JP2821752B2 - Exposure monitoring device and exposure device - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an exposure monitoring device used in a semiconductor device such as an aligner or an etching device using a light source that emits pulsed light.

[Prior Art] An exposure amount monitoring apparatus in an exposure apparatus using pulsed light or the like is known. As this monitoring device, the output signal of the high-speed photoelectric conversion element is integrated at a fixed timing created from the oscillation synchronization signal synchronized with the laser emission, the integration result is read by the CPU to obtain the exposure amount of one pulse, and the pulse emission is performed. The exposure amount is obtained by adding the exposure amount of one pulse in the CPU every time, and the exposure amount is calculated through a process such as comparing the exposure amount with a preset designated exposure amount. Monitoring may be performed.

[Problems to be Solved by the Invention] However, for example, in an excimer laser used as this kind of pulsed light emission source, the timing from the oscillation synchronization signal before emission to the output of the pulsed light waveform from the sensor.
t _X is likely to change over time, to carry out for a long time stable measurement had to take longer integration period t _2. For this reason, there is a disadvantage that noise and offset output from the sensor are also integrated in addition to the emission time of the pulse light, and the measurement accuracy of the exposure amount is reduced.

Especially in excimer lasers, gradually changes the timing t _X from the light emitting front of the oscillation synchronizing signal until the sensor outputs a pulse light waveform due deterioration of gas and discharge electrodes, and when replacing the deteriorated gas and electrodes would be significantly above timing t _X is changed. Therefore, in such a case, it is necessary to readjust, and there is a disadvantage that the number of maintenance steps is increased.

The present invention has been made in view of the above-described conventional problems, and provides an exposure amount monitoring apparatus capable of suppressing a decrease in exposure amount measurement accuracy due to a temporal change of a laser and performing stable exposure amount monitoring. The purpose is to:

[Means and Actions for Solving the Problems] In order to achieve the above object, an exposure monitoring apparatus according to the present invention provides an integration start signal, an integration time, and an integration start signal generated from an oscillation synchronization signal output before pulse emission. And a means for setting an optimum signal time, that is, an integration start signal and an integration time, based on the result of integration of the pulse light by the integration means.

As a result, the integration start time and the integration time are optimized as the light emission timing changes over time, and the exposure amount can be monitored with stable accuracy.

Embodiment An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a configuration diagram of an exposure monitoring apparatus according to one embodiment of the present invention. In FIG. 1, an excimer laser light source 1 is a light source that emits a pulsed laser light in which, for example, KrF or XeCl is sealed. Laser light generated by the excimer laser light source 1 enters the illumination system 2. The illumination system 2 shapes the incident laser light into a desired beam shape, and emits the light with uniform light distribution characteristics. A mask M on which an integrated circuit pattern is formed is arranged in an emission optical path of the illumination system 2, and the integrated circuit pattern is projected and exposed on a wafer W by a projection optical system 3.

On the other hand, a part of the light beam emitted from the illumination system 2
Accordingly, the light is incident on the photoelectric conversion surface of the sensor 5. Since the temporal width of the pulse light emitted from the excimer laser 1 is extremely short, on the order of several tens of ns, it is preferable to use a photoelectric conversion element having a response of several ns or less, such as a PIN photodiode or a biplanar photoelectric tube, as the sensor 5. Further, since the light distribution characteristics of the beam emitted from the illumination system 2 are uniform, the power of the light entering the sensor 5 and the power of the light reaching the wafer W are in a proportional relationship. Therefore, the power of light on the wafer W can be monitored by measuring the electrical output signal of the sensor 5.

The output signal of the sensor 5 is input to the integrator 6, and is electrically and analog-integrated for each pulse of the laser light. And
The integrated value is converted into a digital value by the A / D converter 7 and read into the microcomputer 8. The microcomputer 8 integrates the digitized integral value for each pulse. Then, the integrated value is compared with a preset exposure amount for the wafer W, and when the exposure value is larger than the set exposure amount, a command is issued to the laser control unit 9 to stop the emission of the excimer laser 1. As a result, the exposure amount can be controlled.

The integration control signal generation section 10 is a section that causes the integrator 6 to generate an integration control signal that defines the start and end of integration.

FIG. 2 is a timing chart showing the processing timing for the integrator 6. In the figure, the oscillation synchronizing signal “a” is set to the excimer laser
This is the synchronization signal output from. As the oscillation synchronization signal a, a trigger signal of a thyratron, which triggers laser emission, is usually used. Therefore, delay time (delay time) from the synchronization timing to the actual light emission timing of the laser t _X is not fixed, change over time as described earlier. In addition, it is known that it contains a jitter component of several tens ns.

In this embodiment, with respect to the oscillation synchronizing signal a, now the delay time t _X - start the integration after the jitter amount = t ₁ hour, further the light emission time t _W t ₂ hours that a margin against Later, the integrator output was held. By doing so, the actual emission energy of the laser can be contained in the integration section of the integrator 6.

It is effective that the integration execution time t ₂ is as short as possible in order to increase the accuracy of the integration value. The reason is that noise may be mixed in the output signal of the sensor 5 that converts the light emission of the laser 1 into an electric signal, and the offset level fluctuates minutely. This is for reducing the influence on the integrator 6 which is sensitive for integration.

The integration control signal generator 10 is configured to change the integration start time t ₁ and the integration execution time t ₂ from the start of the integration to the hold by an instruction from the microcomputer 8.
That is, the integration start signal c and the hold signal d are output in synchronization with the oscillation of the excimer laser 1.

FIG. 3 shows the timing when the integration start timing is longer than the light emission delay time, and FIG. 4 shows the timing when the integration start timing is shorter. As it is apparent from these figures, when not set the proper integration time t ₁ to the light emitting delay time t _X, a smaller value than when the integral value is given a proper timing.

Next, with reference to the flowchart of FIG. 6 and FIG. 8, the procedure of the determination of the integration timing t ₁ and the integral running time t _2. The symbols t ₁ , t ₂ , Δt, i, n, L, etc. used in these flowcharts indicate a register in the microcomputer 8 or a storage area secured on a memory (not shown). For example, t ₁ represents such a register that is reserved for storing the value of the integral timing, t ₂ denote the like reserved register for storing the value of the integral execution time.

In step S1 of FIG. 6, the microcomputer 8 performs initial setting of registers and the like. Here the 100, i is t ₂ 0, n is 50, sets the Delta] t 10 ns, L a (0) ~ (n) 0 and. L is an array.

Here, t ₂ when performing the initial setting is setting the expected values it the emission width + jitter component + margin of the pulsed light. The t ₁ is set in consideration of the time light emission delay time is the shortest.

In step S2, specify a value of t ₁ and t ₂ are stored and held in the integration control signal generating unit 10. And step
In S3, the microcomputer 8 outputs a one-pulse light emission command, and reads the integrated value A in step S4. In step S5,
t ₁ is increased by Δt, i is incremented, and L
The integral value A is set in (i). Next, in step S6, i
Is determined to be not less than n, otherwise return to step S2 and repeat steps S2 to S6 again. If i is greater than or equal to n, the process proceeds from step S6 to step S7. Step S7
, The maximum value of L (0) to L (n) is obtained, and is set to M. Then, L (j) <0.8M < L (j + 1) L (k)>0.8M> L (k + 1) and comprising j, a k calculated in step S8, the T ₁ in step S9 Is calculated. This value is stored in the microcomputer.

In the above sequence, the microcomputer reads the integrated value while changing the integration start timing t ₁ , searches for the largest one among the integrated values, finds the time j and k that intersect with 80% of the maximum value, the intermediate value is designated as the T _1.
FIG. 5 is a graph showing the relationship between i and the integral value.

Next, t ₁ and t ₂ are obtained by the procedure shown in the flowchart of FIG. First, in step S11, t _{2 is set} to 0, i is set to 0, and n is set to 5
0 and Δt are initialized to 1 ns, and L (0) to L (n) are initialized to 0. Next, at step S12, T ₁ and t ₂ stored and held
More t ₁ And then, where P indicates the set value of t ₂ which is initialized in step S1. Next, the values of t ₁ and t ₂ are specified to the integral control signal generator 10 in step S13. Then, in step S14, the microcomputer 8 outputs a one-pulse light emission command,
In step S15, the integral value A is read. In step S16, t ₂
Is increased by Δt, i is incremented, and L (i)
Is set to the integral value A. Next, in step S17, i is n
It is determined whether or not the above has been reached, otherwise, step S1
It returns to 2 and repeats in steps S12-S17 again. If i is greater than or equal to n, the process proceeds from step S17 to step S18. Steps
In S18, the maximum value of L (0) to L (n) is obtained, and is set to M. Then, in step S19, the minimum j is determined by j such that L (j) <0.9M <L (j + 1). In step S20, t ₁ and t ₂ are calculated as t ₂ = j × Δt × 1.5. Is calculated.

In the above sequence, the microcomputer reads the integral value while changing the integration execution time t ₂ , searches for the maximum value among the integrated values, and finds the minimum time that intersects 90% of the maximum value. , it is obtained by the t ₂ it by 1.5 times. t ₁ is obtained from the equation. The optimal t ₁ in this way
And t ₂ can be obtained. FIG. 7 is a graph showing the relationship between t ₂ and the integral value.

In the above embodiment, t ₂ = 10 in the initial setting of step S1.
0nS, Δt = 10 ns, and constant values such as 80% in step S8, 90% in step S19, and 1.5 times in step S20 are used. However, the present invention is not limited to this, and can be changed as appropriate and executed. .

By setting the above as obtained in integration start timing t ₁ and the integral running time t ₂ the optimum value to the integration control signal generating unit, accurate integration is can be realized. If this optimum value is set, for example, for each wafer, it is possible to control the exposure amount by absorbing the change over time of the excimer laser.

[Other Embodiments] In the above embodiment, an analog circuit is used as the integrator 6, but it may be implemented by a digital sampling and adding circuit. Further, time course of integration execution time t ₂ is smaller than the integration start timing t _1, it is also possible to t ₂ a constant value.

Further, in the above embodiment, the oscillation synchronizing signal is received from the laser light source 1 side. However, when the laser is externally synchronized, the same effect can be obtained by using the synchronizing signal given to the laser side as the oscillation synchronizing signal. . Further, the integrator 6 may be controlled by an integration window signal in which the integration start timing and the hold timing are set to the rising and falling of one signal. Further, in the embodiment, the microcomputer 8 determines the optimum timing by calculation. However, the calculation method is not limited to the above-described embodiment, and may be realized by an analog circuit.

[Effects of the Invention] As described above, according to the present invention, since the integration timing is optimized based on the output of the integration circuit, the influence of noise, offset drift, and aging of the laser is small, and the measurement accuracy is reduced. And stable exposure amount monitoring can be performed. Further, even if the timing of the synchronizing signal is changed by changing the type of laser as a light source, the exposure amount can be monitored without changing hardware and software.

[Brief description of the drawings]

FIG. 1 is a configuration diagram of an exposure monitoring apparatus according to an embodiment of the present invention, FIGS. 2 to 4 are timing charts for explaining signal timing of the apparatus of the present embodiment, and FIG. a graph showing the relationship between i and integrated value at the time of determining T _1, FIG. 6 is a flowchart for explaining a part of the processing of the microcomputer in the present embodiment, FIG. 7 is, t _1, FIG. 8 is a graph showing a relationship between t ₂ and the integral value when determining t _2. FIG. 8 is a flowchart for explaining a part of the processing of the microcomputer in the present embodiment. 1: excimer laser, 2: illumination system, 3: projection optical system, 4: half mirror, 5: photoelectric conversion element, 6: integrator, 7: A / D converter, 8: microcomputer, 9: laser controller , 10: integral control signal generator, M: mask, W: wafer.

──────────────────────────────────────────────────続 き Continuation of the front page (56) References JP-A-10-11683 (JP, A) JP-A-2-177414 (JP, A) JP-A-61-154128 (JP, A) JP-A-57-167 71132 (JP, A) JP-A-63-23316 (JP, A) JP-A-61-69132 (JP, A) JP-A-57-101839 (JP, A) JP-A-61-216324 (JP, A) JP-A-61-202437 (JP, A) JP-A-60-162258 (JP, A) (58) Fields investigated (Int. Cl. ⁶ , DB name) H01L 21/027 G03F 7/20 521

Claims (2)

(57) [Claims] An apparatus for monitoring an exposure amount when performing exposure using pulsed light emitted intermittently as exposure light, wherein the photoelectric conversion device receives a part of the light beam of the pulsed light and outputs an electric signal. Means for integrating the output of the photoelectric conversion means; and starting integration processing by the integration means after a predetermined delay time from the oscillation synchronization signal output before emission of the pulse light, and performing a predetermined integration execution time. And a means for generating an integration control signal so as to terminate the integration process after the above, and a means for optimizing the delay time and / or the integration execution time based on an output value of the integration means. Exposure monitor. 2. An exposure apparatus for performing exposure using pulsed light emitted intermittently as exposure light, comprising: photoelectric conversion means for receiving a part of the light beam of the pulsed light and outputting an electric signal; Means for integrating the output of the means; starting the integration processing by the integration means after a predetermined delay time from the oscillation synchronization signal output before the emission of the pulse light; and terminating the integration processing after a predetermined integration execution time An exposure apparatus comprising: means for generating an integration control signal; and means for optimizing the delay time and / or the integration execution time based on an output value of the integration means.