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
  • H01J37/32917—Plasma diagnostics
  • H01J37/32935—Monitoring and controlling tubes by information coming from the object and/or discharge
• • 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
  • H10P95/00—Generic processes or apparatus for manufacture or treatments not covered by the other groups of this subclass
• • 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

Definitions

• the present invention relates in general to substrate manufacturing technologies and in particular to methods and apparatus for monitoring a process in a plasma processing system by measuring self-bias voltage.
• a substrate e.g., a semiconductor substrate or a glass panel such as one used in flat panel display manufacturing
• plasma is often employed.
• the substrate is divided into a plurality of dies, or rectangular areas, each of which will become an integrated circuit.
• the substrate is then processed in a series of steps in which materials are selectively removed (etching) and deposited (deposition) in order to form electrical components thereon.
• etching selectively removed
• deposition deposition
• a substrate is coated with a thin film of hardened emulsion (i.e., such as a photoresist mask) prior to etching.
• Areas of the hardened emulsion are then selectively removed, causing components of the underlying layer to become exposed.
• the substrate is then placed in a plasma processing chamber on a substrate support structure comprising a mono-polar or bi-polar electrode, called a chuck or pedestal.
• Appropriate etchant source are then flowed into the chamber and struck to form a plasma to etch exposed areas of the substrate.
• capacitively coupled plasma processing systems may be configured with a single or with two separate RF power sources.
• Source RF generated by source RF generator 134
• bias RF generated by bias RF generator 138
• bias RF is commonly used to control the DC bias and the ion bombardment energy.
• matching network 136 is coupled to source RF generator 134 and bias RF generator 138, that attempts to match the impedance of the RF power sources to that of plasma 110.
• matching network 136 may also include a V/I probe (not shown) that can measure the voltage and impedance of a current transmitted to plasma 110, as well as the ability to modify a generated plasma frequency in order to better optimize the plasma to process conditions.
• Lp set of gases is flowed into chamber 102 through an inlet in a top electrode 104 from gas distribution system 122.
• a cooling system 140 is coupled to electrostatic chuck 116 in order to achieve thermal equilibrium once trie plasma is ignited.
• the cooling system itself is usually comprised of a chiller that pumps a coolant through cavities in within the chuck, and helium gas pumped by pump 111 between the chuck and the substrate (e.g., backside He Flow).
• the helium gas In addition to removing the generated heat, the helium gas also allows the cooling system to rapidly control heat dissipation. That is, increasing helium pressure subsequently also increases the heat transfer rate.
• Most plasma processing systems are also controlled by sophisticated computers comprising operating software programs.
• manufacturing process parameters e.g., voltage, gas flow mix, gas flow rate, pressure, etc.
• dielectric layers are electrically connected by a conductive plug filling a via hole.
• an opening is formed in a dielectric layer, usually lined with a TaN or TiN barrier, and then subsequently filled with a conductive material (e.g., aluminum (Al), copper (Cu), etc.) that allows electrical contact between two sets of conductive patterns.
• a conductive material e.g., aluminum (Al), copper (Cu), etc.
• CMP chemical mechanical polishing
• a blanket layer of silicon nitride is then deposited to cap the copper.
• Contamination in particular, tends to present a substantial problem.
• the degree of contamination is usually dependent on the specific plasma process (e.g., chemistry, power, and temperature) and the initial surface condition of chamber. Since fully removing deposits may be time consuming, a plasma processing system chamber is generally only substantially ciliated UiraCffipM ⁇ bl ⁇ -icbnlsHiination levels reach unacceptable levels, when the plasma processing system must be opened to replace a consumable structure (e.g., edge ring, etc.), or as part of scheduled preventive maintenance (PM).
• a consumable structure e.g., edge ring, etc.
• One solution may be to create a simplified empirical model of the plasma processing system in order to sufficiently capture the behavior of the tool.
• creating an empirical model may be problematic.
• a modified non-operational plasma chamber may be analyzed in order to extract parameters for the simplified empirical.
• the individual components of a plasma processing system may be individually measured using a network analyzer.
• the invention relates, in one embodiment, in a plasma processing system, to a method for in-situ monitoring a process in a plasma processing system having a plasma processing chamber.
• the method includes positioning a substrate in the plasma processing chamber.
• the method also includes striking a plasma within, the plasma processing chamber Iv ⁇ iIe tM ' MM ⁇ fat ' eTKsW& ⁇ i&det' within the plasma processing chamber.
• the method further includes obtaining a measured self-bias voltage that exists after the plasma is struck, the measured self-bias voltage value having a first value when the plasma is absent and at least a second value different from the first value when the plasma is present.
• the method also includes correlating the measured self-bias voltage value with an attribute of the process, if the measured self-bias voltage value is outside of a predefined self-bias voltage value envelope.
• the invention relates, in one embodiment, in a plasma processing system, to an apparatus for in-situ monitoring a process in a plasma processing system having a plasma processing chamber.
• the apparatus includes a means of positioning a substrate in the plasma processing chamber.
• the apparatus further includes a means of striking a plasma within the plasma processing chamber while the substrate is disposed within the plasma processing chamber.
• the apparatus also includes a means of obtaining a measured self-bias voltage that exists after the plasma is struck, the measured self-bias voltage value having a first value when the plasma is absent and at least a second value different from the first value when the plasma is present. If the measured self-bias voltage value is outside of a predefined self-bias voltage value envelope, the apparatus further includes a means of correlating the measured self-bias voltage value with an attribute of the process.
• FIG. 1 shows a simplified diagram of a capacitively coupled plasma processing system
• FIG. 2 shows a simplified statistical process control diagram of a set of blanket oxide etches in a particular same plasma processing system, according to one embodiment of the invention
• FIG. 3 shows the simplified diagram of FIG. 2, with the addition of the backside
• FIG. 5 shows the simplified diagram of FIG. 2, with the addition of the measured impedance for 2 MHz at the V/I probe, according to one embodiment of the invention
• FIG. 6 shows the simplified diagram of FIG. 2, with the addition of the measured frequency for 27 MHz at the V/I probe, according to one embodiment of the invention
• FIG. 7 shows the simplified diagram of FIG. 2, with the addition of the measured impedance phase angle at the V/I probe, according to one embodiment of the invention.
• FIG. 8 shows a simplified diagram of a method for the in-situ monitoring of a process, according to one embodiment of the invention.
• an excursion represents a data point that is outside of an established statistical range or a value envelope. That is, an excursion may be a data point above a statistical upper control limit or below a statistical lower control limit. In a plasma process, any excursion that goes undetected or is not forestalled may place a significant amount of substrate material at risk.
• plasma parameters are expected to remain within a particular range or value envelope (i.e., a set of impedances for each plasma frequency, a set of phase angles for each plasma frequency, a particular frequency range for each plasma frequency, a self-bias voltage, etc.).
• This range is often 3 standard deviations (or 3 ⁇ ) of some target or base line.
• Standard deviation ( ⁇ ) is generally the square root of the variance. It is the most commonly used measure of spread. In general, if the mean and standard deviation of a normal distribution are known, it is possible to compute the percentile rank associated with any given score (i.e., data point, etc.). In a normal distribution, about 68% of the scores are within one standard deviation of the mean, about 95% of the scores are within two standards deviations of the mean and about 99% of the scores are within three standards deviations of the mean, ⁇ - ⁇ (X - ⁇ ) 2 /N (Equation 1) where X is a particular score, ⁇ is the mean, and N is the number of scores.
• plasma processing recipes are optimized for, and hence tend to be very sensitive to, the plasma parameters. Therefore, for a given problem in a plasma processing system, a substrate attribute excursion (i.e., improper etch rate, etc.) can be correlated to a plasma parameter excursion (i.e., impedance value greater than 3 ⁇ for a particular frequency, etc.). That is, a particular problem would also tend to cause a set of excursions in both the plasma as well as the substrate.
• Common plasma processing problems include chamber contamination, plasma structural damage and deterioration, gas pressure leak, gas flow mixture problem, chamber temperature out of specification, bad RF cable, improperly connected cable, etc.
• a correlation can be determined between an excursion in the impedance of an RF power source at a particular frequency and a substrate attribute excursion (e.g., improper photoresist etch rate, etc.).
• a correlation can be determined between an excursion of a frequency in a frequency-tuned plasma system and a substrate attribute excursion (e.g., improper photoresist etch rate, etc.).
• a substrate attribute excursion e.g., improper photoresist etch rate, etc.
• frequency-tuned plasma systems can modify a set of frequencies used to generate the plasma in order to minimize the reflected power during a process. As a result, the frequency changes as a response to the changes in plasma impedance.
• a correlation can be determined between an excursion in a phase angle of an RF power source at a particular frequency and a substrate attribute excursion (e.g., improper photoresist etch rate, etc.).
• a correlation can be determined between an excursion in a self-bias voltage and a substrate attribute excursion (e.g., improper photoresist etch rate, etc.).
• an electric field must be generated just in front of the substrate (e.g., between the substrate and the plasma) which will allow plasma ions of sufficient energy to bombard the substrate.
• self-bias voltage the greater the potential difference between it and the plasma discharge voltage, the greater the tendency of the substrate to attract plasma ions.
• the self-bias voltage must also have a substantially large potential difference to these surfaces. Subsequently, a problem that would tend to affect the plasma, and hence the substrate, would also tend to affect the self-bias voltage.
• plasma processing systems are often powered with some type of RF power source.
• RF power source Often, there is a source RF generator used to generate and control the plasma density, and a bias RF generator commonly used to control the plasma DC bias and the ion bombardment energy.
• These RF sources are commonly coupled to the plasma through a matching network that attempts to match the impedance of the RF power sources to that of plasma.
• matching network may also include a V/I probe that can measure voltage (V), current (I), phase angle ( ⁇ ) between the voltage (V) and current (I) of the plasma, impedance (Z), delivered power, forward power, reflected power, reactive power, reflection coefficient, etc.
• the matching network may also modify a generated plasma frequency within an established range value envelope in order to better optimize the plasma to process conditions.
• a plasma processing system that can modify a set of frequencies used to generate the plasma is generally referred to as a frequency-tuned plasma system.
• Delivered power can generally be derived as follows:
• Impedance a complex number
• V 0 the voltage at fundamental (peak voltage)
• I 0 the current at fundamental (peak current)
• R the real resistance
• j sqrt(-l) (the imaginary part of a complex number)
• X the complex reactance.
• Complex reactance is an expression of the extent to which an electronic component, stores and releases energy as the current and voltage fluctuate with each AC cycle of the generated signal with a angular frequency denoted by ⁇ .
• phase angle of the plasma impedance can be represented in the form of:
• FIG. 2 a simplified statistical process control diagram of a set of blanket oxide etches in a particular same plasma processing system over the course of a few weeks is shown, according to one embodiment of the invention.
• quality in a plasma processing system refers to conformance to requirements.
• Conformance generally refers to the degree to which a substrate meets pre-established requirements or specifications in a recipe, such as targets, tolerances, etc.
• any given plasma process may also include a degree of uncertainty, also known as variance.
• variance a degree of uncertainty
• a decrease in variance is often directly correlated to an corresponding increase in quality.
• Some causes of variance are considered normal or acceptable, and do not necessarily call for action. For example, slight differences in a manufactured substrate caused by running the same process on difference plasma processing systems. That is, in an attempt to match one plasma processing system to another, variations are almost certain to occur.
• Other causes of variance are out of the ordinary or special. They are not an expected part of the process and hence may require some type of corrective action. That is, they exceed the boundaries of normal variation. For example, moisture in a plasma chamber which can destroy a substrate.
• the target is a desired mean etch rate of about 110.52 nm/min, and tolerance refers to maintaining the etch rate within an upper control limit (ER UCL) of about 120.12 nm/min, and a lower control limit (ER LCL) of about 100.91 nm/min.
• ER UCL upper control limit
• ER LCL lower control limit
• Plot 202 reflects the etch rate of the blanket oxide in nanometers per minute (nm/min) over the course of several weeks.
• two excursion points may become apparent: 204 performed on 4/6/2004, and 206 performed on 4/9/2004.
• an excursion represents a data point that is outside of an established statistical range or value envelope, and may be caused by several factors (i.e., chamber contamination, plasma structural damage and deterioration, gas pressure leak, gas flow mixture problem, chamber temperature out of specification, bad RF cable, improperly connected cable, backside He flow, etc.).
• plot 202 reflects the etch rate of the blanket oxide in nanometers per minute (nm/min) over the course of several weeks.
• plot 208 reflects the corresponding measured backside He flow during each etch.
• both etch plot 202 and plot He flow plot 208 show excursions at 204. That is, as the He flow became reduced to about 33.5 SCCM, the etch rate also was substantially reduced to about 33.4 nm/min, substantially outside the 3 ⁇ lower control limit (LCL) of 100.91 nm/min.
• LCL 3 ⁇ lower control limit
• plot 202 reflects the etch rate of the blanket oxide in nanometers per minute (nm/min) over the course of several weeks.
• plot 402 reflects the corresponding measured impedance for 27 MHz.
• the desired target etch rate is about 110.52 nm/min, with an upper control limit (ER UCL) of about 120.12 nm/min and a lower control limit (ER LCL) of about 100.91 nm/min.
• the desired target impedance is about 3.88 Ohms, with an upper control limit (Z UCL) of about 4.02 Ohms and a lower control limit (Z LCL) of about 3.75 Ohms.
• Both etch plot 202 and the measured impedance for 27 MHz 402 show excursions both around 204 on 4/6/2004 and 206a-b on 4/9/2004. Hence, an excursion in the measured impedance (whether above the Z UCL or below the Z LCL) appears to be correlated to a substantial reduction in the etch rate below the E/R LCL (i.e., an attribute excursion).
• the inventor believes that factors that may substantially alter a plasma impedance, may also tend to cause substantial changes in substrate attributes, such as the etch rate. These factors may include the deterioration of chamber materials (e.g., electrode, confinement ring, etc.), excursion of gas flow, gas pressure, or temperature, changes in substrate types, changes in the chuck surface, problems with the RF generator, an RF connection, a bad RF cable, etc.
• chamber materials e.g., electrode, confinement ring, etc.
• excursion of gas flow e.g., gas pressure, or temperature
• changes in substrate types e.g., changes in the chuck surface
• problems with the RF generator e.g., an RF connection, a bad RF cable, etc.
• plot 202 reflects the etch rate of the blanket oxide in nanometers per minute (nm/min) over the course of several weeks.
• plot 502 reflects the corresponding measured impedance for 27 MHz.
• the desired target etch rate is about 110.52 nm/min, with an upper control limit (ER UCL) of about 120.12 nm/min and a lower control limit (ER LCL) of about 100.91 nm/min.
• the desired target impedance is about 145.73 Ohms, with an upper control limit (Z UCL) of about 149.16 Ohms and a lower control limit (Z LCL) of about 142.29 Ohms.
• ' 'fftOl ' Sf ill S G iotnilci pGE2lf2 and the measured impedance for 2 MHz 402 show excursions both around 204a-b on 4/6/2004 and 206 on 4/9/2004.
• an excursion in the measured impedance appears to be correlated to a substantial reduction in the etch rate below the E/R LCL (i.e., an attribute excursion).
• FIG. 6 the simplified diagram of FIG. 2 is shown, with the addition of the measured frequency for 27 MHz at the V/I probe is shown, according to one embodiment of the invention.
• frequency-tuned plasma systems can modify a set of frequencies used to generate the plasma in order to minimize the reflected power during a process. As a result, the frequency changes as a response to the changes in plasma impedance.
• plot 202 reflects the etch rate of the blanket oxide in nanometers per minute (nm/min) over the course of several weeks.
• plot 602 reflects the corresponding measured frequency for 27 MHz.
• the desired target etch rate is about 110.52 nm/min, with an upper control limit (ER UCL) of about 120.12 nm/min and a lower control limit (ER LCL) of about 100.91 nm/min.
• the desired target frequency for 27 MHz is about 27 Al '680 MHz, with an upper control limit (FREQ UCL) of about 27.52331 MHz and a lower control limit (FREQ LCL) of about 27.43029 MHz..
• Both etch plot 202 and the measured frequency for 27 MHz 602 show excursions both around point 204a-b at 4/6/2004 and points 206 at around 4/9/2004.
• an excursion is defined as a point beyond 3 standard deviations (3 ⁇ ) of the plot mean.
• an excursion in the measured frequency appears to be correlated to a substantial reduction in the etch rate below the E/R LCL (i.e., an attribute excursion).
• plot 202 reflects the etch rate of the blanket oxide in nanometers per minute (nm/min) over the course of several weeks.
• plot 702 reflects the corresponding measured phase angle for impedance.
• the desired target etch rate is about 110.52 nm/min, with an upper control limit (ER UCL) of about 120.12 nm/min and a lower control limit (ER LCL) of about 100.91 nm/min.
• the desired target of the measured impedance phase angle is about (ANGLE UCL) of about -58.17°, and a lower control limit (ANGLE LCL) of about -61.16°.
• Both etch plot 202 and the measured phase angle 702 show excursions both around point 204 at 4/6/2004 and point 206a-b at around 4/9/2004.
• an excursion is defined as a point beyond 3 standard deviations (3 ⁇ ) of the plot mean.
• an excursion in the measured phase angle appears to be correlated to a substantial reduction in the etch rate below the E/R LCL (i.e., an attribute excursion).
• FIG. 8 a simplified diagram is shown of a method for the in- situ monitoring of a process in a plasma processing system having a plasma processing chamber, according to one embodiment of the invention.
• a substrate is positioned in the plasma processing chamber, at step 802.
• a plasma is struck within the plasma processing chamber while the substrate is disposed within the plasma processing chamber, at step 804.
• a measured self-bias voltage that exists after the plasma is struck is then obtained, the measured self-bias voltage value having a first value when the plasma is absent and at least a second value different from the first value when the plasma is present, at step 806.
• the measured self-bias voltage value is outside of a predefined self-bias voltage value envelope, at step 808, then the measured self-bias voltage value is correlated with an attribute of the process, at step 810. If not, then the measured self-bias voltage value is not correlated with an attribute of the process, at step 812.
• Advantages of the invention include methods and apparatus for monitoring a process in a plasma processing system by measuring self-bias voltage. Additional advantages include the use of a substantially reliable signal that can be used for diagnostics or monitoring purposes.

Landscapes

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

Priority Applications (3)

Applications Claiming Priority (2)

Publications (2)

Family

ID=36097829

Family Applications (1)

Country Status (6)

Cited By (2)

Families Citing this family (54)

Family Cites Families (19)

• 2004
  • 2004-09-27 US US10/951,553 patent/US7323116B2/en not_active Expired - Lifetime
• 2005
  • 2005-09-23 KR KR1020077009422A patent/KR101164828B1/ko not_active Expired - Fee Related
  • 2005-09-23 WO PCT/US2005/034227 patent/WO2006036821A2/en not_active Ceased
  • 2005-09-23 JP JP2007533669A patent/JP5057980B2/ja not_active Expired - Lifetime
  • 2005-09-23 TW TW094133150A patent/TWI398626B/zh not_active IP Right Cessation
  • 2005-09-23 CN CN2005800397621A patent/CN101088147B/zh not_active Expired - Lifetime

Also Published As

Similar Documents

Legal Events