The present invention relates to a circuit and a method for processing an output signal from an eddy current sensor.
Eddy current, sensors are used for applications such as Mm thickness measurement and displacement measurement. Hereinafter, film thickness measurement will be taken as an example to describe an eddy current sensor. An eddy current sensor for film thickness measurement is used in a semiconductor device fabrication step (polishing step), for example. In the polishing step, the eddy current sensor is used as follows. As the level of integration in semiconductor devices advances, circuit interconnects are becoming finer, and the spacing between interconnects is becoming narrower. Accordingly, it is necessary to planarize the surface of a semi conductor wafer, which is a polishing target. It is therefore a common practice to polish the wafer by a polishing apparatus as a way of planarization.
A polishing apparatus is provided with a polishing table for holding a polishing pad for polishing the polishing target and a top ring for holding and pressing the polishing target against the polishing pad. The polishing table and the top ring are each rotated by a driving unit (a motor, for example). A liquid containing: a polishing agent (slurry) is made to flow onto the polishing pad, and by pressing the polishing target held by the top ring against the polishing pad, the polishing target is polished.
If the polishing of the polishing target by the polishing apparatus is insufficient, circuits will not be insulated front each other, possibly leading to a short circuit. On the other hand, other problems may occur if the polishing target is over-polished, such as a rise in resistance values because of a decrease in the cross-sectional area of the wiring, or the wiring itself may be completely removed and the circuit itself may not be formed. Consequently, there is a demand to detect an optimal polishing endpoint in a polishing apparatus.
Such technology is described in Japanese Patent Laid-Open No. 2011-23579. In this technology, an eddy current sensor using three coils is used to detect a polishing endpoint. As illustrated in FIG. 5 of Japanese Patent Laid-Open No. 2011-23579, among the three coils, a detecting coil and a dummy coil form a series circuit, both ends of which are connected to a resistance bridge circuit including a variable resistance. By adjusting the balance with the resistance bridge circuit, it is possible to adjust the zero point such that, the output of the resistance bridge circuit goes to zero when the film thickness is zero. The output of the resistance bridge circuit is input into a synchronous detector circuit, as illustrated in FIG. 6 of Japanese Patent Laid-Open No. 2011-23579. The synchronous detector circuit extracts a resistance component. (R), a reactance component (X), an amplitude output (Z), and a phase output (tan−1R/X) associated with changes in the film thickness from the input signal. With regard to a detection method using a bridge circuit of the related art, the magnitude of adjustment to the resistance value when adjusting the zero point is extremely small, compared to the magnitude of the entire resistance value forming the bridge circuit. As a result, the magnitude of a temperature change in the entire resistance value is non-negligible compared to the magnitude of the resistance value adjustment when adjusting the zero point. Because of change, in the resistance value due to temperature change, change in the stray capacitance of the resistance, and the tike, the properties of the bridge circuit are highly susceptible to the influence of changes in the surrounding resistance environment. As a result, there is a problem in that the zero point described above shifts easily, and the accuracy of measuring the film thickness fails.
Also, with regard to the method of measuring the film thickness using an eddy current sensor of the related art, the signal output by the eddy current sensor is noisy. For this reason, when detecting tiny interconnects (Cu interconnects, for example), the output signal itself is small, and in some cases the output signal is buried in the noise and the film thickness cannot be measured. Additionally, in the cases where the magnitude of the output signal is somewhat larger than, the noise, the film thickness is measurable, but there is still the problem of low accuracy of the film thickness measurement.
One aspect of the present invention has been devised in order to address such issues, and an object thereof is to provide a circuit and a method for processing an output signal from an eddy current sensor that is less susceptible to the influence of changes in the surrounding environment compared to the related art.
Another aspect of the present invention has been devised in order to address other issues, and an object thereof is to provide a circuit and a method for processing an output signal from an eddy current sensor in which the signal-to-noise ratio (S/N) is improved over the related technology.
To address the above problems, Aspect 1 adopts a configuration of an eddy-current sensor assembly comprising: an eddy current sensor; and an output signal processing circuit, that processes an output signal of the eddy current sensor, wherein the output signal processing circuit includes a mixer circuit that accepts the output signal and a signal of a predetermined frequency as input, multiplies the two input signals, and outputs an output signal obtained by the multiplication, and a low-pass filter that accepts the output signal output by the mixer circuit as input, cuts a high-frequency signal included in the output signal received as input, and outputs at least a direct-current (DC) signal.
Aspect 2 adopts a configuration of an eddy current sensor output signal processing circuit that processes first and second output signals output by an eddy current sensor including first and second coils that respectively output the first and second output signals, the output signal processing circuit comprising: a first mixer circuit that accepts the first, output signal and a signal of a predetermined frequency as input, multiplies the two input signals, and outputs an output signal obtained by the multiplication; a first low-pass filter that accepts the output signal output by the first mixer circuit as input, cuts a high-frequency signal included in the output signal received as input, and outputs at least a DC signal; a second mixer circuit that accepts the second output signal and the signal of the predetermined frequency as input, multiplies the two input signals, and outputs an output signal obtained by the multiplication; a second low-pass filter that accepts the output signal output by the second mixer circuit as input, cuts a high-frequency signal included in the output signal received as input, and outputs at least a DC signal; and a first subtractor circuit that accepts the DC signal output by the first low-pass filter and the DC signal output by the second low-pass filter as input, obtains a difference between the two input DC signals, and outputs the obtained difference.
Aspect 3 adopts a configuration of the eddy current sensor output signal processing circuit according to Aspect 2, wherein the output signal processing circuit includes a first adjustment circuit that accepts the DC signal output by the first low-pass filter as input, adjusts a magnitude of an amplitude of the input DC signal, and outputs an adjusted DC signal, and the first subtractor circuit accepts the DC signal output by the first adjustment circuit and the DC signal output by the second low-pass filter as input, obtains a difference between the two input DC signals, and outputs the obtained difference.
Aspect 4 adopts a configuration of the eddy current sensor output signal processing circuit according to Aspect 2, wherein the output signal processing circuit, includes a first adjustment-circuit that accepts the DC signal output by the first low-pass filter as input, adjusts a magnitude of an amplitude of the input DC signal, and outputs an adjusted. DC signal, and a second adjustment circuit that accepts the DC signal output by the second low-pass filter as input, adjusts a magnitude of an amplitude of the input DC signal, and outputs an adjusted DC signal, and the first subtractor circuit accepts the DC signal output by the first adjustment circuit and the DC signal output by the second adjustment circuit as input, obtains a difference between the two input. DC signals, and outputs the obtained difference.
Aspect 5 adopts a configuration of the eddy current sensor output signal processing circuit according to any one of Aspects 2 to 4, wherein the eddy current sensor includes third and fourth coils that respectively output third and fourth output signals, and the output signal processing circuit includes a third mixer circuit that accepts the third output signal and the signal of the predetermined frequency as input, multiplies the two input signals, and outputs an output signal obtained by the multiplication, a third low-pass filter that accepts the output signal output by the third mixer circuit as input, cuts a high-frequency signal included in the output signal received as input, and outputs at least a DC signal a fourth mixer circuit that accepts the fourth output signal and the signal of the predetermined, frequency as input, multiplies the two input signals, and outputs an output signal obtained by the multiplication, a fourth low-pass filter that, accepts the output, signal output, by the fourth mixer circuit as input, cuts a high-frequency signal included in the output signal received as input, and outputs at least a DC signal, a second subtracter circuit that accepts the DC signal output by the third low-pass filter and the DC signal output by the fourth low-pass filter as input, obtains a difference between the two input DC signals, and outputs the obtained difference, and an adder circuit that accepts the difference output by the first subtracter circuit and the difference output by the second subtractor circuit as input, obtains a sum of the two input differences or a difference between the two input differences, and outputs the obtained sum or difference.
Aspect 6 adopts a configuration of an eddy current sensor output signal processing method comprising: inputting an output, signal of an eddy current sensor and a signal of a predetermined frequency into a mixer circuit, multiplying the two input signals with the mixer circuit, and outputting an output signal obtained by the multiplication, and inputting the output signal output by the mixer circuit into a low-pass filter, cutting a high-frequency signal, and outputting at least a DC signal.
Aspect 7 adopts a configuration of an eddy current sensor output signal processing method that, processes first and second output signals output by an eddy current sensor including first and second coils thru respectively output first and second output signals, the output signal processing method comprising: inputting the first output signal and a signal of a predetermined frequency into a first mixer circuit, multiplying the two input signals with the first mixer circuit and outputting an output signal obtained by the multiplication; inputting the output signal output by the first mixer circuit into a first low-pass filter, cutting a high-frequency signal included in the output signal with the first low-pass filter, and outputting at least a DC signal; inputting the second output signal and the signal of the predetermined-frequency into a second mixer circuit, multiplying the two input signals with the second mixer circuit, and outputting an output signal obtained by the multiplication; inputting the output signal output by the second mixer circuit into a second low-pass filter, cutting a high-frequency signal included in the output signal with the second low-pass filter, and outputting at least a DC signal; and inputting the DC signal output by the first low-pass filter and the DC signal output by the second low-pass filter into a first subtractor circuit, obtaining a difference between the two input DC signals with the first subtractor circuit, and outputting the obtained difference.
Aspect 8 adopts a configuration of an eddy current sensor output signal processing method wherein the eddy current sensor includes third and fourth coils that respectively output third and fourth output signals, and the output signal processing method comprises; inputting the third output signal and the signal of the predetermined frequency into a third mixer circuit, multiplying the two input signals with the third mixer circuit, and outputting an output signal obtained by the multiplication; inputting the output signal of the third mixer circuit into a third low-pass filter, cutting a high-frequency signal included in the output signal with the third low-pass filter, and outputting at least a DC signal; inputting the fourth output signal and the signal of the predetermined frequency into a fourth mixer circuit, multiplying the two input signals with the fourth mixer circuit, and outputting an output signal obtained by the multiplication; inputting the output signal of the fourth mixer circuit into a fourth, low-pass filter, cutting a high-frequency signal included in the output signal with the fourth low-pass filter, and outputting at least a DC signal; inputting the DC signal output by the third low-pass filter and the DC signal output by the fourth low-pass filter into a second subtractor circuit, obtaining a difference between the two input DC signals with the second subtractor circuit, and outputting the obtained difference; and inputting the difference output by the first subtracter circuit and the difference output by the second subtractor circuit into an adder circuit, obtaining a sum of the two differences or a difference between the two differences with the adder circuit, and outputting the obtained sum or difference.
Aspect 9 adopts a configuration of a polishing apparatus comprising: a polishing table to which a polishing pad is attached for polishing a substrate; a driving unit configured to rotationally drive the polishing table; a holding unit configured to hold the substrate and press the substrate against the polishing pad; the eddy current sensor disposed inside the polishing table and configured to detect an eddy current formed in a conductive film formed on the substrate in association with the rotation of the polishing table; the eddy current sensor output signal processing circuit according to any one of Aspects 2 to 5; and an endpoint detection controller configured to compute film thickness data about the substrate from the output of the output signal processing circuit.
To address the above problems, Aspect 10 adopts a configuration of an eddy current sensor assembly comprising: an eddy current sensor including an exciting coil capable of accepting an excitation, signal and generating a magnetic field, and a detecting coil capable of detecting the magnetic field and outputting a detection signal; and an output signal processing circuit that processes the detection signal, wherein the output signal processing circuit includes a generator circuit capable of generating a noise reduction signal for reducing noise from the excitation signal or the detection signal, and an adder circuit capable of adding the noise reduction signal generated in the generator circuit to the detection signal to generate a noise-reduced signal in which noise included in the detection signal is reduced.
Aspect 11 adopts a configuration of an eddy current sensor output signal processing circuit that processes a first detection signal and a first dummy signal output by as eddy current sensor including a first exciting coil capable of accepting an excitation signal as input and generating a first magnetic field, a first detecting coil capable of detecting the first magnetic field and outputting the first detection signal, and a first dummy coil capable of detecting the first magnetic field and outputting the first dummy signal, the output signal processing circuit comprising: a first resistance bridge circuit capable of outputting a difference between the first detection, signal and the first dummy signal as a first difference signal; a first generator circuit capable of generating a first noise reduction signal for reducing noise from any of the excitation signal, the first detection signal, the first dummy signal and the first difference signal; and a first adder circuit capable of adding the first noise reduction signal generated in the first generator circuit to the first difference signal to generate a first noise-reduced signal in which noise included in the first difference signal is reduced.
Aspect 12 adopts a configuration of the eddy current sensor output signal processing circuit according to Aspect 11, wherein the eddy current sensor includes a second exerting coil capable of accepting the excitation signal as input and generating a second magnetic field, a second detecting coil capable of detecting the first magnetic field and the second magnetic field, and outputting a second defection signal, and a second dummy coil capable of detecting the first magnetic field and the second magnetic field, and outputting a second dummy signal, the first detecting coil is capable of detecting the first magnetic field and the second magnetic field to output the first detection signal the first dummy coil is capable of detecting the first magnetic field and the second magnetic field to output the first dummy signal, and the output signal processing circuit, includes a second resistance bridge circuit capable of outputting a difference between the second detection signal and the second dummy signal as a second difference signal, a second generator circuit capable of generating a second noise reduction signal for reducing noise from any of the excitation signal, the second detection signal the second dummy signal, and the second difference signal, a second adder circuit capable of adding the second noise reduction signal generated in the second generator circuit to the second difference signal to generate a second noise-reduced signal in which noise included, in the second difference signal, is reduced, and a third adder circuit capable of adding together the first noise-reduced signal and the second noise-reduced signal.
Aspect 13 adopts a configuration of an eddy current, sensor output signal processing circuit that processes first and second detection signals and first and second dummy signals output by an eddy current sensor including first and second exciting coils capable of accepting an excitation signal as input and respectively generating first and second magnetic fields, first and second detecting coils capable of detecting the first magnetic field and the second magnetic field, and respectively outputting the first and second detection signals, and first and second dummy coils capable of detecting the first magnetic field and the second magnetic field, and respectively outputting the first and second dummy signals, the output signal processing circuit comprising: a first resistance bridge circuit capable of outputting a difference between the first detection signal and the first dummy signal, as a first difference signal; a second resistance bridge circuit capable of outputting a difference between the second detection signal and the second dummy signal as a second difference signal, a third adder circuit capable of adding together the first difference signal and the second difference signal.
Aspect 14 adopts a configuration of an eddy current sensor output si goal processing method that processes a detection signal output by an eddy current sensor including an exciting coil capable of accepting an excitation signal as input and generating a magnetic field, and a detecting coil capable of detecting the magnetic field and generating the detection signal, the output signal processing method comprising: generating a noise reduction signal for reducing noise from the excitation signal or the detection signal; and adding the generated noise reduction signal to the detection signal to generate a noise-reduced signal in which noise included, in the detection signal is reduced.
Aspect 15 adopts a configuration of an eddy current sensor output signal processing method that processes a first detection signal and a first dummy signal output by an eddy current sensor including a first exciting coil capable of accepting an excitation signal as input and generating a first magnetic field, a first detecting coil capable of detecting the first magnetic field and outputting the first detection, signal, and a first dummy coil capable of detecting the first magnetic field and outputting the first dummy signal, the output signal processing method comprising: outputting a difference between the first detection, signal and the first dummy signal as a first difference signal; generating a first noise reduction signal for reducing noise from any of the excitation signal, the first detection signal, the first dummy signal, and the first difference signal; and adding the generated first noise reduction signal to the first difference signal to generate a first noise-reduced signal in which noise included in the first difference signal is reduced.
Aspect 16 adopts a configuration of the eddy current sensor output signal processing method according to Aspect 15, wherein the eddy current sensor includes a second exciting coil capable of accepting the excitation signal as input and generating a second magnetic field, a second detecting coil capable of detecting the first magnetic field and the second magnetic Held, and outputting a second detection signal, and a second dummy coil capable of detecting the first magnetic field and the second magnetic field, and outputting a second dummy signal, the first detecting coil is capable of detecting the first magnetic field and the second magnetic field to output the first detection signal, the first dummy coil is capable of detecting the first magnetic field and the second magnetic field to output the first dummy signal, and the output signal processing method comprises: outputting a difference between the second detection signal and the second dummy signal as a second difference signal; generating a second noise reduction signal for reducing noise from any of the excitation signal, the second detection signal, the second dummy signal, and the second difference signal; adding the generated second noise reduction signal to the second difference signal to generate a second noise-reduced signal in which noise included in the second difference signal is reduced; and adding together the first noise-reduced signal and the second noise-reduced signal.
Aspect 17 adopts a configuration of an eddy current sensor output signal processing method that processes first and second detection signals and first and second dummy signals output by an eddy current sensor including first and second exciting coils capable of accepting an excitation, signal as input and respectively generating first and second magnetic fields, first and second detecting coils capable of detecting the first magnetic field and the second magnetic field, and respectively outputting the first and second detection, signals, and first and second dummy coils capable of detecting the first magnetic field and the second magnetic field, and respectively outputting the first and second dummy signals, the output signal processing method comprising: outputting a difference between the first detection signal and the first dummy signal as a first difference signal; outputting a difference between the second detection signal and the second dummy signal as a second difference signal; adding together the first difference signal and the second difference signal.
Hereinafter, embodiments of the present invention will be described with reference to the drawings. Note that in the following embodiments, identical or corresponding members are denoted with the same signs, and a duplicate description may be omitted. Also, features indicated in each embodiment are also applicable to other embodiments insofar as the embodiments do not contradict each other,
The polishing table 100 is coupled to a motor (not illustrated) that acts as a driving unit disposed underneath through a table spindle 100a, and is capable of rotating about the table spindle 100a. A polishing pad 101 is affixed to the top face of the polishing table 100, and the surface 101a of the polishing pad 101 forms a polishing surface that polishes a semiconductor wafer WH. A polishing liquid supply nozzle 102 is installed above the polishing table 100, such that a polishing liquid Q is supplied onto the polishing pad 101 on the polishing table 100 by the polishing liquid supply nozzle 102, As illustrated in
The top ring 1 basically includes a top ring body 142 that presses the semiconductor wafer WH against the surface 101a and a retainer ring 143 that holds the peripheral edges of the semiconductor wafer WH such, that the semiconductor wafer WH does not fly off the top sing.
The top ring 1 is connected to a top ring shaft 111, and the top ring shaft 111 is configured to move up and down with respect to a top ring head 110 by a raising/lowering mechanism 124. By the up and down movement of the top ring shaft 111, the entire top ring 1 is raised or lowered and positioned with respect to the top ring head 110. Note that a rotary joint 125 is attached to the top end of the top ring shaft 111.
The raising/lowering mechanism 124 that moves the top ring shaft 111 and the top ring 1 up and down is provided with a bridge 128 that rotatably supports the top ring shaft 111 through a bearing 126, a ball screw 132 attached to the bridge 128, a support stand 129 supported by a support column 130, and a servo motor 138 provided on the support stand 129. The support stand 129 that supports the servo motor 138 is secured to the top ring head 110 through the support column 130.
The ball screw 132 is provided with a screw shaft 132a coupled to the servo motor 138 and a nut 132b with which the screw shaft 132a engages. The top ring shaft 1H is configured to move up and down as one with the bridge 128. Consequently, when the servo motor 138 is driven, the bridge 128 moves up and down through the hall screw 132, thereby causing the top ring shaft 111 and the top ring 1 to move up and down.
Additionally, the top ring shaft 111 is coupled to a rotating cylinder 112 through a key (not illustrated). The rotating cylinder 112 is provided with a timing pulley 113 on the outer periphery thereof. A top ring motor 114 is secured to the top ring head 110, and the timing pulley 113 is connected to a timing pulley 116 provided in the top ring motor 114 through the timing belt 115, Consequently, by rotationally driving the top ring motor 114, the rotating cylinder 112 and the top ring shaft 111 rotate as one through, the timing pulley 116, the timing belt 115, and toe timing pulley 113, and the top ring 1 rotates. Mote that the top nog head 110 is supported by a top ring head shaft 117 rotationally supported by a frame (not illustrated).
In the polishing apparatus configured as illustrated in
Next, the eddy current sensor 50 provided in the polishing apparatus according to the present invention will be described in further detail using the attached drawings.
The eddy current sensor 50 is disposed near the metal film (or conductive film) to be detected at a distance of approximately 1.0 mm to 4.0 mm for example. The coil is normally wound around a magnetic material such as ferrite (not illustrated). The eddy current sensor 50 may also be an air-core coil.
The type of signal detection by the eddy current sensor is referred to as the impedance type, in which an eddy current is generated in the metal film (or conductive film) mf to cause a change in the impedance, and the metal film (or conductive film) is detected from the impedance change. In other words, with, the impedance type, changing the eddy current I2 causes the impedance Z to change in the equivalent circuit illustrated in
In an eddy current sensor of the impedance type, it is possible to extract signal outputs X and Y, the phase, and the combined impedance Z (=X+iY). From the impedance components X, Y, and the like, measurement information about the thickness of the metal film (or conductive film) of Cu, Al, Au, or W is obtained. As illustrated in
For the frequency of the eddy current sensor, a single radio wave, AM-modulated radio waves, the sweep output from a function generator, or the like can be used, and it is preferable to select an oscillating frequency and a modulation method with good sensitivity to match the type of metal film.
Hereinafter, an eddy current sensor of the impedance type will be described specifically. The AC signal, source 52 is an oscillator of a fixed frequency approximately from 2 MHz to 3.0 MHz, for which a quartz oscillator is used for example. Additionally, a current I1 flows through the eddy current sensor 50 due to an AC voltage supplied by the AC signal source 52. By causing a current toflow through the eddy current sensor 50 positioned near the metal, film (or conductive film) mf, the magnetic flux links with the metal film (or conductive film) mf toform a mutual inductance M between the two, and an eddy current I2 flows through the metal film (or conductive film) mf Here, R1 is the equivalent resistance on the primary side that includes the eddy current sensor, and L1 is the self-inductance on the primary side that similarly includes the eddy current sensor. On the metal film (or conductive film) mf side, R2 is the equivalent resistance corresponding to eddy current loss, and L2 is the self-inductance thereof. The impedance 2 seen on the eddy current sensor side from terminals a and b of the AC signal source 52 changes depending on the magnitude of the eddy current loss formed, in the metal film (or conductive film) mf.
The eddy current sensor 50 includes an exciting coil for forming an eddy current in the metal film (or conductive film) on the semiconductor wafer WH, and a detecting coil that detects the generated eddy current. For example, the exciting coil and the detecting coil are disposed in the axial direction of a cylindrical ferrite core. The exciting coil is connected to the AC signal source 52, The exciting coil forms an eddy current in the metal film (or conductive film) mf on the semiconductor wafer WH disposed near the eddy current sensor 50 due to a magnetic field formed by the voltage supplied by the AC signal source 52, The detecting coil is disposed on the upper side (the metal flint (or conductive film) side) of the ferrite core, and detects the magnetic field generated by the eddy current formed in the metal film (or conductive film). In
The mixer circuit 178 is an analog multiplier or a digital multiplier. The mixer circuit 178 is a circuit that accepts two voltage signals having the same (or different) frequency components f1 and f2, and performs a multiplication operation, on both signals. The mixer circuit 178 is also referred to by terms such as mixer, mixing circuit, frequency mixer, frequency mixing circuit, MIX, frequency transducer, frequency conversion circuit, and frequency converter. The purpose of the mixer circuit is to multiply two input, signals and extract a sum or a difference of respective frequencies of the two signals (perform frequency conversion), in the case where the respective frequencies of the two input signals are the same, the phase difference between, the two signals can be extracted.
The mixer circuit 178 includes an input port 198, a local oscillator port 200, and an output port 202, The output signal 176 from the eddy current sensor 50 is input into the input port 198. The signal 180 from the AC signal source 52 is input into the local oscillator port 200. The output signal 176 and the signal 180 have the same frequency. The result of multiplying the output signal 176 and the signal 180 in the mixer circuit 178 is output from the output port 202, The output signal 182 output by the output port 202 is input into the low-pass filter 184. By passing through the low-pass filter 184, the amount of phase change (phase difference) between the output signal 176 and the signal 180 detected by the eddy current sensor 50 is extracted as a DC signal (the output signal 186). The cutoff frequency of the low-pass filter can be set in the range from 1 Hz to 1.0 Hz, for example.
The processing performed by the mixer circuit 178 is as follows. Let the output signal 176 be A·sin(ωt+θA). Let the signal 180 be B·sin(ωt+θB). In the above, A and B are the respective amplitudes (mv, for example) of the output signal 176 and the signal 180, ω is the frequency (1/s), t is the time (s), and θA and θB are the respective phases of the output signal 176 and the signal 180. The output signal 182 from the mixer circuit 178 is A·sin(ωt+θA)·B·sin(ωt+θB)=(½)A·B·cos(ωt+θA+ωt+θB)+(½)A·B·cos(ωt+θA−ωt−θB) (½)A·B·cos(2ωt+θA+θB)+(½)A·B·cos(θA−θB). When this signal is input into the low-pass filter 184, the first term above is eliminated, and the output signal 186 from the low-pass filter 184 is the second term (½)·A·B·cos(θA−θB) above, or in other words, a DC signal.
In the present embodiment, the output signal 176 from the eddy current sensor 50 is connected to the mixer circuit 178 without going through a resistance bridge circuit. Because a resistance bridge circuit susceptible to the influence of temperature changes is not used, it is possible to provide the output signal processing circuit 54 of the eddy current sensor 50 that is less susceptible to the influence of changes in the surrounding environment compared to the related art.
Also, as illustrated in FIG. 6 of Japanese Patent Laid-Open No. 2011-23579, a signal detected at the terminal of a resistance bridge circuit passes through a high-frequency amplifier and a phase shift circuit, and the cos component and the sin component of the detection signal are extracted by a synchronous detection unit containing a cos synchronous detector circuit and a sin synchronous detector circuit. In the synchronously detected signals, an unwanted high-frequency component above the signal component is removed by a low-pass filter, Consequently, the high-frequency output that the coil outputs becomes a low-frequency signal after passing through long processing steps. Because there are many steps involving a high-frequency signal the signal is noisy in the related art.
In the present embodiment, the high-frequency output (output signal 176) output by the cod of the eddy current sensor 50 becomes a low-frequency signal (output signal 186) by the mixer circuit 178 and the low-pass filter 184. Because there are few steps involving the output signal 176 which is a high-frequency signal the output Signal 186 contains less noise compared to the related art. In other words, because the signal is converted from a high-frequency signal susceptible to the influence of the surroundings into an easy-to-handle DC signal at an early stage (in few steps), consistent signal measurement can be performed.
In
Next, a different embodiment of the present invention wilt be described.
Here, the exciting coil 72 in the center is connected to the AC signal source 52. The exciting coil 72 forms an eddy current in the metal film (or conductive Him) mf on the semiconductor wafer WH disposed near the eddy current sensor 50 due to a magnetic field formed by the voltage supplied by the AC signal source 52. The detecting coil 73 is disposed on the upper side (the metal film (or conductive film) side) of the ferrite core 71, and detects the magnetic field generated by the eddy current formed in the metal film (or conductive film). Additionally, the dummy coil 74 is disposed on the side of the exciting coil 72 opposite the detecting coil 73, The exciting coil 72, the detecting coif 73, and the dummy coil 74 are coils having the same number of turns (from 1t to 20t), for example. The reason for providing the dummy coil 74 is to enable the output from the output signal processing circuit 54 to be adjusted to zero when a metal film (or conductive film) is not present.
The output signal processing circuit 54 additionally includes a mixer circuit 1782 (second mixer circuit) that accepts the output signal 188 and the signal 180 of the predetermined frequency output by the AC signal source 52 as input, multiplies the output signal 188 and the signal ISO received as input, and outputs an output signal 190 obtained by the multiplication, and a low-pass filter 1842 (second low-pass filter) that accepts the output signal 190 output by the mixer circuit 1782 as input, cuts a high-frequency signal included in the output signal 190 received as input, and outputs an output signal 1862 including at least a DC signal.
The output signal processing circuit 54 additionally includes a first subtracter circuit 1961 (subtracter circuit) that accepts the output signal 1861 output by the low-pass filter 1841 and the output signal 1862 output by the low-pass filter 1842 as input, calculates a difference between the two output signals 1861 and 1862 received as input, and outputs the obtained difference.
Furthermore, the output signal processing circuit. 54 may also include a first adjustment circuit 1941 that accepts the output signal 1861 (DC signal) output by the low-pass filter 1841 as input, adjusts the magnitude of the amplitude of the output signal 1861 received as input, and outputs an adjusted DC signal 1921 and a second adjustment circuit 1942 that accepts the output signal 1862 (DC signal) output by the low-pass filter 1842 as input, adjusts the magnitude of the amplitude of the output signal 1862 received as input, and outputs an adjusted DC signal 1922, and additionally include a first subtracter circuit 1961 (subtracter circuit) that, calculates a difference between the adjusted DC signal 1921 output by the first adjustment circuit 1941 and the adjusted DC signal 1922 output by the second adjustment circuit 1942 as input, and outputs the obtained difference.
Also in the present embodiment, the output signal 176 from the eddy current sensor 50 and the output signal IBS are connected to the miser circuit 1781 and the mixer circuit 1782 without going through a resistance bridge circuit. Because a resistance bridge circuit susceptible to the influence of temperature changes is not used, it is possible to provide the eddy current sensor output signal processing circuit 54 that is less susceptible to the influence of changes in the surrounding environment compared to the related art.
In the present embodiment, the first subtractor circuit 1961 is capable of performing detection similar to the detection method using a bridge circuit of the related art, in which two coils are used to extract a tiny signal change corresponding to a film thickness change. In other words, it is possible to calculate the difference between two DC signals to detect only a tiny signal change corresponding to a film thickness change.
At this point.
The magnitude of adjustment to the resistance values of the variable resistance 206 and the variable resistance 208, or in other words the magnitude of adjustment to the resistance value during zero-point adjustment, is extremely small compared to the magnitude of the entire resistance value forming the bridge circuit. As a result, the magnitude of a temperature change in the entire resistance value is non-negligible compared, to the magnitude of the zero-point adjustment. Because of change in the resistance value due to temperature change, the stray capacitance of the resistance, and the like, the properties of the resistance bridge circuit 204 are highly susceptible to the influence of changes in the surrounding resistance environment. As a result, there is a problem in that the zero point described above shifts easily, and the accuracy of measuring the film thickness falls,
Returning to the description of
With an adjustment circuit and a subtractor circuit, it is possible to maintain substantially the same performance as the detection method using a resistance bridge circuit of the related art, in which two coils (the detecting coil 73 and the dummy coil 74) are used to extract a tiny signal change corresponding to a film thickness change. Namely, the level of the phase detection signal from at least one of the coils (but in the ease of
The first adjustment circuit 1941 and the second adjustment circuit 1942 are attenuators, for example. An attenuator refers to a circuit element or device that attenuates a signal input into the attenuator to a suitable signal level (amplitude). The first adjustment circuit 1941 and the second adjustment circuit 1942 may also be amplifiers, for example. Besides the reasons originating from the coil properties, the following are also reasons for providing the first adjustment circuit 1941 and the second adjustment circuit 1942,
A mixer circuit or a low-pass filter may output an input signal with an amplified or attenuated amplitude in some cases. In such cases, the first adjustment circuit 1941 and the second adjustment circuit 1942 are necessary to adjust the amplitude of the output from the mixer circuit or the low-pass circuit.
Next, a different embodiment of the present invention will be described.
Of the six coils 860, 862, 864, 866, 868, and 870, the coils 860 and 862 in the center are exciting coils connected to the AC signal source 52, The exciting coils 860 and 862 form an eddy current in the metal film (or conductive film) mf on the semiconductor wafer WH disposed nearby due to a magnetic field formed by the voltage supplied by the AC signal source 52. The detecting coils 864 and 866 are disposed on the metal film side of the exciting coils 860 and 862, and detect the magnetic field generated by the eddy current formed in the metal film. The dummy coils 868 and 870 are disposed on the opposite side from the detecting coils 864 and 866 with the exciting coils 860 and 862 in between,
The exciting coil 860 is an inner coil disposed on the outer periphery of the magnetic core part 61b, is capable of generating a magnetic Held, and forms an eddy current in a conductive film. The exciting coil 862 is an outer coil disposed on the outer periphery of the surrounding wall part 61c, is capable of generating a magnetic Held, and forms an eddy current in a conductive film. The detecting coil 864 is disposed on the outer periphery of the magnetic core part 61b, is capable of detecting a magnetic field, and detects an eddy current formed in a conductive film. The detecting coil 866 is disposed on the outer periphery of the surrounding wall part 61c, is capable of detecting a magnetic field, and detects an eddy current formed in a conductive film.
The eddy current sensor includes the dummy coils 868 and 870 that detect an eddy current formed in a conductive film. The dummy coil 868 is disposed on the outer periphery of the magnetic core part 61b, and is capable of detecting a magnetic field. The dummy coil 870 is disposed on the outer periphery of the surrounding wall part 61c, and is capable of detecting a magnetic field. In the present embodiment, the detecting coils and the dummy coils are disposed on the outer periphery of the floor part 61a and the outer periphery of the surrounding wall part 61c, but the detecting coils and the dummy coils may also be disposed on only one of either the outer periphery of the floor part 61a or the outer periphery of the surrounding wall part 61c.
The axial direction of the magnetic core part 61b is orthogonal to the conductive film on the substrate, the detecting coils 864 and 866, the exciting coils 860 and 862, and the dummy coils 868 and 870 are disposed at different positions in the axial direction of the magnetic core part 61b, and additionally, the detecting coils 864 and 866, the exciting coils 860 and 862, and the dummy coils 868 and 870 are disposed to the above order in the axial direction of the magnetic core part. 61b proceeding from a position close to the conductive film on the substrate to a position farther away, A lead line (illustrated in
The conducting wire used in the detecting coils 864 and 866, the exciting coils 860 and 862, and the dummy coils 868 and 870 is copper, Manganin wire, or nichrome wire. As a result of using Manganin wire or nichrome wire, there are fewer temperature changes due to electrical resistance and the like, and the temperature properties are improved,
In the present embodiment, wire rod is wrapped around the outside of the magnetic core part 61b and the outside of the surrounding wall part 61c made of ferrite or the like toform the exciting coils 860 and 862, and therefore the eddy current density flowing through the measurement target can be raised. Additionally, because the detecting coils 864 and 866 are alsoformed on the outside of the magnetic core part 61b and the outside of the surrounding wall part 61c, the generated reverse magnetic field (interlinkage flux) can be collective efficiently.
To raise the eddy current density flowing through the measurement target, in the present embodiment, the exciting coil 860 and the exciting coil 862 are additionally connected in parallel, as illustrated in
As illustrated in
Because the magnetic field 876 and the magnetic field 878 illustrated in the region 880 have the same orientation, the two magnetic fields are added together and become larger. Compared to the case where only the magnetic field 876 generated by the exciting coil 860 exists like in the related art, in the present embodiment, the magnetic field is larger by the extent of the magnetic field 878 generated by the exciting coil 862.
The detecting coil 864, the exciting coil 860, and the dummy coil 868 correspond to the detecting coil 73, the exciting coil 72, and the dummy coil 74 of
An example of the output signal processing circuit 54 corresponding to the eddy current sensor of
The processing by the mixer circuit 1781, the mixer circuit 1782, the low-pass filter 1841, and the low-pass filter 1842 associated with the first output signal and the second output signal is as described above, and therefore a description is omitted.
The output signal processing circuit 54 includes a third mixer circuit 1783 that accepts the third output signal 1761 and the signal 180 of the predetermined frequency output by the AC signal source 52 as input, multiplies the two signals received as input, and outputs an output signal 1821 obtained by the multiplication, and a third low-pass filter 1843 that accepts the output signal 1821 output by the third mixer circuit 1783 as input, cuts a high-frequency signal included, in the output signal 1821 received as input, and outputs an output signal 1863 including at least a DC signal.
The output signal processing circuit 54 includes a fourth mixer circuit 1784 that accepts the fourth output signal 1881 and the signal 180 of the predetermined frequency output by the AC signal source 52 as input, multiplies the two signals received as input, and outputs an output signal 1901 obtained by the multiplication, and a fourth, low-pass filter 1844 that accepts the output signal 1901 output by the fourth mixer circuit 1784 as input, cuts a high-frequency signal included in the output signal 1901 received as input, and outputs an output signal 1864 including at least a DC signal.
An internal configuration of a digital signal processor 216 is illustrated in
In fee present embodiment, fee digital signal processor (DSP) 216 includes the functions of the first, second, third, and fourth adjustment circuits, fee first and second subtractor circuits, and the adder circuit. In other words, the digital signal processor 216 may also include a third adjustment circuit 1943 that receives the output signal 1863 (DC signal) output by fee third low-pass filter 1843 as input, adjusts fee magnitude of the amplitude of fee output signal 1863 received as input, and outputs an adjusted DC signal 1923, and a fourth adjustment circuit 1944 that receives the output signal 1864 (DC signal) output by the fourth low-pass filter 1844 as input, adjusts the magnitude of the amplitude of the output signal 1864 received as Input, and outputs an adjusted DC signal 1924, The digital signal processor 216 is a microprocessor suitable for digital signal processing. At least one of fee first, second, third, and fourth adjustment circuits, the first and second subtractor circuits, and the adder circuit may also be provided independently as an analog circuit or a digital circuit,
The present embodiment includes a coil assembly with two coils in addition to the coil assembly with two coils illustrated in
In the case of taking the difference, one pair is capable of measuring the film thickness over a wider area than the other pair, for example. In this case, by subtracting the output of one pair from the other pair, the influence of the him thickness over a wide area can be reduced and only the change in film thickness over a narrow area can be measured more accurately. In other words, the spatial precision of measurement can be improved,
In the present embodiment, the output signal from the eddy current sensor likewise is connected to the mixer circuit without going through a resistance bridge circuit. Because a resistance bridge circuit susceptible to the influence of temperature changes is not used, it is possible to provide the eddy current sensor output signal processing circuit that is less susceptible to the influence of changes in the surrounding environment compared to the related art,
In the present embodiment, the subtractor circuit is capable of performing detection similar to the detection method using abridge circuit of the related art, in which two coils are used to extract a tiny signal change corresponding to a film thickness change. In other words, it is possible to calculate the difference between two DC signals to detect only a tiny signal change corresponding to a film thickness change.
The multiplier circuit and the subtractor circuit may also be a circuit, that outputs the result of the multiplication and subtraction multiplied by a predetermined constant (where the constant may be 1, less than 1, or greater 1), or in other words, outputs an amplified output signal
Next, a method of processing the output signal from the eddy current sensor will be described, in
Next, a different method of processing the output signal from the eddy current sensor will be described. In
The first output signal 176 and the signal 180 from the AC signal source 52 are input into the first mixer circuit 1781, the two signals are multiplied by the first mixer circuit 1781, and the output signal 182 obtained by the multiplication is output. The output signal 182 output by the first mixer circuit 1781 is input into the first low-pass filter 1841, a high-frequency signal included in the output signal 182 is cut by the Fust low-pass filter 1841, and at least the DC signal 1861 is output.
The second output signal 188 and the signal 180 from the AC signal source 52 are input into the second mixer circuit 1782, the two signals are multiplied by the second mixer circuit 1782, and the output signal 190 obtained by the multiplication is output. The output signal 190 output by the second mixer circuit 1782 is input into the second low-pass filter 1842, a high-frequency signal included in the output signal 190 is cut by the second low-pass filter 1842, and at least the DC signal 1862 is output,
The DC signal 1861 output by the first low-pass filter 1841 and the DC signal 1862 output by the second low-pass filter 1842 are input into the first subtractor circuit 1961, the difference 172 between the two DC signals is obtained by the first subtractor circuit 1961, and the obtained difference 172 is output.
Next, a different method of processing the output signal from the eddy current sensor will be described. In
The output signal 1821 from the third mixer circuit 1783 is input into the third low-pass filter 1843, a high-frequency signal included in the output signal 1821 is cut by the third low-pass filter 1843, and at least, the DC signal 1863 is output. The fourth output signal 1881 and the signal 180 from the AC signal source 52 are input into the fourth mixer circuit 1784, the two signals are multiplied by the fourth mixer circuit 1784, and the output signal 1901 obtained by the multiplication is output,
The output signal 1901 from the fourth mixer circuit 1784 is input into the fourth low-pass filter 1844, a high-frequency signal included in the output signal 1901 is cut by the fourth low-pass filter 1844, and at least the DC signal 1864 is output. The DC signal 1863 output by the third low-pass filter 1843 and the DC signal 1864 output by the fourth, low-pass filter 1844 are input into the second subtractor circuit 1962, the difference 2222 between the two DC signals 1863 and 1864 is obtained by the second subtracter circuit 1962, and the obtained difference 2222 is output. The difference 2221 output by the first subtractor circuit 1961 and the difference 2222 output by the second subtracter circuit 1962 are input into the adder circuit 224. The sum of the two differences 2221 and 2222 or the difference between the two differences 2221 and 2222 is obtained by the adder circuit 224, and the obtained sum 172 or difference 172 is output.
A method of controlling each unit of the polishing apparatus on the basis of the film thickness obtained by the eddy current sensor 50 is described hereinafter. As illustrated in
The polishing table 100 is coupled to a motor (not illustrated) that acts as a driving unit disposed underneath through a table spindle 170, and is capable of rotating about the table spindle 170, A polishing pad 101 is affixed to the top face of the polishing table 100, and the surface 101a of the polishing pad 101 forms a polishing surface that polishes a semiconductor wafer WH. A polishing liquid supply nozzle 102 is installed above the polishing table 100, such that a polishing liquid Q is supplied onto the polishing pad 101 on the polishing table 100 by the polishing liquid supply nozzle 102. As illustrated in
Hereinafter, an eddy current sensor of the impedance type according to the present embodiment will be described specifically. The AC signal source 52 includes an oscillator 260 (see
The AC signal source 52 may also include an amplifier 262 that amplifies an output signal 270 from the oscillator 260, and a filter 264 for reducing noise included in an output signal 272 from the amplifier 262, Because noise is also amplified by the amplifier 262, the filter 264 is installed downstream of the amplifier 262 to reduce the amplified noise. In the present embodiment, the output signal from the filter 264 is the excitation signal 25
The eddy current sensor 50 includes an exciting coil 72 for forming an eddy current in the metal film (or conductive film) on the semiconductor wafer WH, and a detecting coil 73 that detects the generated eddy current. For example, the exciting coil 72 and the detecting coil 73 are disposed in the axial direction of a cylindrical ferrite core. The exciting coil 72 is connected to the AC signal source 52. The exciting coil 72 forms an eddy current in the metal film (or conductive film) mf on the semiconductor wafer WH disposed near the eddy current sensor 50 due to a magnetic field formed by the voltage supplied by the AC signal source 52. The detecting coil 73 is disposed on the upper side (the metal film (or conductive film) side) of the ferrite core, and detects the magnetic field generated by the eddy current formed in the metal film (or conductive film). The eddy current sensor 50 may also include a dummy coil 74 as described later,
The output signal 176 is input into the generator circuit 254 through the filter 266 and the amplifier 268 of the output signal processing circuit 54 An amplifier not illustrated is disposed upstream of the filter 266. The filter 266 is installed to reduce amplified noise. In only the amplifier not illustrated, the output signal 176 may be weak in some cases, and therefore the amplifier 268 is disposed downstream of the filter 266. An output signal 274 from the amplifier 268 is input into the generator circuit 254 and the adder circuit 258.
The generator circuit 254 generates the noise reduction signal 252. The noise reduction signal 252 is generated as follows. The generator circuit 254 includes a band-stop filter 276 and a phase inverter circuit 278. The hand-stop litter 276 attenuates to an extremely low level only frequency signals near a specific frequency signal (for example, 16 MHz) generated by the oscillator 260, The hand-stop filter 276 passes other frequency signals (that is, the noise signal) as-is. An output signal 282 from the band-stop filter 276 is sent to the phase inverter circuit 278, and the phase is inverted by the phase inverter circuit 278. Inverting the phase means advancing the phase 180 degrees.
The noise reduction signal 252 that is the output from the phase inverter circuit 278 is sent, to the adder circuit 258 and added to the output signal 274 in the adder circuit 258. The noise reduction signal 252 is a signal that cancels out only the noise component included in the output signal 274, The noise-reduced signal 256 that is the output from the adder circuit 258 is a signal in which the signals of frequencies other than the specific frequency generated by the oscillator 260 (that is, noise) is reduced. The noise-reduced signal 256 is sent to a detector circuit 280, The detector circuit 280 extracts the signal outputs X and Y, the phase, and the combined impedance Z (=X+iY) described above from the high-frequency signal. These are DC signals. The output from the detector circuit 280 is the output signal 172.
In the present embodiment, because the noise reduction signal 252 is added to the output signal 274 to generate the noise-reduced signal 256 in which the noise included in the output signal 274 is reduced, the S/N is improved over the technology of the related art. In the case of using the eddy current sensor 50 to measure film thickness, the detection accuracy of Cu wiring and the like is Improved over the technology of the related, art by reducing the influence of noise, and the performance of the eddy current sensor 50 is improved,
The output signal 282 is converted to the noise reduction signal 252 by the phase inverter circuit 278. Letting Y2 be the noise reduction signal 252, then Y2=−Y1. Expressed differently, Y2=A sin(ωt+π). On the other hand, the output signal 274 output by the detecting coil 73 contains a noise component Y3=A sin(ωt) which is the same as the above noise component Y1=A sin(ωt).
The adder circuit 258 adds the noise reduction signal 252 to the output Signal 274 and outputs the noise-reduced signal 256. Letting Y4 be the noise component included in the noise-reduced signal 256, then Y4=Y3+Y2, Because Y3 and Y2 have the same amplitude with the phase shifted 180 degrees, Y4=0. The noise-reduced signal 256 is a signal that does not include a noise component.
In
Returning to
As described above, a signal source 52 supplies an AC signal to the eddy current sensor 50 disposed near the semiconductor wafer WH on which the metal film (or conductive film) mf to be detected is formed. The signal source 52 is an oscillator of fixed frequency containing a quartz oscillator. The signal source 52 supplies a voltage at a fixed frequency of 2 MHz, 8 MHz, or 16 MHz, for example. The AC voltage formed by the signal source 52 is supplied to the eddy current, sensor 50 through, a bandpass filter 82. A signal detected at the terminal of the eddy current, sensor 50 passes through a high-frequency amplifier 83 and a phase shift circuit 84, and the cos component and the sin component of the detection signal are extracted by a synchronous detection unit containing a cos synchronous detector circuit 85 and a sin synchronous detector circuit 86. Here, the oscillation signal formed by the signal source 52 is used by the phase shift circuit 84 to synthesize two signals of the in-phase component (0°) and the orthogonal component (90°) of the signal source 52, which are introduced into the cos synchronous detector circuit 85 and the sin synchronous detector circuit 86 respectively to perform the synchronous detection described above.
In the synchronously detected signals, an unwanted high-frequency component above the signal component is removed by low-pass filters 87 and 88 to extract the resistance component (R) output as the cos synchronous detection output and the reactance component (X) output as the sin synchronous detection output, respectively. Additionally, the amplitude output (R2+X2)1/2 is obtained, from the resistance component (R) output and the reactance component (X) output by a vector operator circuit 89. Also, the phase output (tan−1R/X) is obtained similarly from the resistance component output and the reactance component output by a vector operator circuit 90. Here, the main body of the measuring apparatus is provided with various fitters for removing noise, components' from sensor signals. A corresponding cutoff frequency is set for each of the various Oilers, and by setting the cutoff frequency of the Sow-pass filters in the range from 1 Hz to 10 Hz for example, the noise component mixed with the sensor signal being polished is removed, and the metal film (or conductive film) targeted for measurement can be measured accurately.
Next, an exemplary configuration of the coils in the eddy current sensor 50 according to the present embodiment will be described.
Here, the exciting coil 72 in the center is connected to the AC signal source 52. The exciting coil 72 forms an eddy current in the metal film (or conductive film) mf on the semiconductor wafer WH disposed near the eddy current sensor 50 due to a magnetic field formed by the voltage supplied by the AC signal source 52. The detecting coil 73 is disposed on the upper side (the metal film (or conductive film) side) of the ferrite core 71, and detects the magnetic field generated by the eddy current formed in the metal film (or conductive film). Additionally, the dummy coil 74 is disposed on the side of the exciting coil 72 opposite the detecting coil 73. The exciting coil 72 the detecting coil 73, and the detecting coil 73 are coils having the same number of turns (from 1t to 20t), for example. The reason for providing the dummy coil 74 is to enable the output from the output signal processing circuit 54 to be adjusted to zero when a metal film (or conductive film) is not present.
There are various possible methods of processing the output from the detecting coil 73 and the dummy coil 74. For example, as illustrated in
Specifically, a signal line 731 of the detecting coil 73 is connected to a terminal 773 of the resistance bridge circuit 77, and a signal line 732 of the detecting coil 73 is connected to a terminal 771 of the resistance bridge circuit 77, A signal line 741 of the dummy coil 74 is connected to a terminal 772 of the resistance bridge circuit 77, and a signal line 742 of the dummy coil 74 is connected to the terminal 771 of the resistance bridge circuit 77. The terminal 771 is grounded. A terminal 774 of the resistance bridge circuit 77 is the sensor output. The detecting coil 73, the exciting coil 72, and the dummy cod 74 have inductances L1, L2, and L3, respectively,
By connecting the exciting coil 72 to the AC signal source 32 and generating alternating flux, an eddy current is formed in the metal, film (or conductive film) mf disposed nearby. By adjusting the resistance value of the variable resistance 76, the output voltage of the series circuit containing the detecting coil 73 and the dummy coil 74 is adjustable so as to be zero when a metal film (or conductive film) is not present. The L1 and L2 signals are adjusted- to be in the same phase with each other by the variable resistance 76 (VR1 and VR2) inserted hi parallel with each of the detecting coil 73 and the dummy coil 74. In other words, in the equivalent circuit in
VR1-1×(VR2-2+jωL3)=VR1-2×(VR2-1+jωL1) (1)
With this arrangement as illustrated in
Next, a different embodiment of the present invention will be described.
Of the six coils 860, 862, 864, 866, 868, and 870, the coils 860 and 862 in the center are exciting coils connected to the AC signal, source 52. The exciting coils 860 and 862 form an eddy current in the metal film (or conductive film) mf on the semiconductor wafer WH disposed nearby due to a magnetic field, formed by the voltage supplied by the AC signal source 52. The detecting coils 864 and 866 are disposed on the metal film side of the exciting coils 860 and 862, and detect the magnetic field generated by the eddy current formed in the metal, film. The dummy coils 868 and 870 are disposed on the opposite side from the detecting coils 864 and 866 with the exciting coils 860 and 862 in between.
The exciting coil 860 is an Inner coil disposed on the outer periphery of the magnetic core part 61b, is capable of generating a magnetic field, and forms an eddy current in a conductive film. The exciting coil 862 is an outer toil disposed on the outer periphery of the surrounding wall part 61e, is capable of generating a magnetic field, and forms an eddy current hi a conductive film. The detecting coil 864 is disposed on the outer periphery of the magnetic core part 61b, is capable of detecting a magnetic field, and detects an eddy current formed in a conductive film. The detecting coil 866 is disposed on the outer periphery of the surrounding wall part 61c, is capable of detecting a magnetic field, and detects an eddy current formed in a conductive film.
The eddy current sensor includes the dummy coils 868 and 870 that detect an eddy current formed in a conductive film. The dummy coil 868 is disposed on the outer periphery of the magnetic core part 61b, and is capable of detecting a magnetic field. The dummy coil 870 is disposed on the outer periphery of the surrounding wall part 61c, and is capable of detecting a magnetic field. In the present, embodiment, the detecting coils and the dummy coils are disposed on the outer periphery of the floor part 61a and the outer periphery of the surrounding wall part 61c, hot the detecting coils and the dummy coils may al so be disposed on only one of either the outer periphery of the floor part 61a or the outer periphery of the surrounding wall part 61c.
The axial direction of the magnetic core part 61b is orthogonal to the conductive film on the substrate, the detecting coils 864 and 866, the exciting coils 860 and 862, and the dummy coils 868 and 870 are disposed at different positions m the axial direction of the magnetic core part 61b, and additionally, the detecting coils 864 and 866, the exciting coils 860 and 862, and the dummy coils 868 and 870 are disposed in the above order in the axial direction of the magnetic core part 61b proceeding from a position close to the conductive film on the substrate to a position farther away. A lead line (illustrated in
The conducting wire used in the detecting coils 864 and 866, the exciting coils 860 and 862, and the dummy coils 868 and 870 is copper, Manganin wire, or nichrome wire. As a result of using Manganin wire or nichrome wire, there are fewer temperature changes due to electrical resistance and the like, and the temperature properties are improved.
In the present embodiment, wire rod is wrapped around the outside of the magnetic core part 61b and the outside of the surrounding wall part 61c made of ferrite or the like to form the exciting coils 860 and 862, and therefore the eddy current density flowing through the measurement target can be raised. Additionally, because the detecting coils 864 and 866 are also formed, on the outside of the magnetic core part 61b and the outside of the surrounding wall part 61c, the generated reverse magnetic field (interlinkage flux) can be collective efficiently.
To raise the eddy current density flowing through the measurement target, in the present embodiment, the exciting coil 860 and the exciting coil 862 are additionally connected, in parallel, as illustrated in
As illustrated in
Because the magnetic field 876 and the magnetic field 878 illustrated in the region 880 have the same orientation, the two magnetic fields are added together and become larger. Compared to the case, where only the magnetic field 876 generated by the exciting coil 860 exists like in the related art, in the present embodiment, the magnetic field is larger by the extent of the magnetic field 878 generated by the exciting coil 862.
The detecting coil 864, the exciting coil 860, and the dummy cod 868 correspond to the detecting coil 73, the exciting coil 72, and the dummy coil 74 of
An example of the output signal processing circuit 54 corresponding to the eddy-current sensor of
First, the configuration of the comparative example will be described. In the comparative example, a detecting coil 864 of an inner eddy current sensor 286 and a detecting coil 866 of an outer eddy current sensor 288 are connected in series. A dummy coil 868 of the tuner eddy current sensor 286 and a dummy coil 870 of the outer eddy current sensor 288 are connected in series. An exciting coil 860 and an exciting coil 862 of the inner and outer eddy current sensors are connected in parallel to the signal source 52,
The specific connections are as follows. In the inner eddy current sensor 286, the detecting coil 864 includes signal lines 8641 and 8642. The exciting coil. 860 includes signal Hues 8601 and 8602, The dummy coil 868 includes signal lines 8681 and 8682. In the outer eddy current sensor 288, the detecting coil 866 includes signal lines 8661 and 8662, The exciting coil 862 includes signal, lines 8621 and 8622. The dummy coil 870 includes signal lines 8701 and 8702.
The signal line 8641 of the detecting coil 864 of the inner eddy current sensor 286 is connected to the terminal 773 of the resistance bridge circuit 77. The signal line 8642 of the detecting coil 864 is connected to the signal line 8661 of the detecting coil 866 of the outer eddy current sensor 288. The signal line 8662 of the detecting coil 866 is connected to the terminal 771 of the resistance bridge circuit 77. The signal line 8681 of the dummy coil. 868 of the inner eddy current sensor 286 is connected to the terminal 772 of the resistance bridge circuit 77. The signal line 8642 of the dummy coil 868 is connected to the signal line 8701 of the dummy coil 870 of the outer eddy current sensor 288. The signal line 8702 of the dummy coil 870 is connected to the terminal 771 of the resistance bridge circuit 77.
In the comparative example, the output of detecting coil 864 of the inner eddy current sensor 286 and the output of the detecting coil 866 of the outer eddy current sensor 288 are in series, and consequently there is an effect of increased output compared to the case of a single detecting coil. The terminal 774 of the resistance bridge circuit 77 is connected to the detector circuit 280. The output of the detector circuit 280 is the output signal 172, and is connected to the endpoint detection controller 246 illustrated in
The embodiment illustrated in
The generator circuit 254 for the inner eddy current sensor 286 is a generator circuit 2541, while the generator circuit 254 for the outer eddy current sensor 288 is a generator circuit 2542, The generator circuit 2541 and the generator circuit 2542 have the same configuration as the generator circuit 254. The filter 266 and the amplifier 268 are provided upstream of the generator circuit 2541 and the generator circuit 2542, similarly to
The noise-reduced signals 256 obtained by the inner eddy current sensor 286 and the outer eddy current sensor 288 are sent to the detector circuit 280, and the outputs of the two bridge circuits 7710 and 7711 are in series. Thereafter, the signals are added together in an adder circuit 284. Adding the signals together improves the sensitivity.
The configuration of the present embodiment will be described specifically. The inner eddy current sensor 286 includes the exciting coil 860 (First exciting coil) that accepts excitation signals 8601 and 8602 from the signal source 52 as input and is capable of generating the first magnetic field 876 (see
An output signal processing circuit 290 of the present embodiment that processes the signal 8641 and the signal 8681 output by the inner eddy current sensor 286 includes a first resistance bridge circuit 7710 capable of outputting the difference between the signal 8641 and the signal 8681 as a first difference signal 292. The output signal processing circuit 290 includes the generator circuit 2541 (first generator circuit) capable of generating a first noise reduction signal 252 for reducing noise from any of the excitation signal 8601, the signal 8641 (first detection signal), the signal 8681 (first dummy signal), and the first difference signal 292. In the present, embodiment, the first noise reduction signal 252 for reducing noise is generated from the first difference signal 292 that is the output from the terminal 774.
The output signal processing circuit 290 further includes a first adder circuit 2581 capable of adding the first noise reduction signal 252 generated in the generator circuit 2541 to the first difference signal. 292, and thereby generating a first noise-reduced signal 256 in which the noise included in the first difference signal 292 is reduced.
The outer eddy current sensor 288 includes the exciting coil 862 (second exciting coil) that accepts an excitation signal 8621 as input and is capable of generating the second magnetic field 878 (see
The detecting coil 864 of the inner eddy current sensor 286 is capable of detecting the first magnetic field 876 and the second magnetic field 878 and outputting the signal 8641. The dummy cod 868 is capable of detecting the first magnetic field 876 and the second magnetic field 878 and outputting the first dummy signal. The output signal processing circuit 290 includes a second resistance bridge circuit 7711 capable of outputting the difference between the signal 8661 and the signal 8701 as a second difference signal 294,
Here, the meaning of the first and second detecting coils detecting the first magnetic field 876 and the second magnetic field 878 will be described. The detecting coil 864 of the inner eddy current sensor 286 detects the film thickness as follows. First, an eddy current is induced on the metal surface by the high-frequency magnetic Held generated by the exciting coil. In other words, the state of the magnetic field is different depending on whether the wafer (conductive film) is nearby or not. An exciting coil capable of generating a magnetic field on the metal surface by applying a high frequency (2 MHz or higher) to the exciting coil is brought close to the conductive metal surface. The high-frequency magnetic field generates current in eddies on the metal surface. The eddy current flows in a direction that partially cancels out the magnetic field. The detecting coil of the eddy current sensor detects the magnetic field that is partially canceled out. The eddy current sensor uses the correlation between the magnitude of the eddy current and the thickness of the film to measure the thickness of the conductive film, on the metal.
“Detecting the first magnetic field and the second magnetic field” refers to “defecting the combined magnetic field of the first magnetic field and the second magnetic field”, and the combined magnetic field is also different depending on whether the wafer (conductive film) is present or not, “The first and second detecting coils detect the first and second magnetic fields” means that “an eddy current is generated on the metal surface by the combined magnetic field of the first and second magnetic fields, the combined magnetic field is partially canceled out by the generated eddy current, and the partially canceled, out magnetic field is detected”.
Returning to the description of the outer eddy current sensor 288, the output signal processing circuit 290 includes the generator circuit 2542 (second generator circuit) capable of generating a second noise reduction signal 252 for reducing noise from any of the excitation signal 8621, the signal 8661 (second detection signal), the signal 8701 (second dummy signal), and the second difference signal 294. In the present embodiment, the second noise reduction signal 252 for reducing noise is generated from the second difference signal 294 that is the output from the terminal 772,
The output signal processing circuit 290 includes a second adder circuit 2582 capable of adding the second noise reduction signal 252 generated in the generator circuit 2542 to the second difference signal 294 and thereby generating s second noise-reduced signal 256 in which the noise included in the second difference signal 294 is reduced, and a third adder circuit 284 capable of adding together the first noise-reduced signal 256 and the second noise-reduced signal 256,
According to the present embodiment, because the first difference signal and the second difference signal are added together, the output signal from the eddy current sensor is larger than the related art and the accuracy of Him thickness measurement is improved,
A comparison of the effects of the embodiment in
Comparing
Note that as illustrated in
From the embodiment of
Specifically, the output signal processing circuit in this embodiment includes the first resistance bridge circuit. 7710 capable of outputting the difference between the signal 8641 (first detection signal) and the signal. 8681 (first dummy signal) as a first difference signal 292, the second resistance bridge circuit 7711 capable of outputting the difference between the signal. 8661 (second detection signal) and the signal 8701 (second dummy signal) as a second difference signal 294, and the third adder circuit 284 capable of adding together the first difference signal 292 and the second difference signal 294 via the detector circuits 28
) In
Compared to
In the embodiment of
Note that in the eddy current sensor illustrated in
The band-stop filter 276, the phase inverter circuit 278, and the adder circuit 258 can be achieved by a digital signal processor (DSP). A digital signal processor is a microprocessor suitable for digital signal processing. The band-stop filter 276, the phase inverter circuit 278, and the adder circuit 258 may also be analog circuits. The digital signal processor includes an upstream analog-to-digital converter circuit (that is, an electronic circuit that converts an analog electrical signal to a digital electrical signal; also referred to as an A/D converter circuit or an ADC) not illustrated. The digital signal processor includes a downstream digital-to-analog converter circuit (that is, an electronic circuit that converts a digital electrical signal to an analog electrical signal; also referred to as a D/A converter circuit or a DAC) not illustrated. The detector circuits 280 downstream of the adder circuit 258 can also be achieved by a digital signal processor.
Note that in the description of
Note that in fee adder circuit 258 of
Consequently, “adding a signal of inverted phase” and “subtracting a signal of non-inverted phase” have fee same effect from fee viewpoint of noise canceling. In this specification, fee term “adding together” encompasses “addition” and “subtraction” in this sense.
Note that by disposing a temperature sensor inside the eddy current sensor 50, temperature changes Inside fee eddy current sensor 50 can be detected, and the output of the detecting coil 73 cart be corrected. The correction method is as follows, for example. Before polishing the semiconductor wafer WH, the relationship between fee temperature of the eddy current sensor 50 and the output of the detecting coil 73 is measured, and the relationship is created as a correction table. The correction table may have any of various possible configurations. For example, a coefficient by which to multiply the output of the detecting coil 73 for each temperature of the eddy current sensor 50 is created as fee correction table. Alternatively, a polynomial expressing correction coefficients as a function of temperature is created as the correction table (a table displaying fee coefficient of each order in the polynomial). Correcting according to temperature makes it possible to accommodate temperature changes. The film thickness measurement accuracy can be raised, and fee accuracy of detecting fee end of polishing can be improved.
Next, a method of processing the output signal from fee eddy current sensor will be described. In
Next, a different method of processing the output signal from the eddy current sensor will be described. In
In the eddy current sensor output signal processing method that processes the signal 8641 and the signal 8681 output by the eddy current sensor 286, the first resistance bridge circuit 7710 outputs the difference between the signal 8641 and the signal 8681 as the first difference signal 292. The generator circuit 2541 generates the first noise reduction signal 252 for reducing noise from any of the excitation signal 250, the signal 8641, the signal 8681, and the first difference signal 292, In
Next, a different method of processing the output signal from the eddy current sensor will be described. In
The detecting coil 864 detects the first magnetic field 876 and the second magnetic field 878, and outputs the signal 8641 (first detection signal). The dummy coil 868 detects the first magnetic field 876 and the second magnetic field 878, and outputs the signal 8681 (first dummy signal). The second resistance bridge circuit 7711 outputs the difference between the signal 8661 and the signal 8701 as the second, difference signal 294. The generator circuit 2542 generates the second noise reduction signal 252 for reducing noise from any of the excitation signal 250, the signal 8661, the signal 8701, and the second difference Signal 294, The second adder circuit 2582 adds the generated second noise reduction signal 252 to the second difference signal 294 to generate the second noise-reduced signal 2802 in which the noise included in the second difference signal 294 is reduced. The third adder circuit 284 adds together the first noise-reduced signal 281 and the second noise-reduced signal 2802,
Next, a different method of processing the output signal from the eddy current sensor will be described. The present embodiment is obtained by removing the noise canceling, circuit from the embodiment in
In the output, signal processing method of the present embodiment, the signals 8641 and 8661 and the signals 8681 and 8701 output by the eddy current sensors 286 and 288 are processed. The first resistance bridge circuit 7710 outputs the difference between the signal 8641 (first detection signal) and the signal 8681 (first dummy signal) as the first difference signal 292. The second resistance bridge circuit 7711 outputs the difference between the signal 8661 (second detection signal) and the signal 8701 (second dummy signal) as the second difference signal 294. The first difference signal 292 and the second difference signal 294 are converted to DC signals in the detector circuit 280 and then added together in the third adder circuit 284,
A method of controlling each unit of the polishing apparatus on the basis of the film thickness obtained by the eddy current sensor 50 is described hereinafter. As illustrated in
Note that the operations according to the embodiments of the present invention are also achievable using the following software and/or system. For example, the system (polishing apparatus) includes a main controller (control unit) that controls the apparatus overall and a plurality of sub-controllers that respectively control the operations by each unit (driving unit, holding unit, eddy current sensor output signal processing circuit, endpoint detection controller). The main controller and the sub-controllers each include a CPU, memory; a recording medium, and software (a program) stored in the recording medium for causing each unit to operate. The method for processing an output signal from an eddy current sensor according to an embodiment of the present invention may also be executed by the software (program).
The foregoing describes exemplary embodiments of the present invention, but the embodiments described above are for facilitating the understanding of the present invention, and do not limit the present invention. The present invention may be modified and improved without departing from the scope of the invention, and any equivalents obtained through such modification and improvement are obviously included in the present invention. Furthermore, any combination, or omission of the components described in the claims and the specification is possible insofar as at least one or some of the issues described above can be addressed, or insofar as at least one or some of the effects are exhibited.
This application claims the benefit of priority under the Paris Convention for Japanese Patent Application No. 2019-226547 and Japanese Patent Application No. 2019-226549 filed on Dec. 16, 2019 at the Japan Patent Office. The entire disclosure of Japanese Patent Laid-Open No. 2011-23579, including specification, claims, drawings, and abstract, is incorporated by reference as a part of this specification.
Number | Date | Country | Kind |
---|---|---|---|
2019-226547 | Dec 2019 | JP | national |
2019-226549 | Dec 2019 | JP | national |