Dechucking is the controlled release step that frees a wafer from an ESC once a process finishes. Residual clamping is what goes wrong when that release fails. Residual clamping is charge left trapped in the ceramic dielectric after the chuck voltage is switched off. The trapped charge makes the wafer stay stuck to the surface. This article covers the normal release sequence and why charge gets trapped. It also covers why Johnsen–Rahbek ESCs are most affected, the standard fixes, and what persistent sticking tells you about the ESC.
For the full scope of ESC design, materials and failure modes, see our complete guide to ESCs.
What dechucking is and the normal release sequence
Dechucking is the deliberate sequence a process tool runs to remove the electrostatic clamp force and lift the wafer clear without damage. Simply cutting the chuck voltage is rarely enough. Charge does not vanish the instant the supply is switched off. A well-designed dechuck step first neutralises the stored charge. It then confirms the wafer is free, and then allows the lift pins to raise it. On a healthy ESC the whole sequence completes in under 1–5 s.
The standard release sequence runs through several stages in order:
- The process ends. The high-voltage DC clamp supply is ramped down or switched off.
- The power supply applies a controlled neutralisation step, commonly a brief reverse-polarity pulse. This cancels the charge stored in the dielectric.
- If plasma is still present, it helps bleed surface charge to ground while the wafer relaxes off the surface.
- A short timed delay lets any remaining charge decay before the ESC reports the wafer as released.
- The lift pins raise the wafer and the handling robot removes it.
The dechuck behaviour is set by the same physics that produces the clamp in the first place. So it differs sharply between the two clamping regimes covered in Coulombic vs Johnsen–Rahbek ESCs.
What residual clamping is
Residual clamping, also called sticking, is the holding force that remains after the clamp voltage has been removed. Its cause is charge that stays trapped in or on the ceramic dielectric instead of draining away. While that charge remains, the wafer behaves as though the ESC were still powered. The lift pins then meet resistance instead of a free wafer. In a Coulombic ESC the trapped charge sits mainly on the buried electrode, and it decays quickly. In a Johnsen–Rahbek ESC a layer of charge builds up at the ceramic surface during clamping, and it is much slower to dissipate.
A small amount of residual force is normal. The dechuck sequence is designed to clear exactly this. The problem starts when the residual force is too large or too persistent. The standard sequence then cannot neutralise it before the lift pins fire.
Why charge gets trapped after power-off
Several mechanisms leave charge in the dielectric once the clamp voltage is gone. Which one dominates depends on the ceramic and the process history. The most common contributors are:
- Surface charge accumulation: in Johnsen–Rahbek operation, charge migrates through the slightly conductive ceramic to its surface. There it forms a thin clamping layer that takes time to decay.
- Dielectric polarisation: the ceramic itself polarises under sustained DC. It relaxes slowly after the field is removed, so it holds a residual field for a short period.
- Trapped charge at defects: contamination, surface erosion or micro-cracks create sites that capture charge and release it slowly. So an aged ESC holds residual charge longer than a new one.
- Uncompensated dechuck: the neutralisation pulse may be mistuned for the ESC and process. It can then under-cancel or even over-cancel the stored charge, leaving a net clamp.
The dielectric resistivity controls how charge moves and decays. So the same property that defines the clamping regime also defines how readily an ESC traps charge after power-off.
Why Johnsen–Rahbek ESCs stick more than Coulombic ESCs
Johnsen–Rahbek ESCs are inherently more prone to residual clamping. The reason is structural, not a fault. A JR dielectric is deliberately made slightly conductive, with a bulk resistivity of roughly 10¹⁰–10¹³ Ω·cm. So charge migrates to the ceramic surface, and the effective clamping gap collapses to microscopic contact points. That mechanism gives a JR ESC far higher clamping force per volt. But the same surface charge layer is exactly what must be neutralised at dechuck, and it decays slowly.
A Coulombic ESC instead uses a near-perfect insulator above 10¹⁵ Ω·cm. The charge stays on the buried electrode rather than the surface. So the field collapses quickly when the supply switches off, and release is fast. JR ESCs also vary with temperature, because resistivity falls as the ceramic heats. This means the same ESC can dechuck cleanly at one process temperature and stick at another. That is why JR designs need a carefully tuned dechuck sequence, and tighter monitoring of release behaviour over the ESC’s life.
The standard fixes for residual clamping
Most residual clamping is corrected at the tool by tuning the dechuck sequence, not by touching the ESC. The four standard techniques are used together. A controlled dechuck sequence ramps the clamp voltage down in a managed way instead of cutting it abruptly. A reverse-polarity pulse drives a short, opposite-sign voltage to actively cancel the stored charge. Plasma-assisted discharge keeps a low-power plasma present during release, so surface charge can bleed to ground. A timed delay holds the lift pins until the residual field has decayed below the level that would damage the wafer. The table below maps the common causes to the fix that addresses each.
| Cause of residual clamping | Standard fix |
|---|---|
| Surface charge layer on a JR dielectric decaying slowly | Reverse-polarity dechuck pulse tuned to the ESC, plus a timed delay before lift |
| Slow dielectric polarisation relaxation after sustained DC | Controlled voltage ramp-down rather than abrupt cut-off |
| Charge unable to find a path to ground after power-off | Plasma-assisted discharge to bleed surface charge while the wafer relaxes |
| Mistuned neutralisation under- or over-cancelling the clamp | Recalibrate dechuck pulse amplitude and duration for the process recipe |
| Temperature-dependent release on a JR ESC | Set dechuck parameters per process temperature; verify release across the operating range |
| Persistent sticking the sequence cannot clear | Inspect for dielectric ageing, contamination or damage; refurbish or replace |
The reverse-polarity pulse and plasma-assisted discharge are the two most effective measures for JR ESCs. They act on the surface charge layer directly. The detail of how the supply sequences these steps is handled by a dedicated ESC power supply.
The risk of a stuck wafer
An incompletely dechucked wafer is a handling hazard, not just a throughput problem. If the lift pins rise against a wafer that is still clamped, the localised force at the pin tips can crack or shatter it. That leaves fragments and particles inside the chamber, which contaminate the ESC surface and the next several wafers. A wafer that releases suddenly and unevenly can pop, shift or be thrown off the pins. It can then be mishandled or dropped, stalling the tool. Each event costs the wafer, plus chamber downtime for clean-up, plus the risk of further damage to the ESC. This is why the timed delay before the lift pins move is treated as a safety interlock, not an optional tuning parameter.
When residual clamping means the ESC needs refurbishment
Residual clamping that the dechuck sequence used to clear, but no longer does, is a reliable warning. It means the dielectric itself has changed. As an ESC ages, plasma erosion thins and roughens the ceramic surface. Process residues build up, and micro-cracks or embedded contamination create new charge-trapping sites. Each of these makes charge slower to drain. The residual force then rises above what the standard sequence can neutralise. Sticking may worsen over time, return after every recipe change, or force operators to extend dechuck delays to keep wafers intact. At that stage the problem has moved from tuning to hardware.
At that point the options are to replace the ESC or to recondition it. Where the ESC body is otherwise sound, ESC repair and refurbishment can restore the dielectric surface and clamping behaviour to as-new condition. The work includes full functional and particle testing, at a fraction of the cost and lead time of a new OEM part. We restore consistent dechuck performance as part of that work. An ESC that was sticking and stalling the tool then releases cleanly again.
Frequently asked questions
What causes residual clamping on an ESC?
The cause is charge trapped in or on the ceramic dielectric after the clamp voltage is switched off. In Johnsen–Rahbek ESCs it is mainly a surface charge layer that decays slowly. Ageing, contamination and surface damage make it worse by creating sites that hold charge longer.
How do you fix a wafer that will not dechuck?
Tune the dechuck sequence. Ramp the voltage down in a controlled way, and apply a reverse-polarity pulse to cancel the stored charge. Use plasma-assisted discharge to bleed surface charge to ground, and hold a timed delay before the lift pins move. If that no longer clears it, the dielectric likely needs refurbishment.
Why do Johnsen–Rahbek ESCs stick more than Coulombic ESCs?
A JR dielectric is slightly conductive (10¹⁰–10¹³ Ω·cm). Charge builds up at the ceramic surface to generate the clamp, and it is slow to decay after power-off. A Coulombic ESC uses a near-perfect insulator above 10¹⁵ Ω·cm. Its charge stays on the electrode, where the field collapses fast, so it releases more cleanly.