Semiconductor devices, methods of manufacturing the semiconductor device and tools are disclosed herein. Some methods include providing an electrostatic chuck and placing an edge ring adjacent to the electrostatic chuck. The electrostatic chuck includes a first electrode to generate a sheath at a first distance over the electrostatic chuck. The edge ring includes a coil and a second electrode to generate an electric field control to maintain a portion of the sheath over the edge ring in a coplanar orientation with the portion of the sheath over the electrostatic chuck.

A power supply comprises at least one waveform generator that produces a clamp waveform responsive to a clamp signal, and at least one amplifier that amplifies and provides the clamp waveform to an electrostatic chuck. An advisor module receives parameter values for parameters affecting operation of the power supply, uses a neural network to determine whether the parameter values are consistent with trained parameter values, and continuously and automatically modifies weighting of inputs to the neural network when any parameter values are inconsistent with the trained parameter values. A controller provides the clamp signal to the waveform generator, receives reports from the advisor module, and adjusts the clamp signal or provides a status report when any parameter values are inconsistent with the trained parameter values.

A substrate fixing device includes: a base plate; and an electrostatic chuck that is fixed to the base plate and configured to adsorb a substrate by electrostatic force. The electrostatic chuck includes: a ceramic layer that is formed of a ceramic and that is configured to adsorb the substrate in a state that the ceramic layer contacts the substrate; a first heater pattern that is disposed on the ceramic layer and configured to generate heat; an insulating resin layer that is disposed on the first heater pattern to cover the first heater pattern; a second heater pattern that is disposed on the insulating resin layer and configured to generate heat; and an electrically conductive member that penetrates the insulating resin layer such that one end of the electrically conductive member contacts a surface of the first heater pattern and the other end of the electrically conductive member contacts a surface of the second heater pattern.

An electrostatic chuck includes a base plate that is made of a metal; a ceramic plate that is fixed to the base plate and configured to adsorb an object by electrostatic force; and a bonding layer that is provided between the base plate and the ceramic plate to bond the base plate and the ceramic plate to each other. The bonding layer is formed of a composite material including the metal forming the base plate and a ceramic forming the ceramic plate.

Some embodiments include a thermal management system for a nanosecond pulser. In some embodiments, the thermal management system may include a switch cold plates coupled with switches, a core cold plate coupled with one or more transformers, resistor cold plates coupled with resistors, or tubing coupled with the switch cold plates, the core cold plates, and the resistor cold plates. The thermal management system may include a heat exchanger coupled with the resistor cold plates, the core cold plate, the switch cold plate, and the tubing. The heat exchanger may also be coupled with a facility fluid supply.

A mounting table is provided. The mounting table includes a base having a first flow path, a recess, and a second flow path connected to the recess, and a variable control mechanism configured to variably control a contact area between a target object disposed on the base and a mounting surface for mounting thereon the target object by filling and discharging fluid into and from the recess through the second flow path.

A substrate pedestal includes a thermally conductive substrate support including a mesh, a thermally conductive shaft including a plurality of conductive rods therein, each conductive rod having a first end and a second end, and a sensor. The first end of each conductive rod is electrically coupled to the mesh, and the sensor is disposed between the first and second ends of each conductive rod and configured to detect current flow through each conductive rod.

Embodiments of the present disclosure relate to a system for pulsed direct-current (DC) biasing and clamping a substrate. In one embodiment, the system includes a plasma chamber having an electrostatic chuck (ESC) for supporting a substrate. An electrode is embedded in the ESC and is electrically coupled to a biasing and clamping network. The biasing and clamping network includes at least a shaped DC pulse voltage source and a clamping network. The clamping network includes a DC source and a diode, and a resistor. The shaped DC pulse voltage source and the clamping network are connected in parallel. The biasing and clamping network automatically maintains a substantially constant clamping voltage, which is a voltage drop across the electrode and the substrate when the substrate is biased with pulsed DC voltage, leading to improved clamping of the substrate.

Exemplary support assemblies may include an electrostatic chuck body defining a substrate support surface. The substrate support assemblies may include a support stem coupled with the electrostatic chuck body. The substrate support assemblies may include a heater embedded within the electrostatic chuck body. The substrate support assemblies may include a first bipolar electrode embedded within the electrostatic chuck body between the heater and the substrate support surface. The first bipolar electrode may include at least two separated mesh sections, with each mesh section characterized by a circular sector shape. The substrate support assemblies may include a second bipolar electrode embedded within the electrostatic chuck body between the heater and the substrate support surface. The second bipolar electrode may include a continuous mesh extending through the at least two separated mesh sections of the first bipolar electrode.

Methods and systems for reducing substrate particle scratching using machine learning are provided. A machine learning model is trained to predict process recipe settings for a substrate temperature control process to be performed for a current substrate at a manufacturing system. First training data and second training data are generated for the machine learning model. The first training data includes historical data associated with prior process recipe settings for a prior substrate temperature control process performed for a prior substrate at a prior process chamber. The second training data is associated with a historical scratch profile of one or more surfaces of the prior substrate after performance of the prior substrate temperature control process according to the prior process recipe settings. The first training data and the second training data are provided to train the machine learning model to predict which process recipe settings for the substrate temperature control process to be performed for the current substrate correspond to a target scratch profile for one or more surfaces of the current substrate.