JP6003068B2 - Exposure method - Google Patents

Description

  The present invention relates to an exposure method.

  In an integrated circuit manufacturing process such as LSI (Large Scale Integration), a design pattern of an integrated circuit formed on a photomask is transferred to a resist film (photoresist film) by a projection exposure apparatus (for example, a stepper).

  In an integrated circuit which has been miniaturized, a pattern close to the resolution limit of a projection exposure apparatus is projected onto a resist film. Then, the pattern (resist pattern) transferred to the resist film does not coincide with the design pattern due to the optical proximity effect (Optical Proximity Effect).

  Therefore, optical proximity effect correction is performed to correct the shape of the mask pattern in advance so that the shape of the resist pattern substantially matches the shape of the design pattern.

JP 2000-310851 A

  The smaller the minimum dimension of the design pattern, the greater the optical proximity effect. For this reason, every time a design rule that defines the minimum dimensions of elements and wiring is newly determined, OPC parameters used for optical proximity effect correction (OPC) are also newly generated.

  However, the OPC parameter is generated by an enormous amount of work based on an exposure experiment. For this reason, if the design pattern is reduced and the integrated circuit is miniaturized, there is a problem that a great deal of effort is taken to generate a new OPC parameter.

  In order to solve the above problem, according to one aspect of the present method, a resist is formed through a step of forming a photomask having a first pattern, a projection lens having an adjustable numerical aperture, and the photomask. A step of irradiating the film with irradiation light, and the step of forming the photomask includes a step of correcting the design pattern of the integrated circuit based on the first numerical aperture of the projection lens and further reducing the second optical pattern by reducing the optical proximity effect. And a step of forming a mask pattern corresponding to the second pattern on the photomask, and the numerical aperture of the projection lens when the resist film is irradiated with the irradiation light is A second numerical aperture greater than the first numerical aperture is provided.

  According to this method, the reduced integrated circuit design pattern can be transferred to the resist film with almost no damage to the pattern shape by the mask pattern based on the existing OPC parameters.

1 is a structural diagram of a projection exposure apparatus used in an exposure method of an embodiment. It is a flowchart explaining the production | generation method of a mask pattern. It is a pattern figure explaining the production | generation method of a mask pattern. It is a pattern figure explaining the production | generation method of a mask pattern. It is a pattern figure explaining the production | generation method of a mask pattern. It is a flowchart of an exposure process. It is a flowchart explaining the production | generation method of another mask pattern used for transfer of a reduction | decrease design pattern. It is a figure explaining the transmittance | permeability of an OPC pattern along the VIIIA-VIIIA line | wire of FIG. It is a figure explaining the diffracted light of an OPC pattern. It is a figure explaining the diffracted light of a reduced OPC pattern. It is a figure explaining the diffracted light when the numerical aperture of a projection lens is enlarged. It is an example of Ia (x) of the OPC pattern when the projection lens has a first numerical aperture. It is an example of I (x) of the reduced OPC pattern when the projection lens has a first numerical aperture. It is an example of Ia (x) of the OPC pattern when the projection lens has a first numerical aperture. It is an example of I (x) of the reduced OPC pattern when the projection lens has a second numerical aperture.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. However, the technical scope of the present invention is not limited to these embodiments, but extends to the matters described in the claims and equivalents thereof. In addition, the same code | symbol is attached | subjected to the corresponding part even if drawings differ, The description is abbreviate | omitted.

(1) Exposure Apparatus FIG. 1 is a structural diagram of a projection exposure apparatus 2 (for example, a stepper) used in the exposure method of the embodiment. As shown in FIG. 1, the projection exposure apparatus 2 includes a light source (for example, ArF excimer laser) 4, an illumination system diaphragm 6, a condenser lens 8, a projection lens (for example, a reduction projection lens) 10, and an aperture diaphragm. 12.

  FIG. 1 also shows a substrate 16 and a photomask 18 coated with a resist film (photoresist film) 14. The substrate 16 is, for example, an unprocessed semiconductor substrate or a semiconductor substrate on which a transistor or an interlayer insulating film is formed. A mask pattern (first pattern) 22 is provided on one surface of the photomask 18.

  The diaphragm 6 of the illumination system is disposed between the light source 4 and the condenser lens 8 and adjusts the partial coherence factor (so-called σ value) of the illumination system 7 including the condenser lens 8 and the light source 4. The partial coherence factor is a parameter (auxiliary variable) representing the coherence of the light source 4. Hereinafter, both the partial coherence factor as a variable and its value (parameter value) are collectively expressed as a partial coherence factor σ.

  The aperture stop 12 has an opening disposed inside the projection lens 10 (or on the photomask 18 side or the substrate 16 side of the projection lens 10), and adjusts the numerical aperture of the projection lens 10. That is, the aperture stop 12 is a variable aperture stop of the projection lens 10.

  The condenser lens 8 converts the irradiation light (exposure light) generated by the light source 4 and transmitted through the diaphragm 6 of the illumination system into parallel rays. The irradiation light 20 a converted into parallel rays is irradiated onto the resist film 14 through the photomask 18 having the first pattern 22 and the projection lens 10. The adjustable range of the numerical aperture of the projection lens 10 is, for example, 0.6 to 0.93.

(2) Mask Pattern FIG. 2 is a flowchart illustrating a method for generating the mask pattern 22. 3 to 5 are pattern diagrams for explaining a method of generating the mask pattern 22.

(I) Generation of design pattern (step S2)
First, the layout of the integrated circuit is designed based on the first design rule. With this layout design, a design pattern is generated for each layer (S2). FIG. 3 is an example of an enlarged view of a part of the design pattern 24. The pattern in FIG. 3 is a wiring layer design pattern.

  The minimum gate length of a MOS (Metal Oxide Semiconductor) transistor defined in the design rule is, for example, 90 nm. Here, the layout is an arrangement of components (elements, wirings, etc.) of the integrated circuit. The design rule is a design rule that defines minimum values of dimensions and intervals of element components (for example, the gate of a transistor) and wiring. Therefore, a design rule is newly created each time the minimum line width (for example, the minimum value of the gate length) of the integrated circuit is reduced.

  The layout design of the design pattern 24 in the present embodiment is performed based on existing design rules. That is, the first design rule is an existing design rule.

  As shown in FIG. 3, at this stage, the design pattern 24 is not corrected for the optical proximity effect.

(Ii) Optical proximity effect correction (step S4)
Next, the optical proximity effect correction is performed on the design pattern 24 of the integrated circuit based on the first numerical aperture NA ₁ (for example, 0.8) of the projection lens 10 (S4). Thereby, an OPC pattern is generated.

  FIG. 4 is an example of an enlarged view of a part of the OPC pattern 26.

  In the optical proximity effect correction, for example, as shown in FIG. 4, a correction pattern is added to the corner portion of the partial pattern 24 a (see FIG. 3) included in the design pattern 24. Furthermore, as shown in FIG. 4, an auxiliary pattern 28 may be added around the corrected design pattern 24.

  In the optical proximity effect correction, a pattern image formed on the resist film 14 (hereinafter referred to as an imaging pattern) is calculated by simulation. Further, based on the calculated imaging pattern, the region of the resist film 14 that is exposed to the exposure light is simulated. Based on these simulations, a resist pattern formed on the resist film 14 is predicted.

  The position and size of the correction pattern are adjusted so that the predicted shape of the resist pattern substantially matches the design pattern 24.

―OPC parameters―
The shape of the resist pattern is predicted based on OPC parameters including the numerical aperture of the projection lens 10.

  As the numerical aperture of the projection lens 10 increases, the resolution of the projection exposure apparatus 2 increases. On the other hand, when the numerical aperture increases, the depth of focus of the projection exposure apparatus 2 becomes shallow. Therefore, the numerical aperture of the projection lens 10 is designed for each design root in consideration of the balance between the ease of realizing the minimum dimensions (minimum values such as the gate length and wiring interval) specified in the design rules and the ease of focusing. Is determined.

The optical proximity effect correction (S4) of the design pattern 24 (see FIG. 3) is performed based on the _first numerical aperture NA ₁ determined for the design rule (first design rule) of the design pattern 24.

  The shape of the resist pattern is predicted based not only on the numerical aperture of the projection lens 10 but also on other parameters such as the focus shift of the projection exposure apparatus 2 and the partial coherence factor σ of the light source 4. These parameters are included in the OPC parameters.

  The OPC parameter is determined for each design rule that defines the minimum dimensions of transistors and wirings. As described above, the first design rule used for generating the design pattern 24 is an existing design rule. Therefore, the first OPC rule used for the optical proximity effect correction of the design pattern 24 is also an existing OPC rule.

  If the OPC pattern 26 generated based on the first design rule and the first OPC rule already exists, steps 2 and 4 may be omitted.

(Iii) Reduction of OPC pattern (step S6)
Next, the OPC pattern 26 is reduced (shrinked) (S6). Thereby, a reduced OPC pattern (second pattern) is generated. FIG. 5 is an example of an enlarged view of a part of the reduced OPC pattern 30.

  The OPC pattern 26 is reduced by 0.92 times, for example. As a result, the 90 nm line becomes 82.8 nm (= 90 nm × 0.92).

  The reduction rate (shrink rate) s is preferably 0.6 or more and less than 1, and more preferably 0.85 or more and 0.95 or less. If the reduction ratio s is within these ranges, a pattern obtained by reducing the design pattern 24 (see FIG. 3) to s times is transferred to the resist film 14 without substantially losing its shape (an “exposure process” described later). reference).

(Iv) Production of photomask (Step S8, Step S10)
When the projection lens 10 is a reduction projection lens, the reduction OPC pattern 30 (see FIG. 5) is enlarged according to the magnification of the projection lens 10 (S8). For example, when the magnification of the projection lens 10 is 0.25, the reduced OPC pattern 30 is enlarged four times. Thereby, data of the mask pattern 22 is generated.

  Next, a mask pattern 22 corresponding to the reduced OPC pattern 30 is formed on the photomask substrate by electron beam drawing based on the generated data (S10).

  On the other hand, when the projection lens 10 is an equal magnification projection lens, the reduced OPC pattern 30 and the same size mask pattern 22 are formed on the photomask substrate by electron beam drawing.

  Thus, the photomask 18 having the mask pattern 22 is manufactured. The design data 24, the OPC pattern 26, and the reduced OPC pattern 30 are graphic data. The mask pattern 22 is a pattern of a light shielding film (for example, a Cr film) formed on the photomask 18.

  In the example shown in FIG. 2, the reduced OPC pattern 30 is enlarged according to the magnification of the projection lens 10. However, the pattern enlargement according to the magnification of the projection lens 10 may be performed at any stage.

  For example, in step S2, a design pattern enlarged according to the magnification of the projection lens 10 may be generated. In this case, the optical proximity effect correction (S4) is performed considering that the design pattern is enlarged. On the other hand, enlargement of the reduced OPC pattern (S8) is not performed.

  As a result, the same mask pattern 22 generated in steps S2 to S10 in FIG. 2 is formed. The mask pattern formed on the photomask 18 is obtained by enlarging or magnifying the reduced OPC pattern 30 generated in steps S <b> 2 to S <b> 6 according to the magnification of the projection lens 10. It corresponds. However, the procedure for actually forming the mask pattern formed on the photomask 18 may be any.

(3) Exposure Step FIG. 6 is a flowchart of the exposure step.

  As shown in FIG. 1, the photomask 18 having the mask pattern 22 and the substrate 16 coated with the resist film 14 are mounted on the projection exposure apparatus 2 (S12).

  Next, the numerical aperture of the aperture stop 12 is adjusted, and the numerical aperture of the projection lens 10 is set to a second numerical aperture larger than the first numerical aperture (the numerical aperture based on the optical proximity effect correction) (S14). .

  As the second numerical aperture, a numerical aperture (for example, 0.87) obtained by dividing the first numerical aperture (for example, 0.8) by a reduction ratio s (for example, 0.92) for reducing the OPC pattern is preferable. (See “(4) Resist pattern” below.)

  The projection lens 10 has an objective numerical aperture (photomask-side numerical aperture) and an image-side numerical aperture (resist film-side numerical aperture). The numerical aperture of the projection lens 10 is the image-side numerical aperture. That is, the first and second numerical apertures are the image-side numerical apertures of the projection lens 10.

  Since the magnification of the projection lens 10 is constant, the ratio between the objective numerical aperture and the image-side numerical aperture is constant. Therefore, when the image-side numerical aperture of the projection lens 10 is increased, the objective numerical aperture of the projection lens 10 is also increased.

  The partial coherence factor σ is a parameter representing the coherence of irradiation light generated by a light source having a finite size. As described above, the optical proximity effect correction (step S4) is performed based on the first numerical aperture of the projection lens 10. When the numerical aperture of the projection lens 10 is set to a second numerical aperture larger than the first numerical aperture at the time of exposure, the partial coherence factor σ of the irradiation light 20 is the partial coherence factor used for optical proximity effect correction. It becomes smaller than the factor σ (for example, about 0.5).

  Therefore, the diaphragm 6 of the illumination system is adjusted to bring the partial coherence factor σ (hereinafter referred to as σ) of the irradiation light 20a closer to σ used for optical proximity effect correction (S16). Preferably, σ of the irradiation light 20 is matched with σ used for optical proximity effect correction.

  Next, the resist film 14 is irradiated with irradiation light 20 generated by the light source 4 through the photomask 18 having the first pattern 22 and the projection lens 10 (S18).

  Thereafter, the resist film 14 is developed. Then, a resist pattern having substantially the same shape as a pattern (hereinafter referred to as a reduced design pattern) obtained by multiplying the design pattern 30 by s times (for example, 0.92 times) is generated. In addition, you may change suitably the order which performs step S12-S16.

―Another mask pattern―
FIG. 7 is a flowchart for explaining another method of generating a mask pattern used for transferring a reduced design pattern. Even when the mask pattern generated by the method shown in FIG. 7 is used, a pattern (reduced design pattern) obtained by reducing the design pattern 24 can be accurately transferred to the resist film. However, it takes much labor to execute the method shown in FIG. 7 compared to the method shown in FIG.

  In order to generate this mask pattern, an integrated circuit design pattern is first generated based on the first design rule as in step 2 of FIG. 2 (S22). Next, the generated design pattern is reduced to s times (for example, 0.92 times) (S24).

  In parallel with the reduction of the design pattern (S22 to S24), parameters used for the optical proximity effect correction simulation are acquired by an exposure experiment (S26).

  The pattern generated in steps S22 to S24 is the same as the design pattern of the integrated circuit generated based on the second design rule in which the minimum dimension of the first design rule is s times (<1). Therefore, the parameter acquired in step S26 is the second OPC parameter corresponding to the second design rule.

  The OPC parameter includes a first parameter used for the simulation and a second parameter that defines a rule for optical proximity effect correction.

  The first parameters are, for example, the numerical aperture NA of the projection lens 10, the wavelength λ of the light source 4, the partial coherence factor σ of the light source 4, the acid diffusion distance of the chemical multiplication resist, and the focus shift of the projection exposure apparatus 2. The amount of flare (stray light), the thickness of the resist film 14, the surface structure of the substrate 16, and the like. The second parameter is the minimum interval, minimum width, cutting length, maximum moving distance, etc. of the figure to be corrected.

  In step S26, the first parameter is fitted based on an exposure experiment in which the test pattern is transferred to the resist film 14 so that the actually obtained resist pattern matches the resist pattern obtained by the simulation. However, known values are used for the numerical aperture NA, the wavelength λ, and the partial coherence factor σ. Alternatively, these known values are finely adjusted by fitting.

  Next, based on the first parameter, a second parameter that defines a rule for optical proximity effect correction is created (S28).

  Next, based on the second OPC parameter obtained in steps S26 and S28, the reduced design pattern generated in steps S22 to S24 is corrected for the optical proximity effect (S30).

  The pattern subjected to the optical proximity correction is enlarged according to the reduction ratio of the projection lens 10 (S32). Finally, a photomask having an enlarged pattern as a mask pattern is produced by electron beam drawing (S34).

  Even using a photomask manufactured by the above procedure, a resist pattern having substantially the same shape as a reduced design pattern obtained by reducing the design pattern 24 can be generated. The numerical aperture of the projection lens at the time of exposure is a numerical aperture included in the OPC parameter. However, in steps S26 to S28 for generating OPC parameters, an enormous exposure experiment or the like is executed, and a great amount of labor is expended.

  On the other hand, the optical proximity effect correction (step S4) of the present embodiment is performed based on the existing OPC parameters as described above. Therefore, according to the present embodiment, the mask pattern 22 can be easily generated.

(4) Resist Pattern FIGS. 8A to 8C are diagrams for explaining the transmittance 32 of the OPC pattern 26 along the line VIIIA-VIIIA in FIG. In the following description, in order to simplify the description, the projection lens 10 is an equal magnification projection lens.

  As shown in FIG. 8A, the transmittance 32 of the OPC pattern 26 is divided into a first partial transmittance 32a shown in FIG. 8B and a second partial transmittance 32b shown in FIG. 8C. Can be disassembled.

  The first partial transmittance 32a is a pattern that periodically changes at a constant pitch p. The pitch p is, for example, twice the minimum line width (for example, the minimum dimension of the gate length) in the first design rule. On the other hand, the second partial transmittance 32b is a pattern that changes irregularly as shown in FIG.

  Here, consider a case where the OPC pattern 26 shown in FIG. 4 is directly masked without being reduced. When this mask is irradiated with the irradiation light 20, the irradiation light 20 is diffracted along the VIIIA-VIIIA line of the masked OPC pattern 26 due to the periodicity of the first partial transmittance 32a. Further, the irradiation light 20 is scattered by the second partial transmittance 32b.

  The scattered light based on the second partial transmittance 32b is superimposed on the diffracted light based on the first partial transmittance 32a, and the transmitted light of the masked OPC pattern 26 (hereinafter referred to as the transmitted light of the OPC pattern 26). Is generated.

  The irradiation light scattered by the second partial transmittance 32b is broad and weak. Therefore, the transmitted light of the OPC pattern 26 is substantially the same as the diffracted light by the first partial transmittance 32a. The masked OPC pattern 26 similarly generates diffracted light in the direction perpendicular to the VIIIA-VIIIA line.

  FIG. 9 is a diagram for explaining the diffracted light 34 a of the masked OPC pattern 26. An OPC pattern 26 (more precisely, a light shielding film corresponding to the OPC pattern 26) is formed on the photomask 18a. The numerical aperture of the projection lens 10 is the first numerical aperture used for optical proximity effect correction (step S4).

  The diffracted light 34 a of the OPC pattern 26 enters the projection lens 10, and a part thereof passes through the aperture stop 12 and passes through the projection lens 10. The diffracted light 34 a that has passed through the aperture stop 12 is imaged on a resist film (not shown) to generate an image of the OPC pattern 26. The image of the OPC pattern 26 becomes clearer as the diffracted light 34a passing through the aperture stop 12 increases.

  FIG. 10 is a diagram illustrating the diffracted light 34 of the reduced OPC pattern 30 shown in FIG. A reduced OPC pattern 30 is formed on the photomask 18b. The photomask 18b corresponds to the photomask 18 of the present embodiment described with reference to FIG.

  The numerical aperture of the projection lens 10 is the first numerical aperture used for optical proximity effect correction (step S4). That is, the numerical apertures of the projection lens of FIG. 9 and the projection lens of FIG. 10 are the same.

  The pitch of the reduced OPC pattern 30 is narrower than the pitch of the OPC pattern 26 (see FIG. 9). The diffraction angle of the diffracted light is inversely proportional to the pitch of the diffraction grating. Therefore, the diffraction angle θ of the diffracted light 34 by the reduced OPC pattern 30 is larger than the diffraction angle θa of the diffracted light 34 a by the OPC pattern 26.

  For this reason, as shown in FIG. 10, the higher order diffracted light among the diffracted light 34 of the reduced OPC pattern 30 is blocked by the aperture stop 12 and does not pass through the projection lens 10. For this reason, the reduced OPC pattern 30 imaged by the projection exposure lens 10 of FIG. 10 is unclear.

  FIG. 11 is a diagram illustrating the diffracted light 34 when the numerical aperture of the projection lens 10 is increased. As in FIG. 10, a reduced OPC pattern 30 is formed on the photomask 18b.

  The projection lens 10 of FIG. 11 transmits a diffracted light 34 having a higher numerical aperture and a higher order. Therefore, the reduced OPC pattern 30 imaged by the projection dew lens 10 of FIG. 11 is clear. That is, by increasing the numerical aperture of the projection lens 10, it is possible to prevent or suppress blurring of the imaging pattern due to pattern reduction.

  In the exposure step (S18) of the present embodiment, the numerical aperture of the projection lens 10 when the resist film 14 is irradiated with the irradiation light 20 is set to a second numerical aperture that is larger than the first numerical aperture.

  Therefore, according to the present embodiment, it is possible to prevent or suppress blurring of the imaging pattern of the reduced OPC pattern 30 due to pattern reduction. Therefore, the reduced design pattern can be transferred to the resist film with almost no change in the shape.

  Note that the resist pattern may be blurred due to factors other than the blurring of the imaging pattern (for example, photomask manufacturing error or chemical diffusion resist acid diffusion). However, if the reduction ratio of the OPC pattern is 0.6 or more and less than 1 (preferably 0.85 or more and less than 1), it is confirmed by simulation that the blur due to these factors can be ignored.

―Quantitative analysis of resist pattern shape―
Now, the case where the second numerical aperture NA ₂ of the projection lens 10 is equal to the value obtained by dividing the first numerical aperture NA ₁ of the OPC parameter by the reduction ratio s (<1) as shown in the equation (1). Think. Further, it is assumed that the light source 4 can be approximated by a point light source for the sake of explanation.

s is a magnification smaller than 1. Therefore, NA ₂ is greater than NA ₁ .

  As described above, the ratio between the resist-side numerical aperture of the projection lens 10 (that is, the numerical aperture of the projection lens) and the photomask-side numerical aperture is constant. Therefore, equation (2) is immediately derived from equation (1).

na ₁ is the photomask-side numerical aperture of the projection lens 10 corresponding to the first numerical aperture NA ₁ . na ₂ is the photomask-side numerical aperture NA ₂ of the projection lens 10 corresponding to the second numerical aperture.

Here, as shown in FIG. 11, the order of the diffracted light 34 diffracted by the reduced OPC pattern 30 and transmitted through the projection lens 10 is n (= 0, ± 1, ± 2...). Then, the diffraction angle θ _n of the nth-order diffracted light satisfies the expressions (3) and (4).

  Here, λ (for example, 193 nm) is the wavelength of the irradiation light 20. p (for example, 90 nm) is the pitch of the OPC pattern 26. Equation (3) is derived from the diffraction conditions of the diffraction grating. Equation (4) is derived from the definition of the numerical aperture.

  When Expression (2) and Expression (3) are substituted into Expression (4), Expression (5) is obtained.

Equation (5) shows the order n of the diffracted light 34 that passes through the projection lens 10 having a second numerical aperture NA _2.

Equation (6) represents the order n of the diffracted light 34a (see FIG. 9) that passes through the projection lens 10 having the first numerical aperture NA ₁ in the simulation of the optical proximity effect correction (S4).

  As is clear from the equations (5) and (6), the equations (5) and (6) are equivalent.

Therefore, the highest order of the diffracted light 34 of the reduced OPC pattern 30 that transmits the projection lens 10 having the second numerical aperture NA ₂ (> NA ₁ ) and the projection lens 10 having the first numerical aperture NA ₁ are transmitted. The highest order of the diffracted light 34a of the OPC pattern 26 is the same.

  Therefore, if the expression (1) is satisfied, the image formation pattern of the reduced OPC pattern 30 becomes as sharp as the image formation pattern of the OPC pattern 26.

  It is possible to derive this conclusion analytically. However, when trying to explain all OPC patterns, the mathematical formula becomes complicated. Therefore, a case where the OPC pattern 26 periodically changes in one direction as shown in FIG. 8B will be described. In this case, the design pattern has a pattern repeated at a constant pitch.

Equation (7) is the light intensity distribution I _a (x) of the imaging pattern generated by the OPC pattern 26 (see FIG. 9) (Keisuke Iizuka, Optoelectronics, pp. 101-120, Kyoritsu Publishing Co., Ltd.). x is a position coordinate on the resist film.

Here, n is the order of the diffracted light. Nmax is the highest order of the diffracted light 34 a that passes through the projection lens 10. * Represents a complex conjugate. d _n is the electric field strength of the n-th order diffracted light 34a. d _n is (to be exact, the Fourier coefficients of the transmittance of the OPC pattern 26) the Fourier coefficients of the OPC pattern 26 is proportional to. The Fourier coefficient is obtained by Fourier integration.

  Expression (8) is the light intensity distribution I (x) of the imaging pattern generated by the reduced OPC pattern 30 (see FIGS. 10 and 11).

d _n is the electric field intensity of n-th order diffracted light 34 (more precisely, the Fourier coefficients of the transmittance of the reduction OPC pattern 30) the Fourier coefficients of the reduced OPC pattern 30 is proportional to.

  FIG. 12A is an example of Ia (x) of the OPC pattern 26 when the projection lens 10 has the first numerical aperture. The horizontal axis is the coordinate x (the same applies to FIGS. 12B to 13B below). The vertical axis represents the light intensity distribution (the same applies to FIGS. 12B to 13B).

  Here, the OPC pattern 26 generates a large number of diffracted beams 34a that pass through the projection lens 10, as shown in FIG. Therefore, Ia (x) becomes a clear image similar to the OPC pattern 26 as shown in FIG. 12A. However, since not all diffracted light passes through the projection lens 10, Ia (x) is somewhat blurred compared to the OPC pattern 26.

FIG. 12B is an example of I (x) of the reduced OPC pattern 30 when the projection lens 10 has the first numerical aperture NA ₁ (see FIG. 10).

The pitch of the reduced OPC pattern 30 is s times (<1) the pitch p of the OPC pattern 26. Therefore, the diffraction angle θ of the diffracted light 34 is increased. Therefore, the numerical aperture of the projection lens 10 If left first numerical aperture NA _1, the highest degree Nmax of the diffracted light passing through the projection lens 10 is reduced. For this reason, I (x) becomes unclear as shown in FIG. 12B.

  12A and 12B show the photosensitive level 38 of the resist film 14. When the intensity of irradiation light exceeds the photosensitive level 38, the resist film is exposed and removed by the developer. 12A and 12B also show resist film regions (hereinafter referred to as development regions) 40a and 40b that are removed by the developer.

As described above, If you leave the aperture of the aperture stop 12 was kept at the first numerical aperture NA _1, I (x) of the reduced OPC pattern 30 becomes unclear. Then, as shown in FIG. 12B, the development area 40b exposed and removed by I (x) is s times (<1) the development area 40a (see FIG. 12A) of Ia (x) corresponding to the OPC pattern 26. The region 40c becomes narrower.

That is, while maintaining the numerical aperture of projection lens 10 to the first numerical aperture NA _1, be imaged reduced OPC pattern 30 (see FIG. 10) in the resist film 14, reducing the design pattern (see FIG. 3) The resist pattern is not generated.

FIG. 13A is an example of Ia (x) when the projection lens 10 has the first numerical aperture NA ₁ (see FIG. 9), as in FIG. 12A. FIG. 13B is an example of I (x) of the reduced OPC pattern 30 when the projection lens 10 has the second numerical aperture NA ₂ (see FIG. 10).

  The Fourier integration of the mask pattern is constant regardless of the period (pitch) if the ratio of the transmissive part (the part where the transmittance is 1) and the non-transparent part (the part where the transmittance is 0) of the mask pattern is constant. is there. For example, if the ratio of the transmissive part to the non-transmissive part is 1: 1, the 0th-order Fourier coefficient is always 0.5, and the ± 1st-order Fourier coefficient is always 1 / π.

Since even by reducing the pattern does not change the ratio of the transmitted portion and the non-transmissive portion, and _{d n} in the imaging pattern Ia of OPC pattern 26 (x), in the reduced OPC pattern 30 of the imaging pattern I (x) d _n is equal.

Further, as shown in the equations (5) and (6), the highest order Nmax of the diffracted light passing through the projection lens 10 having the first numerical aperture NA ₁ and the second numerical aperture NA ₂ (= NA ₁ / The highest orders Nmax of the diffracted light transmitted through the projection lens 10 having s) are equal.

Therefore, the amplitude 42a of Ia (x) when the projection lens 10 having an amplitude 42b of I (x) when having a second numerical aperture NA _2, the projection lens 10 is first numerical aperture NA ₁ is equal to (See FIGS. 13A and 13B). That is, the image formation pattern I (x) of the reduced OPC pattern 30 becomes as clear as the image formation pattern Ia (x) of the OPC pattern 26.

  As a result, the development area 40d (see FIG. 13B) corresponding to the reduced OPC pattern 30 matches the area 40c (see FIG. 13B) obtained by multiplying the development area 40a (see FIG. 13A) of the OPC pattern 26 by s times (<1). To do. That is, the reduced design pattern of the integrated circuit can be transferred to the resist film 14 without breaking the pattern shape.

  In the above description, the light source 4 is a point light source. However, even when the light source 4 has a finite size, if the optical proximity effect correction is performed based on σ of the irradiation light 20 when the reduced OPC pattern 30 is projected onto the resist film (that is, the exposure step S18). The same conclusion can be obtained. This can be confirmed by simulation.

  Regarding the above embodiment, the following additional notes are disclosed.

(Appendix 1)
A step of forming a photomask having a first pattern, and a step of irradiating the resist film with irradiation light through the projection lens having an adjustable numerical aperture and the photomask,
The step of forming the photomask includes the step of generating a second pattern by correcting the design pattern of the integrated circuit based on the first numerical aperture of the projection lens to further reduce the optical proximity effect, and generating the second pattern. Forming a mask pattern corresponding to the pattern on the photomask,
An exposure method in which the numerical aperture of the projection lens when the resist film is irradiated with the irradiation light is a second numerical aperture that is larger than the first numerical aperture.

(Appendix 2)
In the exposure method according to attachment 1,
When the projection lens is a reduction projection lens, the first pattern is a mask pattern obtained by enlarging the second pattern according to the magnification of the projection lens,
When the projection lens is an equal magnification projection lens, the first pattern is an equal magnification mask pattern to the second pattern.

(Appendix 3)
In the exposure method according to appendix 1 or 2,
2. The exposure method according to claim 1, wherein the second numerical aperture of the projection lens is a numerical aperture obtained by dividing the first numerical aperture by a magnification for reducing the design pattern corrected for the optical proximity effect.

(Appendix 4)
In the exposure method according to any one of appendices 1 to 3,
An exposure method, wherein a magnification for reducing the design pattern corrected for the optical proximity effect is less than 1.0 and 0.6 or more.

(Appendix 5)
In the exposure method according to any one of appendices 1 to 4,
The proximity effect correction is performed based on a partial coherence factor when the irradiation light is irradiated.

(Appendix 6)
In the exposure method according to any one of appendices 1 to 5,
The exposure method, wherein the design pattern includes a pattern repeated at a constant pitch.

2 ... projection exposure apparatus 4 ... light source 10 ... projection lens 12 ... aperture stop 14 ... resist film 16 ... substrate 18 ... photomask 20 ... irradiation light 22 ... First pattern 24 ... design pattern 26 ... OPC pattern 30 ... reduced OPC pattern (second pattern)

Claims (4)

A step of forming a photomask having a first pattern, and a step of irradiating the resist film with irradiation light through the projection lens having an adjustable numerical aperture and the photomask,
The step of forming the photomask includes the steps of generating a second pattern by optical proximity correction on the basis of the overall design pattern of the integrated circuit to a first numerical aperture of the projection lens, said second pattern generating a third pattern by whole further reduced, and a step of forming the first pattern on the photomask by using the third pattern,
An exposure method in which the numerical aperture of the projection lens when the resist film is irradiated with the irradiation light is a second numerical aperture that is larger than the first numerical aperture. The exposure method according to claim 1, wherein
2. The exposure method according to claim 1, wherein the second numerical aperture of the projection lens is a numerical aperture obtained by dividing the first numerical aperture by a magnification for reducing the design pattern corrected for the optical proximity effect. In the exposure method according to claim 1 or 2,
An exposure method, wherein a magnification for reducing the design pattern corrected for the optical proximity effect is less than 1.0 and 0.6 or more. In the exposure method according to any one of claims 1 to 3,
Prior Symbol light proximity effect correction, exposure method, characterized in that it is performed based on the partial coherence factor when irradiating the illumination light.