JP4854098B1 - Film forming method and film forming apparatus using the same - Google Patents

Abstract

【Task】
In the formation of an optical thin film, there is provided a technique capable of bringing the optical performance finally obtained closer to the target optical performance even when the film thickness and refractive index deviate from the design values.
[Solution]
When the extreme value of the light amount is found in a layer having an optical film thickness of λ / 4 or more, the refractive index and film thickness of a thin layer having an optical film thickness of λ / 4 or less previously formed are estimated, The target optical performance is obtained by correcting the film thickness of the layer during film formation and the layers after the layer during film formation using an optimization method.
[Selection figure] None

Description

The present invention relates to a film forming method and a film forming apparatus for controlling a process of forming an optical thin film by vapor deposition or sputtering.

Since the optical performance of the thin film is determined by the refractive index and the film thickness, measuring and controlling the refractive index and the film thickness during film formation is important for obtaining a thin film having desired optical performance. In addition, as for the film thickness in the optical thin film, the optical film thickness (product of the refractive index and the physical film thickness) is more useful information than the physical film thickness. Therefore, in the formation of an optical thin film that requires precise film thickness control, an optical film thickness control method that can obtain information on the refractive index and the optical film thickness is widely used. Examples of the optical film thickness control method include monochromatic photometry, two-color photometry, and multicolor photometry.

Monochromatic photometry is the simplest method, and uses the maximum and minimum values of reflectance or transmittance that appear repeatedly as the optical film thickness increases by λ / 4 (λ is the photometry wavelength). An example using this method is disclosed in Japanese Patent Publication No. 57-24485.

In addition, in monochromatic photometry, an extreme value does not appear when the optical film thickness is λ / 4 or less, and thus there is a problem that film thickness control of λ / 4 or less cannot be performed accurately. An example of enabling more accurate film thickness control even with a film thickness of λ / 4 or less by applying a change in rate to a model formula and performing control is disclosed in Japanese Unexamined Patent Application Publication No. 2010-138463.

Further, the refractive index of the film may differ from the design value due to fluctuations in various conditions such as the evaporation rate of the substance during film formation, the evaporation distribution, the pressure during film formation, the temperature of the substrate, and the like. An example of preventing refractive index deviation by controlling the evaporation rate in monochromatic photometry is disclosed in Japanese Patent Application Laid-Open No. 2005-2462.

Multicolor photometry is a method of controlling film thickness by measuring spectral reflection characteristics or spectral transmission characteristics during film formation using a spectrophotometer or the like. An example using this method is disclosed in Japanese Patent Laid-Open No. 5-255850 and Japanese Patent No. 3754874.

Japanese Patent Publication No.57-24485 JP 2010-138463 A JP 2005-2462 PR JP 5-255850 PR Japanese Patent No. 3754874

However, since various conditions such as the evaporation rate, evaporation distribution, pressure during film formation, and substrate temperature during the film formation fluctuate, a thin layer film having an optical film thickness of λ / 4 or less in monochromatic photometry When thickness control is performed, even if the method described in Patent Document 2 is used, film formation does not always proceed according to the model formula, and there is a difference between the refractive index of the actual film and the refractive index of the film estimated by the model formula. Therefore, there is a problem that accurate film thickness control cannot be performed.

Further, even if a desired optical film thickness is obtained, when the refractive index is different from the design value, there is a problem that an optical thin film having optical performance as designed cannot be obtained.

Even if the evaporation rate is controlled to obtain the refractive index as designed, if the conditions such as evaporation distribution, pressure during film formation, substrate temperature, etc. are different, the refractive index as designed is not necessarily changed. It has the problem that it cannot be obtained.

In addition, the multicolor photometry method has a problem that it is difficult to adopt in terms of cost because a complicated apparatus such as a spectrophotometer is required.

Accordingly, an object of the present invention is to provide a technique capable of bringing the optical performance of an optical thin film closer to the target even when the film thickness and refractive index deviate from the design values in the formation of the optical thin film.

The present invention includes a film thickness control step for controlling the film thickness of each layer of the optical multilayer film by measuring during film formation on the monitor substrate using monitor light;
The optical film measured in the film thickness control step when the optical film thickness of the layer being formed on the monitor substrate is greater than or equal to λ / 4 (λ is the measurement wavelength of the monitor light) Based on the maximum or minimum value of reflectance or transmittance at a position where the thickness is an odd multiple of λ / 4 , the refractive index of the layer and the layer before the layer whose optical film thickness is less than λ / 4 is estimated. Refractive index estimation step;
Film thickness estimation for estimating the film thickness of the layer before the layer whose optical film thickness is less than λ / 4 from the measurement result obtained in the film thickness control process and the refractive index estimated in the refractive index estimation process Process,
Using the refractive index estimation step and the refractive index and film thickness of each layer estimated in the film thickness estimation step, and having a film thickness correction step of correcting the target film thickness of the layer and the subsequent layers. Features.

Further, the film thickness correction method in the film thickness correction step includes the target spectral characteristics, the refractive index estimation step, and the film thickness with the film thicknesses of the layer being formed and the layers after the layer being formed as parameters. It is preferable to use a method of optimizing parameters so that the sum of squares of the difference between the refractive index estimated in the estimation step and the spectral characteristic calculated using the estimated value of the film thickness is minimized.

Here, the optimization methods include simplex method, steepest descent method, Newton method, quasi-Newton method, Gauss-Newton method, gradient method, conjugate gradient method, Levenberg-Marcate method, etc., and annealing. Methods (simulated annealing), genetic algorithms, tabu search, and methods using a neuronetwork are included, but the method is not limited to this, and an appropriate optimization method may be selected.

Moreover, it is preferable that the film thickness measuring method in a film thickness control process measures the light quantity of a single wavelength.

The film forming method of the present invention is preferably a vacuum evaporation method or a sputtering method.

According to the present invention, even when a film thickness or a refractive index deviates from a design value during film formation, the target optical can be obtained by correcting the film thickness of the layer being formed and the subsequent layers. Performance can be obtained.

In optical multilayer films that include thin layers with an optical film thickness of less than λ / 4, the film thickness shifts due to the refractive index deviating from the design value, and the target optical performance cannot be obtained. However, according to the present invention , not only a layer having an optical film thickness of λ / 4 or more but also a refractive index and a film thickness of a layer having an optical film thickness of less than λ / 4 can be estimated. Such a problem can be prevented by correcting the film thickness of the layers after the first layer.

1 is an overall configuration diagram of a film forming apparatus according to an embodiment of the present invention. It is the schematic diagram which looked at the monitor substrate mask 15 of FIG. 1 from the top. It is a figure which shows the spectral reflection characteristic in the design value of the 4 layer antireflection coating which concerns on one Embodiment of this invention. It is a figure which shows the ideal light quantity change curve for film thickness control of the 4 layer antireflection coating which concerns on one Embodiment of this invention. It is a figure explaining the actual light quantity change curve for film thickness control of the 3rd layer of the 4 layer antireflection coat concerning one embodiment of the present invention. It is a figure explaining the actual light quantity change curve for film thickness control of the 1st layer of the 4 layer antireflection coat concerning one embodiment of the present invention. It is a figure which shows the spectral reflection characteristic explaining the film thickness correction effect of the 4 layer antireflection coating which concerns on one Embodiment of this invention.

An embodiment of the present invention will be described with reference to the drawings, but the present invention is not limited to this.

FIG. 1 is a schematic diagram showing an example of a film forming apparatus of the present invention. An evaporation source 12 is provided in the lower part of the vacuum chamber 11, a dome 13 for holding the product substrate is provided above it, a film thickness control monitor substrate holder 14 in the center of the dome 13, and a monitor substrate mask 15 in the lower part of the substrate holder 14. The monitor substrate 16 is disposed inside the substrate holder 14.

The film thickness control step is performed by optically measuring the thickness of the film being formed, and includes a substrate holder 14, a monitor substrate mask 15, a monitor substrate 16, a light projecting / receiving unit 17, and a calculation / control unit 18. Then, the monitor light emitted from the light projecting / receiving unit 17 is applied to the monitor substrate 16, and the reflected light reflected by the upper surface of the monitor substrate 16 and the lower surface of the monitor substrate 16 on which the film is being formed is reflected by the light projecting / receiving unit 17. Received light. The received light quantity is taken into the calculation / control unit 18, where the calculation / control unit 18 calculates the refractive index and thickness of the film, finishes the film formation when the film thickness reaches the target value, and stores the measurement value. Is done.

FIG. 2 is a schematic view of the shape of the monitor substrate mask 15 provided under the substrate holder 14 as viewed from above. A hole is formed in the opening 21, and the film is formed on the monitor substrate 16 only in the opening 21. The thickness of each layer can be individually measured by rotating the substrate holder 14 by a certain angle for each layer. The monitor light is irradiated in the area of the opening 21.

Next, a film forming method when the present invention is applied to a four-layer antireflection coating will be described.

Table 1 shows one design example of the four-layer antireflection coating.

本実施形態では、第１層および第２層がモニター基板上での光学的膜厚がλ／４未満の層であり、第３層および第４層がモニター基板上での光学的膜厚がλ／４以上の層である。

In the present embodiment, the first layer and the second layer are layers having an optical film thickness of less than λ / 4 on the monitor substrate, and the third layer and the fourth layer have an optical film thickness on the monitor substrate. It is a layer of λ / 4 or more.

In this design example, BK-7 manufactured by SCHOTT is used for both the product substrate and the monitor substrate.

In Table 1, the number of layers represents the order of layers from the substrate side , the optical film thickness on the product substrate is the optical film thickness of the corresponding layer on the product substrate , and λ is the design reference wavelength (here, λ = 500 nm). The substance is the name of the substance in the corresponding layer, and the film thickness ratio is the so-called tooling factor. The ratio of the physical film thickness of the layer on the product substrate to the physical film thickness of the layer on the monitor substrate (film thickness ratio = product substrate The physical thickness of the upper layer ÷ the physical thickness of the layer on the monitor substrate), Δn in vacuum is the difference between the refractive index of the layer on the monitor substrate in vacuum and the refractive index of the layer in vacuum (vacuum) The refractive index of the layer on the monitor substrate in the layer = the refractive index of the layer in the atmosphere + Δn in vacuum), the measurement wavelength is the wavelength of the light beam measured in optical film thickness control (unit: nm), and the starting light quantity is film formation Reflected light amount of monitor board without film before start , light on monitor board The academic film thickness represents the optical film thickness of the corresponding layer on the monitor substrate in vacuum .

The film thickness ratio is obtained in advance by measuring the film thickness of the product substrate and the monitor substrate that are formed for each layer.

Δn in vacuum is obtained from the amount of light measured when the optical film thickness λ / 4 is formed on the monitor substrate.

The refractive index of the substrate and the film material generally has a different refractive index for each wavelength. In this embodiment, the refractive index for each wavelength is calculated by the Cauchy dispersion formula shown in Equation 1.

In the above formula, λ is a wavelength and its unit is μm. A ₀ , A ₁ , and A ₂ are coefficients of the dispersion formula, and the values in the following table are used for each substrate and substance.

Note that the optimum dispersion formula and coefficient may be used as appropriate depending on the substrate, the material, and the film forming conditions to be used.

The spectral reflection characteristics of the four-layer antireflection coating calculated from the design values are shown in FIG. A well-known matrix method was used for the calculation.

FIG. 4 is a diagram (film thickness control chart) showing a change in light amount in a film thickness control step when the four-layer antireflection coating of this embodiment is ideally formed according to design values for both refractive index and film thickness. As shown in FIG. Table 3 shows the light amount values of the respective layers at this time.

The peak light intensity is the light intensity at the point where the optical film thickness on the monitor substrate is λ / 4, the stop light intensity is the light intensity at the end of film formation, and the stop light intensity (%) is (peak light intensity−stop light intensity) ÷ (peak light intensity−starting). Light amount) × 100.
The peak light amount and stop light amount (%) of the first layer and the second layer are unknown because the optical film thickness on the monitor substrate does not reach λ / 4.

The curve of the ideal film thickness control chart of FIG. 3 can be derived by the following theoretical formula.

In each of the above formulas, R ₁ is the reflectance on the film formation surface side of the monitor substrate, n is the refractive index of the film in vacuum, _nm is the refractive index of the monitor substrate, and λ is the measurement wavelength of monitor light (unit: nm) , D is the physical film thickness (unit: nm), R ₀ is the reflectance of the surface of the monitor substrate where no film is attached, R _mes is the total reflectance of both surfaces of the monitor substrate being deposited, and R _ini is the film on both surfaces The total reflectance on both sides of the monitor substrate in a state where no mark is attached, R is a light quantity value to be obtained. n and n _m uses the value obtained by the dispersion formula, the n adds the vacuum [Delta] n. Further, d is a design value obtained by dividing the physical film thickness on the product substrate by the film thickness ratio.

When the ideal film thickness control as shown in FIG. 4 is performed, the spectral reflection characteristic as the design value shown in FIG. 3 is obtained.

However, in actual film formation, such an ideal film is rarely formed, and the refractive index of the film varies depending on various conditions such as the evaporation rate of the substance during film formation, the pressure during film formation, and the temperature of the substrate. Depending on the design value. Further, the thin layer (the first layer and the second layer in this example) having an optical film thickness of λ / 4 or less has a different refractive index. As a result, the film thickness also differs from the design value.

For example, when the refractive index of the third layer ZrO ₂ (2.022 in vacuum (λ = 450 nm)) has increased by 0.02, the peak light amount increases from the ideal value of 71.06 to 73.06. FIG. 5 is a film thickness control chart of the third layer when the solid line 51 has an ideal refractive index and the dotted line 52 has a refractive index increased by 0.02.

In general, when the evaporation rate of a substance during film formation is controlled to be constant, the main factors of refractive index fluctuation are fluctuations in pressure during film formation and fluctuations in substrate temperature. The pressure during deposition and the temperature of the substrate are less varied in a series of deposition processes. Therefore, when the refractive index of the third layer ZrO ₂ is increased as in this example, it can be estimated that the refractive index of the first layer ZrO ₂ is also increased.

In the refractive index estimation step of the present invention, when the deviation between the ideal peak light amount and the actual peak light amount is found in this way, the actual refractive index is estimated as follows.

The actual refractive index of the third layer is obtained by the following equation using the peak light quantity value.

In the above formula, Rs is the initial light amount of the relevant layer , Rp is the peak light amount of the relevant layer , Ro is the reflectance of the substrate, and R ′ is the reflectance at the peak of the relevant layer .
That is, the reflectance at the peak of the third layer can be obtained by using the initial light amount Rs of the third layer, the peak light amount Rp of the third layer, and the reflectance Ro of the substrate.

In the above formula, _nm represents the refractive index of the substrate, and n represents the refractive index of the film in vacuum.

When calculated according to the above equation, the refractive index in vacuum of the third layer is n = 2.042 (λ = 450 nm) (increase 0.020). The refractive index in the atmosphere is also estimated to increase by 0.020.

Next, the refractive index of the first layer ZrO ₂ is estimated.
First, the peak light quantity when the first layer is assumed to have a film thickness of λ / 4 or more is estimated from the peak light quantity of the third layer by using the correction formula of tens.

In the above formula, P ′ is the peak light amount of the estimated layer (first layer), P is the peak light amount of the layer having the peak (third layer), and A is a correction coefficient determined for each layer.

Note that this correction formula takes into account the effect that the refractive index of the later layer increases as the pressure in the vacuum chamber decreases as the film formation progresses or the substrate temperature increases due to radiant heat from the evaporation source. However, the present invention is not limited to this, and an optimal correction formula may be adopted as appropriate.
Here, the correction coefficient of the first layer is -0.01.
The estimated peak light amount of the first layer is P ′ = 73.06 + 73.06 × −0.01 = 72.33.

Using Equations 8 and 9, the estimated refractive index in vacuum of the first layer is 2.035 (λ = 450 nm) (increase 0.013). The refractive index in the atmosphere is also estimated to increase by 0.013.

From the above, the refractive indexes of the first layer and the third layer can be estimated from the peak light amount of the third layer.

Next, the film thickness of the first layer is estimated by the film thickness estimation step.

FIG. 6 is a film thickness control chart of the first layer. A solid line 61 is a curve when an ideal film is formed. As shown in Table 3, it stopped at a stop light amount of 69.62. A dotted line 62 is a curve when it is assumed that the film formation is continued without stopping at the stop light amount. The peak light intensity is 71.06.

A solid line 63 is a curve when the refractive index is increased. The stop light amount is 69.62, which is the same as when the ideal film is formed. This is because the film formation is performed assuming that the peak light amount is ideal.

A dotted line 65 is a curve when it is assumed that the film of the solid line 63 has been formed without stopping at the stop light amount. The peak light quantity is the estimated value 72.33. In order for the first layer to have an optical film thickness as designed, considering the difference between the ideal peak light quantity 71.06 and the estimated peak light quantity 72.33, the end light quantity 70. It was necessary to stop the film formation at 86.

This is because the ratio of the stop light amount to the ideal light amount change width is (69.62-25.00) ÷ (71.06-25.00) = 0.969, and therefore the stop light amount in the actual light amount change width. This is because (71.06-25.00) × 0.969 + 25.00 = 70.86 is necessary to achieve the same ratio.

The first layer is thinned by the thickness corresponding to the thick solid line 64.

The film thickness on the monitor substrate on which the first layer is actually formed is calculated by calculating back the number 2 to number 7 using the estimated refractive index, so that the optical film thickness 0.2119λ (λ = 450 nm) = 95. It can be estimated that 355 nm and physical film thickness 95.355 ÷ 2.035 = 46.86 nm.

The film thickness on the product substrate is obtained by multiplying the physical film thickness on the obtained monitor substrate by the film thickness ratio shown in Table 1.
Physical film thickness on product substrate = 46.86 × 0.75 = 35.15 nm, optical film thickness = 35.15 × (2.050 + 0.013) = 72.51 = 0.145λ (λ = 500 nm) . (2.050 is the refractive index of ZrO ₂ at a wavelength of 500 nm calculated using the dispersion formula.)

Thus, the film thickness of the first layer was estimated.

Next, the thicknesses of the third layer and the fourth layer are corrected by the thickness correction step.

In the film thickness correction process, the refractive index and film thickness of the first layer are used as the refractive index and film thickness estimated in the refractive index estimation process and film thickness estimation process, and the refractive index and film thickness of the second layer are designed values. The refractive index of the third layer is the refractive index calculated from the peak light amount, the design value is used for the film thickness of the third layer, and the design value is used for the refractive index and film thickness of the fourth and subsequent layers. Using these refractive index and film thickness as initial values, and using the film thicknesses of the third and fourth layers as parameters, optimization is performed so as to approach the spectral reflection characteristics shown in FIG.

Optimization methods include simplex method, steepest descent method, Newton method, quasi-Newton method, Gauss-Newton method, gradient method, conjugate gradient method, Levenberg-Markert method, annealing method (simulation) Ted Annealing), genetic algorithm, tabu search, and a method using a neuro network are not limited to this, and an appropriate optimization method may be selected.

For example, when using the Levenberg-Marcate method, William H. Press et al. “NUMERICAL RECIPES in C”
Chapter 14 Section 4 P.505 Levenberg-Marquardt method,
Japan, Technical criticism company,
The codes described in May 13, 2006, first edition, 13th edition, ISBN4-87408-560-1, can be used.

As an example, optimization was performed using the Levenberg-Marquardt method, the film thicknesses of the third layer and the fourth layer were corrected as shown in Table 4, and the stop light amount was corrected.

If film formation is performed by correcting the stop light amounts of the third layer and the fourth layer to the values shown in Table 3, a product having spectral characteristics closer to the design values can be obtained as compared with the case where correction is not performed.

FIG. 7 is a diagram showing the spectral reflection characteristics of the product substrate. A thin solid line 71 is a design value, and a thick solid line 72 is a spectral reflection characteristic when the stop light amounts of the third layer and the fourth layer are corrected to the values shown in Table 3. A dotted line 73 is a spectral reflection characteristic when the film is formed without correction. The effect of the correction is clear from this figure, and the spectral reflection characteristic when the correction is performed is closer to the design value than when the correction is not performed.

Further, when the peak shifts in the fourth layer, the refractive index is obtained by the same method, the refractive index and the film thickness of the second layer are estimated, and the stop light amount of the fourth layer is corrected.

In this embodiment, the refractive index and the film thickness are estimated for the same material layer. However, different materials may be used as long as the materials exhibit the same tendency in peak light intensity.

Further, even if the film configuration is such that there is no thin layer and all layers can obtain the peak light amount, the refractive index shift can be corrected by correcting the film thickness by applying the present invention.

Further, in this embodiment, optimization is performed so as to approach the spectral characteristics of the design value. However, for example, when a lower reflectance is desired in the antireflection coating, data having a reflectance lower than the design value is optimized as a target. You may do.

DESCRIPTION OF SYMBOLS 11 Vacuum chamber 12 Evaporation source 13 Product substrate arrangement dome 14 Monitor substrate holder 15 Monitor substrate mask 16 Monitor substrate 17 Projecting and receiving unit 18 Calculation and control unit 21 Monitor substrate mask opening

Claims (5)

A film thickness control step for controlling the film thickness of each layer of the optical multilayer film by measuring the film thickness on the monitor substrate using monitor light;
The optical film measured in the film thickness control step when the optical film thickness of the layer being formed on the monitor substrate is greater than or equal to λ / 4 (λ is the measurement wavelength of the monitor light) Based on the maximum or minimum value of reflectance or transmittance at a position where the thickness is an odd multiple of λ / 4 , the refractive index of the layer and the layer before the layer whose optical film thickness is less than λ / 4 is estimated. Refractive index estimation step;
Film thickness estimation for estimating the film thickness of the layer before the layer whose optical film thickness is less than λ / 4 from the measurement result obtained in the film thickness control process and the refractive index estimated in the refractive index estimation process Process,
Using the refractive index estimation step and the refractive index and film thickness of each layer estimated in the film thickness estimation step, and having a film thickness correction step of correcting the target film thickness of the layer and the subsequent layers. A characteristic film forming method. The film thickness correcting method in the film thickness correcting step according to claim 1, wherein the film thickness of the layer during film formation and the film thickness of the layer after the film formation are used as parameters, and the target spectral characteristics, and the refractive index estimating step, The parameter is optimized so that the sum of squares of the difference between the refractive index estimated in the film thickness estimation step and the spectral characteristic calculated using the estimated value of the film thickness is minimized. The film forming method. 3. The film forming method according to claim 1, wherein the film thickness measuring method in the film thickness controlling step according to claim 1 measures a light amount of a single wavelength. The film forming method according to claim 1, wherein the multilayer film is formed by a vacuum deposition method or a sputtering method. A film forming apparatus equipped with a film thickness monitoring mechanism and a film thickness control mechanism, wherein the film forming method according to claim 1 is used.

Images