Classifications

• • H—ELECTRICITY
  • H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
  • H10P—GENERIC PROCESSES OR APPARATUS FOR THE MANUFACTURE OR TREATMENT OF DEVICES COVERED BY CLASS H10
  • H10P50/00—Etching of wafers, substrates or parts of devices
  • H10P50/20—Dry etching; Plasma etching; Reactive-ion etching
  • H10P50/24—Dry etching; Plasma etching; Reactive-ion etching of semiconductor materials
  • H10P50/242—Dry etching; Plasma etching; Reactive-ion etching of semiconductor materials of Group IV materials
• • H—ELECTRICITY
  • H01—ELECTRIC ELEMENTS
  • H01J—ELECTRIC DISCHARGE TUBES OR DISCHARGE LAMPS
  • H01J37/00—Discharge tubes with provision for introducing objects or material to be exposed to the discharge, e.g. for the purpose of examination or processing thereof
  • H01J37/32—Gas-filled discharge tubes
  • H01J37/32009—Arrangements for generation of plasma specially adapted for examination or treatment of objects, e.g. plasma sources
  • H01J37/32082—Radio frequency generated discharge
  • H01J37/32174—Circuits specially adapted for controlling the RF discharge
• • H—ELECTRICITY
  • H01—ELECTRIC ELEMENTS
  • H01J—ELECTRIC DISCHARGE TUBES OR DISCHARGE LAMPS
  • H01J37/00—Discharge tubes with provision for introducing objects or material to be exposed to the discharge, e.g. for the purpose of examination or processing thereof
  • H01J37/32—Gas-filled discharge tubes
  • H01J37/32009—Arrangements for generation of plasma specially adapted for examination or treatment of objects, e.g. plasma sources
  • H01J37/32018—Glow discharge
  • H01J37/32036—AC powered
• • H—ELECTRICITY
  • H01—ELECTRIC ELEMENTS
  • H01J—ELECTRIC DISCHARGE TUBES OR DISCHARGE LAMPS
  • H01J37/00—Discharge tubes with provision for introducing objects or material to be exposed to the discharge, e.g. for the purpose of examination or processing thereof
  • H01J37/32—Gas-filled discharge tubes
  • H01J37/32009—Arrangements for generation of plasma specially adapted for examination or treatment of objects, e.g. plasma sources
  • H01J37/32018—Glow discharge
  • H01J37/32045—Circuits specially adapted for controlling the glow discharge
• • H—ELECTRICITY
  • H01—ELECTRIC ELEMENTS
  • H01J—ELECTRIC DISCHARGE TUBES OR DISCHARGE LAMPS
  • H01J37/00—Discharge tubes with provision for introducing objects or material to be exposed to the discharge, e.g. for the purpose of examination or processing thereof
  • H01J37/32—Gas-filled discharge tubes
  • H01J37/32009—Arrangements for generation of plasma specially adapted for examination or treatment of objects, e.g. plasma sources
  • H01J37/32082—Radio frequency generated discharge
• • H—ELECTRICITY
  • H01—ELECTRIC ELEMENTS
  • H01J—ELECTRIC DISCHARGE TUBES OR DISCHARGE LAMPS
  • H01J37/00—Discharge tubes with provision for introducing objects or material to be exposed to the discharge, e.g. for the purpose of examination or processing thereof
  • H01J37/32—Gas-filled discharge tubes
  • H01J37/32009—Arrangements for generation of plasma specially adapted for examination or treatment of objects, e.g. plasma sources
  • H01J37/32082—Radio frequency generated discharge
  • H01J37/32137—Radio frequency generated discharge controlling of the discharge by modulation of energy
  • H01J37/32146—Amplitude modulation, includes pulsing
• • H—ELECTRICITY
  • H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
  • H10P—GENERIC PROCESSES OR APPARATUS FOR THE MANUFACTURE OR TREATMENT OF DEVICES COVERED BY CLASS H10
  • H10P50/00—Etching of wafers, substrates or parts of devices
  • H10P50/20—Dry etching; Plasma etching; Reactive-ion etching
  • H10P50/24—Dry etching; Plasma etching; Reactive-ion etching of semiconductor materials
  • H10P50/242—Dry etching; Plasma etching; Reactive-ion etching of semiconductor materials of Group IV materials
  • H10P50/244—Dry etching; Plasma etching; Reactive-ion etching of semiconductor materials of Group IV materials comprising alternated and repeated etching and passivation steps

Definitions

• the present disclosure relates to a nanofabrication process. More specifically, the present disclosure relates to new cyclic process for etching a solid surface with atomic layer precision.
• Atomic layer deposition is a nanofabrication process that has become an important method for the growth of high dielectric constant materials, also known as “high-k materials,” for replacement of silicon oxides (S1O 2 ) as the gate dielectric in metal-oxide- semiconductor field-effect-transistors (MOSFETs).
• Atomic layer etching also known as "digital etching” has developed as an alternate process to ALD.
• ALET was first reported for gallium arsenide (GaAs) etching with alternating chlorine gas (CI 2 ) adsorption and electron beam etching. With the development of these techniques, additional research explored the possibility of ion bombardment to effect ALET of silicon, but the necessary period for each etch cycle exceeds the acceptable limits even at the laboratory scale.
• a complete cycle of the traditional approach to atomic layer etching consists of four steps.
• the chemisorption step including clean substrate exposure to a reactant gas to facilitate the adsorption of the gas onto the surface.
• excess CI 2 gas is purged with an inert gas flow to avoid etching by a gas-phase reactant in the subsequent step.
• the reaction step such as chemical sputtering, is affected between the adsorbed gas and the underlying solid reaction, often via inert gas plasma. Ideally, this process is also self-limiting; ions react only with substrate atoms bonded to the chemisorbed gas. Once the chlorinated layer is removed, further etching by physical sputtering of the substrate must not occur or be
• the evacuation of the reaction chamber exhausts the etching products. If the periods of chemisorption in the first step and the etching third step are for sufficiently extended durations, the etching rate approaches one atomic layer per cycle, where the atomic layer thickness is that of the chlorinated layer, but not necessarily one monolayer of the substrate. Additionally, if the substrate surface remains nearly- atomic ally smooth during repeated ALET cycling, it is possible to achieve the ideal condition of removal of substantially one monolayer of the substrate per cycle.
• a system comprising a pulsed plasma source, comprising: spiral coil electrode disposed around a tube; a Faraday shield disposed between the tube and the spiral coil electrode and cooled by a fluid flow; a counter electrode disposed at the top of the tube and at least partially extending into the tube; a gas inlet disposed in the tube and in fluid communication with a process gas supply; and a reaction chamber in fluid communication with the pulsed plasma source comprising: a substrate support; and a boundary electrode.
• a method for etching a substrate comprising introducing a feed gas comprising a mixture of inert gas and reactant gas, into a plasma chamber; disposing the substrate in the plasma chamber; generating a plasma containing reactants and ions from the feed gas; saturating a substrate surface with the reactants to form a product layer comprising a monolayer of the reactant species and a first monolayer atoms of the substrate; and removing the product layer by exposing the product layer to the ions.
• a method for processing a substrate comprising, directing ions from plasma afterglow toward a substrate surface saturated with a first substance. And in certain embodiments, removing the first substance and a monolayer of substrate atoms with the ions.
• FIG. 1 illustrates traditional Atomic Layer Etching (ALET) process.
• Figure 2 illustrates an exemplary ALET process according to one embodiment of the present disclosure.
• Figure 3 illustrates an exemplary ALET system according to one embodiment of the disclosure.
• Figure 4 illustrates another exemplary ALET process according to another embodiment of the disclosure.
• Figure 5 illustrates another exemplary ALET process according to another embodiment of the present disclosure.
• Figure 6 illustrates another exemplary ALET system according to another embodiment of the disclosure.
• Figure 7 illustrates measured ion energy distributions (IED) obtained by applying DC voltages of 30, 50, 70 and 100 V to a boundary electrode in the afterglow of a pulsed plasma.
• Figure 8 illustrates simulated ion energy distributions (IED) obtained by applying DC voltages of 30, 50, 70 and 100 V to a boundary electrode in the afterglow period of a pulsed plasma.
• IED simulated ion energy distributions
• Figure 9 illustrates ion and electron densities as a function of vertical position along the discharge tube axis.
• Figure 10 illustrates a simulated SiCl and SiBr laser-induced fluorescence above a Si substrate after laser-induced thermal desorption.
• Figure 11 illustrates IEDs at a fixed pressure for different DC bias applied continuously at the boundary electrode.
• Figure 12 illustrates resolved Langmuir probe measurements of electron temperature for different pressures.
• Figure 13 illustrates normalized IEDs with DC bias applied continuous on the boundary electrode.
• Figure 14 illustrates IEDs at different pressures under pulsed plasmas conditions.
• Figure 15 illustrates IEDs with a synchronous DC bias boundary electrode pulses at different times during the afterglow of pulsed plasma, and where (a) is graph for bias starting in the early afterglow and (b) is the graph for bias starting in the late afterglow.
• Figure 18 illustrates the graph of IEDs with a synchronous DC bias boundary electrode pulses during the afterglow of pulsed plasma for different duty cycles.
• ALET atomic layer etching
• a substrate such as silicon (Si)
• CI chlorine
• CI reactant gas
• purging the excess reactant gas from the chamber exposing the adsorbed reactant gas to an energetic flux such as a plasma
• exhausting the chamber of etching products such as silicon chloride radicals (SiCl x ), where x is between about 0 and about 4.
• the first step comprises a chemisorption step (1).
• a clean substrate typically comprising silicon, is exposed to a reactant gas, such as chlorine (CI 2 ).
• CI 2 a reactant gas
• the reactant gas adsorption is self-limiting as the chemisorption stops when all available surface sites are occupied.
• the reactant gas flow is only activated during this chemisorption step.
• the second step (2) is necessary to remove the excess reactant gas that may be in proximity to the substrate or the substrate surface and prevent temporary deposition on the chamber walls. More specifically, purging of excess reactant gas (CI 2 ) may avoid spontaneous etching by gas-phase reactant released from the walls in the subsequent etching step (3). Spontaneous etching caused by excess or lingering reactant gas eliminates the possibility of monolayer precision.
• the surface of the substrate is exposed to an energetic flux, such as ions, electrons, or fast neutrals often via inert gas plasma, such as inductively coupled plasma (ICP) to effect the reaction between the adsorbed gas and the underlying solid.
• an energetic flux such as ions, electrons, or fast neutrals often via inert gas plasma, such as inductively coupled plasma (ICP) to effect the reaction between the adsorbed gas and the underlying solid.
• ICP inductively coupled plasma
• the reaction or chemical sputtering is also self-limiting, because the ions react only with substrate atoms bonded to the chemisorbed gas. Once the chemisorbed layer is removed, additional etching of the substrate is not desirable to maintain approximately single atomic layer etching resolution.
• the chamber is evacuated to remove the etching products and any substrate-reactant gas radicals that may be present.
• this traditional ALET process requires a very long etching cycle that is, for example, about 150 seconds (s) per cycle. Further, extending the periods of chemisorption (1) and etching (3), the etching rate approaches one atomic layer per cycle but at the expense of increased cycle times and decreased process efficiency. If the substrate surface remains at or nearly atomically smooth during repeated ALET cycling, it is possible to achieve the ideal condition of removal of substantially one monolayer of the substrate per cycle. However, if the process overly extended, the atomic layer thickness is that of the chlorinated layer, and not necessarily one monolayer of the substrate, thereby at least partially failing the objective of ALET.
• Novel ALET Overview In the present disclosure, several exemplary embodiments of the techniques and systems for ALET process are disclosed. For the purposes of clarity and simplicity, the disclosure is made focusing on one or more specific exemplary systems and one or more specific particular techniques. Those skilled in the art will recognize that the embodiments are exemplary only. The disclosure is not to be limited in scope by the specific embodiments described herein. Indeed, other various embodiments of and modification to the present disclosure, in addition to those described herein, will be apparent to those of ordinary skill in the art.
• the system and method for new ALET process may be pulsed-plasma and pulsed- electrode bias voltages based process.
• the system may include a plasma source such as ICP source, capacitively coupled plasma (CCP) source, or helicon source.
• the plasma source is an ICP source.
• the plasma source may be provided with DC or radio-frequency (RF) power in a continuous or pulsed current.
• at least one electrode is disposed near the substrate or immersed in the plasma.
• the ICP pulsing system has at least one radio-frequency (RF) power generator for creating rapid RF plasma pulses.
• the rapid ALET system comprises electrodes positioned in the reaction chamber for biasing the chamber, biasing the chamber wall, and biasing the plasma.
• the plasma pulsing system comprises a secondary or auxiliary plasma source to aid in the stabilization of the ICP during pulsing.
• the new ALET process presents a potential means for obviating the traditional ALET rate limiting steps, in a non-limiting example, gas pulsing.
• the new ALET process method may comprise two stages: adsorption stage and etching stage.
• the process may utilize switchable electrical pulses to the ICP source and electrodes positioned in the reaction chamber to control the chemisorption and etching.
• the electrodes may apply bias voltages in the plasma in an approximately synchronous manner with the plasma pulses. Differential control of the plasma pulses and electrode bias voltages may permit fine control of the ion energy distribution impingent upon the substrate.
• the new ALET process uses reduced amounts of process and reactant gases, which may be understood to be toxic and corrosive. Compared to the traditional etching methods, this provides reduced costs for gases, improved safety, and improving environmental implications for the process.
• Novel ALET process Referring to Figure 2, there is shown ALET process 200 according to an embodiment of the present disclosure. The top portion of the figure illustrates
• ALET process whereas the bottom portion of the figure illustrates the process parameter.
• ALET process comprises two stages: adsorption stage 212 and etching stage 252.
• a substrate may be exposed to adsorbate such that the adsorbate may adsorb onto the surface of the substrate.
• the adsorbate may be a reactant.
• the adsorbate may comprise dissociated reactant atoms or dissociated reactant molecules having unpaired electrons or dangling bonds.
• the reactant may comprise, without limitation, halogens, fluorine (F), chlorine (CI), bromine (Br), or iodine (I).
• the reactant may be disassociated chlorine (CI) atoms that are derived from chlorine reactant gas (CI 2 ).
• CI chlorine reactant gas
• the reactant may also be used in the adsorbate.
• the intact or un-dissociated reactant may also be used as the adsorbate on the substrate.
• gas includes vapor generated from a substance in solid or liquid state at room temperature or at standard temperature and pressure, without limitation.
• the adsorbate may be obtained by generating plasma containing the reactants.
• inert gas may be ionized along with the reactant.
• the resulting plasma may contain reactants, reactant gas ions, and inert gas ions.
• argon (Ar) is used as the inert gas. Additionally, a skilled artisan will recognize that any noble gas species or other inert gas species may also be used.
• the concentration of the reactant gas may be between about 0.01% and about 20% by volume; alternatively, the reactant gas concentration may be between about 0.01% and about 15%; and in certain instances, the reactant gas concentration may be between about 0.01% and about 10% by volume of the combined gas. In certain embodiments, the reactant gas may comprise a concentration of less than about 1% by volume.
• the plasma generated may primarily comprise Ar species and a small portion of CI reactant gas species.
• the plasma source is used to generate the reactant.
• Non-limiting exemplary plasma sources may include, inductively coupled plasma (ICP) sources, capacitively coupled plasma (CCP) sources, or helicon sources.
• the plasma source is an ICP source.
• the ICP source may RF powered during the adsorption stage 212.
• the plasma source is not powered through the entire adsorption stage 212.
• the RF power applied to the plasma source may be lowered during the latter portion of the adsorption stage 212.
• the plasma source may be RF
• the adsorption process may occur as described herein.
• a substrate comprising a clean surface, without a passivating layer may include unpaired electron or dangling bonds.
• reactants from the plasma near the substrate surface may then easily bond with the dangling bonds of the surface, such as through chemisorption, to form a product layer.
• the product layer may comprise a monolayer of the reactants and a monolayer of the substrate atoms that are associated.
• the CI reactants are adsorbed onto the surface of an exemplary silicon (Si) substrate to form a product layer comprising SiCl x .
• the product layer may comprise a monolayer of reactant species CI atoms and a monolayer of Si atoms. Adsorption may continue until the substrate surface is saturated with the reactants. Without limitation, saturation is achieved when substantially all available substrate surface-sites, such as unpaired electrons or dangling bonds, are occupied or associated with the reactants. As may be understood by a skilled artisan, in certain instances a portion of the substrate surface is not covered with the reactants. For example, a portion of the substrate surface may contain a passivating layer, such as but not limited to an oxide layer. In non- limiting examples, the passivating layer may not contain available sites, available unpaired electrons or dangling bonds, and as such is not covered with the reactants. In certain instances, the substrate surface is at least partially covered with chemisorbed reactants in the product layer and at least partially covered with a passivating layer.
• the reactant gas ions and/or inert gas ions may be present in the plasma, such that the substrate surface comprising a product layer is exposed to the ions.
• the energy of the ions bombarding the substrate may be selectively controlled to avoid or minimize undesired etching, physical or chemical sputtering.
• the energy required by CI ions to etch Si may be about 10-25 eV
• the energy required by Ar ions to cause sputtering may be about 30-60 eV.
• the energy of the ions bombarding the substrate during the adsorption stage 412 may be controlled to be about 10 eV or less.
• the ion energy may be controlled by, for example, providing an electrostatic shielding (e.g. Faraday shield) of the plasma source and/or performing the process under relatively high pressure in order to minimize undesired etching,
• the etching stage 252 may be performed. During this etching stage 252, ions may bombard the substrate to remove the product layer.
• the ions comprise positively charged ions or negatively charged ions. In instances, positively charged ions are used to remove the product layer.
• the energy of the ions bombarding the substrate during the etching stage 252 may preferably be above the threshold for chemically-assisted sputtering but below the threshold for physical sputtering.
• the ions with selected energy may be directed toward the substrate by controlling the potential difference between the plasma and the substrate.
• the potential difference between may be increased by increasing the plasma potential relative to the substrate potential, decreasing the substrate potential relative to the plasma, or both.
• the potential difference between may be increased by decreasing the plasma potential relative to the substrate potential, increasing the substrate potential relative to the plasma, or both.
• Positive or negative, DC or RF bias may be applied to the plasma and/or the substrate during the etching stage 252.
• continuous bias may be provided to the plasma and/or the substrate as shown in Figure 2.
• a series of pulsed bias may be provided as shown in Figure 4.
• the plasma source may be RF powered during the etching stage 252, as shown in Figure 4.
• the plasma source may be provided with pulsed RF power, where each RF power pulse is provided between the bias pulses noted above.
• a series of pulsed RF power may be applied to the plasma source during the etching stage 252 and a series of pulsed DC or RF bias may be applied to the plasma and/or the substrate.
• Each bias pulse may be provided between the RF power pulses.
• the bias pulse is between about 1 ⁇ 8 and about 20 ⁇ 8; alternatively about 10 ⁇ 8 into the afterglow of each plasma source pulse.
• the product layer comprising the chlorinated product layer in the certain embodiment described here may be removed.
• the monolayer of the substrate atoms associated with the product may be removed from the substrate concurrently.
• the adsorption stage 212 and the etching stage 252 may be repeated to remove additional layers of the substrate atoms one layer at a time.
• ALET system 300 may comprise a plasma chamber 326 having top wall 328, bottom wall 330, and side wall 332.
• ALET system 300 may also comprise a plasma source 302, a shield 304 interposed between the plasma chamber 326 and the plasma source 302, a substrate support 306, a boundary electrode 308, a counter-electrode 310, and an inlet 312.
• the plasma source 302 may be coupled to a pulsing system 314.
• the substrate support 306, meanwhile, may be coupled to a support system 316.
• the support system 316 may be a power supply capable of providing continuous or pulsed DC or RF bias to the substrate support 306. Alternatively, the support system 316 may simply be a ground or a component connected to ground.
• the boundary electrode 308 may be coupled to a first voltage system 318.
• the counter-electrode 310 may be coupled to a second voltage system 320.
• the ALET system may additionally comprise a pump 124 coupled to the plasma chamber 126.
• at least one cooling conduit 336 may be included.
• the substrate support 306 may comprise a differential pumping conduit 334.
• the plasma chamber top 328 may comprise a counter-electrode 110 and the gas inlet 112.
• the system 300 may further comprise an auxiliary plasma chamber 350 coupled to the plasma chamber 326.
• An auxiliary plasma source 352 may be disposed near the auxiliary plasma chamber 350.
• the plasma source 302 and the auxiliary plasma source 352 may be any type of plasma source known to those skilled in the art, including an ICP source, CCP source, helicon source and heat source, without limitation.
• the plasma source 302 may be ICP source 302.
• ICP source 302 may be a planar or a cylindrical ICP source 302 comprising a planar or helical coil.
• the ICP source may have other geometry.
• the portion of the plasma chamber 326 and/or the auxiliary chamber 350 adjacent to the plasma source 302 and/or the auxiliary plasma source 352 may be made out of dielectric material such as, for example, quartz or alumina.
• the ICP source 302 comprises a spiral coil electrode disposed around an alumina or other dielectric discharge tube. In further instances, the ICP source comprises a three-coil spiral electrode.
• the shield 304 may comprise a Faraday shield.
• the Faraday shield comprises any conducting material suitable for preventing external interference with the ICP source 302.
• the shield 304 may comprise copper.
• the shield 304 may be configured to prevent capacitive coupling between the coil of the ICP source 302 and the plasma it generates.
• the shield 304 is configured to prevent any electrostatic signals from exiting the plasma chamber 326.
• the substrate support 306 comprises a support for a semiconductor during etching.
• the substrate support 306 comprises an electrode.
• the substrate support 306 is a ground electrode.
• the substrate support 306 comprises a bias electrode, configured to generate and maintain a bias voltage in response to an RF electromagnetic-field or direct current (DC) pulsing.
• the substrate support 306 enters the plasma chamber 326 via the bottom 330 of the plasma chamber 326.
• the substrate support 306 supports the substrate 301 at or in proximity to the bottom 330 of the plasma chamber 326.
• the boundary electrode 308 comprises an electric conducting material disposed proximal to the substrate support 306.
• the boundary electrode 306 may be disposed concentrically around the substrate support 306 near the bottom 330 of the plasma chamber 326.
• the boundary electrode 308 is configured to apply bias in response to an RF or DC signal applied to the plasma source 350, the auxiliary plasma source 302, and/or the counter electrode 302.
• the counter-electrode 310 may comprise an electrical conducting material disposed vertically opposite from the substrate support 306. In embodiments, the counter-electrode 310 is disposed opposite from the boundary electrode 308 in the chamber 326. In some instances, the counter-electrode 310 is applied with a bias voltage in response to a RF or DC signal applied to the plasma source 302, the auxiliary plasma source 352, and the boundary electrode 308. In certain instances, the counter-electrode 310 generates a bias voltage or a pulsed bias voltage that is opposite to the bias voltage of the boundary electrode 308.
• the inlet 312 comprises a gas conduit into the chamber 126.
• the inlet 112 is proximal to the top of the chamber 126 or through the top 128 or the chamber 126.
• the inlet 312 may introduce inert gas and reactant gas into the plasma chamber 326.
• the inlet 312 provides heated gases to the chamber 326 and the plasma source 302.
• the inlet 312 may introduce non-ionized process and reactant gases to the chamber 326 and the plasma source 302.
• the inlet 312 is in communication with at least one auxiliary plasma source 350, for introducing
• the plasma source 302 may be coupled to a pulsing system 314.
• the pulsing system 314 comprises at least one power supply capable of providing pulsed or continuous RF and/or DC signal to the plasma source 302.
• the pulsing system 314 may comprise at least one RF or DC power supply and an electric power amplifier.
• the pulsing system 314 may comprise a plurality of RF or DC power supply and power amplifiers.
• the pulsing system 314 may be coupled to the plasma source 302 via an impedance-matching (e.g. L-type) network.
• the pulsing system 314 is further configurable to provide electrical power at any frequency to plasma source 302.
• the pulsing system 315 is configured to cut or remove power from the plasma source 302 in periodic pulses.
• the RF or DC power supply may provide the plasma source 302 with a square wave function between zero volts and a predetermined high voltage and at a predetermined frequency. As may be understood by those skilled in the art, removing or altering the RF electric current through the coil removes or enhances the formation of plasma.
• the substrate support 306 is coupled to a support system 316.
• the support system 316 comprises an electric circuit including substrate support 306.
• the support system 316 is a grounded electrode.
• the support system 316 comprises an RF function generator or a DC source.
• the support system 316 is configured for producing a bias voltage at the substrate support 106 in response to electrical pulses from the RF function generator or a DC source.
• the support system 316 receives an RF or DC current from the pulsing system 314 as the bias voltage at substrate support 305. Further, the bias voltage of the substrate support 316 may be pulsed in coordination with other electrodes in the system 300.
• the boundary electrode 308 is coupled to a first voltage system 318.
• the first voltage system 318 comprises an electric circuit including the boundary electrode 318.
• the first voltage system 318 is an electric ground, a RF function generator, or a DC source.
• the first voltage system 318 is configured for producing a bias voltage at the boundary electrode 308 in response to DC source.
• the first voltage system 318 receives an RF or DC current from the pulsing system 314, as the bias voltage at the boundary electrode 308. Further, the bias voltage of the boundary electrode 308 may be pulsed in coordination with other electrodes in the system 300.
• the counter-electrode 310 is coupled to a second voltage system 320.
• the first voltage system 318 comprises an electric circuit including the counter- electrode 310.
• the second voltage system 320 is an electric ground, an RF function generator, or a DC source.
• the second voltage system 320 is configured for producing a bias voltage at the counter-electrode 310 in response to DC source.
• the first second voltage system 320 receives an RF or DC current from the pulsing system 314, as the bias voltage at the counter-electrode 310.
• the bias voltage of the counter-electrode 310 may be pulsed in coordination with other electrodes in the system 300.
• the gas inlet 312 is fluidly connected to a gas source 322.
• the gas source 322 comprises process gas and reactant gas mixture for introduction to the plasma source 302.
• the process gas comprises any inert gas that will be ionized to form plasma at the plasma source 302.
• the process gas comprises a noble gas, nitrogen, hydrogen, oxygen, oxygenated gases, or combinations thereof without limitation.
• the reactant gas comprises any gas that will be chemisorbed by the substrate 301 after partial ionization at the plasma source 302.
• the reactant gas comprises a halogen, a halocarbon, a halide, or other halogenated gases without limitation.
• the process gas and the reactant gas may be any gases suitable for ALET.
• the gas source comprises a concentration of the process gas of greater than about 90% by volume; alternatively, greater than about 95% by volume; and in certain instances, the gas source has a concentration of process gas that is greater then about 99% by volume.
• the thermal conduit 336 is configured to alter the temperature of the gas in the system.
• the cooling conduit may be any conduit in thermal contact with the system 100 and configured for carrying a cooling liquid or gas.
• the cooling conduit 136 is in thermal communication with the cylindrical wall 332, and the shield 304.
• the cooling conduit 336 is disposed in thermal communication with a flange such as chamber bottom 330, which couples the cylindrical wall 332 and the shield 304.
• the pump 324 may be any pump configured to reduce gas pressure in a reaction chamber 326 to about 1 mTorr.
• the pump 324 is configured to lower and maintain the pressure in the plasma chamber 326 to between about 1 mTorr and about 500mTorr; alternatively between about 5 mTorr and about 250 mTorr; and alternatively, between about 10 mTorr and about 100 mTorr.
• the pump 324 operates a pressure between about lOmTorr and about 75mTorr in the chamber 326. In instances, the
• Alternate ALET process Referring again to Figure 4, there is shown an alternate exemplary method for controlling the ALET process according to another embodiment of the present disclosure.
• Figure 4 illustrates the timing sequence of RF/DC power/voltage signals applied various components of the ALET system shown for example in Figure 3.
• the signals may be used to control plasma physics and chemistry during ALET process.
• the plasma source 302 is applied with RF power for approximately 1 second during the etching stage, as in stage 202 in Figure 2, to provide reactants (e.g., CI atoms), to form a chemisorbed layer.
• the plasma source is applied with RF power throughout the entire adsorption stage.
• the plasma may source may be applied with RF power during the beginning portion of the adsorption stage and powered down during the latter portion of the stage.
• the plasma in the plasma chamber 326 may be ignited by the tail-end of a low- power, auxiliary plasma generated in the auxiliary plasma chamber 350.
• the ion bombardment energy may be sufficiently low ( ⁇ 10 eV) to prevent any etching to occur.
• a pulsed ICP period of approximately 0.5s removes the chemisorbed layer (e.g. SiCl x ). Pulsing the plasma source power, as a square wave modulation of 13.56 MHz applied RF voltage, has several benefits described hereinafter.
• the electron energy distribution function (EEDF) cools rapidly during the first several ⁇ 8 of the power OFF portions of the cycle in the afterglow, without a substantial loss of plasma density, for example over a typical about 100 ⁇ 8 OFF time.
• the resulting lower energy time-averaged EEDF offers some level of control of the degree of dissociation of the feed gases.
• a mono-energetic ion flux to the substrate can be generated, as recently demonstrated in this laboratory.
• a pulse of positive DC voltage may be applied to the boundary electrode, raising the plasma potential and pushing positive ions toward the surfaces substrate with lower potential.
• a grounded substrate is bombarded with ions with energy equal to VDCI, as shown in Figures 7 and 8. Since control of the ion energy distribution is critical to effect chemical sputtering of the chemisorbed halogenated layer, without physical sputtering of the underlying substrate, this
• Net positive ion bombardment can cause a positive charge to build up on the substrate. However, after the boundary voltage pulse returns to zero, and the plasma has had a chance to approach its natural V p , any charged surfaces with potentials above ground are the first to receive an excess electron flux over the positive ion flux, bringing their potential back to the floating potential, which is near ground potential.
• a large negative DC bias may be applied to, for example, the counter-electrode 310, while a continuous wave ICP power is ON. This negative voltage may have no effect on V p . However, the resulting high energy ion bombardment of the counter-electrode 310 may generate secondary electrons that are accelerated to the full sheath potential.
• These high energy "ballistic" electrons may have a low scattering cross section and bombard the substrate at nearly normal incidence, compensating positive charge at the bottom of even high aspect ratio insulating structures.
• the ballistic electrons can also have beneficial effects on the bulk plasma, such as the enhanced plasma density and lower bulk T e .
• the method 500 generally comprises two stages: adsorption stage 502 and etching stage 550. As may be understood, within each stage may comprise one or more steps or
• the adsorption stage 502 may comprises substrate positioning step504, reactants forming step 510, and reactant adsorption step 520.
• the etching stage 550 may comprise a potential difference increasing step 570. As noted above, the potential difference between the plasma and the substrate may be increased by applying RF or DC voltage to the plasma or the substrate.
• the etching stage 550 may also comprise substrate charge neutralization step 552, and plasma pulsing step 560, and etched product removing step 580.
• charge neutralization step 552 may be performed by biasing the counter electrode.
• the present ALET process 500 may be considerably faster than conventional ALET process. More specifically, after the substrate positioning step 504, the remaining adsorption steps 520 may require a time between about 0.01s and about 10 seconds; alternatively, between about 0.1 second and about 5 seconds; and in embodiments between about 0.5 second and about 1.5 seconds. Additionally, the etching stage 550 may require a time between about 0.01 second and about 10 seconds; alternatively, between about 0.1 second and about 5 seconds; and in embodiments between about 0.2 second and about 1 second.
• the stages or steps may be repeated in entirety or partially until a desired etch depth is reached.
• charge neutralization step 552, and plasma source pulsing step 560, and the potential difference increasing step 570 may be performed simultaneously, or in the alternative, synchronously.
• the adsorption stage 502 may comprises the steps in the disclosed rapid ALET process suitable for adsorbing reactants on the substrate.
• the first step in the stage comprises the substrate positioning step 504 where the substrate is positioned in a chamber.
• the substrate is mounted a substrate support.
• the substrate support may be an electrode.
• the pressure in the chamber may be reduced.
• the pressure, during the ALET process is maintained between about 1 mTorr and about 500mTorr; alternatively between about 5 mTorr and about 250 mTorr; and alternatively, between about 10 mTorr and about 100 mTorr.
• the pressure is maintained between about lOmTorr and about 75mTorr during the substrate positioning step 504 and maintained there throughout the entire ALET process.
• the pressure may be altered to provide IED control at any time throughout the novel ALET Process. As may be understood by those skilled in the art increase in pressure in the reaction
• feed gas may be introduced into the chamber.
• the feed gas may comprise inert gas and reactant gas.
• the reactant gas may comprise, a reactive species, when ionized.
• the reactant gas may comprise CI2.
• the inert gas may comprise Ar.
• the inert gas may have higher concentration by volume than the reactant gas.
• the reactant gas may comprise a concentration by volume between about 0.01% and about 20%; alternatively, between about 0.01% and about 15%; and alternatively between about 0.01% and about 10% of the mixed gas. In alternate instances, the reactant gas may comprise a concentration anywhere greater than about 0% and below about 5% by volume of the mixed gases.
• the feed gas containing the reactant gas and the inert gas may be ionized by the plasma source to form plasma containing, among others, reactants, reactant gas ions, and inert gas ions.
• various types of plasma source may be used.
• the feed gas may be heated to a temperature of greater than about 200K; alternatively, to a temperature greater than about 400K.
• the gas stream is subjected to further RF electromagnetic fields. Components of this plasma including combinations of excited states of species, radicals, ions, electrons, and photons are injected in to the etching chamber.
• the partially ionized reactant gases are pulled directionally toward or away from the substrate in response to the charge bias within the chamber.
• the reactants are adsorbed or chemisorbed onto the surface of the substrate.
• a voltage bias within the chamber may attract the ionized reactant gases to the substrate.
• the substrate has a limited number of surface sites to adsorb the reactants, such as unpaired electrons or dangling bonds.
• the reactants will continue to adsorb onto the substrate surface until the end of the adsorption stage, when all available surface sites or dangling bonds on the substrate are occupied with the reactants.
• a product layer comprising a monolayer of the reactant atoms and a monolayer of underlying substrate atoms may form.
• the plasma and the ions are maintained at low energy (e.g. 10 eV or less) to avoid or minimize etching during the reactant adsorption step 520.
• etching stage 550 may be performed.
• the etching stage 550 may comprises the potential difference increasing step 570.
• the potential difference between the plasma and the substrate is increased such that ions from the plasma may bombard the substrate at a desired energy range.
• the ion energy may be chosen that is below the physical sputtering threshold but above the threshold for chemically-assisted sputtering.
• the potential difference may be increased by applying DC or RF voltage to the plasma, substrate, or both.
• the applied voltage may be continuous (as shown in Figure 2) or pulsed (as shown in Figure 4).
• RF pulses may be applied to the plasma source, between the voltage pulses.
• applying RF pulses may comprise subjecting plasma source (e.g. ICP source) to a periodic square wave function, where the square wave extends from zero power to a pre-determined power.
• the pre-determined high voltage is capable of creating ions with enough ionic energy to remove the product layer. In certain instances, ions with this energy establish the lower ionic energy limit for an IED. Conversely, it may be understood that the pre-determined high voltage is capable of creating ions with a lower ionic energy that does not damage the substrate. In certain instances, ions with this energy establish the upper ionic energy limit for an IED. More specifically, the high voltage pulsing for the ICP plasma is chosen during the plasma pulsing step 560 such that the IED falls entirely with these parameters.
• a square wave function may pulse the plasma for between about 1 microsecond and about 500 microseconds; alternatively between about 10 microseconds and about 250 microseconds; and in certain instances, the plasma is pulsed for between about 25 microseconds and about 100 microseconds. Further, the square wave function pulses the plasma to about zero voltage for between about 10 microseconds and about 750 microseconds; alternatively between about 50 microseconds and about 500 microseconds; and alternatively between about 100 microseconds and about 250 microseconds.
• an afterglow of ions remains. Without limitation by theory, the afterglow contains ions that are within the IED needed to remove the product layer.
• the counter-electrode may be applied with negative bias voltage.
• the counter-electrode may be applied with negative voltages attract positively charged ions into the counter-electrode.
• the bombardment of the positively charged ions into the counter-electrode may generate high energy secondary electrons which may bombard the substrate at nearly normal incidence.
• the secondary electrons may enhance the plasma density and lower bulk electron temperature T e .
• the boundary electrode may be applied with positive voltage pulses.
• a square wave function DC pulses the boundary electrode to positively-charged voltage bias for between about 10 microseconds and about 750 microseconds; alternatively between about 50 microseconds and about 500 microseconds; and alternatively between about 100 microseconds and about 250 microseconds.
• the positively-charged voltage bias is present only when the high voltage plasma pulse is not.
• the positively-charged voltage bias is present for the complete duration of the etching the product layer 250.
• the substrate support may be grounded, powered with RF, DC or a combination thereof.
• the substrate stage may be pulsed commensurate with the boundary electrode.
• pulsing the substrate support bias provides additional means of controlling the IED as described previously for any electrode in the system. More specifically, the substrate support could be applied with negative DC voltage. Alternatively, a high frequency RF pulse or tailored DC pulse to the substrate support in the case of an insulating substrate or under other selected conditions.
• ALET Pulsing As noted above, the optional plasma pulsing during the etching stage 550 provides the ability to control the disassociation of the feed gases and IED. Providing the plasma pulsing during etching stage 550 may also reduce the angular distribution of ions impacting the substrate. Under collision-less conditions, the angular spread is given by Equation 1:
• ⁇ 3°. This small angular spread, comparable to conventional plasma etching at much higher ion energies, is very desirable for obtaining consistent deep etching through multiple atomic layers, as well as minimizing ion energy transfer to the sidewalls of features from glancing angle collisions and sidewall damage.
• Figure 3 and 6 show schematics of the experimental apparatus used in this study.
• the inductively coupled plasma (ICP) was ignited by a 3 -turn spiral coil in a 17.8 cm long, 8.6 cm inside diameter alumina tube.
• a copper Faraday shield prevented capacitive coupling between the coil and the plasma.
• the discharge tube was
• a stainless steel electrode comprised the top electrode of the plasma source.
• the top electrode had three coaxial cylindrical SS rings welded to the electrode to increase the total surface area to about 300 cm 2 and minimize sputtered metal from coating the chamber.
• the large surface area was found to be necessary during Langmuir probe measurements when the probe was biased close to Vp.
• a large grounded surface was then required to supply an adequate electron current, preventing an artificial increase of Vp.
• Argon gas, with a high purity, 99.999 % was fed into the discharge tube through a 1-mm diameter hole at the center of the top electrode.
• Plasma power at 13.56 MHz was supplied using a function generator (HEWLETT PACKARD ® Model 3325A) feeding a power amplifier (ENI Model A-500).
• the output of the amplified was connected to the coil via an L-type matching network. Forward and reflected powers were monitored by in-line Bird meters placed before the matching network. For typical continuous wave (cw) 300W Argon plasma at 14 mTorr, the reflected power was 1-2W. The actual power dissipated in the plasma is somewhat lower than the net power delivered to the matching box due to power losses.
• the RF pulse was amplitude- modulated by another function generator (BNC Model 645). Waveforms were monitored using a four-channel oscilloscope (TEKTRONIX® Model TDS 2024B).
• Base case conditions for pulsed plasma experiments were 120W time-average forward power, 8W reflected power, 10 kHz power modulation frequency, 20% duty cycle, 14 mTorr pressure, and 40 standard cubic centimeters per minute (seem) argon gas flow rate.
• the applied modulation frequency and duty cycle resulted in 20 ⁇ 8 (microsecond) plasma ON (active glow) time and 80 ⁇ 8 plasma OFF (afterglow) time, during the 100 ⁇ 8 period of a pulse.
• Figures 2 and 4 show examples of a timing sequence that is used to control plasma physics and chemistry.
• an approximately Is (second) continuous-wave main RF ICP is ignited by the tail-end of a low power auxiliary plasma, and provides reactants (e.g., CI) to form a chemisorbed layer.
• reactants e.g., CI
• the ion bombardment energy is too low ( ⁇ 10 eV) for any etching to occur.
• a mono-energetic ion flux to the substrate can be generated, as recently demonstrated in this laboratory.
• a pulse of positive DC voltage is applied to the boundary electrode, raising the plasma potential and "pushing" positive ions to surfaces of lower potential.
• a grounded substrate is bombarded with ions with energy equal to V DCI> as shown in Figures 7 and 8. Since control of the ion energy distribution is critical to effect chemical sputtering of the chemisorbed halogenated layer, without physical sputtering of the underlying substrate, this method of obtaining extremely narrow IEDs, and thus extreme selectivity, is an effective means to achieve ALET with monolayer accuracy.
• This pulsed- main-ICP with synchronous pulsed-immersed-electrode-bias-voltage period is long enough (e.g., 0.5 seconds) to sputter away the halogenated etch product layer.
• Net positive ion bombardment can cause a positive charge to build up on insulating substrates.
• any charged surfaces with potentials above ground are the first to receive an excess electron flux over the positive ion flux, bringing their potential back to the floating potential, which is near ground potential.
• a large negative DC bias may be applied to the counter-electrode, while a continuous wave ICP power is ON, as in Figures 3, 5, and 6. This negative voltage has no effect on V p .
• the resulting high energy ion bombardment of the counter-electrode generates secondary electrons that are accelerated to the full sheath potential.
• ballistic electrons have a low scattering cross section and bombard the substrate at nearly normal incidence, compensating positive charge at the bottom of even high aspect ratio insulating structures.
• the ballistic electrons can also have beneficial effects on the bulk plasma, such as the enhanced plasma density and lower bulk T e .
• Step 1 (lasting typically one second), the sample is exposed to a continuous-wave RF inductively-coupled plasma with the substrate at ground potential.
• the plasma is mostly inert gas with a very small amount ( ⁇ 1 ) of CI2.
• ⁇ 1 the energy of ions impacting the substrate would be less than the chemical sputtering threshold, so no etching would occur during step 1.
• the CI atoms do not etch p-type or moderately doped n-type Si at room temperature.
• CI atoms from dissociation of Cl 2 in the feed gas would allow a saturated layer of chlorinated products (e.g., SiCl x for Si etching) to form in about one second.
• step 2 lasting about 0.5 s, a pulsed main ICP would be used and positive DC bias pulses would be applied synchronously to the boundary electrode about 10 ⁇ 8 into the afterglow of each main ICP pulse, to chemically sputter the product layer.
• the bias in Step 2 could be a negative DC voltage applied to the (conducting) substrate electrode or a high frequency RF pulse or tailored pulse to the (insulating) substrate electrode under selected conditions.
• This step would be monitored by optical emission from etch products, providing fundamental information on chemical sputtering yields and a means of controlling the process. An etching rate of one monolayer in one to several seconds, i.e. quite practical for nanometer scale structures in future devices and much faster than conventional atomic layer etching based on pulsed gas and purge schemes.
• ion energy is chosen that is below the physical sputtering threshold but above the threshold for chemically-assisted sputtering. This regime provides very high selectivity combined with minimum damage, since etching would stop (self- limiting) after the chemisorbed layer of etch products is chemically sputtered away.
• Threshold values for Si are typically 10 - 25 eV under a variety of conditions.
• Langmuir probe A Langmuir probe (Smart Probe, Scientific Systems) was used to measure ion and electron densities (ni and ne), plasma potentials (VP), floating potentials, and electron energy probability functions (EEPF).
• the probe tip had a diameter of 0.19 mm and an exposed length of 40 mm.
• a compensation electrode and RF chokes minimized distortion of the current- voltage (TV) characteristic due to oscillations of the plasma potential. This was not an issue in the present system where, due to the Faraday shield, peak-to-peak plasma potential oscillations were only 1-2 volts.
• the probe was movable along the discharge tube axis to obtain spatially resolved measurements.
• Retarding field energy analyzer A retarding field energy analyzer (RFEA) was constructed to measure the energy distribution of ions passing through a grid on the grounded substrate stage.
• the RFEA was made of a stack of three nickel grids and a stainless steel current collector plate spaced 3 mm apart, as shown in Figure 6 insert.
• the top grid having 50% open with square holes 18 mm on a side, was attached to a grounded SS plate with a 0.3 mm pinhole in contact with the plasma. This grid prevented the plasma sheath from molding over the pinhole.
• the middle and bottom grids were each 85% open with square holes 293 mm on a side.
• the middle grid was biased with -30 V to repel electrons from the plasma, while the bottom grid was biased with a saw-tooth ramp voltage and served as an energy discriminator to measure the ion energy distribution (IED).
• a current amplifier (KEITHLEY ® model 427) was used to measure the ion current on the collector plate.
• a 20 Hz ramp voltage was applied to the discriminator grid using a pulse generator and a power amplifier (AVTECH AVR-3-PS-P-UHF and AV- 112AH-PS).
• the experiment was controlled through a Lab VIEW (NATIONAL INSTRUMENTS ®) program. Noise was reduced by averaging 5000 I-V characteristics resulting in "smooth" IEDs.
• the RFEA was differentially pumped
• Figure 9 shows ion and electron densities as a function of vertical position along the discharge tube axis, measured by the Langmuir probe (LP), for different pressures.
• Charge density reaches a maximum around the middle of the coil and increases with pressure.
• a maximum ion density of 1.5xl012/cm3 is reached for a pressure of 50 mTorr.
• ion-neutral collisions in the probe sheath will cause the positive ion density to be increasingly underestimated at higher pressures; hence the positive ion density could substantially exceed the value recorded at 50 mTorr.
• the electron and ion density are nearly equal for pressures of 3, 7, and 14 mTorr.
• the electron density was lower than the corresponding ion density. This was attributed to the fact that as the probe was biased near Vp, a large electron current was drawn out of the plasma. Apparently, the grounded surface of the boundary electrode in contact with the plasma was not high enough to compensate for the electron loss at these high densities.
• Optical emission spectroscopy for time-resolved detection of etching products: Optical emission spectroscopy can be used to monitor the time-dependence of etching products chemically sputtered from the surface during the energetic ion flux pulses. For Si ALET with chlorine, we anticipate that the emission from Si, SiCl and SiCL. products will be observed, as was found in pulsed laser-induced thermal desorption in Cl 2 plasmas. (Si and SiBr emissions were also found in HBr plasmas). For GaN etching, strong emission is expected from Ga and GaCl. If N2 is a primary product of GaN etching, then it can be easily detected in the plasma via N2 optical emission.
• Emission from e.g., SiCl provides a measure of the chemical sputtering yield as a function of the instantaneous CI coverage, as well as the total amount of material removed per ion pulse. This measurement can be used to control the etching rate in real time (e.g., the ion pulse durations could be adjusted to obtain a constant etching rate).
• Optical emission actinometry can be to measure absolute CI densities, as demonstrated previously in several ICP systems.
• In-situ laser-induced thermal desorption (LITD): In selected experiments, laser- induced thermal desorption are used to monitor instantaneous coverage of CI, Br and perhaps other surface species. This method can detect 1% of monolayer coverage with a time resolution of 10 ns (the laser pulse width) as the substrate is etching in a plasma as in Figure 10. Each laser pulse up to 80 or 5000 pulses/s with the available lasers rapidly heats the surface, resulting in a thermal desorption of typically half the Si-halide (CI or Br) layer formed in the plasma. The surface can thus be probed as a function of time during the chemisorption step and well as during the etching step.
• LITD In-situ laser-induced thermal desorption
• In-situ XPS and in-situ AFM/STM surface roughness measurements After plasma exposure, samples are transferred under vacuum to an ultrahigh vacuum chamber and analyzed by XPS. Angle-resolved measurements is carried out to measure the depth of penetration of reactants such as CI and Br and also to obtain a depth profile of Si-mono, di- and tri-halides, and the " ⁇ Si-" moiety, a Si with 3 bonds to Si and 1 dangling bond. On masked samples, electron shadowing is used to characterize the sidewalls that are exposed to glancing angle ion bombardment. These methods have been used with this system to characterize the surface after Si etching in Cl 2 and HBr plasmas.
• In-situ characterization of the sidewalls is particularly important in the case of GaN.
• XPS provide a wealth of information regarding any changes in surface stoichiometry as a function of ALET process parameters.
• An in-situ AFM-STM instrument allows atomic resolution measurements on processed surfaces without exposure to the atmosphere. Since rapid ALET offers atomic layer accuracy, it is important to avoid even sub-monolayer coverage by atmospheric contaminants, which can distort the experimental findings. These measurements will aid in identifying process parameters that minimize surface roughness after repeated ALET cycling, leading to etching with accuracy down to one monolayer per cycle.
• Figure 11 shows IEDs for 14 mTorr, 300W, and cw-Ar plasmas for different values of DC bias, applied continuously to the boundary electrode.
• the values of VP measured by the Langmuir probe at the location of the RFEA for each DC bias voltage are shown in Figure 11 by a vertical dashed line.
• Vp values are in excellent agreement with the peak energies of the IED.
• Vp is raised, shifting the IED to higher energies.
• negative DC bias there is an initial small drop in Vp, but it saturates as the applied bias becomes more negative.
• the peak of the IED shifts by 3, 7, and 11 eV for applied DC bias of 4, 8, and 12 V, respectively.
• the 1 V difference between the applied bias and the peak ion energy is probably due to a slight gradient of Vp.
• a negative DC bias is applied, the shift in the peak ion energy saturates at 4 V lower than without bias.
• Vp The shift in Vp with the application of a DC bias on the boundary electrode is readily understood.
• a positive bias drains electrons from the plasma raising Vp so that all but the highest energy electrons remain confined in the plasma.
• Vp With the application of a small negative bias (less than a few T e ) Vp becomes less positive as electron current to the boundary electrode is cut off. Larger negative bias on the boundary electrode causes negligible change in the ion current, hardly affecting Vp. The ion current saturates at large enough negative bias, assuming there is no perturbation of the plasma density or T e .
• Pulsed Plasmas To obtain nearly mono-energetic ion bombardment it may be desirable to reduce the energy spread of ions entering the sheath, as well as maintain a constant sheath potential. Since RF oscillations of the plasma potential are eliminated by the Faraday shield, the spread in the energy of ions entering the sheath scales with T e . Hence, lowering T e should reduce the energy spread. T e can be lowered by modulating the plasma power, such as pulsed plasma. When a DC bias is applied to the boundary electrode under these conditions, ions can be accelerated to a desired energy with a narrow energy spread.
• Figure 12 shows time resolved Langmuir probe measurements of electron temperature for different pressures.
• the T e increases rapidly after the plasma is turned ON, overshoots, and then reaches a quasi steady-state value.
• the steady-state T e decreases with increasing pressure, as expected.
• T e decreases at a progressively slower rate longer into the afterglow.
• T e decays faster at lower pressure.
• diffusion to the walls is the dominant cooling mechanism during the afterglow for electrons with energies below the lowest excited state (the 3 P 2 metastable state at 11.55 eV).
• Lower pressure results in faster diffusion rates, and therefore a faster decay of Te in the afterglow.
• Figure 13 shows IEDs under pulsed plasma conditions, when a DC bias was continuously applied to the boundary electrode. For each value of the DC bias, the IED has two peaks. The broader peaks at higher energy correspond to ions bombarding the substrate when the plasma is ON. The shape and energy
• the pressure can be chosen so that the low energy peak produces no etching.
• the DC bias can be chosen such that the high energy peak lies between the thresholds of etching the film and etching the substrate, assuming there is sufficient separation between these two thresholds.
• the fraction of ions under each peak can also be optimized by varying the duty cycle of the pulsed plasma and/or the length of time in the afterglow during which the DC bias is applied, as discussed next.
• IEDs in the afterglow were also measured with a synchronous DC bias (+24.4 V) applied to the boundary electrode for different start times (tb) and time windows (Atb).
• the pulsed plasma was generated with 120 W average power at 10 kHz and 20% duty cycle, 14mTorr, and 40 seem Ar flow rate. IEDs with the DC bias
• the bias starting time tb was varied while keeping a constant Atb of 50 ⁇ 8 or 15 ⁇ 8.
• the average power into the pulsed plasma was 120W.
• the biasing window is long, 50 ⁇ 8, compared to the T e decay time, -10 ⁇ 8, the biasing starting time hardly affects the ion energy distribution as send in Figure 16(a). This is because the average T e over these bias windows is low and roughly equal.
• the width of the IED diminishes progressively, as tb is shifted to later times in the afterglow. Again, the width of the IED correlates with T e during the corresponding biasing window.

Landscapes

• Engineering & Computer Science (AREA)
• Physics & Mathematics (AREA)
• Plasma & Fusion (AREA)
• Chemical & Material Sciences (AREA)
• Analytical Chemistry (AREA)
• Power Engineering (AREA)
• Drying Of Semiconductors (AREA)
• Plasma Technology (AREA)

Priority Applications (3)

Applications Claiming Priority (2)

Publications (2)

Family

ID=44121419

Family Applications (1)

Country Status (6)

Cited By (25)

Families Citing this family (569)