate in the porous NiSnP layer between the columnar two-phase (Ni₃P+Sn) layer and the (Cu,Ni)₆Sn₅ intermetallic layer. The fact that the Ni(P)|Au interconnections fail at the PWB side, even though higher stresses are generated on the component side, underlines the brittle nature of the reaction layer.     

Interfacial reaction study on a solder joint with Sn-4Ag-0.5Cu solder ball and Sn-7Zn-Al (30 ppm) solder paste in a lead-free wafer level chip scale package

The interfacial reactions of solder joints between the Sn-4Ag-0.5Cu solder ball and the Sn-7Zn-Al (30 ppm) presoldered paste were investigated in a wafer level chip scale package (WLCSP). After appropriate surface mount technology (SMT) reflow process on the printed circuit board (PCB) with organic solderability preservative (Cu/OSP) and Cu/Ni/Au surface finish, samples were subjected to 150°C high-temperature storage (HTS), 1,000 h aging. Sequentially, the cross-sectional analysis is scrutinized using a scanning electron microscope (SEM)/energy-dispersive spectrometer (EDS) and energy probe microanalysis (EPMA) to observe the metallurgical evolution in the interface and solder buck itself. It was found that Zn-enriched intermetallic compounds (IMCs) without Sn were formed and migrated from the presolder paste region into the solder after reflow and 150°C HTS test.     

Application of the Taguchi method to chip scale package (CSP)design

A three-dimensional (3-D) nonlinear finite element model of an overmolded chip scale package (CSP) on flex-tape carrier has been developed by using ANSYS^TM finite element simulation code. The model has been used to optimize the package for robust design and to determine design rules to keep package warpage within acceptable Joint Electron Device Engineering Council (JEDEC) limits. An L₁₈ Taguchi matrix has been developed to investigate the effect of die thickness and die size, mold compound material and thickness, flex-tape thickness, die attach epoxy and copper trace thicknesses, and solder bail collapsed stand-off height on the reliability of the package during temperature cycling. For package failures, simulations performed represent temperature cycling 125°C to -40°C. This condition is approximated by cooling the package which is mounted on a multilayer printed circuit board (PCB) from 125°C to -40°C. For solder ball coplanarity analysis, simulations have been performed without the PCB and the lowest temperature of the cycle is changed to 25°C. Predicted results indicate that for an optimum design, that is low stress in the package and low package warpage, the package should have smaller die with thicker overmold. In addition to the optimization analysis, plastic strain distribution on each solder ball has been determined to predict the location of solder ball with the highest strain level. The results indicate that the highest strain levels are attained in solder balls located at the edge of the die. The strain levels could then be used to predict the fatigue life of individual solder balls     

Guidelines to select underfills for flip chip on board assemblies and compliant interposers for chip scale package assemblies

The effect of thermomechanical properties of underfill and compliant interposer materials, such as coefficient of thermal expansion (CTE) and stiffness (Young's modulus) on reliability of flip chip on board (FCOB) and chip scale packages (CSPs) under thermal cycling stresses is investigated in this study. Quasi-three-dimensional viscoplastic stress analysis using finite element modeling (FEM) is combined with an energy partitioning (EP) model for creep-fatigue damage accumulation to predict the fatigue durability for a given thermal cycle. Parametric FEM simulations are performed for five different CTEs and five different stiffnesses of the underfill and compliant interposer materials. The creep work dissipation due to thermal cycling is estimated with quasi 3-D model, while 3-D model is used to estimate the hydrostatic stresses. To minimize the computational effort, the 3-D analysis is conducted only for the extreme values of the two parameters (CTE and stiffness) and the results are interpolated for intermediate values. The results show that the stiffness of the underfill material as well as the CTE play important role in influencing the fatigue life of FCOB assemblies. The fatigue durability increases as underfill stiffness and CTE increase. In the case of compliant interposers, the reverse is true and durability increases as interposer stiffness decreases. Furthermore, the interposer CTE affects the fatigue durability more significantly than underfill CTE, with durability increasing as CTE decreases. The eventual goal is to define the optimum design parameters of the FCOB underfill and CSP interposer, in order to maximize the fatigue endurance of the solder joints under cyclic thermal loading environments.     

Wafer level packaging of a tape flip-chip chip scale packages

The advent of chip scale packages (CSPs) within the semiconductor community has led to the development of wafer scale assembly (WSA) or wafer level packaging (WLP) manufacturing in order to raise assembly efficiencies and lower operating costs. Texas Instruments (TI) has developed a unique WLP process for forming flip-chip, ball grid array packages. The die inputs and outputs of the TI CSP are connected through solder bumps to a polyimide film interposer. Solder balls on the other side of the interposer complete the electrical connection to a customer’s printed circuit board. A wafer-sized array of interposers designed to match the pattern of dies on a wafer is aligned and reflowed to a bumped wafer. The TI WLP process is completed by singulating the CSPs from the wafer using standard wafer saw equipment.Attachment of the interposer to the die as well as applying the die and board level solder bumps are carried out in wafer form using a new bumping technology called Tacky Dots™. Tacky Dots uses an array of sticky dots formed in a photosensitive coating laminated to a polyimide film for transferring and attaching solder spheres to semiconductor substrates. A populated film containing one solder sphere per Tacky Dot is positioned over the wafer or interposer and lowered until the spheres contact the pads. A reflow process transfers the spheres from the film to the wafer or interposer and the film is removed once the spheres have frozen.This paper illustrates the process steps and custom equipment developed for forming the TI CSP. The strategic use of finite element modeling for optimizing the design of the package is outlined. The paper concludes by summarizing the current package level reliability results.