Estimated reading time: 5 minutes

The Chemical Connection: Back to Basic Cupric Chloride Etching

This month’s theme is every layer interconnect (ELIC) or every layer microvias (ELMV), a subject that, I must confess, I have little practical knowledge of. The process problems here are mostly related to registration during multiple lamination steps and how to minimize or correct any misalignment. I know this is very difficult, but I suspect that if you are already doing this or considering it, you’ve probably figured it out on your own.

So, when casting about for an alternative subject, it occurred to me that we have received several calls in the past few months from younger process people who have been thrust into more senior process positions as their older colleagues retire. They’ve realized they lack the practical knowledge their predecessors have accumulated over the years. They know what their various process parameters are, but not why those particular parameters were in place.

I thought that writing a Frequently Asked Questions (FAQ) column would be a good idea, where I could cover some of the simple practical process questions that some of us lofty personages with decades of experience forget to put in our tutorials.

However, when I checked with our field service team and others in contact with customers, I found that the most frequently asked questions have been about cupric chloride etch chemistries. Therefore, I will cover the very basics of cupric chloride etching; not so much the technical facts but why we give the advice we do.

The Old Need for Speed
Back in the almost prehistoric days when I started in the industry, the mantra was speed, speed, speed. Get as many square feet through etching per hour as possible, because all those TVs, refrigerators, stereos, newfangled micro-computers, etc., needed circuit boards. State-of-the-art line and width spacings for circuit traces were 0.010" (10 mil or 254 µm) for densely packed circuits, and most of the production circuit designs were most likely double this. A sophisticated PCB would be double-sided with plated through-holes on 0.064" (1.65 mm) thick epoxy fiberglass laminate.

There were no such things as multilayers, other than a gleam in a PCB engineer’s eye. Components (resistors, capacitors, etc.) were much larger than today (you could actually see a resistor with the naked eye), so a circuit board had to be quite large to fit all the components. Hence, the need to cram as many panels per hour as possible through the etcher to meet the demand for boards. Nowadays, of course, with multilayer boards, chip-on-board, and many other miniaturization technologies, modern boards can be much, much smaller in area to accomplish the same tasks as early boards.

The need for raw etch speed has diminished since boards are now much smaller and multi-functional, and we don’t need as many square feet per month to meet demand, even though extra square feet are needed for multilayer production. I have to admit, however, that I still hesitate to suggest or try new things that might improve etch quality at the expense of etch speed, so ingrained is the need for speed from my early career.

The Quality Factor
A major concern today is etch quality, which, in the case of using cupric chloride to etch copper circuit lines, means the rate of sideways etch to the rate of downward etch (etch factor) as the line is formed, or how straight we can make the sides of the circuit trace. Ideally, the sides of the trace would be straight up and down, but in reality, they are slanted, with the trace being wider at the bottom.

When the space between traces is 0.020" (500 µm), the shape of the sidewall doesn’t matter, but when the space is 0.002" (50 µm) or less, as they tend to be these days, the shape of the sidewall is extremely important to avoid shorts and crosstalk. The main contributors to both etch speed and etch quality are the concentration of cupric chloride in the solution (specific gravity) and the amount of free acid (hydrochloric acid or HCl) in the mix.

The best range of concentrations for cupric chloride etching copper is between 275 grams per liter (specific gravity at 1.2501 or 29°Be) and 370 gpl (sg at 1.3302 or 36°Be). We use specific gravity to define the concentration of cupric chloride as it is easier to measure and control. Degrees Beaume is an old French unit used since the 1600s to measure specific gravity, and it has somehow survived to describe the specific gravity of cupric chloride, ferric chloride, and alkaline etchants. So, the best range of specific gravity for cupric and etching copper is between 29 and 36°Be with 28°Be giving the fastest etch rate downward, but also the fastest sideways etch. Running at an sg of 36°Be gives a slower etch rate but a much-reduced sideways etch.

Recommended free acid levels are between 1N and 3N where N is the normality of the acid in the cupric chloride solution and describes the concentration of the acid. A 1N solution is 36.5 gpl HCL. The free acid level is important for two reasons. One is to keep the cuprous chloride byproduct of the etch reaction in solution so an accurate reading of the ORP for regeneration can be made.

The second reason is that the free acid prevents the formation of oxides on the surface of the copper during etching. The etch rate would be halved without the free acid. Within the recommended range of free acid concentration, increasing the concentration gives faster downward etch rates but also faster sideways etch rates. The free acid level is important, and many smaller shops I have visited don’t have the means to monitor it; they just estimate.

The equipment to measure it is not expensive, and it doesn’t take a trained chemist to use it. I always recommend that a corner somewhere in the shop be cleared and a bench set up to measure free acid at least once a shift.

Running at the lowest recommended specific gravity and highest recommended free acid level will give the highest etch rate, at around 0.002"/min. (50 µm) but a sideways-to-downwards etch ratio of about 1 to 1. Even in the old days this ratio was unacceptable. Running at the highest specific gravity and lowest free acid concentration drops the etch rate to about 0.001"/min (25 µm), but the sideways-to-downward etch ratio improves to 1 to 4.

That etch rate was too slow for the old days, so a compromise was reached and the recommended specific gravity and free acid levels were stated as 32°Be and between 1N and 2N. This gave an etch rate of around 0.0015"/min. (38 µm) at the higher acid concentration and an etch ratio of around 1 to 3 sideways to downward.

The 32°Be, 1N to 2N recommendation for cupric chloride operating parameters is what is recommended in most etching guides, but still favors etch speed over etch quality. I feel that nowadays etch quality is at least as important, if not more so, than etch speed, so I run my cupric etcher at 36°Be specific gravity and 0.8N free acid and accept the lower etch rate in favor of the improved etch quality. This is what I recommend to those who ask. Higher specific gravities (40°Be is close to the saturation point) slow the etch rate without improving the etch ratio, while acid levels below 0.8N also slow the etch rate without improving etch ratios very much and, in addition, become harder to control.

In answer to the next anticipated question: Yes, alkaline etchant is reputed to have better etch ratios than cupric, and this is true if the etchant parameters are kept in the middle of the recommended range for pH, chloride concentration, and specific gravity. One can expect etch ratios of 1 to 4 and etch rates of around 0.0024"/min. (60 µm).

By running at the lower end of the ranges for pH and chlorides and higher end for specific gravity you can get etch ratios of about 1 to 5 and a slower etch speed of 0.0018"/min (45 µu). It should be noted, however, that etchant control gets very tricky at these settings.

At the end of the day, there is no magic chemistry that eliminates the tradeoff between etch speed and etch quality. The old mindset of “faster is better” worked well when circuit traces were large and throughput was everything. Today, with much finer features and tighter tolerances, quality deserves equal consideration.

Careful control of specific gravity and free acid levels is not simply housekeeping; it directly affects the balance between productivity and circuit quality. Perhaps that is the real lesson from revisiting these older chemistries. Even with all the advances in PCB technology, the fundamentals still matter, probably more now than ever.

This column originally appeared in the June 2026 issue of I-Connect007 Magazine.