Keeping legacy semiconductor tools running when parts go obsolete
Old controllers, discontinued boards and lost manuals retire most legacy tools. How to plan spares, backups and retrofits before a failure forces it.
A parallel-plate RIE covers most teaching and dielectric work. An ICP-RIE opens up deep, anisotropic and III-V etching. How to decide which to buy used.
Most university cleanrooms end up owning at least one plasma etcher, and the first purchase is often a choice between a reactive ion etcher (RIE) and an inductively coupled plasma etcher (ICP-RIE).
Both are widely available used, from compact open-load RIE systems to load-locked ICP tools that once ran in production. The right one depends on what your users etch, how deep they go, which gases your building can handle, and who will keep the tool running. This guide compares the two from the point of view of a lab manager buying a used or refurbished etcher.
A conventional RIE is a parallel-plate system. The wafer sits on the powered lower electrode, usually driven at 13.56 MHz, and the same RF power both creates the plasma and sets the energy of the ions striking the wafer. Turn up the power to etch faster and you also increase ion bombardment, which affects selectivity, damage and resist erosion.
An ICP-RIE adds a separate inductive coil around or above the chamber. The coil generates a dense plasma, while a second RF supply on the wafer electrode (the platen or table) controls ion energy independently. That separation is the main reason ICP tools reach higher etch rates with good anisotropy and acceptable selectivity at lower pressures.
In practice:
Start by collecting the list of materials and depths your groups need over the next five years, not just this year. Typical patterns:
If the list is dominated by the first group, an RIE is the sensible first tool. If any of the second or third groups are central to your research, an ICP will be needed sooner or later.
Gas chemistry often decides the question before process performance does.
Fluorine chemistries (CF4, CHF3, SF6) with O2 and Ar are comparatively easy to support. Chlorine chemistries (Cl2, BCl3, HBr) are corrosive and toxic, and need ventilated gas cabinets, leak detection, corrosion-resistant lines, and in many buildings an exhaust gas abatement system. If your building has never handled chlorine, adding it can cost as much time and effort as the etcher.
Typical facilities, depending on configuration:
An open-load RIE vents the chamber to air each time a wafer goes in. That is simple, but it lets moisture into the chamber and makes chlorine processes harder to keep stable. A load lock keeps the process chamber under vacuum, which improves repeatability and safety for corrosive gases.
RIE systems are mechanically simple and forgiving. The main wear items are electrode surfaces, O-rings, the RF match and the pumps. Many universities run older RIE tools for decades with little more than preventive maintenance.
ICP tools have more to look after: the coil and its ceramic window or liner, two RF generators and matches, a helium backside cooling circuit with its clamp, the table chiller, the load lock robot and more complex software. Each of these is a known wear point on a used ICP tool and should appear on the inspection report. Chamber condition matters too, because ICP processes leave more varied deposits. Our chamber refurbishment page lists what we strip, clean and replace.
Endpoint detection is worth considering on either type, especially for shared tools where many users etch thin films on different stacks. Optical emission or laser interferometry endpoint helps students stop on the right layer without timed over-etch. See endpoint detection for the options.
For a used RIE, such as an Oxford Plasmalab 80 Plus, focus on:
For a used ICP-RIE, such as an Oxford Plasmalab System 100, add:
In both cases, ask for an etch test on a known material: rate, uniformity and, if possible, a cross-section or profile. That tells you more than hardware checks alone.
For a new or growing teaching cleanroom, buy a solid RIE first, ideally with a load lock if your budget and space allow. It will carry most coursework and many research projects, and it is easier to install and maintain.
Add an ICP-RIE when your research groups have a clear need for III-V, GaN, metals or deep silicon, and when the building can support the gases. At that point, plan the gas safety system and exhaust abatement alongside the tool rather than after it arrives. Many labs end up with both: an RIE as the general workhorse and an ICP dedicated to the harder chemistries so contamination between processes stays under control.
Wafer size deserves a thought too. Research groups often etch pieces and 50 or 100 mm wafers, while a used ICP may arrive set up for 150 or 200 mm. Pieces can be run on a carrier wafer with thermal paste or oil for heat transfer, but the carrier changes the process and adds a cleaning step. Check which clamp and table sizes come with the tool, and whether a different size can be fitted during refurbishment.
We inspect, refurbish and install RIE and ICP-RIE systems for research cleanrooms, and we can add endpoint detection and replace worn chamber parts as part of the refurbishment. Send us your list of materials and gases, and we will suggest which platform fits and what your building will need.
Old controllers, discontinued boards and lost manuals retire most legacy tools. How to plan spares, backups and retrofits before a failure forces it.
Why a second-hand etcher or furnace can need an export licence, which rules usually apply in the EU, and what buyers and sellers should prepare before shipping.
How the Bosch process alternates etch and passivation to cut deep, vertical features in silicon, what limits it, and what to check on a used DRIE tool.
Tell us the process, wafer size and facilities. We shortlist tools that fit and send an inspection report with every offer.