PCB Tech - How to ensure high aspect ratio and small aperture conductivity

When the diameter of the via hole gets smaller and the thickness-to-diameter ratio gets higher and higher, it becomes more difficult to ensure good metal coverage in the hole. It is also very challenging to ensure the uniformity of the metal in the hole and protect the metal in the hole from being etched away during pattern plating and subsequent masking and etching. This article lists many causes of voids in the copper layer of through holes, discusses how to identify the underlying problems, and proposes some suggestions for the production process to avoid these problems.

Holes in the conductive layer in the vias are caused by different reasons and exhibit different characteristics, but one thing is in common, that is, the metal coverage of the conductive layer in the hole is insufficient or there is no metal coverage. Theoretically, the problem is caused by two situations: insufficient metal deposited, or after a sufficient amount of metal is deposited, some of the metal is lost for some reason. Insufficient metal deposition may be caused by improper plating parameters, such as bath chemical composition, cathode movement, current, current density distribution, or plating time, etc. This may also be caused by foreign matter on the surface of the hole that prevents metal deposition, such as bubbles, dust, cotton fiber or organic film, and dirt. If the surface of the hole wall is not properly treated, it is not conducive to the deposition of the plating solution, and may also lead to poor metal deposition, such as rough drilling, crack formation, or "pink circles". The "eating" of copper from the through hole may be due to chemical factors, such as etching, or mechanism, such as blow-holing, cracks, or peeling of the deposited layer.

This article analyzes the defects and causes in the order of the process steps of the via hole metallization process to study where the problem may occur, and the steps leading to the hole in the hole. And learn from the useful factors of classic problem analysis and solution, such as identifying the shape and location of the cavity, and point out the method to correct the problem.

1. Factors that may lead to voids in the holes in the previous steps of metallization:

A. Drilling

Worn drill bits or other improper drilling parameters may tear the copper foil and dielectric layer and form cracks. The glass fiber may also be torn rather than cut. Whether the copper foil will tear from the resin depends not only on the quality of the drilling, but also on the bonding strength of the copper foil and the resin. A typical example is: the bond between the oxide layer and the prepreg in the multilayer board is often weaker than the bond between the dielectric substrate and the copper foil, so most of the tears occur on the surface of the oxide layer on the multilayer board. In the gold plate, the tearing occurs on the smoother side of the copper foil, unless the "revers treated copper foil" (revers treated foil) is used. The oxidized surface is not firmly bonded to the prepreg, and it may also lead to a worse "pink circle", that is, the copper oxide layer is dissolved in acid. Rough hole walls or rough hole walls with pink circles will lead to voids in the multilayer joints, which are called wedge woids or blow holes. The "wedge woids" are initially located at the joint interface. The name also implies: the shape is like a "wedge", which retracts to form a cavity, which can usually be covered by an electroplated layer. If the copper layer covers these grooves, there will often be moisture behind the copper layer. In subsequent processes, such as hot air leveling and high temperature treatment, evaporation of moisture (moisture) and wedge-shaped voids usually appear together. According to the position and shape, it is easy to identify and distinguish from other types of cavities.

B. Decontamination/etching

The decontamination step is to chemically remove the resin greasy dirt on the inner copper layer. This greasiness was originally caused by drilling. Etching is the further deepening of decontamination, which is about to remove more resin, so that the copper "protrudes" from the resin, and forms a "three-point bond" or "three-sided bond" with the copper plating layer to improve interconnection reliability. Permanganate is used to oxidize resin and "etch" it. First, the resin needs to be swelled to facilitate the permanganate treatment. The neutralization step can remove the manganate residue. The glass fiber etching uses a different chemical method, usually hydrofluoric acid. Improper decontamination can cause two types of cavities: The rough resin sticking on the hole wall may contain liquid, which can lead to "blowing holes". The residual dirt on the inner layer of copper will hinder the good combination of copper/copper plating layer, resulting in "hole wall pullaway" (hole wall pullaway), etc., such as in high temperature processing or related tests, copper plating layer Separated from the hole wall. The resin separation may cause the hole wall to pull off and cracks and voids on the copper plating layer. If the residue of potassium manganate salt is not completely removed in the neutralization step (5 to be precise, when it is in the reduction reaction), it may also cause voids. The reduction reaction often uses reducing agents, such as hydrazine or hydroxylamine.

C. Catalytic steps before electroless copper deposition

The mismatch between decontamination/etching/electroless copper deposition and the inadequate optimization of the individual steps are also issues worthy of consideration. Those who study the voids in the vias strongly agree with the uniform integrity of chemical treatment. The traditional pre-treatment sequence for copper sinking is cleaning, adjustment, activation (catalysis), acceleration (post-activation), and then into cleaning (leaching) washing, pre-soaking, which is completely suitable for Murpiy principle. For example, conditioning agent, a cationic polyester Electrolyte is used to neutralize the negative charge on the glass fiber, and it must be applied correctly to get the required positive charge: too little modifier, the activation layer and adhesion are not good; too much modifier will form a film and lead to copper deposition Not good; so that the hole wall is pulled off. The conditioning agent is not sufficiently covered, and it is most likely to appear on the glass head. In metallography, the void opening is manifested in the poor copper coverage at the glass fiber, or no copper. Others cause voids in the glass The reasons are: insufficient glass etching, excessive resin etching, excessive glass etching, insufficient catalysis, or poor activity of the copper sink. Other factors that affect the coverage of the Pd active layer on the pore wall are: activation temperature, activation time, concentration, etc. If the cavities are on the resin, there may be the following reasons: manganate residues in the decontamination step, plasma residues, insufficient adjustment or activation, and low activity of the copper sink.

2. Hollow holes related to chemical copper deposition

When looking at the hole in the hole, always check whether there is a problem with the chemical bath, and also look at the pre-treatment bath for electroless copper, but also cover the common problems of chemical copper, copper electroplating, and lead/tin bath. In general, we can understand that bubbles, solid matter (dust, cotton) or organic matter sticky, dry film may hinder the deposition of plating solution or activation solution. The bubbles are welcome, there are external and internally generated bubbles. Sometimes foreign air bubbles may enter the slot or the through hole when the board is oscillating. The inherent bubbles are caused by hydrogen generated by the reaction in the chemical copper precipitation solution, or hydrogen generated by the cathode or oxygen generated by the anode in the electroplating solution. The voids caused by bubbles have their own characteristics: they are often located in the center of the hole and are symmetrically distributed in the metallography, that is, there is no copper within the same width of the face wall. If there are bubbles on the surface of the hole wall, it will appear as small pits, and the surrounding holes will be spike-shaped. The cavities caused by dust, cotton or oily film are extremely irregular in shape. Some particles that prevent electroplating or activated deposition will also be wrapped by the plating metal. Non-organic particles can be analyzed by EDX, and organic substances can be checked by FTIR.

The research on avoiding the entrapment of bubbles has been quite in-depth. There are many influencing factors: the swing amplitude of the cathode movement, the spacing between the plates, the vibration swing and so on. The most effective way to prevent bubbles from entering the hole is vibration and collision. It is also very important to increase the space between the plates and the moving distance of the cathode. The air stirring in the electroless copper precipitation tank and the impact or vibration of the activation tank are almost useless. In addition, it is also very important to increase the wettability of electroless copper, and to avoid air bubbles in the pre-treatment tide. The surface energy of the plating solution is related to the size of the hydrogen gas bubbles before they escape or burst. Obviously, it is hoped that the bubbles will be excluded from the hole before they become larger, so as not to hinder the solution exchange.

3. Hollow holes related to dry film

A. Characteristic description

Rim voids (Rim voids), that is, the voids are located closer to the board surface. They are often caused by the resist in the holes. They are about 50-70 microns wide and 50-70 microns away from the board surface, with rim voids It may be on one or both sides of the board, which may cause a complete or partial open circuit. The voids caused by chemical copper, copper electroplating, and lead/tin plating are mostly located in the center of the hole. The voids caused by barrel cracks are also different in physical characteristics from the voids caused by dry film.

B. Defect mechanism

The hole or hole edge void is because the resist enters the hole and is not removed during development. It hinders copper, tin, and solder plating. The resist is removed when the film is removed, and the chemical copper is etched away. Generally, it is difficult to find the resist in the hole after development. The location of the hole and the width of the defect are the main basis for judging the hole and the hole at the edge. Why does the resist flow into the hole? The air pressure in the hole covered by the resist is 20% lower than the atmospheric pressure. The air in the hole is hot when the film is applied, and the air pressure decreases when the air is cooled to room temperature. The air pressure causes the resist to slowly flow into the hole until it develops.

There are three main factors that cause the depth of resist flow velocity, namely:

(1) There is water or vapor in the front hole of the film.

(2) Small holes with high aspect ratio, take 0.5mm holes as an example.

(3) The filming and developing time is too long.

The main reason why water vapor stays in the hole is that the water can reduce the viscosity of the resist and make it flow into the hole faster. Small holes with high thickness-to-diameter ratio are more prone to voiding problems, because such holes are more difficult to dry. The resist in the pinholes is also more difficult to develop. The longer time before development also allows more resist to flow into the holes. The surface treatment and the automatic filming connection are more prone to problems.

C. Avoid holes or holes around holes

The best and simple way to avoid holes or holes around holes is to increase the degree of drying after surface treatment. If the holes are dry, no holes or cavities around the holes will occur. No matter how long the storage time and poor development, it will not cause holes or holes at the edges. After adding drying, try to keep the time between the film and the developer as short as possible, but the stability problem should be considered. If the following situations occur, the orifice or the edge of the hole is empty

Holes may occur (not previously):

(1) After installation of new surface treatment equipment and drying equipment.

(2) The surface treatment equipment and the drying section are malfunctioning.

(3) Production of small orifice plates with high thickness-to-diameter ratio.

(4) Change of resist or change to thick dry film.

(5) Use of vacuum film sticking machine.

The worst and rare case is that the resist forms a masking layer in the hole. It appears that the mask layer is pushed into the hole 50-70 microns deep. Since the mask will prevent the solution from entering, it will appear as a general edge cavity at one end of the hole, and the cavity will extend to most of the holes, starting from the other end of the hole., The thickness of the plating layer is thinner as it approaches the center of the hole.

Many printed board factories have switched to the direct electroplating process, which is sometimes connected to the pasting machine. If the subsequent drying is not sufficient, holes and holes at the edges may occur. To make the small holes fully dry, the drying section needs to be very sufficient.

4. Holes related to masking

In the masking process, if the mask is not good, the etchant will enter the hole to etch the deposited copper. The mechanical damage of the mask occurs dynamically, and the upper and lower masks have less holes together. Similarly, the mask is very weak, resulting in negative pressure in the hole, which eventually leads to defects in the mask. This layer of mask can reduce the negative pressure, and the opposite mask is easier to survive. The mask on one side is broken, the etchant enters the hole, and the copper on the side of the broken mask is first etched away. On the other side, the mask blocks the exit of the etchant, and the exchange of the etchant is too little, so the cavity pattern is also more symmetrical, showing that one end is thick with copper and the other end is thin. Depending on the degree of mask damage, the situation is different. In extreme cases, all the through-hole copper is etched away.

5. Direct plating

Direct electroplating avoids the traditional chemical copper deposition, but there are three types of pretreatment process steps; such as: palladium matrix process, carbon film process, and organic conductive film process. Any situation that can affect the deposition of catalysts, or when the polymer conductive film is deposited, monomer deposition and polymer composition deposition can form voids. Most carbon membrane, graphite and palladium membrane processes rely on proper pore wall adjustment, using polymer electrolyte cations and organic catalytic layers containing opposite charges. In order to achieve better catalytic adsorption. Naturally, chemical copper deposition has been proved to be a good process step in practice, such as hole wall cleaning, adjustment, catalytic deposition, etc., are properly applied in the direct electroplating process. Of course, special problems in the electroless copper bath, such as hydrogen generation, will not occur here.

When using the direct electroplating process, if it is not carried out according to the conditions recommended by the potion supplier, some special problems will often occur. For example, in the carbon film process, it is generally not recommended to scrub the board surface after the carbon film is deposited, because the brush will remove the carbon film particles on the edge of the hole. In this case, it is difficult for the electroplating process to enter the center of the hole from the copper surface in time, or even not at all. If the orifice carbon film on one side of the board is brushed off, electroplating can also be performed from the opposite side. However, the result of electroplating is gradually weakened, and the electroplated copper may not be able to communicate with the copper surface on the other side. The result is similar to the mask cracking in the masking process. If in the carbon film or graphite process, pumice powder is sprayed after the catalytic deposition, voids will also occur. The jetted pumice powder particles may enter the holes at a high speed and wash away the catalyst layer particles. On the other hand, the graphite process seems to be able to withstand the pumice stucco treatment.

6. Holes related to copper electroplating and lead-tin electroplating (to pure tin)

A. The internal cause of bubbles

Fortunately, the acid copper plating bath has a very high cell efficiency, so hydrogen generation in a better bath is a small problem. What needs to be avoided is the conditions that are likely to cause hydrogen generation, such as high current density and rectifier fluctuations that cause short-term large current density drift. Some tin/lead baths or tin baths are less efficient than copper baths. Hydrogen generation becomes an important issue. An interesting development in avoiding the generation of hydrogen fractionation is the addition of "antipititting additives". These organic compounds, such as caprolactam derivatives, may participate in redox reactions, taking away atoms before forming hydrogen molecules. The state of hydrogen prevents the generation of bubbles. The reduced "anti-pit additive"' is re-oxidized at the anode and transferred to the cathode to restart the cycle.

B. External causes of bubbles

The most obvious external cause of bubbles is the bubbles filled in the holes before the board is immersed in the solution. In order to expel the air in the hole before the board is immersed in the bath, some electroplating fixture designers have experimented with forming a certain angle between the board and the fixture. Paddle agitation can generate enough pressure differential to drive bubbles out of the holes. Using compressed air through a sprayer to agitate the liquid (air sparging) through the surface of the plate also helps to drive away air bubbles. Of course, the spray stirring itself is also a kind of gas, mixed into the tank, the air enters the circulating filter pump to produce a supersaturated liquid flow, which will form bubbles at the gathering position, and also form bubbles at the defects of the hole wall. Some manufacturers are troubled by this problem and turn to airless agitation (solution spraying).

In addition to resist residues and bubbles that hinder electroplating, other obvious problems that cause electroplating voids are: poor penetration and foreign body blockage. Poor penetration of the bath will cause no copper in the middle, but this is a very extreme situation. Usually, the copper thickness in the center of the hole is not enough to meet the acceptance standard. In the acid copper plating bath, the poor penetration is caused by the following reasons: improper copper/acid ratio, bath contamination, low or insufficient organic additives, poor current distribution, blocking effect or stirring, etc. If particle contamination is found, it is mostly caused by a failure of the circulation or filter pump, the frequency of tank reversal is too low, the anode bag is damaged or the cathode membrane is defective.

7. Voids caused by copper being etched

If there is any problem with the electroplated metal resist, the copper in the via will be exposed to the etchant, resulting in voids. In this case, the voids are caused by copper being etched away rather than undeposited copper. This is a bit contrary to the order of priority. Here, it is still necessary to emphasize that the copper is etched away, which causes the void.

The first possible condition that may cause copper loss is that if there is residual moisture in the hole during electroless copper deposition, or if it is left for too long before the next operation, or the corrosive atmosphere, the copper will be oxidized. The copper is dissolved in the prepreg step before. Another possibility is excessive microetching before plating. Secondly, the copper of electroless copper may fall off. It can be seen if it is directly metallographic or thermally shocked after chemical copper deposition. The reasons for such voids are: improper composition of the electroless copper bath, entrainment of the treatment solution, poor adhesion of the electroless copper due to improper adjustment of decontamination, catalysis, or accelerators.

When wave soldering, hot air leveling, or other high-temperature reflow soldering steps or simulated thermal stress testing, copper defects (cracks, peeling) on the hole wall occur. The root cause of such problems often needs to be traced back to the hole wall pretreatment and the initial The metallization step of the hole. The hole wall can have many causes. According to the manufacturing process, it can be traced back to the previous steps such as drilling, or it can only occur during lead/tin plating. However, the shape and location of the cavity can often provide us with some Clues query the root of the problem. Hole wall voids are often caused by the mutual influence of multiple process conditions. They may act at the same time, or they may have a sequence. Only by carefully analyzing the defect characterization along the process steps can it be possible to find the root point sharply.