Ion Source Assisted Ignition of Inductively Coupled RF Plasma

Inductively coupled plasma (ICP) is a common method of ionizing the residual gas in a vacuum chamber. A RF source (commonly 13.56mhz) is connected to a coil (usually outside of the chamber) which induces currents in the nearby gas volume because of the time-varying magnetic fields.

I used a 10 meter (28mhz) 100 watt ham radio transmitter at 60mTorr with multiple coil designs to create an inductively coupled plasma. Even while minimizing the reflected power with a matching network (antenna tuner) the plasma was difficult to ignite. By briefly powering an anode layer ion source in proximity to the ICP coil I was able to ignite and sustain the ICP at lower power levels.

Once the ICP is sustained, the ALIS is powered off and the RF ICP remains on.

Paper written by Samuel Zeloof and Joshua Newstadt: A Brief Analysis of the Interactions Observed in Ion Clouds Created by Electric and Magnetic Fields From Multiple RF and DC Ion Sources

Anode Layer Ion Source Testing


Preliminary testing of the anode layer ion source (ALIS) from RTFtechnologies.

Tests were run at 50millitorr and between 800-1000v. Air plasma can been seen as low as 300v. Typical current at 1kv is 4ma. When the source is current limited, you can see a de-focusing effect as seen in the 2nd picture. This is because the ion source is not operating through the closed drift mechanism but is rather operating closer to a modified glow discharge under the influence of a local magnetic field.

Above pictures taken at much lower pressure and higher voltage (5e-4 torr and 4kv+)

ALIS mounted on 2.75″ CF flange via 10-32 standoff and electrical connection via MHV connector. As suggested and fabricated by J.D Guerard machining Co., a ditch was formed around the hole drilled for MHV feedthrough on the interior side of the flange which created a wall matching in thickness to the MHV connector feedthrough to facilitate with fusing. This can be seen in the first picture. When welded, the two surfaces of identical thickness are fused together easier, with lower current (this weld was done at only 20amps), and with consequently with no warping of the entire flange because of the localized heat.

Using the anode layer ion source as a crude ion pump (far from ideal conditions,  stainless steel chamber walls do not have high getter/sputter capacity)

Ion source power up sweep 400v – 4000v

Ion Source and Faraday Cup Biasing Power Supply (dual 0-350v and 0-4000v)

I made this dual high voltage power supply for an ion source (0-4kv) and also to bias the collector in a Faraday cup to remove secondary electrons from measurement (0-350v, typically 90v). There are 2 boost converters for the high voltage and one buck converter to power the 0-4kv boost, which is an Emco F40. It can provide 4kv @ 2ma and is connected to the MHV connector on the back panel. The Faraday Cup bias supply outputs through the BNC, and the main power connector is a 12v in through the barrel jack. There is a 10 turn potentiometer on the front panel with locking knob for precise adjustment of the high voltage.

IMG_0124 IMG_0125


Spellman High Voltage Reference Manual

Faraday Cup design


Cup Drawing

CAD (autodesk) files: Faraday Cup

Bias schematic: Diagram

A Faraday cup is an instrument that can be used to measure ion beam currents. I designed this one for testing the anode layer ion source from It has 2 permanent magnet rods (3/16″ dia x .5″) in the first part of the cup to act as a trap for incoming electrons and the cup can be positively  biased to suppress secondary electrons so that the measured current is just the ion beam.

Between the section with the magnets and the collector, there is an interchangeable aperture so you can use different sizes. This was designed to be hand made on a lathe; there is no tolerancing so do not attempt to make it unmodified on a CNC.  The body, aperture, and collector are made from stainless steel and the ring which holds the magnets and insulates the collector from the body is plastic (ideally PEEK). The backside of the cup has two blind holes, the center for mounting and the offset hole for BNC connection.

Anode Layer Ion Source (RTFtechnologies)

ALIS acquired from Andrew ( CF flange, MHV feedthrough, and HV PSU are on order.

Ion source links:

Abstract: Anode layer ion sources are one type of gridless (no acceleration or focusing grids/screens) closed-drift (electron and plasma movement is confined due to E x B force) ion sources and are therefore a simple way to make a confined ion/plasma stream. These sources commonly operate under partial vacuum in O2/N2/Ar environments. They have many applications, one of which is a high power directional plasma cleaning source which has many advantages over glow discharge or inductively/capacitively coupled plasmas. Their cylindrical geometry can be explained because it allows the E x B currents to converge and close in on themselves rather than being swept to one side in non-cylindrical designs. They are similar to inverted magnetrons and end-Hall ion thrusters in construction and in operation except they often do not require external neutralization for the ion beam.

Neutralizing electrons come from the from the surroundings and collisions of ions with the discharge channel walls because the anode is held at a higher potential than that of a hall effect source and they therefore operate in the self-sustained discharge area. Thus, some of the electrons from the cathode escape into the ion beam and neutralize the space charge on the ions. The extent of space charge neutralization can be related to the electric field strength and therefore the voltage applied to anode. These high voltages are typically in the 800v-2000v range and ALIS devices can operate up into the tens of millitorr down to the 1e-5 torr range at higher voltages.  Continue reading Anode Layer Ion Source (RTFtechnologies)

6mhz TTL Oscillator Used as Quartz Microbalance


A 6mhz oscillator can was cut up with pliers and the top was removed to expose the quartz crystal. The output is AC coupled (high pass filtered) to the positive (+5v) rail so that a single coax cable can be used to connect this to the thickness monitor as the power is basically just a DC offset on the signal. If you buy a pack of crystals from somewhere like Lesker they will (depending on your thickness monitor) most likely require the usage of an external oscillator. Such a circuit can be made by a complimentary pair or NPN and PNP transistors and couple passive components. Thanks to Ben from Applied Science for the idea.

Abstract: The quartz microbalance is a quick way of monitoring thin film thicknesses in real time, a bare quartz crystal which oscillates at a known frequency is placed in the chamber near the substrate involved in deposition. As the film accumulates on the crystal, it gains mass and the resonant frequency is lowered. The cumulative frequency change is analyzed and interpreted as the rate of accumulation the rate is integrated into a total thickness. Note that accumulation rate is the proper term for deposition rate, as in many cases (sputtering) the actual deposition rate is higher but some atoms are resputtered off the substrate due to high energy molecules in the chamber, substrate temperature, etc. and the total film thickness can momentarily decrease. The net result of the deposition and resputtering is known as accumulation


+/- 0.1nm resolution!