JPH10135112A - Measurement of resist photosensitive parameter and semiconductor lithography based thereon - Google Patents

Abstract

PROBLEM TO BE SOLVED: To accurately measure resist photosensitive parameters easily. SOLUTION: Light having a photosensitive wavelength is illuminated on a resist film L, formed on a substrate to continuously detect a reflected light intensity distribution on a plane of the pupil 17 and to calculate transient changes in the thickness, refractive index and attenuation coefficient of the resist film on the basis of the detected result. On the basis of transient change characteristics of the calculated film thickness and refractive index, a theoretical value of a mean attenuation coefficient within the resist film L is calculated, and the theoretical value is collated with a transient change characteristic of a mean attenuation coefficient to calculate a photosensitive parameter. The photosensitive parameter is used to find a solid configuration of a latent image within the resist film L, and to control the exposure conditions of an aligner on the basis of the solid configuration.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a lithography technique for forming device patterns such as circuits and wirings constituting a semiconductor integrated circuit on a substrate, and more particularly, to measuring a photosensitive parameter of a positive type photoresist film and measuring the measurement result. The present invention relates to a semiconductor lithography technique capable of controlling a semiconductor lithography process using the same.

[0002]

2. Description of the Related Art In the manufacture of semiconductor integrated circuit devices,
2. Description of the Related Art A photolithography technique for forming a resist pattern corresponding to a device pattern such as a circuit or a wiring constituting a semiconductor integrated circuit on a substrate on which a photosensitive photoresist film is formed has been applied.

When a device pattern is formed on a semiconductor wafer, a resist solution is applied to the wafer to form a resist film, and after prebaking is completed, the device pattern is formed on the resist film using an exposure apparatus. The device pattern is formed by baking and treating the chemically changed resist film with a developing solution to elute unnecessary portions. In such a photolithography technique, it is an important issue in pattern miniaturization whether a device pattern can be formed on a substrate with desired dimensional accuracy.

[0004] Therefore, in the semiconductor lithography process,
Various works are performed to maintain dimensional accuracy,
One of them is the management of the film thickness. When the resist film thickness changes, the effective absorption efficiency of exposure energy changes, and a dimensional change called a swing curve occurs. Therefore, in order to check the operation of the rotary resist coating device and the baking furnace, the resist film thickness is measured and managed using an optical measuring device such as a spectral reflection type or an ellipsometric type.

In recent formation of submicron device patterns, observation and length measurement of developed resist patterns with a scanning electron microscope (SEM) are the most important methods for improving and managing dimensional accuracy. It has become. Based on the measurement results of the SEM, initial settings of exposure conditions such as the exposure amount and focus in the exposure apparatus are performed, and dimensional accuracy is periodically inspected. Also,
The pattern size is controlled by increasing or decreasing the exposure amount of the exposure apparatus based on the measurement result of the SEM with reference to a relationship chart between the exposure amount and the size acquired in advance. Since this relationship diagram varies depending on the type of pattern, the difference in the exposure apparatus, and the type of the resist solution, it is necessary to prepare each combination as needed.

Factors that degrade the dimensional accuracy of a pattern include variations in the thickness of a thin film layer such as an oxide film formed on a substrate, the resist material, and the temperature of a developing solution, in addition to the resist film thickness. Here, the optical properties of the resist material include a refractive index, an extinction coefficient, and photosensitive parameters A and B.
And C. These exposure parameters A, B and C
Is the input value for simulating the three-dimensional shape of the photoresist pattern.The concept, measurement principle and outline of the measurement device are described in IEEE Trans.Electron Devices.ED.
-22,7. Freder in pp445-452 (issued July 1975)
characterization of Positive P by ick H. Dill et al.
hotoresist ", and IEICE Transactions C-II Vol.J77-C-II No.12 pp.555-563 (1994
Sekiguchi et al., Published in December, entitled "Development of Photosensitivity Parameter (A, B, C) Measurement Apparatus for Photoresist".

Here, the photosensitive parameter A represents the light absorption coefficient of the photosensitive agent before exposure, the photosensitive parameter B represents the light absorption coefficient of the base resin, and the photosensitive parameter C represents the decomposition speed (resist sensitivity) of the photosensitive agent. Represent.

FIG. 15 is a schematic view showing a conventional measuring device for measuring the photosensitive parameters A, B and C. Conventionally, the photosensitive parameters A, B are detected by detecting light transmitted through a resist film L. And C are measured. In this measuring device, the transparent substrate T on which the resist film L is formed is arranged between a collimator lens 52 that converts light from the light source 51 into parallel light and a detector 53. With the transparent substrate T arranged, the shutter 54 is opened and the light from the light source 51 converted into parallel light by the collimator lens 52 is converted into light having the same wavelength as that of the pattern exposure by the monochromatic filter 55, and this is converted into a resist film L.
, The light transmitted through the transparent substrate T is detected by the detector 53, and a transient change characteristic of the transmittance as shown in FIG. 16 is obtained from the output signal of the detector 53.

[0009] Photoresists include negative resists and positive resists, and when exposed to light, the negative resist undergoes a polymerization or crosslinking reaction to become insoluble or hardly soluble in a developer and remains on the substrate surface until after development. The positive resist is a photosensitive material having characteristics such that it is decomposed when irradiated with light, becomes soluble in a developing solution, and is removed from the substrate surface during development.

A positive resist is composed of three components, a base resin, a photosensitizer and an organic solvent. The photosensitizer undergoes a chemical reaction upon irradiation with ultraviolet rays to solubilize the resist film.
At this time, since the light absorption coefficient of the photosensitive agent decreases, FIG.
As shown in (2), the transmittance of the resist film increases, that is, bleaching occurs. When the chemical reaction of the photosensitive agent is completed, the transmittance does not increase, and this state is called bleach-out. Assuming that the transmittance before exposure is T ₀ , the transmittance during bleach-out is T∞, the light intensity on the resist film surface is 1 and the resist film pressure is d, the photosensitive parameters A, B and C are given by the following equations. .

[0011]

(Equation 1)

In order to simulate the shape of a photoresist pattern, first, the intensity of an optical image projected on a resist film is calculated, and then a latent image, that is, a resist image using a photosensitive parameter and a calculated value of a standing wave in the resist film. The photosensitizer concentration distribution in the film is calculated, and further, the development using the resist dissolution rate parameter, that is, the development parameter is used to calculate the resist pattern shape after development. The photosensitive parameters A, B and C are input parameters required for calculating the latent image.

The outline of the simulation method is described in SPIE Vol.5.
38 Optical Microlithography IV.pp207-220 (issued in 1985) by Chris A. Mack, "PROLITH: a compre
hensive optical lithography model,
And IEICE Transactions C-II Vol.J76-C-II No.8
pp 562-570 (issued in August 1993) by Minami et al. in a document entitled "Study of Defocus Simulation Using Actual Dissolution Rate in Photolithography". The simulation technique is an effective method for examining the selection of a photoresist, an exposure amount, a focus margin, and the like at a process design stage.

[0014]

However, the simulation of the photoresist shape involves controlling the lithography process,
For example, it is not very used for determining operating conditions of an exposure apparatus or the like. The reason is that unless accurate values of various parameters in the manufacturing process are input, the simulation result does not match the actual resist shape, and the various parameters cannot be easily and accurately obtained in the manufacturing process.

For example, when the measuring apparatus shown in FIG. 15 is used, light is transmitted through the resist film L to measure a transient change in the transmittance, and the photosensitive parameters A, B and C are obtained by the above-described formula. In this method, there is no reflection at the interface S1 between the resist film L and the transparent substrate T, and no reflection at the surface S2 of the transparent substrate T on the detector side. Therefore, for strict measurement, both surfaces S1 and S2 must be non-reflective surfaces. To make the surface S1 non-reflective, the transparent substrate T
Must be matched with the resist film L or an antireflection film must be provided on the surface S1. However, the refractive index of the substrate material is discrete, and it is difficult to realize an antireflection film that satisfies the non-reflection condition due to material restrictions, and the surface S2 has the same problem. In addition, a mismatch between the refractive indexes due to the type of the resist film L, a difference in the manufacturing lot, and the like, and a change in the refractive index of the resist during exposure also cause measurement errors.

The thickness d of the resist film L on the transparent substrate T needs to be measured in advance by another device in order to input a fixed value. The use must be a factor in reducing productivity.

On the other hand, the measurement by SEM is an optimal method for strictly determining whether or not the dimensions of a device pattern formed on a wafer are within an allowable range due to its high resolution performance. However, it is difficult to identify a cause such as whether the dimensional variation is caused by non-uniformity of the substrate or a variation in resist sensitivity, and to quantitatively determine a correction amount of a manufacturing process. In addition, since the SEM requires evacuation time and the like, the throughput is slower and the apparatus price is higher than that of optical equipment. For this reason, mass production of semiconductor devices on the premise of heavy use of SEM is not a desirable mode.

An object of the present invention is to make it possible to easily measure resist exposure parameters with high accuracy.

Another object of the present invention is to measure a resist film formed on a substrate made of a material optically opaque to an exposure wavelength, such as a silicon wafer, and to measure a photosensitive parameter of the resist film. Is to do so.

Still another object of the present invention is to measure a photosensitive parameter of a resist film formed on a substrate for forming a semiconductor device pattern and control exposure conditions in a lithography process to improve the productivity of a semiconductor integrated circuit device. Is to improve.

Still another object of the present invention is to enable control of a lithography process by measuring a photosensitive parameter.

The above and other objects and novel features of the present invention will become apparent from the description of the present specification and the accompanying drawings.

[0023]

SUMMARY OF THE INVENTION Among the inventions disclosed in the present application, the outline of a representative one will be briefly described.
It is as follows.

That is, in the method of measuring the resist exposure parameter of the present invention, the resist film formed on the substrate is condensed and irradiated with light having a photosensitive wavelength, the reflected light after the start of illumination is captured by a lens, and the pupil of the lens is captured. A change with time of the reflected light intensity distribution on the surface is detected, and a photosensitive parameter is calculated based on the detected value.

In the method for measuring a photosensitive parameter according to the present invention, the film thickness, the refractive index, and the extinction coefficient of the resist film are continuously detected to calculate the transient characteristics of the resist film. The theoretical value of the average extinction coefficient in the resist film is calculated based on the transient change characteristic of the resist film, and the theoretical value is compared with the transient change characteristic of the extinction coefficient to calculate the photosensitive parameter.

Further, in the method for measuring a photosensitive parameter according to the present invention, the photosensitive parameter of a resist film formed on a substrate opaque to the photosensitive wavelength may be measured or the photosensitive parameter formed on the substrate may be measured. A photosensitive parameter of a resist film formed on a substrate film opaque to a wavelength is measured.

The reflected light may be detected after passing through a polarizing element, or the reflected light may be detected after passing through a stop provided on an image forming surface of a resist film.

As described above, in the method for measuring the resist photosensitivity parameter of the present invention, the reflected light from the resist film is detected without measuring the photosensitivity parameter based on the light transmitted through the resist film. Therefore, the photosensitive parameter can be detected with high accuracy for a resist film formed on a semiconductor wafer or the like actually used as a substrate.

According to the semiconductor lithography method of the present invention, there is provided a resist coating step of forming a resist film by applying a resist solution to a substrate, an exposure step of exposing a pattern to the resist film, and a developing step of eluting an unnecessary portion of the resist film. A semiconductor lithography method comprising the steps of:
The refractive index and the extinction coefficient are continuously detected, the transient characteristics thereof are calculated, the resist film parameters of the resist film are calculated by the reflected light of the light applied to the resist film, and the resist film parameters and pattern information are calculated. Calculating a latent image in the resist film based on the stepper optical conditions, calculating a three-dimensional shape of the latent image formed in the resist film based on the calculated latent image and the development parameters, and controlling exposure conditions in the exposure step. It is characterized by doing.

Further, the semiconductor lithography method according to the present invention is characterized in that a resist film parameter, pattern information, a substrate film formation parameter and a stepper optical condition of the resist film formed on the substrate via the substrate film formation are used. Calculate the latent image. The resist film parameter is a film thickness, a refractive index and a photosensitive parameter, the development parameter is a resist dissolution rate, and the substrate film formation parameter is a film thickness and a thickness of a thin film formed between the substrate and the resist film. It is an optical constant.

Further, according to the semiconductor lithography method of the present invention, any one of the resist film parameter, the development parameter and the substrate deposition parameter is measured, and it is determined whether the measured value is within the specified value. An abnormal part in the lithography process is detected from the parameter.

As described above, in the semiconductor lithography method of the present invention, since the photosensitive parameters can be detected and collected in the production line of the semiconductor integrated circuit device, the exposure parameters can be controlled by utilizing the parameters. A semiconductor integrated circuit can be manufactured efficiently.

[0033]

Embodiments of the present invention will be described below in detail with reference to the drawings.

FIG. 1 is a schematic view showing an apparatus for measuring a photosensitive parameter of a positive photoresist according to an embodiment of the present invention. This apparatus comprises a support stage for supporting a semiconductor wafer W having a resist film L formed thereon. The resist film L is irradiated with light from an illumination system. The illumination system has a light source 11, a collector lens 12, a monochromatic filter 13, a shutter 14, a half mirror 15, and an objective lens 16. The light emitted from the light source 11 includes light having a wavelength sensitive to the resist, and the monochromatic filter 13 extracts light equivalent to monochromatic light used for transferring a device pattern of a semiconductor exposure apparatus.
The monochromatic light is condensed by the objective lens 16 and is irradiated on the resist film L.

The detection system that receives the reflected light from the resist film L and detects the light intensity distribution on the pupil plane 17 of the objective lens 16 has the objective lens 16 and the detector 21.
A lens 22, a stop 23 and a polarizing plate 24 are arranged between them, and the light intensity distribution on the pupil plane 17 is imaged on the detector 21. The detector 21 is formed by one-dimensional or two-dimensional arrangement of photoelectric conversion elements.

FIG. 2 is a block diagram showing a control circuit of the measuring apparatus shown in FIG. 1, and the control circuit comprises a central processing unit CP.
U31, and a Z-axis motor 3 for operating the support stage 10 in the optical axis direction by a signal from the CPU 31.
2. The operations of the shutter 14 and the polarizing plate 24 of the illumination system are controlled. A signal corresponding to the light intensity distribution from the detector 21 is sent to the CPU 31.
The analog signal from 1 is converted into a digital signal by an A / D converter (not shown) and sent to the CPU 31. Members such as the above-described shutter 14 that constitute the measuring device include:
The operation is controlled according to a procedure stored in the ROM 33 in advance.

The control unit has a waveform memory 34 for storing a detection waveform corresponding to the light intensity distribution from the detector 21. Since the resist film L changes by photochemical reaction when irradiated with light, The detection signal is stored in the waveform memory 34 in synchronization with the opening of the shutter 14, and a change in the light intensity distribution on the pupil plane 17 from the start of light irradiation to the resist film L to the start of exposure is detected. Further, the control circuit has a reflectance calculator 35 for calculating the reflectance of the resist film L from the waveform of the light intensity distribution stored in the waveform memory 34, and a transient change calculator 36. The unit 36 calculates the thickness d of the resist film L, the refractive index n, and the extinction coefficient k from the reflectance obtained by the reflectance calculation unit 35, and performs the same for the detected waveform at each time after the start of illumination. By repeating the process, the transient characteristics of the film thickness d, the refractive index n, and the extinction coefficient k are obtained. A photosensitive parameter calculator 37 is connected to the CPU 31 to calculate the photosensitive parameters A, B, and C based on the transient characteristics.

Next, the principle of calculating the reflectance from the light intensity distribution on the pupil plane 17 to obtain the optical characteristics n, k, and d of the resist film L will be described. In this embodiment, the resist film L formed on the silicon semiconductor wafer W is measured, and the resist film L is considered to be a homogeneous and isotropic single-layer thin film sandwiched between parallel planes. In the following description, the symbol i indicates an imaginary unit, and the suffix j identifies a medium. The air is j = 0, the resist film L is j = 1, and the semiconductor wafer W is j = 2. Further, a thin film d _j , a refractive index n _j ,
Expressed as extinction coefficient k _j , complex refractive index N _j = n _j -ik _j , wavelength λ, incident angle θ _j , and amplitude reflectance r. d _j ,
n _j , k _j , and λ are real numbers, and N _j , θ _j , and r are parameters of complex numbers.

As shown in FIG. 3, one light beam 5 incident on the resist film L is amplitude-divided at the boundary, and is repeatedly reflected at the air-resist boundary and the resist-wafer boundary. Then, a large number of parallel light rays whose amplitude gradually decreases are reflected from the surface of the resist film L. When the film thickness d ₁ is in the order of wavelength, these reflected light beams have coherence, and the result of amplitude addition corresponds to the reflectance. The amplitude addition is equivalent to converging the reflected light flux to one point on the pupil plane 17 by the objective lens 16, and the amplitude of the convergence point P is obtained by multiplying the amplitude of the incident light by the amplitude reflectance r. The amplitude reflectance r is given by the following equation.

[0040]

(Equation 2)

Here, δ is the optical path difference caused by one reciprocation of the light beam through the resist film L, and r _{j, j + 1} is the medium j to the medium j
The amplitude reflectance of the boundary surface when light is incident on +1 and is given by the Fresnel equation. This depends on the polarization state, and for P-polarized light,

[0042]

(Equation 3)

For S-polarized light,

[0044]

(Equation 4)

Is as follows.

Further, according to Snell's law, the following relation holds.

[0047]

(Equation 5)

Accordingly, the intensity reflectance R is represented by the following equation.

[0049]

(Equation 6)

Here, r ^* is a conjugate complex number of r, and g is a function.

In this embodiment, n ₀ , k ₀ , n ₂ , k
₂ and λ are fixed values, and d ₁ , n ₁ and k ₁ are determined. Therefore, the reflectance characteristic R (θ ₀ ) when the incident angle θ ₀ is changed is detected. Then, three parameters are determined by performing the reflectance matching calculation by the least square method. This can be done by examining the degree of coincidence between the theoretically calculated reflectance characteristic _Rth and the detected reflectance characteristic R. That gives the evaluation function M ₁ degree of matching for example by the following equation.

[0052]

(Equation 7)

By searching for a case where the evaluation function M ₁ is minimized by this equation, the resist film parameters d ₁ , refractive index n ₁ , and extinction coefficient k ₁ can be determined.

Next, a method for measuring the reflectance characteristic will be described.

In FIG. 3, the objective lens 16 satisfies the sine condition. At this time, the position where the reflected light beam converges on the pupil plane 17 is given by the following equation.

[0056]

(Equation 8)

Here, X is the distance from the optical axis of the objective lens 16 to the convergence point P, θ _L is the incident angle of the optical axis of the objective lens 16, and f is the focal length of the objective lens 16. The intensity at the point P corresponds to the reflectance of the resist film L at the incident angle θ ₀ .

As shown in FIG. 4, considering that the resist film L is illuminated with a convergent light beam parallel to the paper surface sandwiched between the incident light beams 5a and 6a, the convergent light beam is a collection of light beams over a certain incident angle range. And the respective reflected rays 5b, 6b
Has the intensity reflectance according to the equation (2a), and the convergence position according to the equation (8). That is, the pupil plane 17 has a reflectance characteristic R
A change in brightness corresponding to (θ ₀ ) occurs. This is called equi-tilt interference fringes. Then, if the equi-tilt angle interference fringes are imaged by the array sensor, that is, the detector 21, the reflectance characteristic R (θ ₀ ) can be detected at once.

When monochromatic light having a photosensitive wavelength is used as the illumination light, the illumination causes a chemical reaction in the resist film L, and the optical characteristic value gradually changes. Therefore, the equi-tilt interference fringes also change. Therefore, if the intensity of the pupil plane 17 is continuously imaged by the detector 21, a transient change R (θ ₀ , t) of the reflectance characteristic during exposure as shown in FIG. 4 can be detected. Then, by performing the reflectance comparison calculation at each time point, the transient change characteristics d ₁ (t), n ₁ (t), and k ₁ (t) of the film thickness, the refractive index, and the extinction coefficient can be calculated.

As shown in FIG. 5, in the equi-tilt interference fringe, the optical axis of the objective lens 16 is perpendicular to the sample (θ _L = 0).
But it can be detected. In this case, the position where the optical axis of the objective lens 16 is perpendicular to the sample surface and the reflected light beam converges on the pupil surface 17 is given by the following equation.

[0061]

(Equation 9)

The maximum incident angle is limited by the numerical aperture (NA) of the objective lens 16. FIG. 5 shows that a ray passing through Pin on the pupil plane 17 is reflected on the sample surface in the field of view, and the axially symmetric position Pout
FIG. Concentric circular oblique interference fringes are formed on the pupil plane 17, and the intensity distribution in the radial direction from the center O corresponds to the reflectance characteristic. As shown in FIG.
→ If Pout is parallel to the direction of polarization, the reflectance characteristic is P-polarized light, and if Pout is perpendicular, it corresponds to the characteristic of S-polarized light. This can be achieved by adjusting the attachment of the polarizing plate 24 and the detector 21 shown in FIG. By using the polarizing plate 24, there is no need to strictly set the polarization characteristics of the optical components such as the half mirror 15, and there is an advantage that the optical system can be easily manufactured. Further, if detection is performed in the state of linear polarization, the above equation (3) is obtained.
Theoretical calculation in the reflectance matching according to (4) can also be reduced.

By the way, as shown in FIG. 6, it is difficult to illuminate with a uniform illuminance over a wide field of view with an ordinary optical device. Therefore, in the present embodiment, the visual field to be detected is limited, and only the reflected light is detected from a region where the illuminance is uniform.
For this purpose, a stop 23 is provided in the detection optical system shown in FIG. The diaphragm 23 is composed of the objective lens 16 and the lens 22.
Thus, the sample surface is placed at a position in the optical axis direction where the image is formed. Thereby, the intensity distribution of the pupil plane 17 to be detected can be limited to the reflected light from the region where the illuminance on the sample surface is uniform. In other words, it can be limited to light reflected from a region where the progress of the chemical change of the resist film is uniform.

FIG. 7 shows a measurement example of the transient change characteristic of the intensity reflectance, and it can be seen that the reflectance increases as the exposure time elapses. This is because the resist film was bleached by the exposure and the transparency was increased. FIG. 8 shows the result of calculating the optical characteristic value by repeating the reflectance collation calculation of the equation (7), and the calculation result of the transient change of the extinction coefficient k. The bleaching phenomenon is quantitatively expressed by an optical constant. It was done. The extinction coefficient k represents the degree to which light is absorbed. As the extinction coefficient k increases, light is absorbed and the transparency decreases.

In FIG. 8, the photosensitive parameter A corresponds to a value obtained by multiplying the extinction coefficient k when the exposure time is 0 by a predetermined coefficient because this is the light absorption coefficient of the photosensitive agent before exposure. Since this is the light absorption coefficient of the base resin, the photosensitive parameter B corresponds to a value obtained by multiplying the extinction coefficient when the exposure time is infinite by a predetermined coefficient.
Further, since the photosensitive parameter C is the decomposition speed of the photosensitive agent, it corresponds to the speed at which the extinction coefficient k decreases.

Next, the procedure for calculating the photosensitive parameters A, B and C will be described. When the thin film is irradiated with monochromatic light,
Since light traveling in the vertical direction in the film interferes, the light intensity varies depending on the depth. The light intensity fluctuates periodically in the depth direction, which is called a standing wave. FIG. 9 shows a calculation example of a standing wave in a resist film having a film thickness d. The horizontal axis represents the depth Z from the resist film surface (resist-air interface), and the vertical axis represents the standing wave intensity.

Z = 0 means the resist surface, and Z = d means the resist-semiconductor wafer interface. t = 0 indicates a state before the exposure, t = t ₁ , t ₂ indicates a state after the start of the exposure, and t ₁ <t ₂ . The reason why the standing wave intensity decreases as the depth increases is that the resist film absorbs light.

FIG. 10 is a theoretical calculation example of the extinction coefficient distribution in the thickness direction in the resist film. Before exposure (t = 0), the extinction coefficient is constant because the material is uniform. However, since the standing wave is generated as shown in FIG. 9 after the start of the exposure, the way of the chemical change of the resist film is different in the depth direction. Therefore, the extinction coefficient also fluctuates periodically in the depth direction.

The strength of the standing wave in FIG. 9 and the theoretical calculated value k _th of the extinction coefficient in FIG. 10 are the same as those in SPIE Vol.
"PROLITH: a comprehensive opti" by Chris A. Mack in crolithography IV, pp.207-220 (issued in 1985)
The method may be performed by a method described in a document titled “cal lithography model”. To summarize this method, the exposure time t,
Using the illuminance 1, the film thickness d, the refractive index n, the photosensitive parameters A, B and C, and the optical constants of air and the substrate as input variables, the calculation of the standing wave and the calculation of the extinction coefficient are performed alternately, and the change in the time direction is obtained. It is to be calculated. That is, the extinction coefficient distribution k _th in the film thickness direction is expressed by the following function.

[0070]

(Equation 10)

Here, in order to calculate the photosensitive parameters A, B and C, the average extinction coefficient k _{th ave} (the average value of the extinction coefficient in the film thickness direction) represented by the following equation is newly defined as follows. .

[0072]

[Equation 11]

The above-described methods for measuring the transient changes in the film thickness, refractive index, and extinction coefficient assume that the resist film is a homogeneous and isotropic thin film sandwiched between parallel planes. That is, the transient change characteristic k ₁ (t) of the extinction coefficient obtained by this method is the average extinction coefficient k
_{th ave} (t, A, B, C). Therefore, A, B
To calculate the C and C parameters, the following evaluation function M
_The least squares method may be applied using ₂ .

[0074]

(Equation 12)

The photosensitive parameter calculator 37 receives the measurement results of the transient changes in the film thickness and the refractive index and theoretically calculates the transient changes in the average extinction coefficient k _{th ave} . Then, the photosensitive parameters A, B, and C are obtained by collating and calculating the transient change measurement value k _{1 of} the extinction coefficient using the equation (12).

Next, a procedure for measuring a photosensitive parameter of the resist film L by the measuring apparatus shown in FIG. 1 will be described with reference to a flowchart shown in FIG.

First, the pupil plane waveforms of the reference plane and the dark level are detected as shown in steps S1 and S2. As the reference surface, for example, a material whose reflectance characteristic can be calculated by theoretical calculation, such as a metal reflecting surface having a known refractive index and extinction coefficient, such as a silicon semiconductor wafer, is used. In order to obtain the dark level waveform, the sample may be retracted from below the objective lens 16 and detected, and these values are stored in the memory.

Next, the wafer W on which the resist film L to be measured is formed is set under the objective lens 16, and the waveform of the pupil plane 17 is continuously detected after the start of illumination, and the waveform memory shown in FIG. 34 (steps S3 and S4). From the detected waveforms of the film to be measured and the reference surface, step S
In 5, the dark level is corrected. In this dark level correction, as shown in FIG. 12, the dark level waveform is subtracted from the waveform of the film to be measured, and the dark level waveform is also subtracted from the waveform of the reference surface, so that stray light of the measuring optical system and the detector. This is performed to remove an offset caused by an error in an electric circuit.

On the basis of the waveform signals I and Iref corresponding to the intensity of the reflected light after the dark level correction, the reflectance calculator 35 shown in FIG. 2 calculates the reflectance characteristic R as shown in step S6. The calculation procedure is as shown in FIG. 12, where the waveform signal I of the film to be measured is a value obtained by multiplying the reflectance R by the illuminance distribution D, and the waveform signal Iref of the reference surface is the theoretical reflectance R of the reference surface.
Since ref is a value obtained by multiplying the illuminance distribution D, these values are divided to remove the illuminance distribution term D to obtain the value of I ₂ first.
Since this value I ₂ is represented by R / Rref, the reflectance characteristic R is obtained by multiplying I ₂ by the value of the theoretical reflectance Rref.

After the reflectance characteristic R is obtained in this manner, the transient change calculator 36 performs a reflectance collation calculation in step S7 to obtain a film thickness d and a refractive index, which are optical characteristic values of the film to be measured. The rate n and the extinction coefficient k are calculated. The calculation of these resist film characteristic values is repeatedly performed until it is determined in step S8 that the calculation is completed up to a predetermined number of times.
When each of the optical characteristic values at each time point is obtained, as shown in step S9, the transient change characteristics d ₁ (t), n ₁ (t), and k ₁ (t) of the film thickness, the refractive index, and the extinction coefficient are obtained. Can be calculated.

Next, the values of the transient change characteristics d ₁ (t) and n ₁ (t) obtained in step S 9 are substituted into the above-mentioned equation (11), and the photosensitive parameter calculation unit 37 executes step S 1
As shown in FIG. 0, the transient change of the average extinction coefficient k _{th ave} is theoretically calculated. In step S11, the average extinction coefficient k _{th ave} and the measured value k (t) of the extinction coefficient are substituted into equation (12), and the photosensitive parameters A, B, and C are calculated by the least square method.

In the illustrated embodiment, the parameters A, B and C are calculated using the transient characteristics of the film thickness, refractive index and extinction coefficient detected by the optical system shown in FIG. Note that first, the sample is illuminated simultaneously by light rays of different angles of incidence, and then that the sample is illuminated by all light rays that have passed through the circular pupil plane.

For this reason, it is necessary to perform a theoretical calculation of a standing wave in consideration of the incident angle of the illumination light beam and its weight. FIG. 13 shows a pupil plane 17 and a sequence of sampling points for standing wave calculation.
The sample is illuminated with all light passing through pupil plane 17. At this time, the light passing through the ring portion in the figure illuminates the sample at the same incident angle. However, if the radius differs, the incident angle changes, and the area of the ring portion corresponding to the light intensity also changes. Therefore, a standing wave is calculated using the area of the ring portion as the weight of the light beam at each incident angle.

In the actual calculation, the intensity of the standing wave due to the light beam that has passed through the sampling point sequence discrete in the radial direction from the center O is calculated. Therefore, as shown in FIG.
When the sampling of the standing wave calculation is Δr on the pupil plane 17, the radius of the ring inner diameter is r−Δr / 2, and the outer radius is r.
+ Δr / 2, and the weight is calculated from the area of the ring portion.
This is multiplied by the intensity of the standing wave at each incident angle, and the intensity of the standing wave is calculated by integrating all the incident angles at all incident angles. Based on this, the theoretical calculated value of the average extinction coefficient k _{th ave}
Is calculated, and the photosensitive parameters A and B are calculated by the equation (12).
And C may be obtained.

In the description so far, the resist film L applied and formed on the surface of the silicon wafer W was measured. However, several thin films such as insulating films, that is, substrate films are formed on the surface of the silicon wafer W. Further, the present invention can be applied to the measurement of the resist film L formed thereon. In this case, the film thickness, the refractive index, and the extinction coefficient of the resist film are calculated in consideration of the optical influence of the formation of the substrate under the resist film L, and the theoretical calculation of the standing wave in the resist film is performed. Just do it. That is, the theoretical calculation of the standing wave of the reflectance distribution may be performed by setting the film thickness and the optical constant of the substrate film formation to known values. In this case, there is no change in the detection of the reflectance distribution on the pupil plane 17, and the photosensitive parameters A, B and C can be measured by the measuring device shown in FIG.

Next, a description will be given of a semiconductor lithography method using the above-described method of measuring the photosensitive parameter of a resist film.

FIG. 14 is a view showing a semiconductor lithography process. In this process, a resist solution is applied onto a silicon wafer W using a resist coating device 41a such as a spin coater which is a rotary resist coating device, and a resist film L is formed. A resist coating process 41 for forming a mask, an exposure process 42 for exposing the mask pattern formed on the photomask to the resist film L using a projection exposure device 42a such as a stepper, and an unnecessary portion of the resist film L is eluted by a developing device 43a. And a developing step 43 for etching a specific portion of the thin film formed on the surface of the wafer W, a resist removing step, and the like.

Step 41 of Applying Resist to Silicon Wafer W
The resist film parameters including the film thickness, the refractive index, and the photosensitive parameters A, B, and C of the resist film L formed by the above are obtained by the measuring device 44 as described above, and the obtained resist film parameters are sent to the simulation section 45. It is supposed to be. Exposure parameters A, B and C
May be used as the resist film L for detecting the resist film parameters including the semiconductor film W, which is formed of the same material as the semiconductor wafer W. A resist film L may be formed on a sample semiconductor wafer W as a substrate, and the resist film L may be used to measure resist film parameters such as a photosensitive parameter.

A control signal is sent from the control unit for controlling the operation of the developing device 43a to the developing device 43a. The control unit sends a developing parameter representing the resist dissolution rate to the simulation unit 45. It has become. To measure the resist dissolution rate given as the development parameter, a conventional method that regards a decrease in resist film during development as a change in interference intensity can be applied.

The simulation section 45 receives pattern information, substrate deposition parameters, stepper optical conditions, and the like. Pattern information is information on the shape of the device pattern drawn on the photomask,
It is obtained by searching for mask information corresponding to the product name and process of the semiconductor integrated circuit device. The substrate film formation parameter, that is, the substrate film formation information is a design value of each thin film laminated on the surface of the silicon wafer W, an optical constant, or an actually measured value collected on a production line. Can be calculated. However, if a resist film is formed directly on the surface of the wafer W, no substrate deposition parameters are required. Further, the stepper optical conditions are values required for calculating optical imaging performance, such as the wavelength for operating the exposure device 42a, the number of NAs of the projection exposure lens, the coherency coefficient, and the illuminance.

The latent image data is calculated from the resist film parameters including the film thickness, refractive index, and photosensitive parameters A, B, and C sent from the measuring device 44, the pattern information, the stepper optical conditions, and the substrate film forming parameters. ,
The resist three-dimensional shape is obtained by sequentially calculating the interface dissolution phenomenon based on the latent image data and the development parameters.

As described above, if the three-dimensional shape of the resist can be calculated, it is possible to set the optimum value of the exposure condition such as the exposure amount, which is the dimension adjustment parameter of the exposure device 42a. Also,
Due to variations in the production lot of the resist material,
When the photosensitive parameters A, B, and C, which are the resist film parameters, change, the optimum exposure amount can be corrected in a short time.

Conventionally, photosensitive parameters A, B and C
There is no means for simply and accurately measuring the resist parameters, and it is difficult to apply the resist shape simulation technology to the management of the lithography process. However, according to the present invention, the photosensitive parameters A and B are obtained from the reflected light of the resist film. Since C and C can be measured, these parameters can be easily collected on a production line, so that exposure conditions for obtaining a desired resist three-dimensional shape can be set.

In another embodiment of the semiconductor lithography method, a resist film parameter, a development parameter, and a substrate deposition parameter may be monitored to manage the flow of exposure work. In other words, each parameter for detecting an abnormality in the photolithography process is measured, and if these measured values do not deviate from the standard values, it is possible to omit the dimension checking operation by SEM before exposing a large number of wafers. If it becomes possible and the dimensional inspection by the SEM can be omitted, the productivity can be improved. Further, when the measured value deviates from the standard value, the production is interrupted, an abnormal portion of the measured data is examined, and it is possible to quickly determine which of the film formation, the resist coating and the development is the cause.

Although the invention made by the present inventor has been specifically described based on the embodiments, the present invention is not limited to the above embodiments and can be variously modified without departing from the gist thereof. Needless to say,

For example, in the case shown, the photosensitive parameters A, B and C of a resist film formed on a silicon wafer are measured using a silicon wafer as a substrate. However, when a photomask is manufactured, it is formed on the substrate of the photomask. The present invention may be applied to measure the photosensitive parameters A, B, and C of the resist film obtained.

[0097]

Advantageous effects obtained by typical ones of the inventions disclosed in the present application will be briefly described.
It is as follows.

(1) Since the photosensitive parameter is calculated by detecting the reflected light of the light of the photosensitive wavelength irradiated on the resist film from the resist film, it is opaque to the illumination wavelength such as a semiconductor wafer. The resist film formed on the substrate can be measured.

(2) This eliminates the need for a transparent substrate dedicated to measurement, and reduces the number of measurement error factors. Therefore, the resist exposure parameters can be measured with high accuracy without increasing the inspection cost. Reliability can be improved.

(3) Since the film thickness can be measured simultaneously with the detection of the photosensitive parameter, the measurement of the resist film thickness can be performed, and the refractive index can be measured. The film parameters can be managed, and the fluctuation factors of the semiconductor lithography process can be quantitatively monitored in more items, so that the productivity of the semiconductor integrated circuit device can be improved.

(4) Since input parameters for simulating the three-dimensional shape of the resist formed on the developed resist film can be collected on the production line, it is possible to control the exposure conditions based on the collected data. As a result, it is possible to strictly control and monitor the dimensions of the semiconductor device pattern, and it is possible to efficiently produce a highly accurate device pattern.

[Brief description of the drawings]

FIG. 1 is a schematic configuration diagram showing a measuring apparatus to which a method for measuring a resist exposure parameter according to an embodiment of the present invention is applied.

FIG. 2 is a block diagram showing a control circuit of the measuring device shown in FIG.

FIG. 3 is a principle diagram showing a state of repeated reflection interference on a resist film.

FIG. 4 is a principle diagram of a method for measuring a resist exposure parameter.

FIG. 5 is a schematic diagram showing an optical system of the measuring device shown in FIG.

FIG. 6 is an explanatory diagram of uniforming the illuminance distribution by limiting the detection field of view in the measuring device shown in FIG. 1;

FIG. 7 is a graph showing an example of a detection waveform from a resist film.

FIG. 8 is a graph showing a measurement example of a transient change of an extinction coefficient.

FIG. 9 is a graph showing a calculation example of a standing wave in a resist film.

FIG. 10 is a graph showing a calculation example of an extinction coefficient in a resist film.

FIG. 11 is a flowchart showing a procedure for calculating resist exposure parameters A, B, and C;

FIG. 12 is a flowchart showing an algorithm for calculating resist exposure parameters A, B, and C;

FIG. 13 is a schematic diagram showing the concept of weight in standing wave calculation.

FIG. 14 is a process chart showing a semiconductor lithography method according to an embodiment of the present invention.

FIG. 15 is a schematic view showing a conventional resist exposure parameter measuring device.

FIG. 16 is a graph showing measurement results obtained by the measurement device shown in FIG.

[Explanation of symbols]

 L resist film W wafer 5a, 6a incident light 5b, 6b reflected light 11 light source 12 collector lens 13 monochromatic filter 14 shutter 15 half mirror 16 objective lens 17 pupil plane 21 detector 22 lens 23 aperture 24 polarizing plate 31 CPU 32 Z axis motor 34 Waveform Memory 35 Reflectivity Calculator 36 Transient Change Calculator 37 Photosensitive Parameter Calculator 41 Resist Coating Process 42 Exposure Process 43 Development Process 44 Measuring Device 45 Simulation Unit

 ──────────────────────────────────────────────────の Continuing on the front page (72) Inventor Shinji Kuniyoshi 5-2-1, Josuihonmachi, Kodaira-shi, Tokyo In the semiconductor division of Hitachi, Ltd.

Claims (10)

[Claims] 1. A resist film formed on a substrate is condensed and irradiated with light having a photosensitive wavelength, reflected light after the start of illumination is captured by a lens, and a change with time of a reflected light intensity distribution on a pupil plane of the lens. And calculating a photosensitive parameter based on the detected value. 2. The method for measuring resist photosensitivity parameters according to claim 1, wherein a film thickness, a refractive index, and an extinction coefficient of the resist film are continuously detected to calculate a transient change characteristic thereof. The theoretical value of the average extinction coefficient in the resist film is calculated based on the transient change characteristics of the film thickness and the refractive index, and the theoretical value is compared with the transient change characteristic of the extinction coefficient to calculate a photosensitive parameter. A method for measuring a resist exposure parameter, characterized in that: 3. A method according to claim 1, wherein a photosensitive parameter of a resist film formed on a substrate that is opaque to the photosensitive wavelength is measured. A method for measuring photosensitive parameters. 4. The method of measuring resist photosensitivity parameters according to claim 1, wherein a photosensitivity parameter of a resist film formed on a substrate film opaque to the photosensitivity wavelength formed on the substrate is measured. A method for measuring a resist exposure parameter, the method comprising: 5. The method for measuring a resist exposure parameter according to claim 1, wherein said reflected light is detected after passing through a polarizing element. 6. The method according to claim 1, wherein said reflected light is detected after passing through a stop provided on an image forming surface of the resist film. How to measure parameters. 7. A semiconductor lithography method comprising the steps of: applying a resist solution to a substrate to form a resist film; forming a resist film; exposing the resist film to a pattern; exposing the resist film to an unnecessary portion; A method comprising: continuously detecting a thickness, a refractive index, and an extinction coefficient of the resist film based on reflected light of light applied to the resist film, calculating a transient change characteristic thereof, and irradiating the resist film. Calculating a resist film parameter of the resist film based on the reflected light of the light, calculating a latent image in the resist film based on the resist film parameter, the pattern information, and the stepper optical condition; The three-dimensional shape of the latent image formed in the resist film is calculated, and the exposure conditions in the exposure step are controlled. Conductor lithography method. 8. The semiconductor lithography method according to claim 7, wherein a resist film parameter, a pattern information, a substrate film formation parameter, and a stepper optical condition of a resist film formed on the substrate through substrate film formation are used. A semiconductor lithography method, wherein a latent image in a resist film is calculated. 9. The semiconductor lithography method according to claim 7, wherein one of a resist film parameter, a development parameter, and a substrate deposition parameter is measured, and whether the measured value is within a standard value is compared, A semiconductor lithography method characterized by detecting an abnormal part in a lithography process from a parameter out of a standard value. 10. The semiconductor lithography method according to claim 8, wherein the resist film parameter is a film thickness, a refractive index, and a photosensitive parameter, the development parameter is a resist dissolution rate, and the substrate deposition parameter is A semiconductor lithography method characterized by a film thickness and an optical constant of a thin film formed between a substrate and the resist film.