In the 1990's, silicon integrated circuit technology will continue to rapidly advance from VLSI towards ULSI. For example, 64 × 106 circuit microprocessor chips, 106 gate CMOS logic chips, and 105 circuit bipolar logic chips, and beyond, will make up the electronics systems of the 1990's. To support these silicon technology advances in density, performance, and cost reduction, packaging technology will also have to advance rapidly to meet packaged electronics systems requirements. Packaging solutions will be required for machines ranging from supercomputers and mainframes to minis and workstations, and for consumer electronics. Silicon technology advances will continue to enable the unique rates of change in performance, function, and cost per function that characterize the information technology industry, including the computer, communications, and consumer electronics industries.

Advances in semiconductor memory storage technology are making possible the development of high density memory cards for hand held and portable product applications. This in turn is driving development of ever denser, ultra-thin component packages. Both materials and package design must be more robust in order for next generation packages to withstand surface mount reflow soldering stresses. Trade-offs among ercapsulant properties may be possible in view of new interconnect, die surface protection and leadframe design technologies which are being adopted for high density packaging. Materials and packaging engineers are challenged to capitalize on opportunities offered by these changes in order to optimize strength and moisture resistarnce in developing encapsulating materials for the new generation of packages.

Multi-chip packages (MCP) are analyzed for high density CMOS packaging using package system simulation, a new systematic methodology for design and trade-off studies of electronic packages. Technology parameters for a representative CMOS chip, multi-chip package, and printed wiring board technology are defined. The effect of these technology parameters on the characteristics of the MCP are examined. The MCP's are optimized using simulated annealing. A large set of optimum MCP's has been found. Based on this set, MCP technology parameters are ranked and their implications on future high density MCP's discussed.

This paper introduces the Polylmide-Ceramic substrate for NEC SX Supercomputers. In case of high performance system such as supercomputers and top end machines in general purpose computer, sophisticated packaging technologies are essential to achieve fastest operations as well as to use highestspeed, highly integrated LSIs.

Wiring substrate which mounts and interconnects LSIs is the key to back up LSI's higher logical-operations.

The high speed interconnection wirings and high density LSI mounting are requested for substrate.

The Polyimide-Ceramic substrate had been developed to meet these demands and have many features of high density thin film wiring, high power supply, high thermal conductivity and huge number of I/Os, in addition to high speed wiring.

25μm wide 75μm center-to-center spacing, two signal layers, 6ns/m signal transmissions, 2.5W/cm2 high power density, 2177 I/Os on a 100mm square substrate have been achieved by using this super substrate technology.

The packaging hierarchy, the first level packaging of TAB LSI, the second level of multi-chip packaging by using Polyimide-Ceramic substrate and liquid cooling module, and the third level of board assemblies are introduced.

Glass-ceramics appear to constitute a new generation in fine ceramic materials for microelectronic packaging. The distinction between glass-ceramics derived via internal nucleation and crystallization of bulk glass and those formed by sintering and surface devitrification of glass particulates is discussed.

The interconnection of many processors in order to increase the computing power of the ensemble is a growing theme in the data processing industry [1]. These processors may be within modules on a common board, on seperate boards within a common frame, or in seperate frames distributed in a room or building. A key element of this “computer complex” is the network which allows efficient data transfers. These data processing networks are differentiated from data communications networks by a demand for fast data transfer, so that the processors and memory at the nodes can interact in times measured in about 1–1000 machine cycles. Machine cycle times might be from tens to hundreds of nanoseconds, and the amount of data transferred might be measured in Kilobytes [2]. This implies both a fast and a high bandwidth technology for implementing these interconnections, as well as a limited link distance. The VLSI IC technology and associated dense electrical chip packaging which has made such powerful computing nodes possible, also implies a requirement for dense packaging of the optical link adapter, for compatability [3]. The multiprocessor complex has very high requirements on reliability, leading to low tolerance for component failure or erroneous data transmission. Eight year life, with failure rates less than 0.01%/Khr, and link bit error rates less than 1 in 1015 are not uncommon requirements [4]. The performance and cost of these networks are often determined by the wiring technology chosen, specifically the electronics and opto-electronics of the interfaces at the network's nodes. Optical interconnections are potentially attractive for this application, but the requirements on optical, electrical and associated packaging technologies are significantly different than the technology which has been developed for the data communications applications. [4, 5].

Femtosecond time domain measurements can provide a powerful approach for characterizing both the nonlinear as well as linear optical properties of waveguide devices. We describe a new time division interferometry technique to perform direct measurements of the nonresonant nonlinear index of refraction in AIGaAs waveguides. In addition, femtosecond time of flight measurements permit linear properties such as absorption and dispersion to be characterized using time domain techniques. These techniques permit a comprehensive characterization of linear and nonlinear waveguide properties.

Fiber optics has had a major impact on the technology of the telecommunications industry but has made little impact on the computer industry. One reason for this is the incompatability of present day optoelectronic component packaging technology with the needs of the computer industry. This problem is most evident in high-performance computer packaging.

Recent progress in InP-based photonic integrated circuits (PIC's) is reviewed, with an emphasis on the new level of design complexity and fabrication flexibility afforded by advances in metal-organic-vapor-phase-epitaxy(MOVPE) growth techniques.

A novel photonic interconnect architecture is proposed which can provide extremely high dimensionality. The proposed architecture, which resembles a collapsed network, avoids the difficulties associated with the use of optical crosspoints. This is accomplished by providing a dedicated path for all input- and output-port connections on a common transmission medium. This eliminates the restrictions imposed by 2×2 switching elements in classical space-division switching architectures.

A widely used technique to metallize substrates for electronic packaging applications is thick film technology. The interaction between commercial thick film materials and aluminum nitride (AlN) has been undertaken and an in depth investigation of the glass/ceramic reactions which occur in frit bonded films. The experimental results have been correlated with thermodynamic predictions of the reaction processes. The standard glasses were found to react with the substrate causing blistering, foaming or dewetting. Following from these studies a model glass system was developed and reactions of this glass with AlN have been investigated using scanning electron microscopical techniques and hot stage microscopy. Thick film inks have been developed using this glass system which are compatible with AlN substrates.

Photosensitive polyimides are used as insulation and protection layers for microelectronics. They can easily give fine-patterned films with excellent characteristics of polyimides by photolithographic procedure.

Photosensitive groups such as double bonds, azides, o-nitrobenzyl and o-naphthoquinonediazides. These photosensitive groups are incorporated to polymer chains through covalent bonds or acid-base ion bonds. Some polyimides have photosensitivity even without intentionally introduced photosensitive groups. Most of photosensitive polyimides are negative working, and a few of them are positive working.

Characteristics of photosensitive polyimides are determined by two factors, the way of introducing photosensitive groups and the structures of polyimide backbone chains. Photosensitivity, resolution, purity and easiness of imidization mainly depends on the former factor. The film properties after curing are mainly determined by the latter. The film properties, however, are affected by the former when imide cyclization is imperfect.

Photosensitive polyimides are widely used as protection and insulation layers of VLSI, multi-chip modules for computers, telecommunication, linephotosensors, thermal-heads, etc.

The role of organic materials in electronic packages is constantly being expanded. A fairly recent application is the use of these materials for encapsulating high lead count IC dice, like TAB chips that are bonded directly to the board or housed in plastic packages. Since these packages are non-hermetic, some special constraints are placed in the selection of these die coat materials. Reliability issues that have to be addressed include compatibility with the underlying dielectric medium, chemical resistance, low moisture absorption, corrosion protection and resistance to thermal stresses. This paper discusses some of these concerns and reviews the suitability of different categories of die coat materials for this application.

Teflon® AF is a family of amorphous fluoropolymers with glass transition temperatures as high as 300° based on bis-2,2-trfluoromethyl-4,5-difluoro-1,3-dioxole, (I), which has unusual properties [1–3] (Figure 1). The family retains the superior electrical, chemical resistance and thermal properties associated with fluoropolymers. In addition the polymers have high optical clarity, limited solubility in some commercially available perfluorinated ethers such as Fluorinert® FC-75 and improved physical properties below their glass transition temperatures (Figure 2). Teflon® AF polymers may be either solution cast into clear micron thin films or melt processed into a variety of forms.

Polyimides are finding increased use as dielectric materials in multi-level metallization technology, which is the key to high-density packaging applications for microelectronics.

Newly developed photosensitive polyimidesiloxane (SIM2000XL-RTS) has been evaluated in terms of thermal stability, photosensitivity and lithography as well as dielectric breakdown properties. The sensitivity to N2 laser exposure has been measured under optimized process conditions. We also find that SIM2000XL-RTS can be processed up to 350°C without significant decomposition. Micron-scale contact images can be successfully patterned by both N2 laser and conventional UV lamp exposure. The dielectrical breakdown field strength of SIM2000XL-RTS is about 4 MV/cm by sandwich measurements.

Currently polyimides are used for a multitude of applications in the electronics field. No where else is it's use growing faster than in the multi-chip module market. It's inherent features in process ability, planarity, heat resistance, and patternability make it ideal for production use in multi chip modules.

We have investigated new polymers of poly(phenylene vinylene), PPV and poly(phenylene), PP. Like polyimides, these polymers can be spin-coated from soluble intermediates and converted to their thermally stable forms by elimination of small groups during a heating cycle. As such, these polymers are insulating. Both poly(phenylene vinylene) and poly(phenylene) also have the potential for metallic conductivity when properly doped. During curing, the complete removal of the leaving groups is important in determining the final behavior of the aromatized polymer film. Elimination of the leaving groups was followed by means of Rutherford Backscattering (RBS) and Forward Recoil Spectrometry (FRES) analysis for various curing conditions and environments. We will describe the curing and film-forming properties of these polymers, and discuss both their dielectric characteristics and thermal stability as a function of the curing conditions.

The fracture behavior of metal-polymer line structures as a function of dimensions was investigated using a stretch-deformation technique. The effects of line orientation, line width and film thickness are reported in this paper. When the line orientation is parallel to the stretching direction, only formation of cracks normal to the lines is observed. However, when the line is perpendicular to the stretching direction, delamination becomes the dominant mode of fracture. Wide lines (16μm) exhibit larger shear stress at the edge of the metal-polymer interface, thus delaminate earlier than narrow lines (4μm). By decreasing the metal film thickness, the depth of stress penetration at the interface decreases, making the propagation of cracks more difficult in thin films than in thick films.

Finite element analysis was carried out to account for the experimental observations and good agreement was obtained. In the analysis, the plastic deformation characteristics of the metal and the polymer have been specifically taken into account. In comparison with a linear elastic analysis, the linear model predicts significantly higher stress levels and local concentrations than the nonlinear model.

Interfacial reactions of polyimide with several metals have been investigated by XPS and TEM to determine their effects on adhesion and long term stability. It has been found that the polyimide-on-metal interface and the metal-on-polyimide interface are intrinsically different; in the case of PI-on-metal interface, the precursor of polyimide, polyamic acid, reacts with the metal resulting in a strong chemical bond and therefore, higher adhesion strength than the corresponding metal-on-PI interface. Both interfaces are found susceptible to T/H environment, resulting in significant adhesion loss. The mechanical and electrical properties of polyimide may be altered as a result of the interaction with metals and therefore, great care is necessary to insure a stable interface and the reliability of devices.

The absorption of several solvents including methylene chloride (MC), methyl chloroform (MCF), gamma-butyrolactone (BLO), dibromomethane (DBRO), 1,2-dichloroethane (DCE), and 1,2-dichloropropane (DCP) in Kapton H*, Upilex S**, and Upilex R** polyimide film was measured gravimetrically. A significant difference in the rate and amount of absorption of these chemicals by the various films was observed. Complimentary Rutherford Backscattering Spectrometry (RBS) measurements provided additional information regarding the amount of solvent absorbed and mode of transport.