DEVICE AND METHOD FOR COATING SUBSTRATES HAVING PLANAR OR SHAPED SURFACES BY MEANS OF MAGNETRON SPUTTERING
20220213591 · 2022-07-07
Assignee
Inventors
- Michael VERGÖHL (Braunschweig, DE)
- Andreas PFLUG (Braunschweig, DE)
- Stefan BRUNS (Braunschweig, DE)
- André KAISER (Braunschweig, DE)
- Thomas MELZIG (Braunschweig, DE)
- Tobias ZICKENROTT (Braunschweig, DE)
Cpc classification
C23C14/044
CHEMISTRY; METALLURGY
International classification
C23C14/35
CHEMISTRY; METALLURGY
C23C14/54
CHEMISTRY; METALLURGY
Abstract
According to the invention, a device is provided for coating substrates having planar or shaped surfaces by means of magnetron sputtering, by means of which device surfaces having any shape, for examples lenses, aspheres or freeform surfaces which have an adjustable layer-thickness profile, can be coated such that a layer function is maintained on the substantially complete surface. A method for coating substrates having planar or shaped surfaces by means of magnetron sputtering is also provided.
Claims
1-17. (canceled)
18. A device for coating substrates having planar or shaped surfaces by magnetron sputtering comprising a) a vacuum chamber; b) at least one magnetron source having at least one magnetron electrode as coating source; c) a turntable having at least one substrate holder, with the turntable enabling a first rotational movement of the substrate and the at least one substrate holder enabling a second rotational movement; d) at least one controllable motor that is coupled to the at least one substrate holder and that effects the second rotational movement of the substrate; and e) a gradient mask having a first region that has a geometry for the inhomogeneous coating of the substrate, with the gradient mask having at least one local elevated portion or at least one local depression in profile, with the layer thickness gradient on the substrate being settable via the pitch at the flank of the elevated portion or depression; and a second region having a geometry that effects a homogeneous coating of a reference substrate.
19. The device in accordance with claim 18, wherein the gradient mask has at least one local elevated portion in profile for the coating of convexly shaped substrates and the film gradient on the substrate is settable via the pitch of the flank of the elevated portion.
20. The device in accordance with claim 18, wherein the gradient mask has at least one local depression in profile for the coating of concavely shaped substrates and the film gradient on the substrate is settable via the pitch of the flank of the depression.
21. The device in accordance with claim 18, wherein the gradient mask has a plurality of regions having different geometries that enable a coating of the substrate with different gradients.
22. The device in accordance with claim 18, wherein the controllable motor is reversibly or irreversibly fixed to the substrate holder, with the controllable motor being able to be arranged inside or outside the vacuum chamber.
23. The device in accordance with claim 18, wherein the coating source is a linear, annular, or tubular magnetron source.
24. The device in accordance with claim 18, wherein the substrate consists of a glass or a plastic.
25. The device in accordance with claim 18, wherein the substrate is an optical component.
26. The device in accordance with claim 25, wherein the optical component is selected from the group consisting of lenses, aspheres, and freeform optics.
27. The device in accordance with claim 26, wherein the lens is planar, convex, or concave.
28. A method of coating substrates having planar or shaped surfaces by magnetron sputtering, in which a) at least one substrate is arranged in an associated substrate holder on a turntable in a vacuum chamber, with the turntable enabling a first rotational movement of the substrate and the at least one substrate holder enabling a second rotational movement, with the second rotational movement taking place via a controllable motor coupled to the substrate holder; b) at least one film is deposited on the at least one substrate in a coating cycle by utilizing at least one magnetron source having at least one magnetron electrode, with the layers of source material of the magnetron electrodes being formed by sputter gas and optionally reactive gas, with a gradient mask having a first region that has a geometry for the inhomogeneous coating of the substrate, with the gradient mask having at least one local elevated portion or at least one local depression in profile, with the layer thickness gradient on the substrate being settable via the pitch at the flank of the elevated portion or depression; and a second region having a geometry that effects a homogeneous coating of a reference substrate, being used; and c) the first rotational movement being coordinated with the second rotational movement such that the second rotational movement only takes place outside the coating cycle.
29. The method in accordance with claim 28, wherein the first rotational movement of takes place at a speed of 30 to 300 r.p.m.
30. The method in accordance with claim 28, wherein the deposited film thickness per film is in the range from 1 to 1000 nm, and the number of layers is in the range from 1 to 1000.
31. The method in accordance with claim 28, wherein a continuous first rotational movement takes place for every deposited film at a speed of 10 to 300 r.p.m, and a step-wise second rotational movement takes place with n steps and a movement by an angle of 360°/n, where n=3, 6, 9, or 12.
32. The method in accordance with claim 28, wherein the gradient mask has at least one local elevated portion in profile for the coating of convexly shaped substrates and the film thickness gradient on the substrate is set via the pitch of the flank of the elevated portion.
33. The method in accordance with claim 28, wherein the gradient mask has at least one local depression in profile for the coating of concavely shaped substrates and the film thickness gradient on the substrate is set via the pitch of the flank of the depression.
34. The method in accordance with claim 28, wherein the shape of the gradient mask is detected via the local distribution of the coating rate, with the coating rate being determined via the shape of the lens in dependence on the mask shape.
Description
[0084] The subject matter in accordance with the invention will be explained in more detail with reference to the following Figures and examples without intending to restrict it to the specific embodiments shown here.
[0085]
[0086]
[0087]
[0088]
[0089]
[0090]
[0091]
[0092]
[0093]
[0094]
[0095]
[0096]
[0097]
[0098]
[0099]
[0100]
[0101]
[0102]
[0103]
[0104]
[0105]
[0106]
EXAMPLE 1
[0107] An SiO.sub.2 film having a thickness of 100 m is applied in a layer stack. The dynamic rate amounts to 0.4 nm/sec with a rotating table having a rotation frequency of 240 r.p.m., i.e., 0.1 mm per revolution. 1000 runs are then required for the 100 nm film. In this example, a rotation by 90° is carried after 250, after 500, and after 750 revolutions.
EXAMPLE 2
[0108] In a second example a rotation is carried out by 9° after every 25 revolutions to achieve a finer division.
EXAMPLE 3
Determining the Geometry of the Gradient Mask
[0109] The aim of the coating of lenses with optical filters is that the filter works the same over the total lens surface, i.e., e.g., the same transmission spectrum and reflection spectrum should result over the total lens surface for every light ray through different points of the lens. Two effects are taken into account here: [0110] The lens surface has a different surface inclination at different radii depending on the lens shape. [0111] Depending on the design of the optics, the light rays are incident on the different radial positions at a different angle, e.g., it makes a difference whether the lens is impinged by collimated or divergent light.
[0112] An “angular spectrum” is corresponding typically specified for a lens, i.e., the mean angle of incidence of the light as a function of the radial distance from the lens center. Depending on the angle of incidence, more or less pronounced spectral shifts result for an optical filter and this has to be compensated by a variation of the film thickness.
[0113] Due to an optical modeling of the filter for different angles of incidence of light, the relative film thickness as a function of the radial position from the lens center (fixed at 100% film thickness at the lens center) is ultimately obtained as the target function.
[0114] The mask form ensures an implementation of this target function that is as good as possible. The mask shape can be different depending on the lens shape, the geometry of the coating process, and also on the optical filter design.
[0115] The determination of the mask shape takes place in the following manner on the basis of this knowledge:
[0116] There is no simple, intuitive relationship between the mask shape and the resulting film thickness profile on the lens. A “digital twin” of the process was developed by which the effect of the mask shape on the film thickness distribution can be simulated. A mask shape adapted to the target function can also be determined while using an optimization algorithm. This has the advantage that the mask shape does not have to be determined in a large number of iterative experiments. The mask shape for the example of a bandpass filter on a spherical lens was determined with the aid of the digital twin and then no longer had to be reworked for the experiments.
[0117] The mask shape is implemented by consecutive arcs in this concept. Coordinate points {x.sub.k, y.sub.k}.sub.k=1 . . . N and circle radii {r.sub.k}.sub.k=1 . . . N−1 were considered by means of software. The X coordinate here faces outward from the turntable center in the radial direction; the Y coordinate faces transversely thereto, i.e., in the direction of movement of the lens. The circle radii are positive for a concave arc and negative for a convex arc.
[0118] The circle center (xm.sub.k, ym.sub.k) of the respective arc is first calculated with reference to the radius r.sub.k between two consecutive points (x.sub.k, y.sub.k), y.sub.k+1). The shape of the mask is subsequently shown as the function y(x):
y(x)=ym.sub.k±√{square root over (r.sub.k.sup.2−(x−xm.sub.k).sup.2)};x∈[xm.sub.k,xm.sub.k+1]
[0119] The positive sign is selected if r.sub.k<0, otherwise the negative sign is selected.
[0120] The digital twin calculates the resulting film thickness distribution on a lens moving through the coating compartment using this parameterization. A fit algorithm is used for the determination of the mask shape in which some of the variables of both the coordinate points {x.sub.k, y.sub.k} and the circle radii {r.sub.k} are deallocated and varied until the calculated film thickness profile corresponds as exactly as possible to the target function.
[0121] With a convex lens, the film thickness should increase toward the margin; this then produces a characteristic mask shape in which a locally serrated thickened portion of the mask impinges on the center of the lens rotating past.