Studies on design, fabrication and reliability assessment of embedded passives on a high-density interconnect (HDI) organic substrate using a sequential build-up process

One of the foremost design considerations in microelectronics miniaturization is the use of embedded passives which provide practical solution. In a typical circuit, over 80 percent of the electronic components are passives such as resistors, inductors, and capacitors that could take up to almost 50 percent of the entire printed circuit board area. By integrating passive components within the substrate instead of being on the surface, embedded passives reduce the system real estate, eliminate the need for discrete and assembly, enhance electrical performance and reliability, and potentially reduce the overall cost. Moreover, it is lead free. Even with these advantages, embedded passive technology is at a relatively immature stage and more characterization and optimization are needed for practical applications leading to its commercialization.This paper presents an entire process from design and fabrication to electrical characterization and reliability test of embedded passives on multilayered microvia organic substrate. Two test vehicles focusing on resistors and capacitors have been designed and fabricated. Embedded capacitors in this study are made with polymer/ceramic nanocomposite (BaTiO3) material to take advantage of low processing temperature of polymers and relatively high dielectric constant of ceramics and the values of these capacitors range from 50 pF to 1.5 nF with capacitance per area of approximately 1.5 nF/cm(2). Limited high frequency measurement of these capacitors was performed. Furthermore, reliability assessments of thermal shock and temperature humidity tests based on JEDEC standards were carried out. Resistors used in this work have been of three types: 1) carbon ink based polymer thick film (PTF), 2) resistor foils with known sheet resistivities which are laminated to printed wiring board (PWB) during a sequential build-up (SBU) process and 3) thin-film resistor plating by electroless method. Realization of embedded resistors on conventional board-level high-loss epoxy (similar to 0.015 at 1 GHz) and proposed low-loss BCB dielectric (similar to 0.0008 at > 40 GHz) has been explored in this study. Ni-P and Ni-W-P alloys were plated using conventional electroless plating, and NiCr and NiCrAlSi foils were used for the foil transfer process. For the first time, Benzocyclobutene (BCB) has been proposed as a board level dielectric for advanced System-on-Package (SOP) module primarily due to its attractive low-loss (for RF application) and thin film (for high density wiring) properties.Although embedded passives are more reliable by eliminating solder joint interconnects, they also introduce other concerns such as cracks, delamination and component instability. More layers may be needed to accommodate the embedded passives, and various materials within the substrate may cause significant thermo -mechanical stress due to coefficient of thermal expansion (CTE) mismatch. In this work, numerical models of embedded capacitors have been developed to qualitatively examine the effects of process conditions and electrical performance due to thermo-mechanical deformations.Also, a prototype working product with the board level design including features of embedded resistors and capacitors are underway. Preliminary results of these are presented.

Phase diagram and dielectric studies of binary organic materials

The phase equilibrium studies of organic system, involving resorcinol (R) and p-dimethylaminobenzaldehyde (DMAB), reveal the formation of a 1:1 molecular complex with two eutectics. The heat of mixing, entropy of fusion, roughness parameter, interfacial energy, and the excess thermodynamic functions were calculated based on enthalpy of fusion data determined via differential scanning calorimetric (DSC) method. X-ray powder diffraction studies confirm that the eutectics are not simple mechanical mixture of the components under investigation. The spectroscopic investigations (IR and NMR) suggest the occurrence of hydrogen bonding between the components forming the molecular complex. The dielectric measurements, carried out on hot-pressed addition compound (molecular complex), show higher dielectric constant at 320 K than that of individual components. The microstructural investigations of eutectic and addition compound indicate dendritic and faceted morphological features. (C) 2000 Elsevier Science B.V. All rights reserved.

Casting and heat treatment variables of Al–7Si–Mg alloy

A review of the research work that has been carried out thus far relating the casting and heat treatment variables to the structure and mechanical properties of Al–7Si–Mg (wt-%) is presented here. Although specifications recommend a wide range of magnesium contents and a fairly high content of iron, a narrow range of magnesium contents, closer to either the upper or lower specified limits depending on the properties desired, and a low iron content will have to be maintained to obtain optimum and consistent mechanical properties. A few studies have revealed that the modification of eutectic silicon slightly increases ductility and fracture toughness and also that the effect of modification is predominant at low iron content. Generally, higher solidification rates give superior mechanical properties. Delayed aging (the time elapsed between quenching and artificial aging during precipitation hardening) severely affects the strength of the alloy. The mechanism of delayed aging can be explained on the basis of Pashley's kinetic model. It has been reported that certain trace additions (cadmium, indium, tin, etc.) neutralise the detrimental effect of delayed aging. In particular, it should be noted that delayed aging is not mentioned in any of the specifications. With reference to the mechanism by which trace additions neutralise the detrimental effect of delayed aging, various hypotheses have been postulated, of which impurity–vacancy interaction appears to be the most widely accepted.

Electrochemical preparation of few layer-graphene nanosheets via reduction of oriented exfoliated graphene oxide thin films in acetamide-urea-ammonium nitrate melt under ambient conditions

Electrochemical reduction of exfoliated graphene oxide, prepared from pre-exfoliated graphite, in acetamide-urea-ammonium nitrate ternary eutectic melt results in few layer-graphene thin films. Negatively charged exfoliated graphene oxide is attached to positively charged cystamine monolyer self-assembled on a gold surface. Electrochemical reduction of the oriented graphene oxide film is carried out in a room temperature, ternary molten electrolyte. The reduced film is characterized by atomic force microscopy (AFM), conductive AFM, Fourier-transform infrared spectroscopy and Raman spectroscopy. Ternary eutectic melt is found to be a suitable medium for the regulated reduction of graphene oxide to reduced graphene oxide-based sheets on conducting surfaces. (C) 2010 Elsevier B.V. All rights reserved.

A study of the influence of mischmetal additions to Al-7Si-0.3Mg (LM 25/356) alloy

This article deals with the effect of 0.25-1.5 wt pct mischmetal (MM) addition on the mechanical properties, microstructure, electrical conductivity, and fracture behavior of cast Al-7Si-0.3Mg (LM 25/356) alloy. Modification of eutectic silicon by MM is compared with strontium modification in terms of microstructure, mechanical properties, and fading behavior. Loss of magnesium encountered on holding the molten alloy and its resultant effect on mechanical properties of alloys modified with MM and Sr are compared with those in the unmodified alloy.

The formation of ?-FeSiAl5 and Be---Fe phases in Al---7Si---0.3Mg alloy containing Be

Thermal analysis and interrupted quench experiments have been carried out to study the formation of beta-FeSiAl5 and (Be-Fe)-BeSiFe2Al8 phases in Al-7Si-0.3Mg alloy with and without Be addition. In the base alloy with 0.6% Fe (without Be addition), a needle- and plate-shaped beta-phase is present in the interdendritic regions and is formed by a ternary eutectic reaction. In the Be- added alloy with 0.6% Fe, a Be-Fe phase of Chinese script and polygon shapes grows along with the primary alpha-Al dendrites, leading to superior mechanical properties. It is proposed that this Be-Fe phase is formed by a peritectic reaction. Be addition has also resulted in some grain refinement.

Effect of Constant and Cyclic Current Stressing on the Evolution of Intermetallic Compound Layers

In this paper the effects of constant and cyclic power loads on the evolution of interfacial reaction layers in lead-free solder interconnections are presented. Firstly, the differences in the growth behavior of intermetallic compound (IMC) layers at the cathode and anode sides of the interconnections are rationalized. This is done by considering the changes in the intrinsic fluxes of elements owing to electromigration as well as taking into account the fact that the growth of Cu3Sn and Cu6Sn5 are coupled via interfacial reactions. In this way, better understanding of the effect of electron flux on the growth of each individual layer in the Cu-Sn system can be achieved. Secondly, it is shown that there is a distinct difference between steady-state current stressing (constant current, constant temperature) and power cycling with alternating on- and off-cycle periods (accompanied by a change of temperature). The reasons behind the observed differences are subsequently discussed. Finally, special care is taken to ensure that the current densities are chosen in such a way that there is no risk for even partial melting of the solder interconnections.

Synthesis of nanodispersed phases during rapid solidification and crystallization of glasses in Ti75Ni25 alloys

The formation of the metallic glass and crystalline phases and related microstructures and the decomposition behavior of rapidly solidified Ti75Ni25 alloys obtained under different processing conditions have been investigated in detail. The competition between glass transition and nucleation of beta-Ti during rapid solidification leads to the possibility of synthesizing the nanocomposites of beta-Ti and glass. Additionally, it is shown that the presence of a small amount of Si also promotes simultaneous nucleation of fine Ti2Ni intermetallic compound. Thermodynamic calculation of the metastable phase diagram indicates the presence of a metastable eutectic reaction between alpha-Ti and Ti2Ni. Evidence of this reaction at lower cooling rates has been presented. On heating, the glass decomposes through this reaction. Finally, on the basis of understanding of the microstructural evolution during decomposition, a new approach has been adopted to synthesize a nanodispersed composite of alpha-Ti in the crystalline Ti2Ni matrix with a narrow size distribution by controlling the devitrification heat treatment of the metallic glass.

Morphology and thermal characteristics of nano-sized Pb-Sn inclusions in Al

Nanoembedded aluminum alloys with bimetallic dispersoids of Sn and Pb of compositions Sn-82-Pb-18,Pb- Sn-64-Pb-36, and Sn-54-Pb-46 were synthesized by rapid solidification. The two phases, face-centered-cubic Pb and tetragonal Sn solid-solution, coexist in all the particles. The crystallographic relation between the two phases and the matrix depends upon the solidification pathways adopted by the particles. For Al-(Sn-82-Pb-18), we report a new orientation relation given by [011]Al//[010]Sn and (011)Al//(101)Sn. Pb exhibits a cube-on-cube orientation with Al in few particles, while in others no orientation relationship could be observed. In contrast, Pb in Sn-64-Pb-36 and Sn-54-Pb-46 particles always exhibits cube-on-cube orientation with the matrix. Sn does not show any orientation relationship with Al or Pb in these cases. Differential scanning calorimetry studies revealed melting at eutectic temperature for all compositions, although solidification pathways are different. Attempts were made to correlate these with the melting and heterogeneous nucleation. characteristics.

The effect of mischmetal addition on the structure and mechanical properties of a cast Al-7Si-0.3Mg alloy containing excess iron (up to 0.6 Pct)

The effect of Fe content (0.2 to 0.6 pct) on the microstructure and mechanical properties of a cast Al-7Si-0.3Mg (LM 25/356) alloy has been investigated. Further, 1 pct mischmetal (MM) additions (a mixture of rare-earth (RE) elements) were made to these alloys, and their mechanical properties at room and at elevated temperatures (up to 200 degreesC) were evaluated. A structure-property correlation on this alloy was attempted using optical microstructure analysis, fractographs, X-ray diffraction, energy-dispersive analysis of X-rays (EDX), and quantitative metallography by image analysis. An increase in Fe content increased the volume percentage of Fe-bearing intermetallic compounds (beta and pi phases), contributing to the lower yield strength (YS), ultimate tensile strength (UTS), percentage elongation, and higher hardness. An addition of 1 pct MM to the alloys containing 0.2 and 0.6 pct Fe was found to refine the microstructure; modify the eutectic silicon and La, Ce, and Nd present in the MM; form different intermetallic compounds with Al, Si, Fe, and Mg; and improve the mechanical properties of the alloys both at room and elevated temperatures.

Grain Floatation During Equiaxed Solidification of an Al-Cu Alloy in a Side-Cooled Cavity: Part II-Numerical Studies

In this article, a single-phase, one-domain macroscopic model is developed for studying binary alloy solidification with moving equiaxed solid phase, along with the associated transport phenomena. In this model, issues such as thermosolutal convection, motion of solid phase relative to liquid and viscosity variations of the solid-liquid mixture with solid fraction in the mobile zone are taken into account. Using the model, the associated transport phenomena during solidification of Al-Cu alloys in a rectangular cavity are predicted. The results for temperature variation, segregation patterns, and eutectic fraction distribution are compared with data from in-house experiments. The model predictions compare well with the experimental results. To highlight the influence of solid phase movement on convection and final macrosegregation, the results of the current model are also compared with those obtained from the conventional solidification model with stationary solid phase. By including the independent movement of the solid phase into the fluid transport model, better predictions of macrosegregation, microstructure, and even shrinkage locations were obtained. Mechanical property prediction models based on microstructure will benefit from the improved accuracy of this model.

Morphology and thermal characteristics of nano-sized Pb–Sn inclusions in Al

Nanoembedded aluminum alloys with bimetallic dispersoids of Sn and Pb of compositions Sn82–Pb18, Sn64–Pb36, and Sn54–Pb46 were synthesized by rapid solidification. The two phases, face-centered-cubic Pb and tetragonal Sn solid-solution, coexist in all the particles. The crystallographic relation between the two phases and the matrix depends upon the solidification pathways adopted by the particles. For Al–(Sn82–Pb18), we report a new orientation relation given by [011]Al//[010]Sn and (o11)A1//(101)Sn. Pb exhibits a cube-on-cube orientation with Al in few particles, while in others no orientation relationship could be observed. In contrast, Pb in Sn64–Pb36 and Sn54–Pb46 particles always exhibits cube-on-cube orientation with the matrix. Sn does not show any orientation relationship with Al or Pb in these cases. Differential scanning calorimetry studies revealed melting at eutectic temperature for all compositions, although solidification pathways are different. Attempts were made to correlate these with the melting and heterogeneous nucleation characteristics.

Synthesis and Microstructure of Laser Surface Alloyed Al–Sn–Si Layer on Commercial Aluminum Substrate

A new type of bearing alloy containing ultrafine sized tin and silicon dispersions in aluminum was designed using laser surface alloying and laser remelting techniques. The microstructures of these non-equilibrium processed alloys were studied in detail using scanning and transmission electron microscopy. The microstructures revealed three distinct morphologies of tin particles namely elongated particles co-existing with silicon, globular particles, and very fine particles. Our detailed analyses using cellular growth theories showed that the formation of these globular tin particles was due to the pinching off of the tin rich liquid in the inter-cellular space by the growth of aluminum secondary dendrite arms. Evidence of fine recrystallized aluminum grains at the top layer due to constrained solidification was shown. Thermal analyses suggested that melting of the spherical shaped tin particles was controlled by the binary aluminum-tin eutectic reaction, whereas non-spherical tin particles melted via the tin-silicon eutectic reaction.

Photo-Defined Electrically Assisted Etching Method for Metal Stencil Fabrication

Metal stencils are well known in electronics printing application such as for dispensing solder paste for surface mounting, printing embedded passive elements in multilayer structures, etc. For microprinting applications using stencils, the print quality depends on the smoothness of the stencil aperture and its dimensional accuracy, which in turn are invariably related to the method used to manufacture the stencils. In this paper, fabrication of metal stencils using a photo-defined electrically assisted etching method is described. Apertures in the stencil were made in neutral electrolyte using three different types of impressed current, namely, dc, pulsed dc, and periodic pulse reverse (PPR). Dimensional accuracy and wall smoothness of the etched apertures in each of the current waveforms were compared. Finally, paste transfer efficiency of the stencil obtained using PPR was calculated and compared with those of a laser-cut electropolished stencil. It is observed that the stencil fabricated using current in PPR waveform has better dimensional accuracy and aperture wall smoothness than those obtained with dc and pulsed dc. From the paste transfer efficiency experiment, it is concluded that photo-defined electrically assisted etching method can provide an alternate route for fabrication of metal stencils for future microelectronics printing applications.

Solubilities of Pb+2, Bi+3, Sb+3 and Cu+1 sulphides in (NaCl KCl Na2S) melts

Solubilities of common metal sulfides have been determined in the (NaCl+KCl) eutectic melt with and without Na2S. A novel gas-phase equilibrium technique has been used for PbS, Bi2S3, and So2S3, and an improved liquid phase equilibrium technique for Cu2S, which eliminates the errors due to physical entrapment of the sulfide phase and segregation on quenching, enabling precise measurements to be made. Solubilities in the (NaCl+KCl) eutectic melt were determined as a function of temperature in the rante 700° to 950°C, and were found to be small. The partial molar heats of mixing of the sulfides in the eutectic melt have been calculated from the solubility measurements, to be 13.3, 31.4, 37.1, and 49.0 kcal for PbSs), Sb2S2(l), and Cu2S(s), respectively. Sodium sulfide addition was observed to enhance these solubilities, the effect being largest for Cu2S followed by Sb2S3, Bi2S3, and PbS. This effect is explained qualitatively. It was observed that PbS and Sb2S3 obey Henry's law up to saturation in (NaCl+KCl+Na2S) melts.