Electroplating was developed as a combination of Direct Current (DC) and a chemical bath. It was understood that this simple waveform and bath composition had considerable limitations. Numerous innovations followed to optimize the plating process for the desired deposit characteristics. There were advances in cell geometries, anode materials, temperature controls, monitoring, instrumentation and numerous others.

A key advancement was the use and optimization of chemical additives for the DC electroplating bath. We recognize the need to change our bath (electrochemical process) based on the required deposit characteristic (i.e. throwing power, flatter deposit, conductivity, etc.). Additives change the process parameters and mediate the desired results.

Performing the electroplating operation as a process (sequence of steps) enables us to properly level complex parts (such as PCBs) and achieve otherwise difficult deposit characteristics. For example, at a specific point during electroplating, we need to utilize an exact quantity of a specific additive. A second example is to stop the plating process, mask the section which was just plated, unmask another section to be plated in the next step of the process, and continue on. The point is that electroplating was developed and successfully utilized by creating processes. A process allows the user to accomplish the task by performing multiple measurable steps each specifically defined to yield a desired result. The sum of the incremental steps is a completed process with a desired result.

The timeless truth I referred to in the subhead is that “electroplating needs to be viewed and executed in terms of processes, regardless if it involves DC or pulse waves. This is especially true as our work pieces are composed of multiple geometries (fine-line traces, vias, PTHs) requiring multiple deposit characteristics.”

Somewhere in our search for continuous improvement, six-sigma quality, and reduced cycle time, we discovered pulse plating and forgot that timeless truth. Thus began our search for the “magic pulse waveform.” You know the one. You set your dials on this waveform and electroplate fine-lines, plated though holes (PTHs) and blind vias with bright deposits and high conductivity using any proprietary chemistry and additives developed for DC. The only reason we cannot find the magic pulse waveform is that those who have it will not share the information with us. We keep looking and waiting for someone to demonstrate it so that we can upgrade our plating operations.

We would not attempt to drill different sized holes with a single drill bit. Nor should we attempt to electrochemically fill different sized holes with a single pulse waveform. This approach is similar to our ancestors frustrated search for the mythical fountain of youth; they returned after several centuries with increased value placed on the old knowledge. Prior to the search for the fountain of youth, that knowledge consisted of a healthy diet, genetics, hygiene and exercise, to name a few.

Today we know to value diet, genetics, hygiene and exercise in the pursuit of a long and prosperous life. The pursuit of such ideal solutions is not itself bad; it motivates us to pursue a noble goal. What we find, however, may not be what we expected. In the pursuit of the fountain of youth, through medicine, technology and other efforts, we have in fact extended our life expectancy considerably, eliminated many deadly plagues, and are now stronger and healthier than at any other time in history. Clearly, we’ve not reached the ultimate goal, nor are our methods that which Juan Ponce de León expected to find in the 15th century, but it all adds up to a longer and more prosperous life.

Similarly with electroplating, as we search for a magic pulse waveform, we advance technology and solutions. This is not as effective as if we had the end goal in mind at the start, but it is progress, and these advances benefit our manufacturing capability and bottom line.
There are many modern examples to indicate that we will not find the magic pulse waveform, but rather, the need to incorporate new pulse technology with the old wisdom of performing the job as a process or sequence of steps. The sum of these tangible steps can yield a more efficient and higher performance process while providing time and cost reductions.

Faraday Technology, during IPC 2000, demonstrated a single waveform was not optimally able to geometrically level both PTHs and blind vias. This is important if we hope to reduce process time and cost by eliminating multiple plating baths.

fast cash

First things First
Quick solutions to problems on an electroplating line can save thousands of dollars. Experience is by far the most valuable tool a troubleshooter can possess. So successful troubleshooting begins when the line is at its best! Build your experience by walking the line when everything is running smoothly. Take note of solution color, smells, gage settings, etc. Intimate knowledge of your plating line and its idiosyncrasies will expedite the solution to future problems.

Now that a problem has developed, you must walk the line looking at temperature gages, current, anode baskets, pumps, etc. You must rule out the cleaning section of the line and the post plate section of the line. These steps may take some time but they must be done. Ninety five percent of plating problems have nothing to do with the plating bath. Occasionally a problem develops, though, which persists despite experience.

Into the Lab
Once the problem has been determined to be a result of the plating solution or the material being coated, the troubleshooting should be done in the laboratory.

Before starting a laboratory investigation, the first thing to do is ship samples of the plating bath and reject work to your supplier. Suppliers often have sophisticated labs with experienced people. Follow up with a phone call to your supplier. Speak directly with a technical service representative and discuss your problem and investigation.

Once in the laboratory, define the condition of the bath with a routine analysis and a routine hull cell. Correct any chemistry problems found by the routine analysis. The routine hull cell should be one you re used to looking at. Suggested conditions for a routine panel are:

1. A two amp, five minute, unagitated panel for acid zinc.
2. A one amp, ten minute, agitated panel for alkaline zinc.
Compare the routine hull cell panel to ones from when the problem was not present. Measure thicknesses across the panel and again compare them to past hull cell panels. If the routine hull cell panel appears normal, chances are your problem lies outside of the plating bath. Re-walk the line and review your observations. Pay particular attention to the electrical portion of the plating bath, as poor electrical connections will make the plating bath appear to be at fault. Investigate the material of the parts exhibiting the problem. Again, material problems will not manifest themselves in the lab. If you are still convinced that the plating bath is the source of the problem, continue with the lab investigation.

The next step is to use a hull cell to generate the problem. Make sure the conditions and time all hull cells were run are clearly marked on the resulting panel. Vary the conditions in the hull cell to give yourself the best opportunity to produce the problem. Some variations, which may prove useful, are:
1. Panels run at low amperage
2. Panels run at high amperage
3. Bent panels to create a shelf area
4. Bent panels to create an extreme low current density area
5. Panels run at a high temperature
6. Panels run for thirty minutes then scribed with an exacto knife (to reproduce blistering Knowing how your bath appears when operating normally will make the interpretation of these hull cells easier. Once the problem has been produced, we can proceed to the next step.

Target Identified
With the ability to produce the problem, one now needs to know how to remove the problem. The problem probably will fall into one of several broad categories:
1. Organic contamination
2. Metallic contamination
3. Poor filtration
4. Imbalance of proprietary chemicals
5. Unknown

With an unlimited supply of solution, take the opportunity to begin multiple treatments. After each of the following treatments, rerun the hull cell test, which produced the problem. First, for organic contamination, treat three hundred milliliters of solution with two grams of activated carbon. Mix the solution continuously for at least thirty minutes, then filter and run the hull cell. Second, for metallic contamination, treat three hundred milliliters of solution with one-half gram of zinc dust. Again, mix the solution continuously for at least thirty minutes, then filter and run the hull cell. Third, filter the solution thoroughly. The solution must be clear after this step. Use a filter aid if necessary. Fourth, if the solution is an acid bath, metallic or organic contamination may be affected by adding one tenth of a gram of potassium permanganate to three hundred milliliters and mix the solution for five minutes. Filter thoroughly and run the hull cell. For an alkaline solution, freeze out carbonates by putting three hundred milliliters in a lab refrigerator. Cool the solution to 30….F for fifteen minutes. Decant the solution, raise the temperature, and run the hull cell test. If one of these treatments affects the problem, you may be well on your way to solving the problem. Give yourself a pat on the back! Not too fast though. You now must translate the lab results to the production line. You must also locate and eliminate the source of the contamination. The quick results in the lab may take a couple of days to accomplish in production. But don t give up. Plug away until the job is finished.

Unknown
If the above treatments did nothing to affect the problem, things just got a lot tougher. Get on the phone to your supplier and ask for their assistance on-sight. Review their analysis of your bath. Is your bath low on carrier, high in brightener, out of balance, etc? Many suppliers have doctor solutions. By comparing notes with your supplier, they will be able to send in a new arsenal of weapons along with technical assistance. Meanwhile, there are still a few hull cells to run:

For an alkaline bath try:
1. Adding 1% sodium
hypochlorite to the hull cell
2. Adding one-half ounce per gallon of EDTA or Rochelle salts
3. Diluting the bath by 25% with virgin solution
For an acid bath try:
1. Heating the bath above the cloud point, then carbon treat
2. Reduce the pH of the solution with 50% hydrochloric acid to kick-out most organics.
Usually a pH of 2.5 is sufficient. Filter the bath, raise the pH, and add carrier.
3. Diluting the bath by 25% with virgin solution
Out of all the tests you have now run, hopefully something you can build on has emerged. If not, you are into the rare problem, which falls into the unknown classification. This type of problem will take time and effort to resolve. Calling in electricians, sending samples to outside laboratories, etc. are examples of the steps that may be necessary to solve the problem. In this case, the economics of dumping the bath and making a new one must also be considered.

Conclusion
Troubleshooting a zinc-electroplating bath will be much easier if one takes the time to observe the line when things are running well. When a problem develops, split the line into a pre-cleaning section, the plating bath, and a post-plate section. Run tests to isolate the problem to one of the three sections. When the plating bath is identified, follow these steps:
1. Send samples and reject parts to your supplier.
2. Use the hull cell tests outlined above to treat the problem. Remember, even if you can treat the problem, you will also have to eliminate the source of the problem.
3. Demand prompt service from your supplier.
4. Label or identify all tests run during the troubleshooting process.
Once the problem is solved, go back and review the troubleshooting process and learn from it. This will build your troubleshooting skills and shorten the duration of future problems

This article describes a number of key factors affecting copper electroplating for microvia filling and the levels of performance that are currently available to meet the needs of this important market.

Bath Chemistry Parameters Affecting Via Fill
The vast majority of via fill electroplating baths are based on electrolytes consisting of copper sulfate and sulfuric acid. Combining low cost and convenient operation, these sulfate based systems are a well established technology, having now been used in the PCB industry for over 50 years and for via fill applications for over 10 years.

A typical acid sulfate system contains copper sulfate (the primary source of cupric ions), sulfuric acid (for solution conductivity) and chloride ion (as a co-suppressor). Of these components, copper sulfate, typically at concentrations above 200 g/L, has the most significant affect on via filling ability.

Acid copper sulfate system operated without additives typically yield deposits of poor physical properties. Organic additives, typically consisting of materials described as brighteners, suppressors and levelers, are therefore used to further refine deposit characteristics.
Carriers are typically large molecular weight polymers that work in conjunction with small amounts of chloride to form a surface film on the plating surface, which retards the plating reaction. This limits the lifetime of individual growing grains, causing the deposit grain size to become smaller than that obtained without carrier. Carriers are present in relatively high concentration (500 to 3,000 g/L) and show relatively low sensitivity to variations in the rate of mass transfer to the surface. However, in the absence of additional additives, deposits from such formulations do not have smooth, bright surfaces.

Brighteners are typically small, molecular weight sulfur-containing compounds that locally increase the plating reaction by displacing adsorbed carrier. The impacts of brightener additions occur preferentially at points of lower field density, typically in surface recesses or at the bottoms of vias or trenches. The function of the brightener is to locally accelerate the rate of the copper plating reaction and further refine the grain size of the deposit.

Levelers, a further class of additives, act as selective suppressors and typically operate at low concentration (< 10 ppm). At these low concentrations, the activity of levelers is much more mass transfer dependent then that of carriers, with the consequence that less isolated locations (such as the panel surface) are more suppressed than more isolated locations, such as the interior surfaces of vias and recesses within via hole walls.

Bottom-Up Fill Mechanism
For blind vias to be filled with a high quality continuous copper deposit, the plating rate within an individual via must vary. The plating rate at the base of the via must be substantially faster than that that of the remaining areas to avoid premature closure of the mouth of the via opening and the consequent formation of voids or seams.

Accelerated bottom-up filling has been attributed to the mode of action of the organic additive system (1). The suppressor or carrier forms a current inhibiting film on the Cu surface. This film forms uniformly at all locations, assisted by the high solution concentration of suppressor. The accelerated bottom-up filling (i.e. “superfilling”) is believed to be driven by brightener concentration enhancement at the base of the feature (via or trench) during the plating process. Progressive reductions in surface area of via bottoms during deposition “squeeze” the brightener into ever decreasing areas. This localized concentration of brightener further accelerates the plating rate relative to the surface. The leveler acts to suppress the plating at the corners of vias, and aid in reducing the formation of a void. To maintain bottom-up filling behavior, brightener concentration must be controlled within specified limits.

Process Parameters Affecting Via Fill
In addition to process chemistry formulation and bath composition, the key process factors affecting via filling are substrate condition, solution flow, current density and the pretreatment process.
Via profile, thickness and uniformity of the initial conductive layer, degree of surface oxidation and type of dielectric material have a significant impact on via filling ability. A ‘V’-shaped via, with uniform sidewalls free of overhang or protruding glass fibers, promotes consistent seed layer formation and enhances subsequent via fill. Accordingly, non-reinforced dielectric materials are generally easier to fill. A thin or discontinuous seed layer will significantly degrade via fill performance.

While lower levels of solution flow will generally improve via filling performance, particularly of large (100 µm or above) vias, this improvement comes at the price of increased risk of improperly filled small (75 µm or less) diameter vias. Improper fill may manifest itself as defects ranging from seams within the plated deposit, to completely voided vias. The consequence of this behavior is that equipment parameters must be optimized to achieve acceptable levels of fill and plating quality for the specific applications being run.
The effects of current density are somewhat less confounded, as lower current density will both enhance via filling performance and also produce product with lower levels of improperly filled vias. However, the impact of current density is strongest at the very early stages of via filling. Once vias have partially filled, higher current densities can be applied without adverse effects.

This program consists of an Operator Interface (OI) for parameter entry and display and a Control Server (CS) to control and monitor the rectifiers. These two applications use a Windows socket connection to interchange data using the Internet protocol. This two-part design allows DynaComm II to be installed on a single computer for a stand alone application or to be installed on separate computers where the operator’s and rectifier’s positions can be separated by feet or miles.

This dual application design also permits any application program that will support the Internet protocol to be connected to the Control Server. The company can create custom additions to the Operator Interface to control and monitor other plating line functions, such as temperature, pH, and so on, whether servers exist or need to be written.

DynaComm II allows multiple monitoring and controlling options. More than one Operator Interface can be connected to a CS for multiple monitoring/control locations for a single line. Or, multiple Control Servers can be connected to a single OI for monitoring of multiple stations from a single supervisory office workstation.

Data logging is an optional feature, and when enabled, this function automatically stores data accumulated during the plating operation. This data can be downloaded to an electronic spreadsheet for further analysis, providing users with a written record to verify rectifier performance for all parts and batch runs against specifications and tolerances required.

To prevent loss or corruption of data, DynaComm II also features UPS power monitoring and control. If AC power is interrupted, it senses the line loss or brownout, closes the open files, closes itself, shuts down Windows and disables the UPS output. The minimum system requirements include a 233 MHz Pentium PC, 32 MB RAM, 4 MB Video RAM, 6 MB free hard drive space, two RS232 serial ports and a 10/100 Base T Ethernet port. Additional hard drive space will be required for plating recipes and data logging storage.