Micro Etch: 5 Rules for Inner Layer Bonding

Micro etch is the controlled chemical removal of a thin layer of copper from the surface of a panel before it is treated and laminated. The amount removed is measured in tenths of a micron, and that small cut produces the roughness that the oxide and the prepreg grip when the stack is pressed.

It is one of the quiet steps in fabrication. Nothing visible comes out of the tank, and the effect is only confirmed much later, at the press or at the electrical test, when a delaminated layer or a lifted pad reveals what the surface was actually like.

What Micro Etch Does to a Copper Surface

The chemistry attacks the copper evenly at first and then preferentially along grain boundaries, which leaves a microscopically rough surface rather than a polished one. That roughness increases the contact area with the resin and gives the bond a mechanical grip.

The surface is also chemically activated. Fresh copper is far more reactive than copper that has sat in air, so the treatment that follows should happen within a defined time rather than after a break in the shift.

Removal Rate and the Chemistry Behind It

The removal rate is the amount of copper removed per unit of time, and it is the parameter that the bath is controlled to. It depends on the chemistry, the concentration, the temperature and the immersion time, and all four have to be inside their own limits for the rate to be predictable.

Rate is measured by a weight loss test on a copper coupon, which is simple but needs discipline: the coupon must be dried in a defined way and weighed on a balance that can resolve the small difference. A rate that is drifting is the first sign that a bath is exhausting.

Oxide Formation and the Bond to Prepreg

The oxide treatment that follows the etch grows a thin, controlled layer of copper oxide on the surface. It forms the interface with the resin, and its thickness and colour are indicators of how the chemistry is running rather than an end in themselves.

An oxide that is too thin gives a weak bond, while one that is too thick becomes brittle and can flake off as a powder during handling. Our brown oxide notes cover the control of that layer and the checks that go with it.

Inner Layer Preparation Sequence

The sequence is a chain, and each link has to hold: clean, etch, rinse, oxide, rinse, dry. Where a step is shortened or a rinse is weak, the effect is carried forward into the press, and it appears as a bond failure that no later step can repair.

The handling between the steps matters as much as the chemistry. A panel touched by bare hands after the etch carries oil into the oxide tank, and a panel left wet on a rack oxidises in the wrong way. Our black oxide notes describe the same sequence for the alternative treatment.

Rinse, Dry and Handling Between Steps

Rinsing removes the chemistry that would otherwise keep working on the panel after it leaves the tank. A rinse that is short leaves acid in the surface texture, and that acid continues to attack the copper while the panel waits, so the surface that reaches the press is not the surface that was measured.

Drying has to be complete but not hot enough to oxidise the copper further. Where panels are dried in a hot tunnel, the time and the temperature should be set with the oxide chemistry in mind rather than for speed alone.

Bath Control, Analysis and Dosing

The bath is analysed on a schedule for its active component and for the dissolved copper that builds up as panels are processed. Dissolved copper changes the rate, and a bath that is dosed only on the number of panels rather than on an analysis will drift outside its window.

Dosing should be recorded with the analysis, because the two together are what allow a bond failure to be traced. A bath that was topped up twice in a shift is telling a story, and that story is easier to read from a log than from memory.

Measuring the Surface Before Lamination

The practical checks are the removal rate from the coupon, the oxide weight or thickness, and a visual comparison against a boundary sample. The boundary sample is what makes the visual judgement repeatable between operators and shifts, and it should be as valuable as the instrument readings.

Where the process runs close to a limit, the coupon should be taken more often. Sampling frequency is a compromise, and the compromise should follow the stability of the bath rather than a figure chosen once and never reviewed.

Faults That Show Up After Pressing

A delamination appears as a blister or a separation between layers, and it can be caused by the etch, the oxide, the prepreg or the press. Locating it requires sectioning the board, and the position of the separation is the clue to which interface failed.

Pink ring, where the resin does not bond around a hole, is a related defect that appears after drilling. Both faults are expensive, and our desmear notes explain the related step that prepares a drilled hole for plating.

Waste Treatment and Process Limits

The spent chemistry carries dissolved copper and has to be treated before discharge. That requirement sets a practical limit on how long a bath can be run, because the copper concentration rises with use and eventually forces a dump and a remake.

Planning the dump around the analysis rather than around the failure keeps the process stable. A bath that is replaced during a busy period, when the analysis is already drifting, produces a change in the surface at exactly the moment the shop can least afford one.

Records for a Bonded Panel

The record should hold the analysis, the dosing, the removal rate, the oxide measurement and the panel identity. When a delamination claim arrives months later, that record is the only way to establish whether the surface was inside its specification.

Where the same defect repeats across several lots, the records allow the common factor to be identified, and it is often a material batch rather than a chemistry problem. The IPC materials standards describe the tests that a supplier claim would be based on.

FAQ

Does a rough surface always bond better? Up to a point. The texture has to be fine enough for the resin to wet into it, and a heavily etched surface that looks impressive can be too fragile to survive handling. The target is a controlled roughness rather than the roughest one achievable.

How long can a treated panel wait before pressing? The answer comes from the process qualification rather than from a general rule, because the surface changes with the environment. Most shops set a maximum queue time and rework or re treat a panel that exceeds it.

Can a delamination be repaired? A local repair is not practical once the layers have separated, because the bond is the structure. The panel is normally scrapped, and the effort goes into the investigation and the process change rather than the salvage.