Troubleshooting part 3

Plating Voids - Part ii Part 3 in the series "Trouble in your tank?" by Michael Carano	Back to articles index

Introduction In the last article, we began our exploration of PTH voids. In this insertion, we will discuss many of the other causes of PTH voids. Pre-electroless causes of voids It is often said by seasoned engineers that there are so many process steps and variables that will have an influence on the success or failure of getting a continuous void free copper deposit in the hole. One of the obvious causes of voids or at least poor copper coverage is the result of poor drilling. Figure 1 below shows voids on the glass bundle fibers. A closer look as this section will show a very poorly drilled hole wall, with glass bundles protruding from the resin. Figure 1: Glass fiber bundles are torn out exposing a deep gouge in the resin. Astonishingly, many feel the plating process should always compensate for such a travesty. A surface such as the one shown here makes it very difficult to catalyze with the palladium-based activator from the electroless copper process. One should immediately investigate the drilling operation, looking at the following: Drill bit quality-how many hits and the overall condition of the drill tool? Spindle feeds and speeds-is the drill tool punching its way through the stack? Slow up-feed-causing the torn out glass bundles-should be 2x the in feed rate Check chip loads-experiment with different chip loads to improve quality Stack height-how many pwb's in the drill stack-consider reducing number to improve quality Figure 2 shows a horizontal section of a PTH. Note again the poor quality in the hole wall and its effect on the plating process. Figure 2: The desmear operation is another potential cause of PTH voids. The most effective desmear process available today is based on alkaline permanganate. As a matter of fact, the permanganate system is considered an industry standard. The use of alkaline permanganate for multiplayer pwb's has had a dramatic effect on copper coverage and adhesion in the plated through hole. However, this process, if not controlled, can be a major cause of voids. First, the process consists of three main chemical process steps: A solvent conditioner designed to penetrate the polymer matrix of the resin system and weaken the polymer-polymer bonds of the cross linked chain. The alkaline permanganate solution consisting of sodium or potassium permanganate and its corresponding hydroxide salt. A neutralizer (possibly in combination with glass etch) for removing manganese residues. Glass etch as a separate step to lightly roughen (frost) or more aggressively remove glass fibers that may be protruding into the hole due to etch back. Any of these four (or three steps) if not controlled can lead to voids. If this is the case, even the most robust of electroless copper processes will not be able to compensate. For example, the effective alkaline permanganate process not only removes drill smear, but will micro-roughen the resin as shown in figure 3. This texturing is often referred to as the honeycombed appearance. This is desirable for two reasons: The textured surface promotes palladium catalyst adsorption which in turn promotes electroless copper deposition and coverage. Surface improves copper adhesion. This in turn promotes improved solderability by minimizing or eliminating the occurrence of blow holes seen in the wave soldering operation. More on blow holes in a future column. Regardless, if the texturing is inadequate, the possibility of voids exists. Then main reason is that the lack of texturing will reduce catalyst adsorption, leading to voids. Electroless copper processes require an adequate amount of catalyst on the hole wall in order to promote the electroless copper deposition. Key questions to ask in the brainstorming session: Is the solvent conditioner (swellant) making sufficient penetration into the resin matrix and is the solvent system compatible with the resin-note that some of the higher Tg resins do not react the same as standard FR-4. The higher degree of cross-linking of the polymer resin makes it more difficult for the solvent to penetrate and thus weaken the polymer-polymer bonds in the resin. This in turn reduces the action of the permanganate solution in terms of resin removal and texturing. Check the operating temperature of the alkaline permanganate solution, as well as the hydroxide and actual permanganate content. The hydroxide content aids in promoting the aggressiveness of the permanganate attack on the resin. The permanganate in the +7 oxidation state will perform the actual oxidization and breakdown of the resin. Manganate (+4) is a by-product of this reaction and will not participate in the resin oxidation. Manganate residue should be controlled and maintained below 20-25 grams/liter. Otherwise, as manganate builds up over time, the rate of resin removal and texturing will diminish. If solvent compatibility and degree of penetration is sufficient, then consider utilizing a higher concentration of permanganate. Sodium permanganate will allow for more actual permanganate in solution over the corresponding potassium salt. And finally Review the neutralizer and glass-etch. A Major cause of voiding is due to manganate residues remaining on the hole wall resin and glass. One should frequently renew this solution to insure fresh working chemistry. Also, check the acid concentration. Lower concentration reduces the effectiveness of the neutralization. Often, in order to reduce process steps, fabricators will combine the neutralizer and glass etch in the same process tank. (This writer prefers separate steps as the process is more effective). As the glass etch weakens due to continued use, the ability of the glass etch to roughen the glass fibers (referred to as " frosting") is reduced. This in turn reduces the ability of the catalyst to adhere to glass. When this happens, voids are the result.