Classifications

• • 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
  • H01J37/32155—Frequency modulation
  • H01J37/32165—Plural frequencies
• • 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
  • H01J37/32431—Constructional details of the reactor
• • H—ELECTRICITY
  • H01—ELECTRIC ELEMENTS
  • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
  • H01L21/00—Processes or apparatus adapted for the manufacture or treatment of semiconductor or solid state devices or of parts thereof
  • H01L21/67—Apparatus specially adapted for handling semiconductor or electric solid state devices during manufacture or treatment thereof; Apparatus specially adapted for handling wafers during manufacture or treatment of semiconductor or electric solid state devices or components ; Apparatus not specifically provided for elsewhere
  • H01L21/67005—Apparatus not specifically provided for elsewhere
  • H01L21/67011—Apparatus for manufacture or treatment
  • H01L21/67017—Apparatus for fluid treatment
  • H01L21/67063—Apparatus for fluid treatment for etching
  • H01L21/67069—Apparatus for fluid treatment for etching for drying etching

Definitions

• the present invention generally relates to semiconductor substrate processing systems and, more particularly, to plasma generation and control using a dual frequency RF source.
• Plasma enhanced semiconductor processing chambers are widely used to manufacture integrated circuit devices.
• multiple radio frequency (RF) sources are utilized to form and control the plasma.
• RF radio frequency
• an RF source with a high frequency is typically used for plasma formation and ion dissociation.
• an RF source with a lower frequency is often used to modulate the plasma sheath in order to control the accumulation of a DO voltage (i.e., bias voltage) on the substrate.
• DO voltage i.e., bias voltage
• a method for controlling a plasma in a semiconductor substrate processing chamber includes the steps of supplying a first RF signal to a first electrode within the processing chamber at a first frequency selected to cause plasma sheath oscillation at the first frequency; and supplying a second RF signal from the source to the first electrode at a second frequency selected to cause plasma sheath oscillation at the second frequency, wherein the second frequency is different from the first frequency by a differential equal to a desired frequency selected to cause plasma sheath oscillation at the desired frequency.
• a method for controlling a plasma in a plasma enhanced semiconductor substrate processing chamber includes the steps of supplying, to a first electrode within the processing chamber, a first and a second RF signal at a first and a second frequency, respectively, selected to selected to cause a response much faster than a transit time for an ion to traverse a sheath of the plasma when accelerated from a bulk of the plasma; and controlling a frequency differential between the first and the second frequencies equal to a desired frequency selected to control an ion energy distribution within the plasma.
• Figure 1 depicts an exemplary block diagram of a plasma enhanced semiconductor processing chamber having a dual frequency RF source
• Figure 2 is a graph of the frequency spectrum of an input waveform on the electrode
• Figure 3 is a graph of the frequency spectrum of the sheath voltage
• Figure 4A is a graph of reflection coefficient magnitude versus frequency
• Figure 4B is a Smith chart for the match element model
• Figure 5 is a simplified schematic of one embodiment of a processing chamber having a dual frequency RF source
• FIG. 6 is a simplified schematic of another embodiment of a processing chamber having a dual frequency RF source.
• Figures 7A and B are graphs comparing the ion energy distribution of one embodiment of a process chamber having dual frequency source to another process chamber having separate frequency sources coupled to separate electrodes in the chamber.
• the present invention is a method and apparatus for forming and controlling characteristics of a plasma in a plasma enhanced semiconductor processing chamber using a dual frequency RF source.
• Plasma enhanced semiconductor processing chambers generally utilize power fed at two frequencies: a high frequency for plasma excitation and ion dissociation; and a low frequency for plasma sheath modulation.
• the present invention utilizes two high frequency inputs generated from a single RF source to generate the plasma.
• One or more characteristics of the plasma, for example, the sheath modulation is controlled by utilization of a wave packet phenomenon that creates a low frequency component in the plasma sheath equivalent to the difference in frequency between the two input signals.
• FIG. 1 depicts a simplified block diagram of a plasma enhanced semiconductor processing chamber having a dual frequency RF source.
• a plasma enhanced processing chamber 100 includes a chamber 102, a dual frequency RF power source 104, and a single matching circuit 106.
• the chamber 102 includes a powered electrode 108 that is coupled to the source 104 through the matching circuit 106 and a grounded electrode 110.
• the chamber 102 is otherwise similar to a conventional plasma enhanced processing chamber with the exception of the reduced RF infrastructure.
• the source 104 is an RF generator with dual frequency excitation.
• the source 104 is generally capable of generating two frequencies in the range of from about 100KHz to about 200MHz.
• the source 104 is also generally able to produce up to 5000W of either continuous or pulsed power. In one specific embodiment, the source produces both about 13MHz and about 14MHz, each frequency at about 3KW.
• the two frequencies generated by the source 104 generally have about the same magnitude at the output of the source 104 and are selected such that each of the individual frequencies control a defined subset of plasma characteristics, such as high frequency plasma excitation, ion dissociation, the ion energy distribution function, and the like.
• the frequencies are also selected to be in close proximity to each other to allow for common matching through the matching circuit 106, as discussed further below.
• the difference between the two selected frequencies produces a wave packet effect that may be utilized to control a second subset of plasma characteristics, such as low frequency plasma excitation, sheath voltage modulation, and the like.
• This wave packet effect is illustrated in Figures 2 and 3.
• the ratio between the magnitude of the two frequency signals generated by the source 104 may be varied in order to control the magnitude of the wave packet effect created by the difference between the two frequencies.
• the plasma characteristics controlled by the dual frequencies and the wave packet effect created from their difference may overlap (i.e., some or all of the characteristics controlled, such as plasma excitation, the ion energy distribution function, sheath modulation, and the like, may be at least partially controlled by both the dual frequencies provided by the source 104 and by the resulting wave packet effect created by the difference between the two frequencies.)
• the same or other plasma characteristics or subsets of plasma characteristics may be controlled by additional RF signals coupled to the plasma.
• a third subset of plasma characteristics may be controlled by an RF signal provided by another RF source coupled to another electrode disposed in the chamber, as discussed with reference to Figure 6, below.
• Figure 2 depicts the magnitude 210 of the Fourier components for the input waveform incident on the powered electrode 108 of the plasma chamber 102 as a function of frequency 220
• Figure 3 depicts a frequency analysis of the sheath voltage waveform 300, showing the amplitude 310 of the resulting frequency components in the sheath as a function of frequency 320.
• the frequency spectrum of the input waveform 200 includes two expected peaks 202, 204, corresponding to frequency 1 and frequency 2 being generated by the source 104.
• a low frequency term (peak 306) is generated in the sheath in addition to the primary drive frequencies (peaks 302 and 304) from the source 104 and their expected harmonics.
• the sheath has a non-linear characteristic curve that affects the RF signals in the same manner as an RF mixer diode affects input RF signals, i.e., an inner modulation frequency component is formed that equals the difference between the two input frequencies.
• the low frequency term (peak 306) is equal to the difference between the two frequencies generated by the source 104.
• the maximum difference in drive frequencies generated by source 104 is determined by the characteristics of the matching circuit 106. Particularly, the resonant bandwidth of the matching circuit 106 is centered between the drive frequencies.
• the matching circuit 106 must have a Q that defines a bandwidth that efficiently couples both frequencies to the electrode 108 and the plasma. As long as the two drive frequencies fall substantially within the bandwidth of the matching circuit, the RF infrastructure is capable of supporting the two separate frequencies. Frequency differences on the order of the low frequency range typically employed for such process enhancement are available with current match technology.
• Figure 4A shows a plot 400 of the reflection coefficient in terms of magnitude for a typical L-type match and depicts the frequency in MHz along an axis 402 versus a reflection coefficient magnitude in decibels along an axis 404.
• the frequencies selected should be such that the reflection coefficient magnitude is small and substantially equal, as depicted by dots 406 and 408, for frequencies 1 and 2 generated by the source 104.
• Figure 4B shows the location of the frequencies in real/imaginary space for a typical L-type match.
• the exemplary Smith chart 450 depicted in Figure 4B denotes a 50 ⁇ load such as is generally present in standard RF power sources.
• the frequencies should be selected such that they fall as close to the center 452 of the Smith chart 450 as possible, so as to provide reasonably low reflected power, as depicted by the exemplary frequencies 454, 456 in Figure 4B.
• the two frequencies fed from the source 104 to the powered electrode 108 may be selected to control the same plasma characteristic as is controlled by the inner modulation frequency component (i.e., the difference between the two primary frequencies tunes the same plasma parameter as the primary frequencies generated by the source 104).
• the ion energy distribution function may be controlled by both the primary frequencies and the inner modulation frequency component. That is, the ion energy distribution function may be controlled by both the selection of the primary frequencies and the control of the difference between the primary frequencies.
• the two frequencies fed from the source 104 to the electrode 108 cause the sheath to oscillate respectively at the two frequencies.
• the sheath oscillation also has a characteristic frequency at the difference between the two frequencies (i.e., due to the inner modulation frequency component).
• the ion reacts to the time average, or DC, component of the sheath and the amount of deflection in trajectory that an ion experiences because of the change in potential of the sheath due to RF oscillation is equal to the inverse of the sheath frequency (e.g., the higher the sheath oscillation frequency, the less modulation of the velocity of an ion traversing the sheath).
• the sheath oscillation at the frequency differential can add time variability to the final energy of the ion.
• the ion energy distribution function may be controlled by controlling oscillation frequencies in the sheath that straddle the ion transit time in the sheath (e.g., oscillation frequencies in the sheath are both greater than and less than the inverse of the ion transit time in the sheath).
• the primary frequencies may be selected to be greater than the inverse of the ion transit time.
• the difference between the primary frequencies may be selected to be less than the inverse of the ion transit time. Controlling both the primary frequencies and the difference between the primary frequencies allows control of the ion energy distribution function.
• the source 104 may generate two signals having primary frequencies of about 13.56 and about 13.86 MHz to the electrode 108. This causes the plasma sheath to oscillate at 13.56 and 13.86 MHz, respectively. In addition, the sheath will have a characteristic oscillation at 300 KHz due to the inner modulation frequency component generated at the difference between the two primary frequencies.
• the ion energy distribution function 700 (depicted in Figure 7A) for a process chamber having a dual frequency source providing primary frequencies of about 13.56 and about 13.86 MHz, as described above, is very similar to the ion energy distribution function 710 (depicted in Figure 7B) resulting from a process chamber having two separate RF sources.
• the dual-frequency source configuration of the present invention advantageously allows for similar control of the ion energy distribution function as compared to chambers having separate RF sources coupled to top and bottom electrodes in the process chamber, thereby significantly reducing the complexity and hardware cost associated with the process chamber.
• the relative magnitude of the primary frequencies and the inner modulation frequency component may further be utilized to control the ion energy distribution function.
• the relative current or the relative power of the two primary frequencies may be selectively provided to control the magnitude of the inner modulation frequency component. Specifically, when the two primary frequencies are applied with the same current, the magnitude of the inner modulation frequency component will be maximized. Reducing the current of one of the two primary frequencies with respect to the other will reduce the magnitude of the inner modulation frequency component until, at zero current applied ⁇ i.e., no second frequency), the inner modulation frequency component disappears.
• the relative magnitude of the inner modulation frequency component vis-a-vis the magnitude of the current or power of the primary frequencies adds further variability to the ion energy distribution and, therefore, allows control of the ion energy distribution function over a range of relative magnitudes.
• Examples of plasma enhanced semiconductor processing chambers that may be adapted to benefit from the present invention include, but are not limited to, the Decoupled Plasma Source (DPS ® , DPS ® II), EMAXTM, MXP ® , and ENABLERTM processing chambers, all available from Applied Materials, Inc. of Santa Clara, California.
• the EMAXTM processing chamber is described in United States Patent No. 6,113,731, issued September 5, 2000 to Shan et al.
• the MXP ® processing chamber is described in United States Patent No. 5,534,108, issued July 9, 1669 to Qian et al., and United States Patent No. 5,674,321 , issued October 7, 1997 to Pu et al.
• the ENABLERTM processing chamber is described in United States Patent No. 6,528,751, issued March 4, 2003 to Hoffman et al.
• FIG. 5 depicts a schematic diagram of one illustrative embodiment of an etch reactor 500 suitable for use with the present invention.
• the reactor 500 includes a process chamber 502 having a substrate support pedestal 512 within a conductive body (wall) 534, and a controller 536.
• the support pedestal 512 is coupled through the matching circuit 106 to the dual frequency RF power source 104.
• the controller 536 includes a central processing unit (CPU) 540, a memory 542, and support circuits 544 for the CPU 540 and facilitates control of the components of the process chamber 502 and, as such, of the etch process, as discussed below in further detail.
• CPU central processing unit
• the controller 536 may be one of any form of general-purpose computer processor that can be used in an industrial setting for controlling various chambers and sub-processors.
• the memory, or computer-readable medium, 542 of the CPU 540 may be one or more of readily available memory such as random access memory (RAM), read only memory (ROM), floppy disk, hard disk, or any other form of digital storage, local or remote.
• the support circuits 544 are coupled to the CPU 540 for supporting the processor in a conventional manner. These circuits include cache, power supplies, clock circuits, input/output circuitry and subsystems, and the like.
• the inventive method is generally stored in the memory 542 as a software routine.
• the software routine may also be stored and/or executed by a second CPU (not shown) that is remotely located from the hardware being controlled by the CPU 540.
• a substrate 514 is placed on the pedestal 512 and process gases are supplied from a gas panel 518 through entry ports 516 and form a gaseous mixture 546.
• the gaseous mixture 546 is ignited into a plasma 548 in the chamber 502 by applying power from the dual RF source 104 to the substrate support pedestal 512.
• the two frequencies generated by the source 104 are generally selected in the high frequency range in order to promote plasma excitation and ion dissociation.
• the two frequencies generated by the source 104 are further differentiated by a predetermined span equal to a particular low frequency desirable, for example, for modulating the plasma sheath.
• the chamber wall 534 is coupled to an electrical ground 552.
• the pressure within the interior of the chamber 502 is controlled using a throttle valve 550 and a vacuum pump 520.
• the temperature of the wall 534 is controlled using liquid- containing conduits (not shown) that run through the wall 534.
• etch chambers may be used to practice the invention, including chambers with remote plasma sources, microwave plasma chambers, electron cyclotron resonance (ECR) plasma chambers, and the like.
• ECR electron cyclotron resonance
• This technique is likewise not limited to etch but to any other plasma assisted process, including deposition, annealing, nitridation, implantation, and the like.
• Figure 6 depicts a schematic diagram of another illustrative embodiment of an etch reactor 600 suitable for use with the present invention.
• the reactor 600 includes a process chamber 602 having a substrate support pedestal 612 within a conductive body (wall) 634, and a controller 636.
• the controller 636 is similar to controller 536 described above with reference to Figure 5.
• the support pedestal 612 is coupled through a matching circuit 106 to a dual frequency RF power source 104. (The source 104 and matching circuit 106 are described above with respect to Figure 1.)
• the chamber 602 further includes a dielectric ceiling 610 with at least one inductive coil 638 disposed thereabove (two coils 638 are depicted in the embodiment of Figure 6).
• the inductive coils 638 are coupled to a second power source 662 through a second match element 660.
• the source 662 is generally a single RF generator capable of producing up to 3000W at a tunable frequency in a range of from about 5OkHz to about 13.56MHz.
• the frequency range given for the source 662 is the range generally used in processing chambers having an inductively coupled top source. Other ranges may be appropriate and may be used with the present invention in other types of processing chambers.
• the source in a chamber with a capacitively coupled top source, the source generally produces signals with frequencies as high as 200MHz.
• microwave sources may produce frequencies as high as 5GHz.
• a substrate 614 is placed on the pedestal 612 and process gases are supplied from a gas panel 618 through entry ports 616 and form a gaseous mixture 646.
• the gaseous mixture 646 is ignited into a plasma 648 in the chamber 602 by applying power from the source 662 to the inductive coils 638 and by applying power from the dual RF source 104 to the substrate support pedestal 612.
• the two frequencies generated by the source 104 are generally selected in the high frequency range in order to promote plasma excitation and ion dissociation. However, the power may not be high enough to accomplish this and the source 104 may be utilized mainly to bias the substrate 614.
• the two frequencies generated by the source 104 are further differentiated by a predetermined span equal to a particular low frequency desirable, for example, for modulating the plasma sheath.
• the utilization of the dual frequency RF source 104 in combination with the source 662 provides more process control variables to control processing of the semiconductor substrate 614 in the chamber 602.
• the chamber wall 634 is coupled to an electrical ground 652.
• the pressure within the interior of the chamber 602 is controlled using a throttle valve 650 and a vacuum pump 620.
• the temperature of the wall 634 is controlled using liquid- containing conduits (not shown) that run through the wall 634.

Landscapes

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

Priority Applications (1)

Applications Claiming Priority (4)

Publications (1)

Family

ID=37396865

Family Applications (1)

Country Status (5)

Cited By (1)

Families Citing this family (18)

Citations (2)

Family Cites Families (35)

• 2006
  • 2006-05-02 US US11/416,468 patent/US7510665B2/en not_active Expired - Lifetime
  • 2006-05-04 JP JP2008511180A patent/JP2008544480A/ja active Pending
  • 2006-05-04 WO PCT/US2006/017075 patent/WO2006121744A1/en not_active Ceased
  • 2006-05-04 KR KR1020077021009A patent/KR20070102623A/ko not_active Ceased
  • 2006-05-05 TW TW095116153A patent/TWI344321B/zh not_active IP Right Cessation

Patent Citations (2)

Also Published As

Similar Documents

Legal Events