By Tom Karlinski and Craig Lazinsky, Speedline Technologies, CAMALOT Division, Haverhill, Mass. Dispensing can be broadly defined as the transfer or displacement of a liquid or fluid (simple or composite) from a storage reservoir to a target substrate in pre-defined patterns of calculated volumes. Numerous methods have evolved to keep pace with the rapidly developing chemistries and increasing engineering use of, and manufacturing need for, adhesives and electrically conductive fluids in a wide array of assembly industries.1 Dispensing Applications and Materials The following is a brief survey of some common assembly and packaging applications: In the electronics industry, the application of adhesives, epoxies and solder paste to printed circuit boards is an integral step in doubled-sided, through-hole assembly. (2, 3) SMD Adhesive During the SMD assembly process, adhesives function to hold the devices first in proper orientation after placement, and then to maintain their position during the reflow cycle. For example, in a mixed-technology assembly, through-hole components are first inserted and clinched. The substrate is then inverted and adhesive is applied, surface mount devices are placed and the adhesive is cured. Once firmly in place, the substrate is again inverted and wave soldered. The ideal SMD adhesive is a one-part material with a long shelf life and an extended working life during use. The rheology (flow characteristics) must be compatible with the application technique (clog free, no stringing), as well as the assembly requirements (appropriate profile heights to contact components, green strength). Cure times need to be short (5 minutes at 125°C) and the material should be electrically insulating and stable under stress from processing chemicals and reflow temperature (26°C). Solder Paste In the electronics circuit board assembly process, solder forms the electrical interconnection between components and the etch patterns of the PC board. Solder paste is a suspension of solder powder in a flux (a mixture of rosins, solvents, binders and additives used in combination to optimize storage, flow and reflow characteristics). When heated to the proper temperature, the particles melt to form a solid mass of metal, creating an electrical path between component and PC board. Solder paste flow characteristics are important for effective application. The properties of viscosity and thixotropy (a property that causes certain gels to liquefy when disturbed and to solidify when left standing) depend on the percentage amount of metal, the size distribution of solder particles and the proprietary additives used to thicken and stabilize the paste. Solder sphere size must be matched to the dispensing method, as well as to the minimum dispense volume required (mesh size, aperture size and needle dimension). Metal content, which affects viscosity, must also be appropriate for the application method. Semiconductor Packaging New packaging designs have sought to increase I/O densities and decrease device response times, while reducing on-board real estate requirements. Common to these new designs is the placement of bare silicon die directly onto the circuit assembly or interconnect platform rather than prepackaging the devices into discrete, sealed components. The silicon chip thus placed has advanced the semiconductor packaging industry, but the bare die alone would quickly fail under constant exposure to the environment without further protection. This is overcome by packaging, or encapsulating, the devices in place, using a variety of barrier materials. Typical liquid encapsulants are a mixture of polymer precursors and fillers. In these epoxy systems, a hardener and epoxy resin, accelerated by heat, react together to produce long polymer chemical chains that are solid, cannot melt and provide good protection against most environmental damage. Fillers, typically simple silica (SiO2), are added to reduce the coefficient of thermal expansion. (The coefficient of thermal expansion, CTE indicates how much a material shape changes for each degree of temperature change.) Wetting and flow agents may also be added to assist in dispersing the encapsulant over the device. Liquid encapsulants are commonly supplied as a premix of hardener and resin that is reactive even at room temperature. Raising the temperature to about 150°C will cause the polymerization to complete in 30 minutes or even less. Conversely, cooling the encapsulant to freezer temperatures (-40 to -50°C) will retard the reaction and provide a storage lifetime up to a year. Stored in freezers until needed, the material must be brought to room temperature before dispensing. The liquid starting point of these encapsulants means that they must be dispensed in place and then hardened to a permanent solid at a later stage. Following are brief descriptions of dispensing requirements for several common packaging schemes.
Next, first-level electrical connections are made between chip pads and substrate by wirebonding. The system is now ready to function, but it is unprotected from the environment and could not survive very long in service. An encapsulant must be added around and over the die, surrounding the wirebonded structure.
Without some restraining feature, these encapsulants would flow out and away from the components. In some packages, it is the substrate itself that forms the barrier; while for others, a second material of very high viscosity is dispensed to provide a retaining dam.
Miscellaneous Applications
Deposition Methods Pin Transfer This method involves dipping an array of vertically held pins into a reservoir of liquid material, withdrawing the pins at a controlled rate, then transferring the adhering liquid to a substrate by making physical contact. This method is confined to "dot" printing and is used for applying SMD adhesives. This mass transfer method is also limited to flat substrates and is tooling intensive. Stencil Printing Stencil printing is perhaps the fastest and most efficient method of applying adhesive or paste to a large area in long production runs. In this method, a stencil is created with laser-cut holes, called apertures, that match the component layout. The stencil is then registered over the substrate and brought into contact. A squeegee forces (or pumps) the material through these apertures onto the substrate and subsequently levels the deposits. Once the entire deposition area has been swiped, the stencil and substrate are separated, leaving behind a pattern of material matching the aperture pattern. Faithful replication of the aperture pattern requires a clean release of material from the aperture sidewalls. This can be complicated by side wall imperfections. In fact, for very-fine-pitch components, the apertures become so small that the relative influence of the sidewalls becomes too great, and materials can no longer be printed effectively. Stencil printing poses other limitations, including the cost of tooling new stencils plus turnaround time, as well as dedicated stencil storage space. Further-more, any change in board configuration or component placement requires retooling a new stencil.
Fluid Jetting Jetting is one of the newer methods and is somewhat analogous to inkjet printing. Fluid is propelled from a "gun" to the substrate in a non-contact process. Although this method applies droplets one at a time, deposition rates can be very high. Droplets are typically restricted to one discrete size during a run and larger volumes are delivered by stacking multiple dots on top of one another. By moving the jetting head during deposition, patterns can be formed. The process tolerates substrate curvature and can deal with some components already in place, although there can be restrictions. One major limitation of jetting is that of material characteristics: At this time, solder paste and other viscous materials are not suitable, and only relatively low-viscosity, highly thixotropic materials work reasonably well. Nozzle (or Needle) Dispensing In this method, the dispenser, fitted at the output with an appropriate nozzle or needle, is positioned over the substrate by means of a computer-controlled XYZ gantry system. Dot size and bead dimensions are determined by the amount of time the dispenser is activated. A principal advantage of this method is that it is a data-driven process. Board size, component size and standoff from the board may all be changed with simple programming adjustments on the dispenser. Material, packed in disposable syringes, is dispensed accurately on each board with little or no waste, in a pitch as small as 12 mils. Dot profile and size may be customized for each component, thereby greatly reducing setup time. This capability is ideal for prototyping and for short runs where the cost and delay of a stencil is prohibitive. Another valuable feature is that material can be applied in the presence of components or even on 3-D structures. In fact, many applications involve placing material next to, into or onto components already in place. Dispensers can manage a wide variety of materials with viscosities spanning many orders of magnitude, from solder paste, conductive adhesives and damming compounds (high viscosity) to fluxes and underfills (low viscosity).(4) The most significant limitation of needle dispensing has been its modest throughput, when compared to methods like printing and pin transfer. Dispensing Pumps The majority of dispensing pumps can be described using three important principles: time-pressure, rotary displacement (auger) and positive displacement (piston). Time Pressure In a time-pressure system, a syringe or other suitable material reservoir is fitted at its output with a restricting needle or nozzle. Material is dispensed by pressurizing the reservoir for a controlled amount of time. When pressure is applied, material flows and when pressure is released, flow stops. Simple on-off needle valves or pinch valves can be added to the system to facilitate material control. Correction for reservoir depletion is necessary, and the accuracy is influenced by material rheology and temperature. Rotary Pumps As with time pressure, the rotary or auger pump (Figure 1) is likewise fitted at its output with a needle. However, unlike time/pressure pumps, the material reservoir is under constant, low pressure. Material is gently fed in at the top of the auger. When the auger rotates, material is moved down the auger and out the needle. When rotation stops, internal pressures are balanced so that material flow ceases. As with time/pressure, accuracy is affected by rheology and temperature, but to a much lesser degree.
|
Copyright © 2001