JPH04315561A - Optical element and working method thereof - Google Patents

Abstract

PURPOSE:To provide the working method for an optical element capable of accurately finishing face and angle of respective faces with simple operation, capable of mass-producing same shaped optical elements, and having a good yield, at working an optical element having a plurality of flat surfaces. CONSTITUTION:This is the working method for an optical element to transcribe the form of a master 12 formed into the same form as the finished form of the optical element to a blank 14 to be worked. The blank 14 having a blank master flat plane G1 finished into nearly same accuracy as the finished accuracy of the optical element is joined to a master having a first and a second master flat planes G2, D2 so as to closely contact the blank master flat plane G1 with the master flat plane G2, and worked so as to form a flat surface D1 in parallel with the second master flat plane D2 on the blank 14.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to optical elements such as corner cube reflectors, and a processing method for processing these with high precision.

0002

[Prior Art] In optical elements that have a plurality of planes and change the optical path by transmitting light through these planes or reflecting light on these planes, it is difficult to accurately change the optical path. In order to achieve this, the flatness of these planes or the angular accuracy between the planes must be finished with extremely high precision.

[0003] As an optical element that has a plurality of planes and changes the optical path as described above, there is, for example, something called a corner cube reflector. This corner cube reflector (hereinafter referred to as C.C.R.) is made of optical glass or the like, and has a shape that looks like one corner of a regular hexahedron has been cut off. In other words, it has the shape of a triangular pyramid with the corner that was one corner of the regular hexahedron as the top (in reality, each corner of the triangle that defines the plane of incidence facing the top is cut out, so 11). Then, as shown in Figure 11, this C
．． C. R50 has a property of reflecting the incident light beam in the direction exactly opposite to the direction of incidence, regardless of the angle of inclination of the incident light beam with respect to the bottom surface, when the light beam is incident from the entrance surface G opposite to the top thereof. Taking advantage of this property, they can be used as movable mirrors in interferometric length measuring machines, or they can be arranged in large numbers on a flat surface and used as reflecting mirrors that return a beam of light incident from a distance to its original exit point.

[0004] This C. C. 12 and 13 are diagrams showing a conventional processing method for R50. In addition, C. which will be discussed in the following explanation. C. R50 has an angular deviation of the output angle relative to the incident angle of about 2 seconds, and a wavefront aberration of wavelength λ=632.
It has extremely high accuracy of about λ/10 for 8 nm light. The flow of the conventional processing method will be briefly explained with reference to FIGS. 12 and 13. First, according to the procedure shown in FIG. C. Before processing R50, the regular hexahedral block 54 is processed. This regular hexahedral block 54
, as shown in FIG. 13(a), a triangular pyramidal block 56
Cut out the three vertices 56a, 56b, 56c.
By cutting out, as shown in FIG. 13(b),
C. C. R50 is completed.

To explain this in detail, first, for example, B
Optical glass such as K7 or transparent quartz glass is molded to produce a regular hexahedral blank 52 as shown in FIG. 12(a). Next, one side of this blank 52, for example, side A, is finished using a flat lapping machine.
Finish to a flatness of approximately 0.03 μm. Next, the surface D, which is the surface opposite to the surface A, is similarly polished using a flat lapping machine so that the flatness is 0.03 μm and it is finished so that it is exactly parallel to the surface A. The parallelism at this time is about 0.4 seconds. FIG. 12(b) shows the finished state of sides A and D.

Next, surface B is finished so that it is perpendicular to surface A. The procedure at this time is to first polish side B with a flat lapping machine to achieve flatness, then measure the perpendicularity of side A and B with an autocollimator, and if the specified perpendicularity is not achieved. If not, polish again using a flat lapping machine. After polishing, measure the perpendicularity between sides A and B again, and if the perpendicularity still does not reach the specified value, polish again using a flat lapping machine until the perpendicularity has reached the specified value. Repeat this operation until you are within range. The perpendicularity required at this time is about 0.4 seconds. After the processing of the B surface is completed, the E surface, which is the surface opposite to the B surface, is polished parallel to the B surface while maintaining flatness, in the same manner as the D surface was processed. The parallelism at this time is also about 0.4 seconds. Figure 12(c) shows the state when this processing is completed.
It is.

Next, the C side is polished. At this time, the C side is polished with a flat lapping machine, and the A of the C side is polished.
Repeat the operation of measuring the perpendicularity to both the plane and the B plane. After the polishing of the C surface is completed, the F surface is polished so as to be parallel to the C surface, and the regular hexahedral block 54 is polished.
Finish processing. Next, as shown in FIG. 13(a), a triangular pyramidal block 56 having each corner of the regular hexahedral block 54 as a top is cut out from this regular hexahedral block 54 (
Four triangular pyramid blocks including 56 are cut out along the line indicated by the chain double-dashed line in the figure. Then, the portions of the vertices 56a, 56b, and 56c that define the G1 plane facing the apex 56d of the triangular pyramidal block 56 are rounded. After that, by polishing the G1 surface with a flat lapping machine,
C.3 as shown in 3(b). C. R50 is completed. In this state, C. C. The flatness of the G1 plane, which is the incident surface of R50, is 0.03 μm, and the flatness of the surfaces other than the G1 plane is 0.03 μm.
03 μm, and the perpendicularity of each surface is 0.4 seconds.

[0008]

However, in the conventional processing method described above, when finishing the regular hexahedral block 54, the flatness of each surface and the angular accuracy of the adjacent surface are finished at the same time. , requires advanced technology. In particular, in the above explanation, when processing C-plane, it is necessary to process C-plane so that it is perpendicular to both A-plane and B-plane, and it is also necessary to ensure the flatness of C-plane itself. This makes the process extremely difficult and takes a long time.

Furthermore, after polishing the regular hexahedral block 54, when cutting out the triangular pyramidal block 56, or when rounding the vertices 56a, 56b, and 56c, chipping defects tend to occur at the cut and edge portions. , there is a problem that the yield is poor. Furthermore, if it is attempted to process a large number of optical elements of the same shape, each one will be processed in the same process, which will take a very long time and increase costs.

[0010] These problems are solved by C. C. Similar problems occur not only in the processing of R, but also in the processing of many other types of optical elements that require surface accuracy of each surface and angular accuracy between those surfaces. Therefore, the present invention has been made in view of the above-mentioned problems, and its purpose is to make it possible to precisely finish the surface accuracy and angular accuracy of each surface when processing an optical element having multiple planes. However, it is an object of the present invention to provide a method for processing optical elements that can process a large number of optical elements of the same shape with simple operations and has a high yield.

[0011]

[Means for Solving the Problems] In order to solve the above-mentioned problems and achieve the objects, the optical element processing method of the present invention includes the following steps:
An optical element processing method for transferring the shape of a master made into the same shape as the completed shape of the optical element onto a blank to be processed into the optical element, the method having an accuracy substantially equal to the completion accuracy of the optical element. joining the blank having a blank reference plane finished to a master having first and second master reference planes with the blank reference plane and the first master reference plane in close contact; The blank is characterized in that the blank is processed so as to form a plane parallel to the second master reference plane.

Another method of processing an optical element of the present invention is to precisely finish the surface accuracy of the plurality of planes and the angular accuracy between the plurality of planes when processing an optical element having a plurality of planes. A method for processing an optical element, which has a plurality of processing planes corresponding to the plurality of planes, and is roughly processed into a shape substantially equal to the completed shape of the optical element in order to be finished into the optical element. The blank is attached to a master that has the same shape as the completed optical element and is finished with a precision higher than the shape accuracy of the completed shape, and the blank has a plurality of planes corresponding to each of the plurality of planes of the master. A state in which a processing plane of the blank corresponding to the bonding surface is in close contact with a predetermined bonding surface that is one of the reference planes of
and in a state in which each of the plurality of reference planes and each of the plurality of processing planes corresponding to each of the plurality of reference planes are approximately point symmetrical with respect to a certain point on the joint surface. joining each of the plurality of processing planes of the blank,
The method is characterized in that the joining is released after finish processing is performed so as to be parallel to the respective reference planes of the master corresponding to the respective processing planes.

[0013]

[Operation] As described above, the optical element processing method according to the present invention is configured so that the optical element has the same shape as the completed shape and is parallel to each surface of the highly precisely finished master. In addition, by simply polishing each side of the blank, it is possible to finish an optical element to an accuracy close to that of the master, making it possible to process a large number of optical elements of the same shape with a simple operation.

[0014] Furthermore, since there is no cutting or rounding process after each surface is finished, there is no need to produce defective products due to chipping defects after performing the troublesome surface leveling work.
Yield is improved.

[0015]

[Example] Hereinafter, a method for processing an optical element according to an example of the present invention will be described. C. A case in which the present invention is applied to processing an R (corner cube reflector) will be described in detail with reference to the attached drawings. Note that in this embodiment as well, the C.I. C. R50 has an angular deviation of the output angle relative to the incident angle of about 2 seconds, and a wavefront aberration of wavelength λ = 632.8n.
It is assumed that the accuracy is extremely high on the order of λ/10 for light of m.

FIG. 1 shows C. C. Master 12 of R and now C. C. Blank 1 to be finished to R50
4 is a diagram showing a state in which they are pasted together. Also, Figure 2
is a view of FIG. 1 viewed from the direction of the arrow. In Figure 1,
Master 12 was processed using the conventional processing method described above, and C. C. The same shape as the completed R50, and C. C. It is processed to a precision that exceeds the precision required for R50 products. Specifically, the flatness of the joint surface G2 is approximately 0.03 μm, and the flatness of each surface B2, C2, and D2 is approximately 0.01 μm;
The perpendicularity between C2 and D2 is approximately 0.2 seconds.

On the other hand, since the shape of the blank 14 leaves a finishing allowance, C. C. It is made slightly larger than the completed shape of R50, and each surface B1, C1, and D1 are still in the state of being roughly rubbed. However, the entrance surface G1, which is the joint surface with the master 12, has already been processed and finished to a flatness of 0.1 μm, which is the required accuracy. Note that the blank 14 and the master 12 are, for example, BK7 (
Refractive index nd = 1.52, Abbe number νd = 64.1,
Thermal expansion coefficient = 74 x 10-7 cm/cm°c) or transparent quartz glass (thermal expansion coefficient = 5 x 10-7 cm/cm
The material is optical glass such as cm°c). here,
Originally Master 12 was C. C. It is important that it has the same shape as R50, and the material is not limited to optical glass, but for convenience of processing inspection, it is desirable that it be transparent.

The master 12 also has a joint surface G2.
A cylindrical chamfered portion 12b is formed on the outer periphery of the master 12, but this portion is used because the corners of the G2 surface are optically unnecessary and when polishing each surface of the blank 14, which will be described later. It serves as an escape part to prevent itself from being cut down. Then, a chamfered portion 14 corresponding to the blank 14 is also provided.
b is formed.

The master 12 and the blank 14, which have been machined to the above-mentioned precision, are connected to the joint surface G of the master 12.
2 and the entrance surface G1 of the blank 14 are bonded together by vacuum bonding. At this time, blank 14 and master 1
As shown in the figure, the corresponding surfaces of the blank 14 and the master 12 are approximately symmetrical with respect to a point near the approximate center on the bonding surface G2. Here, the above-mentioned vacuum bonding is a process in which after the bonding surface G2 and the entrance surface G1 are brought into close contact with each other, a slight amount of liquid such as water is soaked between these surfaces to create a stronger bond. This is the adhesion method to obtain.

After joining the blank 14 and the master 12, each surface B1, C1,
D1, each surface B2, C2, D2 of the corresponding master 12
Process them so that they are exactly parallel to each other. By doing this, the angular accuracy between the surfaces B2, C2, and D2 of the master 12 can be maintained as they are between the surfaces B1, C1, and D1 of the blank 14.
will be transcribed into. That is, each surface B of the blank 14 has the same angular accuracy as that of the master 12.
1, C1, and D1 can be completed.

[0021] However, each side B2, C2,
Since it is impossible to machine each surface B1, C1, D1 of the blank 14 parallel to D2 without error, in reality,
The angular accuracy between each surface B1, C1, D1 of the finished blank 14 is slightly inferior to the angular accuracy between each surface A2, B2, C2 of the master 12. Therefore, in consideration of this processing error, each surface B2, C2, D2 of the master 12 is
The angular accuracy and flatness between them are finished to the above-mentioned accuracy. Finishing the master 12 with such high precision uses a conventional processing method, and as explained in the conventional example, it is a very labor-intensive task. However, once several masters 12 are made, the precision of the masters 12 can be easily transferred to the blank 14 using the method described above, so C. C. When processing a large amount of R50, it is possible to significantly save time and effort.

Next, referring to FIGS. 3 to 6, master 1
A specific method for finishing the corresponding surfaces B1, C1, D1 of the blank 14 so as to be parallel to the respective surfaces B2, C2, D2 of the blank 14 will be described. First, as shown in FIG. 3, the workpiece 16, which is a blank 14 and the master 12 bonded together as shown in FIG. They are glued together. At this time, each vertex 14a of the blank 14 is arranged so as to face inward. Further, the lower surface of the master 12 (in this case, the D2 surface) and the upper surface 18a of the glass substrate 18 are bonded together by vacuum bonding, and a liquid such as water may also be used to bond the blank 14 and the master 12. The same is true for adhesion. Note that this glass substrate 18 also has the above-mentioned B.
The material is optical glass such as K7 or transparent quartz glass, and the flatness is finished to about 0.03 μm.

FIG. 4 is a side view of FIG. 3 seen from below. In order to finish the D1 side of the blank 14, the master 1 is
This figure shows the state in which the D2 surface of No. 2 is brought into close contact with the glass substrate 18 and adhered. Therefore, it is sufficient to finish the D1 surface of the blank 14 so that it is parallel to the upper surface 18a of the glass substrate 18. However, at this time, if the D2 surface of the master 12 is even slightly raised with respect to the upper surface 18a of the glass substrate 18, the angular accuracy of the D2 surface of the master 12 with respect to the other two surfaces B2 and C2 will be lower than that of the blank 14. The image will not be accurately transferred to the D1 side of the image. Therefore, the D2 surface is firmly connected to the upper surface 18 of the glass substrate 18.
Carefully place the master 12 on the glass substrate 1 so that it is in close contact with the glass substrate 1.
It is glued to 8. Then, this D2 surface and the glass substrate 1
For example, the close contact state with 8 is the light reflected by the D2 surface,
This can be investigated by observing the state of interference with the light reflected by the upper surface 18a of the glass substrate 18. Then, set the master 12 so that this interference fringe is minimized.
By adhering with the master 12 positioned, the D2 surface of the master 12 and the upper surface 18a of the glass substrate 18 can be aligned with very high precision.

Similarly, the remaining three works 16 are transferred to D2.
With the surface in close contact with the upper surface 18a of the glass substrate 18,
It is adhered to the glass substrate 18. Here, four works 16
are adhered to the glass substrate 18, the heights of the four workpieces 16, that is, the heights of the D1 surfaces of the four blanks 14 may not match due to errors in adhesion or slight differences in the sizes of the blanks 14. It doesn't necessarily mean that If the heights of the D1 surfaces of these four blanks 14 do not match, the plane lapping machine that finishes the D1 surface cannot perform a large enough shaving to make these heights even. . Therefore, before applying these to a plane lapping machine, the four works 16 are processed using a plane grinding machine so that the heights of these four works 16 are made to be the same. However, normally, only four workpieces 16 are not bonded to the glass substrate 18 and processed, but several sets of these four workpieces 16 are bonded to the glass substrate 18. , FIG. 3, and FIG. 4 show a part of it.

Next, the glass substrate 18 and the workpiece 16 bonded together in the state shown in FIGS. 3 and 4 are placed on a flat lapping machine to form the upper surface D1 of the blank 14 in the figure.
The surface will be finished. At this time, this glass substrate 1
8 and the workpiece 16 are placed upside down on a flat lapping machine. FIG. 5 shows this state. Each D1 surface of the blank 14 is in contact with the upper surface 62a of the polishing disk 62 on the lower side in the figure, and the polishing disk 62
By moving relative to the blank 14, each D1 surface of the blank 14 is polished.

[0026] Here, the structure of this flat lapping machine 60 will be briefly explained. FIG. 6 is a plan view of the flat lapping machine 60 viewed from above. A polishing disc 62 is disposed on the flat lapping machine 60 and is rotatably held with respect to a base 61 in the direction of arrow a in the figure. And this polishing disk 6
A ring 64 for holding the object to be polished is placed on top of 2, and since this ring 64 is only placed on the polishing disk 62, it cannot rotate on the polishing disk 62. be. However, since it is supported by an arm attached to the base 61 and is prevented from moving in the horizontal direction, the base 61
It is configured not to move relative to 1. Then, a core 66 having a hole 66a for accommodating the glass substrate 18 at an eccentric position with respect to the ring 64 is installed in the ring 64, and the blank 14 and the master 2 are inserted into the hole 66a. Place the bonded glass substrate 18. Here, a state in which four sets of four works 16 are attached to the glass substrate 18 is shown. In this state, when the polishing disk 62 is rotated in the direction of arrow a, the ring 64
rotates in the direction of arrow b, and along with this, the glass substrate 1
8 rotates in the direction of arrow c while revolving around the center of the ring 64. And each D1 side of each blank 14 is
It will be polished while being rubbed by the polishing disk 62.

In this way, each D1 of each blank 14
The surface is once polished. At this time, at the stage where the first polishing is completed, although the flatness of the D1 surface is secured,
The upper surface 18a of the glass substrate 18 and D1 of each blank 14
The surfaces are not necessarily parallel. Therefore, after performing the above polishing once, the upper surface 18 of the glass substrate 18 is
a and the parallelism of the D1 plane of the blank 14 is measured.

This measurement method uses an interferometer to observe the parallelism between the upper surface 18a of the glass substrate 18 and the D1 plane of the blank 14. At this time, the required flatness and parallelism can be measured simultaneously by the number of interference fringes.
4 can be considered to be the same plane, so the range from end to end of the plane formed by these 16 blank D1 surfaces is approximately λ/2 (λ = 632.8 nm).
The flatness is about 0.03 μm and the parallelism is about 0.4 seconds). Based on this measurement, if the accuracy is not within this range, polishing is performed again using the flat lapping machine 60. This operation is then repeated until a predetermined accuracy is secured.

Here, as mentioned above, the blank 14 and
By arranging a large number of works 16 with masters 12 glued on them, you can see the number of interference fringes over a wider area than looking at the flatness of one D1 surface. , the measurement accuracy will improve. Further, by making the area of the polished surface as wide as possible compared to the height of the workpiece 16, the edge portions of each surface of the blank 14 can be processed more sharply. In view of these points and in order to improve work efficiency, it is preferable to bond a large number of works 16 to the glass substrate 18 for one lapping process.

When the polishing of the D1 surface of each blank 14 is completed in this way, the blank 14 and the master 1 are then polished so that the next processed surface B1 is the upper surface, as shown in FIG.
The workpiece 16, which is the bonded body with the workpiece 2, is reattached to the glass substrate 18. At this time, the work 16 is placed on the glass substrate 18.
The method for separating the master 12 from the master 12 is to heat the glass substrate 18 from below and use the difference between the coefficient of thermal expansion of the material of the glass substrate 18 and the material of the master 12 to separate it. Therefore, the material of the glass substrate 18 and the material of the master 12 are made of materials having different coefficients of thermal expansion. Furthermore, for the same reason, the material of the blank 14 and the material of the master 12 are also made of materials having different coefficients of thermal expansion. That is, for example, when using BK7 for the blank 14, transparent quartz glass is used for the master 12,
Different materials are used for the glass substrate 18, such as using BK7 again. Conversely, when transparent quartz glass is used for the blank 14 and the glass substrate 18, BK7 is used for the master 12.

The subsequent procedure for bonding the workpiece 16 to the glass substrate 18 is as described above. Further, the orientation of each blank 14 at this time is such that its apex faces inward, as in the case of processing the D1 side. Then, the glass substrate 18 in the state shown in FIG. 7 is placed upside down again on the flat lapping machine 60. Then, in the same way as when processing the D1 surface, finishing is performed while measuring the surface accuracy.

Next, as shown in FIG. 8, the blank 14 and
The bonded body 16 with the master 12 is bonded again to the glass substrate 18, and the C1 side is similarly finished. The blank 14 is now in a substantially completed state. next,
The adhesion between the glass substrate 18 and the master 12 and the adhesion between the blank 14 and the master 12 are removed, but at this time, the materials of the blank 14, the master 12, and the glass substrate 3 are selected as described above. Because
These can be easily peeled off by heating the adhesive and utilizing the difference in thermal expansion between the glass substrate 18, the master 12, and the blank 14.

[0033] Then, the blank 14, which has become a single unit, is
Next, grinding is performed so that the distance from the apex 14a to the entrance surface G1 becomes a predetermined value, and then plane lapping is performed.C. C. R
50 is completed. As explained above, blank 14 and
The respective corresponding surfaces of the master 12 are joined so as to be approximately point symmetrical with respect to a certain point near the center on the joint surface G2,
By polishing the corresponding surfaces of the blank 14 so that they are parallel to each surface of the master 12, the C.I. C. It becomes possible to process R50.

Next, this C. C. An example in which R50 is applied to a length measuring machine for an X-Y table of semiconductor manufacturing equipment (stepper) will be described. FIG. 9 is a diagram showing a measuring optical system of a length measuring machine for an X-Y table of a stepper. In the figure,
Reflection mirror 7 installed on the X-Y table of the stepper
A laser head (not shown) that oscillates a helium-neon laser is provided at a position facing 2, and the helium-neon laser is irradiated toward the reflecting mirror 72 from this laser head. A measurement optical system is placed on the irradiation optical axis 74 of this laser head, sandwiched between the laser head and a reflection mirror 72, and the laser beam passes through this measurement optical system and is placed on the X-Y table. bar mirror 77
It is designed to reach.

A polarizing beam splitter 76 is disposed at the center of the measuring optical system 80 to split the laser beam into two optical paths. This polarizing beam splitter 76
has a bonded surface 76a that reflects or transmits laser light depending on its polarization state, and this bonded surface 76a
is arranged at an angle of 45 degrees with respect to the irradiation optical axis 74 of the laser beam. Therefore, a part of the polarized light component of the laser light is reflected in a direction perpendicular to the irradiation optical axis 74 at this joint surface 76a, and the remaining polarized light component is
The laser light is transmitted in a direction along the irradiation optical axis 74 of the laser light. Further, on the optical axis of the laser beam of the component reflected in the direction orthogonal to the irradiation optical axis 74 of the laser beam, there is a C. C. R50 is placed. A λ/4 plate is arranged between the polarizing beam splitter 76, the reflecting mirror 72, and the bar mirror 77.

In the measurement optical system 80 arranged in this way,
When the laser beam emitted from the laser head enters, the polarization state of the laser beam is changed in each optical element as is well known, and the laser beam is reflected onto the optical axis 79 of the optical element and reflected by the bar mirror 77 of the X-Y stage. The measurement light (shown by the solid line) that has been returned to C. C. Reference light reflected by R50 (
(shown with broken lines) is output. Then, by observing the interference state between the reference light and the measurement light with a receiver (not shown) placed on the output optical axis 79, and measuring the number of changes in the interference state, the wavelength of the laser light can be determined. From this, the moving distance of the bar mirror 77, that is, the moving distance of the XY table can be calculated.

Here, when the laser beam is reflected at the position C, the component reflected by the bar mirror 77 and this C. C. In order to cause interference between the components reflected by R50, it is necessary for both lights to be incident on the receiver at exactly the same position. The simplest way to achieve this is to reflect the laser beam exactly 180 degrees in the opposite direction at the position C, and then reflect the incident light exactly 180 degrees at the bar mirror 77 as well. In other words, if the lights return to each other accurately, their positions will match.

For this reason, if a reflecting mirror is placed at position C, for example, depending on the angle of this reflecting mirror,
Since the direction in which light is reflected changes, it is necessary to accurately position the angle of this reflecting mirror. And X-Y
Similarly, the mounting angle of the bar mirror 77 on the table side must be accurately positioned with respect to the incident beam. Therefore, the reflecting mirror must be accurately positioned at two locations.

At that point, the reflection at the position of C. C. R5
0, this C. C. The positioning of R50 is performed by C.
C. It is only necessary to make sure that the laser beam is incident within the effective diameter of R50, and the accuracy of the mounting angle is significantly reduced compared to the case where a reflecting mirror is used. and,
It is only necessary to precisely adjust the angle of the bar mirror 77 of the XY table, and therefore the reflecting mirror only needs to be positioned at one location.

Furthermore, as shown in FIG. 10, a C.I. C. If R50 is used, positioning of the reflecting mirror 72 is also unnecessary. Note that the present invention can be applied to modifications or variations of the above embodiments without departing from the spirit thereof. For example, in the above example,
Although the case where optical glass is used as the master material has been described, it may be replaced with a metal material or the like.

[0041]

[Effects of the Invention] As explained above, according to the optical element and the processing method thereof of the present invention, the optical element has the same shape as the completed shape of the optical element and is parallel to each surface of the highly precisely finished master. Another advantage is that by simply polishing each side of the blank, it is possible to finish an optical element to an accuracy close to that of the master, making it possible to process large quantities of optical elements of the same shape with simple work. There is.

[0042] Furthermore, since there is no cutting or rounding process after each surface is finished, there is no need to produce defective products due to chipping defects after performing the troublesome surface leveling work.
This has the effect of improving yield.

[Brief explanation of the drawing]

[Figure 1]C. C. FIG. 3 is a diagram showing a state in which an R master and a blank are pasted together.

FIG. 2 is a diagram of FIG. 1 viewed from the direction of the arrow.

FIG. 3 is a diagram showing a state in which a blank and a master are attached to a glass substrate.

FIG. 4 is a side view of FIG. 3;

FIG. 5 is a cross-sectional view showing a state in which a glass substrate is placed on a flat lapping machine.

FIG. 6 is a plan view showing the structure of a flat lapping machine.

FIG. 7 is a side view showing a state in which a blank and a master are attached to a glass substrate.

FIG. 8 is a side view showing a state in which a blank and a master are attached to a glass substrate.

FIG. 9C. C. FIG. 3 is a diagram showing an example in which R is applied to a length measuring machine of an X-Y table of a stepper.

FIG. 10 is a diagram showing a modification of the length measuring machine.

FIG. 11C. C. It is a figure showing the reflection state of R.

FIG. 12 is a diagram showing a processing procedure for a regular hexahedral block.

FIG. 13: C. from regular hexahedral block. C. It is a figure showing the procedure of cutting out R.

[Explanation of symbols] 12 Master 14 Blank 16 Work 18 Glass substrate 50 Corner cube reflector (C.
C. R) 54 Regular hexahedral block 56 Triangular pyramid block 60 Plane lapping machine 62 Polishing machine 64 Ring 66 Core 72 Reflection mirror 74 Irradiation optical axis 76 Polarizing beam splitter 77
Bar mirror 78 λ/4 plate

Claims (17)

[Claims] 1. An optical element processing method for transferring a master shape that is made into the same shape as the completed optical element onto a blank to be processed into the optical element, the method comprising: The blank having a blank reference plane finished with an accuracy substantially equal to the accuracy is brought into close contact with a master having first and second master reference planes, and the blank reference plane and the first master reference plane are brought into close contact with each other. 1. A method of processing an optical element, characterized in that the blank is bonded in a state in which the blank is joined, and the blank is processed so as to form a plane parallel to the second master reference plane on the blank. 2. The blank has a processing plane that defines a second angle with the blank reference plane that is substantially equal to a first angle formed by the first and second master reference planes. , joining the blank and the master so that the first angle and the second angle form a substantially illusion angle;
2. The method of processing an optical element according to claim 1, wherein the processing is performed so that the processing plane is parallel to the second master reference plane. 3. A bonded body in which the blank and the master are bonded is bonded to a base having one bonding plane with the second master reference plane and the bonding plane in close contact with each other. 2. The method of processing an optical element according to claim 1, further comprising processing the blank so as to form a plane parallel to the bonding plane on the blank. 4. A plurality of bonded bodies are bonded to the base, and each of the blanks constituting each of the plurality of bonded bodies is bonded to the base so that a plane parallel to the bonded plane is formed. 4. The method of processing an optical element according to claim 3, further comprising processing a bonded body of. 5. The method of processing an optical element according to claim 1, wherein the master has a relief portion from the processing plane. 6. The blank and the master are bonded together using vacuum bonding.
The method for processing an optical element described in . 7. The method for processing an optical element according to claim 6, wherein a liquid is used for the vacuum bonding. 8. The blank and the master are made of materials having different coefficients of thermal expansion, and the dimensional change due to thermal expansion of the blank and the master, which is caused by heating the joined body of the blank and the master, is utilized. 2. The method of processing an optical element according to claim 1, further comprising: releasing the bonding. 9. An optical element processed using the optical element processing method according to claim 1. 10. A method for processing an optical element for precisely finishing the surface accuracy of the plurality of planes and the angular accuracy between the plurality of planes when processing an optical element having a plurality of planes, the method comprising: A blank having a plurality of processing planes corresponding to the planes of and roughly machined into a shape substantially equal to the completed shape of the optical element in order to be finished into the optical element, is the same as the completed shape of the optical element. A predetermined joint that is one of a plurality of reference planes corresponding to each of the plurality of planes of the master, which has a shape and is finished with a precision higher than the shape accuracy of the completed shape. a state in which a processing plane of the blank corresponding to the joint surface is in close contact with the surface, and each of the plurality of reference planes and each of the plurality of processing planes corresponding to each of the reference planes,
The blanks are joined in a state where they are approximately symmetrical with respect to a certain point on the joining surface, and each of the plurality of machining planes of the blank is set to each reference of the master corresponding to each of the machining planes. A method for processing an optical element, characterized in that the bonding is released after finish processing to make the optical element parallel to a plane. 11. A bonded body in which the blank and the master are bonded is bonded to a base having one bonding plane, with each of the plurality of reference planes sequentially brought into close contact with the bonding surface. 11. Processing of an optical element according to claim 10, wherein each of the plurality of processing planes corresponding to each of the plurality of reference planes is sequentially finished so as to be parallel to the bonding plane. Method. 12. A plurality of the joined bodies are joined to the base, and the plurality of processing planes of each of the blanks constituting each of the plurality of joined bodies correspond to each of the plurality of processing planes. 12. The method of processing an optical element according to claim 11, wherein the finishing process is performed so that the master is parallel to a reference plane of each of the masters. 13. The method of processing an optical element according to claim 10, wherein the master has a relief portion from each processing plane of the blank. 14. The method of processing an optical element according to claim 10, wherein the blank and the master are bonded using vacuum bonding. 15. The method of processing an optical element according to claim 14, wherein a liquid is used for the vacuum bonding. 16. The blank and the master are made of materials having different coefficients of thermal expansion, and the dimensional change of the blank and the master due to thermal expansion is caused by heating the joined body of the blank and the master. Claim 1, characterized in that the bond is released by using
0. The method for processing an optical element according to 0. 17. An optical element processed using the optical element processing method according to claim 10.