CN1187884A

CN1187884A - Optical multiplexing device and method

Info

Publication number: CN1187884A
Application number: CN96194715A
Authority: CN
Inventors: 迈克尔·A·司考比; 玻尔·斯图彼克
Original assignee: Optical Corp of America
Current assignee: Optical Corp of America
Priority date: 1995-06-15
Filing date: 1996-06-10
Publication date: 1998-07-15

Abstract

An optical multiplexing device spatially disburses collimated light from a fiber optic waveguide into individual wavelength bands, or multiplexes such individual wavelength bands to a common fiber optic waveguide or other destination. The optical multiplexing device has application for dense channel wavelength division multiplexing (WDM) systems for fiber optic telecommunications, as well as compact optical instrument design. Multiple wavelength light traveling in a fiber optic waveguide is separated into multiple narrow spectral bands directed to individual fiber optic carriers or detectors. An optical block has an optical port for passing the aforesaid multiple wavelength collimated light, and multiple ports arrayed in spaced relation to each other along a multiport surface of the optical block. A continuous, variable thickness, multi-cavity interference filter extends on the multiport surface of the optical block over the aforesaid multiple ports. At each of the multiple ports the continuous interference filter transmits a different wavelength sub-range of the multiple wavelength collimated light passed by the optical port, and reflects other wavelengths. Multicolor light passed to the optical block from the optical port is directed to a first one of the multiple ports on an opposite surface of the optical block. The wavelength sub-range which is 'in-band' of such first one of the multiple ports is transmitted through that port by the local portion of the continuous, variable thickness interference filter there, and all other wavelengths are reflected. The light not transmitted through the first port is reflected to strike a second port, at which a second (different) wavelength band is transmitted and all other light again reflected. The reflected optical signals thus cascades in a 'multiple-bounce' sequence down the optical block of the multiplexing device, sequentially removing each channel of the multiplexed signal.

Description

Optical multiplexer and its manufacturing method

RELATED APPLICATIONS

The present invention is a continuation of part of the U.S. patent application No. 08/490,829 to co-pending, entitled "optical multiplexer" filed on 6/15 of 1995, and is also a continuation of the U.S. patent application No. 08/300,741 to co-pending, entitled "low pressure reactive magnetron sputtering apparatus and method of manufacturing the same", filed on 9/2 of 1994 of Michael a.

Technical Field

The present invention relates to an optical multiplexer that spatially distributes multiple wavelength collimated light from a fiber optic waveguide into wavelength bands, each of which leads to a separate fiber optic waveguide output line, photodetector, etc., or multiplexes these separate wavelength bands into a common fiber optic waveguide or other destination component. In certain preferred embodiments, the improved multiplexer of the present invention is particularly suitable for use in dense channel wavelength division multiplexing (DWDM) systems for fiber optic communication systems.

Background

While fiber optic cables are being used extensively for data transmission and other communications applications, the cost of newly installing fiber optic cables is an obstacle to increasing the carrying capacity. Wavelength Division Multiplexing (WDM) allows different wavelengths to be transmitted on a common fiber optic waveguide. Presently preferred wavelength bands for optical fiber transmission media include those centered at 1.3 mu and 1.5 mu. The latter wavelength band is particularly popular because of its minimal absorption and the commercially available erbium doped fiber amplifiers. It has a useful bandwidth of about 10-40 nanometers, depending on the application. Wavelength division multiplexing techniques may divide the bandwidth into multiple channels. For example, in an ideal case, a 1.55 μ wavelength band may be separated into multiple discrete channels (e.g., 8 channels, 16 channels, or even up to 32 channels) by a technique known as dense channel wavelength division multiplexing (DWDM). The technique is low cost and substantially increases long distance traffic over existing optical fiber transmission lines. Wavelength division multiplexing may be used to provide video on demand and other existing or planned multimedia interactive services. But require techniques and equipment for multiplexing different discrete carrier wavelengths. That is, the individual optical signals must be combined into a common fiber optic waveguide and then split into individual signals or channels at the other end of the fiber optic cable. Thus, the ability to efficiently combine and then separate a single wavelength (or band of wavelengths) from a broader spectral source is increasingly important for the field of fiber optic communications, as well as other fields using optical instruments.

Optical multiplexers are known for use in optical spectrum analyzers and can combine or separate optical signals in wavelength division multiplexed fiber optic communication systems. Known devices for achieving this have been used, such as diffraction gratings, prisms and various types of fixed or adjustable filters. Gratings and prisms generally require complex and bulky collimating systems and have been found to be less efficient and stable under environmentally variable conditions. Fixed wavelength filters such as interference coatings (interference coating) can be made substantially more stable, but transmit only a single wavelength or band of wavelengths. In this regard, interference coatings made of metal oxides such as niobium oxide (niobia) and silicon dioxide, which have greatly improved properties, can be produced using commercially known plasma deposition techniques such as ion-assisted electron beam evaporation, ion beam sputtering, reactive magnetron sputtering, and the like, as disclosed, for example, in U.S. patent No. 4,851,095 to Scobey et al. Such coating methods allow the manufacture of interference cavity filters composed of stacked dielectric optical coatings, which have the advantages of being dense and stable, low scattering and absorption by the thin film, and insensitive to temperature variations and ambient humidity. Figure 1 shows theoretically the spectral characteristics of a stable triple cavity filter (tilted 12) made by any of the advanced deposition methods. It can be seen that the spectral curve meets stringent application specifications.

In order to overcome the above-mentioned disadvantages of interference filters, i.e. that they transmit only a single wavelength or a single wavelength range, it has been proposed to combine or combine a plurality of filters on a common parallelogram prism or other common substrate. For example in the multiplexer of U.K. patent application GB 2,014,752a, optical filters are combined together to separate the different wavelengths of light transmitted on a common optical waveguide. At least two transmission filters are adjacently secured to a transparent dielectric substrate, each transmission filter transmitting a separate, different wavelength of light and reflecting other wavelengths of light. The optical filters are arranged such that each filter in turn partially transmits and partially reflects a light beam, creating a zigzag light path. Light of a particular wavelength is incorporated or separated at each filter depending on whether the element is used as a multiplexer or demultiplexer. Similarly, in the european patent application No. 85102054.5 filed by Oki electrical industries, ltd, a so-called hybrid optical wavelength division multiplexer-demultiplexer is proposed in which a plurality of discrete interference filters of different transmittances are mounted on the side of a glass block. United states patent No. 5,005,935 to Kunikani et al proposes a related method in which a wavelength division multiplexing optical transmission system used for bidirectional fiber optic communication between a central telephone exchange and subscribers mounts a plurality of discrete filter elements on respective surfaces of a parallelogram prism. For example, U.S. patent No. 4,768,849 issued to Hicks, Jr proposes another method of extracting a selected wavelength from a trunk carrier optical signal over multiple wavelength bands. In this patent, a plurality of filter taps are arranged to separate a series of wavelength bands or channels from the trunk line, with each filter directing a light signal with a dielectric mirror and lens.

Mounting a plurality of discrete filter elements on the surface of a prism or other optical substrate has significant disadvantages in terms of assembly cost and complexity. In addition, the exact wavelength of the filter element is not determined when it is manufactured, thereby creating a significant problem associated with wavelength division multiplexers and similar devices that use multiple discrete interference filter elements. That is, in the manufacture of multiplexers, each bandpass filter element is manufactured separately, for example, a device using 8 individual bandpass filters will typically require much more than 8 coating blocks to make the 8 necessary suitable filter elements. Bandpass filters (especially in the infrared range) are very thick and require complicated and expensive vacuum deposition equipment and techniques. Thus, producing each coating block is expensive and difficult. For this reason, apparatus for manufacturing 8-channel WDM devices using, for example, 8 discrete interference filter elements, are quite expensive and have not been fully accepted by the market.

Another problem associated with optical multiplexers that use multiple individual bandpass filter elements is the need to mount the elements in close to perfect parallelism on an optical substrate. The filter elements are very small, typically around 1.5 mm in diameter, and therefore difficult to operate accurately. Improper fixation of the filter element can greatly reduce the optical accuracy and thermal stability of the device. A related problem is that an adhesive medium has to be added between the filter element and the surface of the optical substrate. The optical signal path passes through the adhesive, causing degradation in system performance. In an optical multiplexer intended for the telecommunications industry, it is desirable that the epoxy adhesive in the optical signal path be as low as possible.

It is an object of the present invention to provide an improved optical multiplexer which reduces or completely overcomes some or all of the above-mentioned difficulties inherent in prior known devices. Particular objects and advantages of the invention will be apparent to those skilled in the art, that is, those who are knowledgeable or experienced in this field of technology, in view of the following disclosure of the invention and detailed description of certain preferred embodiments.

Disclosure of Invention

In accordance with a first aspect, an optical multiplexing device includes an optical component which may be a solid optical substrate such as glass or fused silica, or may be a closed chamber which is hollow (meaning evacuated or filled with air or other optically transparent dielectric). The optical element has an optical port for transmitting collimated light of multiple wavelengths. Depending on the application of the optical multiplexer, multiple wavelengths of collimated light may be passed through an optical port, into an optical component, demultiplexed, or input as a multiplexed signal from the optical component into a fiber optic transmission line or other destination. The plurality of ports are spaced apart from one another along a multiport surface of the optical element. As described below in connection with some preferred embodiments, the optical element may have more than one such multiport surface. Each of these multiple ports may transmit a channel of optical signals. Thus, each port transmits one wavelength subrange of the multiple wavelength collimated light transmitted by the optical port. In the case of the optical multiplexer being applied to a multi-channel communication system, each port of the plurality of ports will typically pass through a single discrete channel and combine, the channels forming the above-described multi-wavelength collimated light transmitted by the optical port. An interference filter of continuously variable thickness, preferably a multi-cavity interference filter, is disposed on the multiport surface of the optical element to provide the plurality of ports. Since the continuous interference filter extending along the multiport surface has a different optical thickness at each of the plurality of ports, the filter will transmit at each such port at a different wavelength (or range of wavelengths) than the remaining ports. Thus, it is preferred that a single film deposited directly on the surface of the optical element transmits the optical signal of one of the plurality of channels and reflects other wavelengths at separate locations. As mentioned above, the optical element may be a solid or hollow chamber. In the case of a solid optical element, the multiport surface carrying the interference filter of continuously variable thickness is generally the outer surface of the element. As discussed further below, each port of the multiport surface may be a bandpass filter, preferably a narrow bandpass filter, which is capable of passing a wavelength sub-range spaced from the next adjacent port by as little as 2 nanometers, or for DWDM, the wavelength band spacing may be even smaller. Alternatively, some or all of the plurality of ports are dichroic, i.e., long pass or short pass edge filters (short pass edge filters), preferably with very sharp transition points. The transition point for each port is set at a slightly longer (e.g., 2 nm) (slightly shorter) boundary wavelength. In the case of demultiplexing, since all light at the shorter (or longer) wavelength has been separated, each port will in turn pass or transmit only the optical signals in the increasing range outside the boundary wavelength of the previous port. In accordance with the above described principle of operation, light outside the new port boundary wavelength will be reflected.

The optical multiplexer further comprises means for stepping light within the optical element from one of the plurality of ports to another along a multi-drop transmission path. In the case of demultiplexing, optical signals will be input to the optical element at the optical port and transmitted along the multipoint transmission path to a plurality of ports (in this case, functioning as output ports). The signal of each individual channel is transmitted out of the optical element at its respective port; other wavelengths are reflected or folded back to further step-wise propagate along the optical transmission path within the optical element. It will be appreciated that at the final output there is no longer any light left to reflect. It will also be appreciated from the above description that the optical multiplexer may operate in reverse, or bi-directional. The ladder transmission device preferably includes a reflective film disposed on the second surface of the optical element, the reflective film being a continuous coating covering a multi-point transmission path of the ladder transmission optical signal, or a plurality of discrete reflective elements. The most common optical element is rectilinear and has a reflective film on a second surface of the optical element opposite and parallel to the multiport surface carrying the continuous interference filter. The second film may be a broadband highly reflective film, i.e. a thin film coating that is highly reflective of all channel wavelengths (which in combination may form the multi-wavelength collimated light), or may act as a second interference filter, passing optical signals of one or more channels at spaced apart locations (i.e. at some or each reflection point). In either case, the interference filter and the reflective film on the spacer surface of the optical element cause the optical signal passing through the optical element to propagate along a multiple reflection sequence ladder starting (or ending) at the optical port through which the multiple wavelength collimated light is transmitted. This multiple reflection step transmission is further described below in conjunction with certain preferred embodiments.

Brief description of the drawings

Some preferred embodiments of the invention will be discussed below with reference to the accompanying drawings, in which:

FIG. 1 is a graph showing theoretical characteristics of a high quality multi-cavity dielectric optical interference filter;

FIG. 2 is a schematic diagram of a first preferred embodiment of an optical multiplexer, in particular, a dense channel wavelength division multiplexer for use in an 8-channel fiber optic communication system or similar application;

FIG. 3 is a schematic diagram of another preferred embodiment of an optical multiplexer according to the present invention, and in particular, a dense channel wavelength division multiplexer for use in an 8-channel fiber optic communication system or similar application;

FIG. 4 is a schematic diagram of another preferred embodiment of an optical multiplexer in accordance with the present invention, and in particular, a dense channel wavelength division multiplexer for use in an 8-channel fiber optic communication system or similar application;

FIGS. 5, 6 and 7 are schematic cross-sectional views of three-cavity interference filters of continuously variable thickness in the optical multiplexer of FIG. 2;

FIG. 8 is a schematic cross-sectional view of an apparatus according to a preferred embodiment of the invention;

FIG. 9 is a schematic cross-sectional view of a magnetron apparatus including a source or target and an inert gas shield in accordance with a preferred embodiment of the present invention;

FIG. 10 is a schematic cross-sectional view of an apparatus having multiple magnetron sputtering assemblies in accordance with a preferred embodiment of the invention;

FIG. 11 is a graph showing the relationship between intracavity pressure and intracavity pumping speed, assuming that magnetron pressure is 0.7 microns and the conductance of the magnetron device (C)_M) 3000 l/s;

FIG. 12 is a graph showing the relationship between the intracavity pressure and the intracavity pumping speed, assuming that the magnetron pressure is 0.4 microns and the conductance (C) of the magnetron device_M) 3000 l/s; and

fig. 13 and 14 are front and plan schematic views respectively showing a device according to another preferred embodiment of the present invention.

It should be understood that the optical multiplexer and interference filter shown in the figures are not necessarily drawn in their different sizes or angular relationships. In view of the foregoing disclosure, and the following detailed description of the preferred embodiments, those skilled in the art will be able to select appropriate size and angular relationships for such devices.

Detailed description of the preferred embodiments

As mentioned above, optical multiplexers may have many applications, for example in optical fibre communication systems. The optical multiplexer of the present design is particularly useful in, for example, test equipment and the like, as well as laboratory instruments. For purposes of illustration, the preferred embodiment, which will be described in detail below, is a dense channel wavelength division multiplexer that solves or alleviates the above-mentioned problems of mounting a plurality of filter elements on a single optical substrate to obtain each individual signal channel, the cost and complexity problems associated with multiple coatings for making the individual filter elements, and the associated problems of uncertainty in filter wavelength.

As discussed below in connection with the figures, a graded-wavelength all-dielectric narrow bandpass filter is placed on at least one side of an optical element, preferably a polished parallel plate of a specified thickness. Such a filter, which may be referred to as a cavity filter for short, forms a continuous coating on at least a portion of the surface of the optical element, is preferably a multi-cavity, preferably triple cavity, thin film stack fabricated behind a Fabry-Perot etalon. A thicker cavity layer separates the two dielectric film stacks which themselves form the reflective film for the wavelength of light in question. This structure is then repeated one or more times to produce a filter with enhanced blocking and improved flatness of transmission in-band. The effect is to create a narrow band transmission filter that transmits light in-band and reflects light out-of-band. As mentioned above, dichroic filters may also be used. The improved filter performance provides commercially acceptable dense channel wavelength division multiplexing techniques for optical multiplexer fiber optic communications applications. It allows for less cross talk between channels and allows for a reasonably high number of channels within a given bandwidth. An excessive number of cavities will adversely affect the transmission of the flat in-band wavelengths and increase the production cost of the optical multiplexer beyond commercially acceptable levels.

Using the deposition techniques described above, multi-cavity interference filters of continuously variable thickness can be fabricated using dense, stable metal oxide stacks. Such filters have been shown to have excellent thermal stability (e.g., 0.004 nm/c at 1550 nm) and ultra-narrow bandwidths spaced as small as 2 nm or even 1 nm. Suitable variable thickness filters have been used in other applications such as U.S. patent No. 4,957,371 to Pellicori et al. Stable ultra-narrow band filters have also been proposed in SPIE journal of month 7, 1994. In particular, e.g. authorisingHigh quality interference filters comprising a stack of metal oxides such as niobium oxide and silicon dioxide can be produced using commercially known plasma deposition techniques such as ion assisted electron beam evaporation, ion beam sputtering and reactive magnetron sputtering as disclosed in U.S. patent No. 4,851,095 to Scobey et al, the contents of which are incorporated herein by reference. These coating methods allow the manufacture of interference cavity filters composed of laminated dielectric optical coatings, which are dense and stable, have low film scattering, low absorption, and are insensitive to temperature variations and ambient humidity. The spectral profile of such coatings meets stringent application specifications. By such techniques, a multi-cavity narrow bandpass filter can be fabricated that can transmit a range of wavelengths that is spaced from adjacent wavelength ranges by 2 nanometers or less. One suitable deposition technique is low pressure magnetron sputtering in which the vacuum chamber of a conventional magnetron sputtering system is equipped with a high speed vacuum pump. Gas lines around the magnetron and the target material confine an inert working gas, typically argon, in the vicinity of the magnetron. Since the gas diffuses and diffuses outwardly from the magnetron region, at high speeds, the unusually high rate of evacuation can expel the diffused gas from the vacuum chamber. Thus, the pressure of the inert gas in the vacuum chamber is a function of the pumping speed of the vacuum pump and the holding efficiency of the magnetron shield, and is preferably 5X 10^-5torr-1.5X 10^-4In the range of torr. The reactant gas is fed into the vacuum chamber through an ion gun which ionizes the gas and directs it toward the substrate. The effect is to reduce the amount of gas required to provide the proper stoichiometry to the film and to reduce the reactive gas at the magnetron. A throw of 16 inches and longer can be achieved. Such deposition techniques are discussed in pending U.S. patent application No. 07/791,773, filed on 13/11/1991 and in pending U.S. patent application No. 07/300,741, filed on 21/9/1994, the contents of which are incorporated herein by reference.

As described above, the filter preferably comprises a multi-cavity coating in which a cavity layer separates two dielectric film stacks, which themselves form reflective films for unwanted wavelengths. This structure is then repeated one or more times to produce the multi-cavity filter described above, with improved blocking and improved flatness of in-band transmission. The effect is to create a narrow band transmission filter that transmits in-band light and reflects out-of-band light. In the preferred three-chamber embodiment fabricated using the above-described deposition techniques, excellent thermal stability (e.g., 0.004 nm or better per degree celsius at 1550 nm) and ultra-narrow bandwidths as small as 2 nm or even 1 nm apart have been achieved with a dense, stable metal oxide thin film stack. The thickness of the interference filter is preferably linearly continuously variable. The thickness of the continuous filter may alternatively be varied discontinuously, for example, substantially uniformly at each of a plurality of ports of an optical element associated with separate channels of the fiber optic system.

Interference filters generally comprise two materials, one of which is a high refractive index material such as niobium pentoxide, titanium dioxide, tantalum pentoxide, and/or mixtures thereof, e.g., niobium oxide (niobia) and titanium oxide (titania), and the like. The refractive index of these materials has a value of approximately 2.1-2.3 at a wavelength of 1.5 microns. The lower index material is typically silicon dioxide, which has a refractive index of about 1.43. Interference filters present an "optical thickness" which is the digital product of the actual thickness and the refractive index. Of course, the optical thickness of the continuously variable thickness multi-cavity interference filter used in the optical multiplexing device disclosed herein varies with the actual thickness of the filter at various points along the surface of the optical element. The optical thickness of the interference filter is tuned to transmit a desired sub-range of wavelengths at each of a plurality of ports of the optical element associated with the individual signal channels. In view of the above, it will be apparent to those skilled in the art that the thickness and composition of each coating of the continuous filter can be selected to accommodate any desired spectral line profile for the application of the optical multiplexer. It will also be seen that the thickness of the continuous filter may vary continuously, either linearly or non-linearly, or discontinuously. In certain preferred embodiments, the thickness of the filter at each port is substantially constant, with the thickness increasing (or decreasing) only between one port and the next.

The continuously variable thickness multiple cavity interference filter used in the optical multiplexer disclosed herein has a number of advantages over prior known filters. In the manufacture of the optical multiplexer described above, the entire working portion of one surface of the optical element may be coated in a single coating step, wherein the coating operation (e.g., by appropriate movement of associated lens arrangements, collimators, etc.) is tunable at each "reflection point" to provide a wavelength of exactly ± 0.1 nm. They are stable with respect to time and humidity when made of durable materials to produce dense coatings with a nearly uniform packing density. A single coating operation can coat many optical elements with interference filters simultaneously, thereby substantially reducing the cost of the optical multiplexer. They are easily manufactured to include multiple cavities coherently coupled together with a quarter-wavelength thick coating in accordance with well-known techniques. As described above, the use of multiple cavities has the effect of creating a filter with an increased slope of side frequency (spectral skip) and a wider transmission area. As mentioned above, these two effects provide many benefits over other types of filters such as etalons and diffraction gratings. Since the optical filter can be formed by deposition directly onto one surface of the optical element, it is not necessary to fix the optical filter with epoxy, so that epoxy is not present in the path of transmitting the optical signal. Since the filter is formed on the optical element and a separate fixing operation for positioning and aligning the filter is not required, the stability of the filter can be improved. As described above, the center wavelength of each of the multiple signal channels can be tuned by simply moving the GRIN lens collimator or similar device associated with each signal channel a slight distance in the direction of the thickness variation of the continuous filter. By so moving the associated lens arrangement, it can be aligned with the desired signal wavelength. In this way the disadvantage of not being able to obtain the correct centre wavelength when manufacturing the discrete filter elements can be substantially overcome.

Figure 1 shows a dense channel wavelength division multiplexer using a multi-cavity interference filter with continuously variable thickness to form an ultra-narrow bandpass filter at each of 8 discrete ports on an optical element. The multiplexer can multiplex individually separate wavelength signals onto a common optical fiber transmission line and/or demultiplex such signals. Since the related multiplexing function is easily understood by those skilled in the art, only the demultiplexing function is described in detail herein for the sake of simplicity of explanation. That is, those skilled in the art will recognize that optical signals from each channel may be multiplexed using the same device in reverse. The general specifications of the optical multiplexer according to the preferred embodiment shown in fig. 2 include the data provided in table a.

TABLE A

Number of channels 8

Channel wavelength 1544-

Channel spacing of 2 nm + -0.2 nm

Minimum isolation (isolation) 20-35 db

Insertion loss (total) less than 6 dB

Optical fiber type single mode, 1 meter pigtail

Working temperature range of-20 ℃ to +50 DEG C

The optical multiplexer of fig. 2 satisfies the specifications of table a and, as can be seen, includes an optical element 10, which is preferably a stabilized glass substrate. A means for projecting collimated light, such as a fiber graded index (GRIN) lens collimator 12, couples highly collimated light 14 at a slight angle to an optical element through an aperture or facet on the surface 16 of the optical element. In accordance with a preferred embodiment, the optical element has a thickness "a" of 5 mm, a length "b" of 14.1 mm or more, and a refractive index of about 1.5. The divergence of the collimated light is preferably no more than about 0.15, and the inclination angle "c" of the collimated light incidence optical element is about 15. Thus, the lens arrangement 12 collimates and directs multi-color or multi-wavelength light carried by a fiber optic (preferably single mode) carrier into an optical port 18 on a face 16 of the optical element 10, from which face 16 the light then passes through the optical element to an opposite face 20. On the optical element surface 20, a graded-wavelength all-dielectric narrow bandpass filter 22 is mounted. In particular, the filter 22 is a multi-cavity interference filter of continuously variable thickness as described above, and preferably is a linear continuously variable filter. Light incident on the optical element at optical port 18 first exits at output end 24 and opposite face 20. At the output end 24, the filter 22 transmits a sub-range of wavelengths contained in the collimated light 14. Specifically, light 26 passes through port 24 of the optical element and enters a collimating lens arrangement 28 associated with the first signal channel. The optical signal carried by port 24 is thus transmitted as a demultiplexed signal to an optical fiber, preferably a single mode optical fiber 30.

The continuous filter 22 at port 24 reflects wavelengths that are not "in-band" wavelengths of the filter at that location. This reflected light 32 is reflected from the optical element surface 20 back to the surface 16. The surface 16 is provided with a broadband highly reflective film or coating 34. The highly reflective film 34 does not obscure the optical port 18 from blocking the passage of collimated light 14 from that location into the optical element. At surface 16, reflective film 34 reflects reflected light 32 from first output end 24 back toward optical element surface 20. At optical port 18, collimated light 14 is incident on the optical element at an angle of about 15, is refracted at an angle of about 9.9 according to Snell's law, and is then reflected between the relatively

parallel surfaces

16 and 20 of the optical element. Thus, light 32 reflects off of the reflective film 34 and strikes the optical element surface 20 at a second location 36 corresponding to a second output end of the optical element. At the output end 36, the multi-cavity interference filter 22, which has a continuously variable thickness, is transparent to a different wavelength or sub-range of wavelengths than at the output end 24. For dense channel wavelength division multiplexing applications, the wavelength spacing between each of the multiple ports linearly spaced along the optical element surface 20 is preferably about 2 nanometers or less. Thus, at output 36, the optical signal corresponding to the second channel is transmitted through filter 22 to collimating lens 38 and thence into fiber carrier 40. The interference filter 22 at the output end 36 reflects the non-in-band light at that location, as at the first output end 24. Thus, the remaining portion 42 of the collimated light 14 that first enters the optical element at optical port 18 is reflected from port 36 back to the highly reflective layer 34 on the opposite face 16 of the optical element. And thus reflects it back to the third output 44. The reflected wavelengths then continue in a staircase-like fashion down the optical element along a "zigzag" or "multi-reflection" path in the same manner, with successive reflections at the surface 20 of the optical element separating the optical signals of each individual channel.

Thus, as shown in FIG. 2, the zigzag path traveled by the light through the optical element 10 causes the reflected wavelengths to impinge in sequence on the lower additional output ends 46, 48, 50, 52, and 54. At each of these ports, the demultiplexed optical signals are transmitted to an associated collimating lens, and each collimating lens communicates with a respective signal carrying line or other destination. Although the filter 22 is preferably capable of reflecting all wavelengths on each output that are not in-band, in some instances it may be necessary to reflect only those wavelengths of the optical signal that are not extracted on the superior output (i.e., the output that was encountered earlier in the multi-reflection stepped sequence). Also, as will be understood by those skilled in the art from this description, the optical multiplexer of fig. 2 is equally suitable for combining optical signals of 8 separate channels. Thus, the multiple ports on surface 20 will be input ends, while optical ports 18 will be output ends. The staircase propagation process will then proceed upwards from the bottom of the optical element (as shown in figure 2).

As described above for the optical element 10 of fig. 2, the linear interval (TAN [9.9] × 2 × 5 mm) of each output end was 1.76mm (8 × 1.76mm) for the optical element having a thickness of 5 mm, a tilt angle of 15 ° and a reflection angle of 9.9 ° in the optical element. Thus, the length of the continuous interference filter 22 on the optical element surface 20 should be at least 14.1 millimeters. The total distance traveled by the optical signal associated with the last of the 8 channels (5 mm x 8 channels x 2 reflections) was 80 mm. The total beam spread (80 mm TAN-1[ SIN-1] [ SIN ] [0.15/1.5]) was approximately 0.138 mm. Thus, the total loss for a 0.5 millimeter beam is approximately 1.9 decibels. Thus, those skilled in the art will appreciate that the optical multiplexer of fig. 2, as described above, is adapted to demultiplex a number of individual wavelength channels from a beam of incident light in a very efficient manner, since the resulting beam divergence is minimal. The preferred embodiment described above has a total beam expansion of about 40% of a half millimeter beam, which only produces the above-mentioned 1.9 db loss, or less than 0.25 db per channel stepped through the optical multiplexer. In particular, those skilled in the art will appreciate that the multi-reflection ladder transmission technique achieved with a continuously variable thickness multi-cavity interference filter deposited directly on the surface of an optical element provides an optical multiplexer having performance characteristics including cost and simplicity of construction, performance reliability and compactness, etc., which are greatly improved over prior known devices.

In another preferred embodiment shown in fig. 3, collimated light 60 from a lens device 62 in communication with a single mode optical fiber 64 is transmitted into an optical element 66 at an optical port 68, substantially in accordance with the embodiment of fig. 2 described above. Thus, light passes through the optical element 66 to the opposing multiport surface 70 in the optical element. A multi-cavity interference filter 74 of continuously variable thickness extends across surface 70 to provide a narrow bandpass filter at each

output end

72, 76, 78 and 80. As with the embodiment of fig. 2, at each such port, the optical filter 74 transmits a different wavelength, thereby transmitting a single optical signal for the respective channels 1, 3, 5 and 7 into the corresponding lens device and fiber optic waveguide, respectively. On the surface 82 of the optical element, a reflective film 84 is provided which, in cooperation with the interference filter 74 on the surface 70, achieves a multi-reflection step transmission within the optical element. However, in accordance with the preferred embodiment, the reflective film 84 also forms a narrow band filter at each reflective location. Thus, each reflection location on the optical element surface 82 is an additional output end at which an optical signal associated with an additional channel is transmitted to an associated lens arrangement and fiber carrier line. In particular, a reflective film 84, also preferably a multi-cavity interference filter of continuously variable thickness, and preferably a linear continuously variable interference filter, transmits the optical signal wavelengths of channel 2 at output 86 and reflects the remaining wavelengths. Similarly, at output 88, reflective film 84 transmits the optical signal of channel 4 and reflects the remaining wavelengths again at that location. Output 90 is transparent to the optical signal of channel 6 and finally output 92 is transparent to the optical signal of channel 8.

Those skilled in the art will appreciate that the optical multiplexer shown in fig. 3 may provide efficient and compact multiplexing and demultiplexing functions. For collimated light having a divergence of 0.15 and incident on optical port 68 at an angle of about 12, the optical element can advantageously be made of fused silica, and has a width of about 10.361 mm. The linear spacing of the output ends on each of the

surfaces

70 and 82 is preferably about 3.067 mm, resulting in a total linear length of the optical element of about 15-20 mm. Generally, in the type of device discussed herein, it is desirable to have the angle of incidence or tilt of light through the optical port small (angle 0, light perpendicular to the surface of the optical element) (measuring the angle of light outside the optical element). Smaller angles of incidence may reduce polarization dependent effects. The adverse effect of collimated light divergence on filter performance can also be reduced, since a smaller angle of incidence will result in closer spacing of the reflecting points within the optical element and a shorter path for light to travel. Typically, the angle of incidence is less than 30 °, preferably 4 ° to 15 °, more preferably 6 ° to 10 °, and most preferably about 8 °.

FIG. 4 illustrates another preferred embodiment in which the reflective film on the second surface 82 of the optical element 66 includes a plurality of discrete cells 120-126. Other features and elements are the same as the corresponding features and elements in the embodiment of fig. 3 and are labeled the same. The individual

reflective film units

120 and 126 may be deposited directly onto the optical element surface 82, for example by a sputtering process or the like, or onto a separate carrier substrate that is positioned and secured separately from the optical element.

The reflective element may be secured with an epoxy or other adhesive. Each reflective film may be a broadband reflective layer that operates in substantially the same manner as reflective film 34 in the embodiment of fig. 1. Alternatively, they may operate as a plurality of additional ports, i.e., bandpass filters or dichroic filters, substantially in accordance with the principles of the reflective film 84 of the optical multiplexer of FIG. 3.

Additional alternative embodiments will be apparent to those skilled in the art in view of the description herein, including, for example, optical multiplexers coated with two (or more) layers of solid optical substrate, one or both (or all) of which are provided with interference filters of continuously variable thickness, to form a plurality of ports on a single planar surface, as described above, which are then brought together to form an optical element.

FIGS. 5 and 6 show thin film stack configurations for a continuously variable thickness multi-cavity interference filter 22 in the preferred embodiment of FIG. 2. It is desirable to precisely control the thickness of each alternating layer (e.g., alternating layers made of niobium pentoxide and silicon dioxide) and the overall thickness of the thin film stack, preferably within 0.01% or 0.2 mm in a few square inches. In addition, the thin film should be deposited as a film with very low absorption and scattering and a nearly uniform bulk density to prevent water induced filter shifting. The ultra-narrow multi-cavity bandpass filter has the following excellent performance characteristics, including: temperature and environmental stability; a narrow bandwidth; high transmittance of the desired optical signal and high reflectance of other wavelengths; steep edges, i.e., high selective transmission (especially in designs using three or more cavities); as well as a relatively low cost and simple construction. As shown in fig. 5, the filter is a three-cavity filter in which one cavity, the "first cavity", is next to the glass substrate. The second chamber is directly superimposed on the first chamber, and the third chamber is directly superimposed on the second chamber and is generally exposed to the surrounding atmosphere. In fig. 6, the structure of the "first cavity" is further shown. A sequence of stacked films is deposited to form the first reflective layer, wherein the stacked films are preferably about 5-15 films of alternating high and low index materials. The first film next to the substrate surface is preferably a layer of higher refractive index material, followed by a layer of lower refractive index material, and so on. Each high index layer 90 is an odd multiple of a quarter-wave optical thickness (QWOT), preferably one-quarter or three-quarter wave, or other odd multiple of QWOT. The thickness of the low index layers 92 interleaved with the high index layers 90 is again a quarter wave optical thickness or other odd multiple of QWOT. For example, there may be about six sets of high and low index layers forming the bottom most dielectric reflective layer 94. Although the cavity isolation layer 96 is illustrated as a layer, it typically comprises 1-4 alternating thin films of high and low index materials, wherein each thin film has a thickness that is an even multiple of QWOT, i.e., an integer multiple of the half wavelength optical thickness. The second dielectric reflective layer is preferably substantially the same as the dielectric reflective layer 94 described above. The second and third chambers are in turn deposited directly on the first chamber and are preferably substantially identical in shape. As described above, the thickness of the interference filter layer varies along the length of the multiport surface of the optical element. Thus, the actual thickness of the QWOT varies along the multiport surface. Various alternative suitable thin film stack structures may be used and will be apparent to those skilled in the art in view of the description herein.

FIG. 7 shows another thin film stack in which the upper and lower

reflective layers

94 and 98 are the same as described above for the embodiment of FIGS. 5 and 6. The cavity isolation layer 97 is shown as being formed of four films, two high index films 97a alternating with two low index films 97 b. Each film has a thickness of 2 or one-half of the QWOT wavelength. Various other alternative suitable thin film stack structures may also be used and will be apparent to those skilled in the art in view of the description herein.

The preferred embodiments disclosed herein differ from existing systems in several key areas and are a significant improvement over existing systems. The ion gun, which directs the ionized reactant gas toward the substrate during deposition, acts to reduce the total reactant gas pressure, prevent poisoning and breakdown (arcing) targets, and the ionization increases the reactivity of the gas, improving the stoichiometric composition of the film. The energy provided by the ion source helps to densify and improve the film quality. The preferred embodiments also require that the lines are not disposed in the plane of the substrate and that the target or sputtering source be tilted to increase the coating rate. The preferred embodiments disclosed herein do not perform the "gas isolation process" indicated by bond between the substrate and the magnetron, but rely on a high speed pumping system to reduce the concentration of inert and reactive gases, thereby further reducing arcing. Higher pumping speeds allow for greater inert gas flow at the magnetron without increasing background pressure. This in turn allows for increased sputtering rates at correspondingly higher power levels. The coating rate is typically 3-6 angstroms/second and the travel distance is 30-35 inches. (the effective coating rate or throughput drops off as the square of the distance.) even the surprisingly high deposition rates here, due to the energy and reactivity of the ionized reactant gases, cause the film to react completely and to have a fully dense packed structure.

Variable thickness filter coatings on multiport surfaces can be prepared according to the following preferred embodiments. Variable thickness can be achieved by positioning the substrate to be coated in the vacuum chamber at an angle relative to the magnetron. The substrate may also be masked locally and/or intermittently. The masking or masking means may comprise a masking member located in the vacuum chamber between the multiport surface of the optical element (i.e. the surface to be coated) and the source of sputtering material of the magnetron. The masking member is preferably closer to the multiport surface than the target source material of the magnetron. For example, the screen or covering may take the form of a substantially flat screen member that is positioned less than 0.5 inches from the multiport surface. The flat masking member may be rotated, wherein the term "rotate" includes rotation about its axis or orbital movement substantially in the plane of the flat member. Typically, a plurality of optical elements are fabricated as a unitary coated substrate, which is cut or singulated after the coating process. Most commonly, the substrate being coated comprises a unitary optical glass disk that is spun in a direction opposite to the direction of rotation of the flat masking member. Alternatively, the optical element, which is still referred to as a disk substrate that will eventually be divided into a plurality of optical elements, is mounted in a vacuum chamber at a position laterally offset from the magnetron. The masking means may also be used to achieve the variable thickness required for the optical element, as will be discussed further below.

The preferred embodiments described below enable the preparation of high quality coatings on substrates using DC reactive magnetron sputtering systems instead of IBS, for example to form mirrors that can be used in fiber optic systems, ring laser gyroscopes, and the like. Such films have properties comparable to IBS films, i.e., they have an extremely high packing density, as well as a smooth surface and low scattering. For example, the total loss of a high reflectivity laser mirror fabricated in accordance with embodiments of the methods disclosed herein is substantially less than 0.01% or 100 ppm.

Fig. 8 and 9 illustrate the method and apparatus of the preferred embodiment. It should be understood that the substrate referred to herein is typically a flat optical glass disk or similar component, e.g., 8 inches, 20 inches, etc., in diameter. One or both surfaces of the disk are coated to form many (possibly hundreds) optical elements simultaneously. That is, the coated disk is cut into a number of individual optical elements, each having a variable thickness filter as described above on one surface to form the desired multiport surface. The housing 110 forms a vacuum chamber 111, and the vacuum chamber 111 contains a low-pressure magnetron device 112 and a planetary substrate holder 113 with a plurality of rotatable planetary turntables 114. Each planetary turntable 114 holds a substrate facing the magnetron device 112. In this embodiment, the distance between the top of the magnetron device 112 and the planet is 16 inches. The magnetron apparatus 112 is connected to a source of working gas 116 via line 117. In the present embodiment, the outer shroud 110 is shown as spherical with a radius of 48 inches, but other configurations are equally suitable.

The housing 110 has a lower sleeve 118 which opens into the vacuum chamber 111 and contains a high speed vacuum pump 120 with a door 121 located between the vacuum pump and the vacuum chamber 111. The vacuum pump is of course used to reduce and maintain the pressure in the vacuum chamber so that the pressure of the inert gas is very low, at 5 x 10^-5torr-2.0X 10^-4In the range of torr.

In this regard, the present invention is completely different from known prior art magnetron sputtering techniques and conventional ion beam techniques. The method is characterized in that the indoor pressure is extremely low, including the pressure of reaction gas is extremely low, and the pressure of inert gas is extremely low. (measured at the surface of the coated substrate) such as O₂、N₂The pressure of the reaction gas with NO is preferably 2.0X 10^-5torr-1.5X 10^-4In the range of torr, preferably 3X 10^-5torr-9X 10^-5In the range of torr. This advantageously reduces or avoids magnetron arcing and source poisoning by the reactant gases. In the preferred embodiment, inert gases such as argon, krypton and xenon are directed primarily to the magnetron. The pressure at which a sharp drop is established for the inert gas (measured at the substrate being coated) is preferably 5.0X 10^-5torr-2.0X 10^-4In the range of torr, but at 5X 10^-5torr-1.5X 10^-4More preferably in the torr range. Such a low chamber pressure provides a longer Mean Free Path (MFP) and, correspondingly, advantageously results in a longer throw without excessive collisions between the chamber gases and the sputtered material. Favorable coating uniformity can be achieved with longer throw, preferably greater than 12 inches, but more preferably 20 inches or more. The extremely low chamber pressure allows the use of a longer throw. That is, despite the use of this generally long range, advantageously higher coating deposition rates can be achieved with correspondingly higher magnetron power levels. By innovatively using extremely low chamber pressures, losses in film or coating quality that are normally thought to be caused by higher magnetron power levels and longer range can be avoided. Thus, the preferred embodiment of the present invention extends to several major operating conditions of IBS (e.g., operating in the same pressure range as described above), but using a DC magnetron sputtering system. This new system, based essentially on magnetron sputtering, increases the coating speed and, correspondingly, the cost and throughput of depositing high quality thin film coatings.

A typical high-speed vacuum pump in the present invention is a 16 inch cryopump or a 16 inch diffusion pump. The pumping speed of these pumps was 5000 liters for a 16 inch cryopumpAbout one second (nitrogen) and about 10000 liters per second for a 16 inch diffusion pump (see the Leybold product and vacuum technology reference manual, 1993). Larger pumps, such as 20 inch pumps, may be used with a pumping speed of 10000 liters/second (N) for cryogenic pumping₂) And 17500 liters/second (N) for diffusion pump pumping speed₂) (see the Varian vacuum catalog of products 1991-1992). The pumping speed referred to above is located at the throat of the pump.

The magnetron device 112 is aligned with a rotation axis (main center line 122) of the planetary substrate holder 113 and a holder for monitoring a reference point (wireless chip)123 in a longitudinal direction. In this embodiment, the range between the magnetron device and the planetary turntable is 16 inches. Each planetary turntable and its substrate rotate about their own centerline 124. Planetary mounts are conventional and need not be described further except for the data noted below. In this example, the planetary disks have a diameter of 15 inches and the substrates have a diameter of 15 inches or any dimension less than 15 inches, and each planetary disk centerline is 14 inches from centerline 122 to accommodate larger substrates. Larger planetary rotors, such as 24 inch planetary rotors, may be used with a corresponding increase in substrate size and throw, and further increases in throughput may be achieved. A masking means may be used which is preferably a substantially flat masking member, for example about 0.5 inches from the substrate. The masking member is gradually moved during the deposition process, i.e. while the magnetron is in operation, to obtain a film thickness having the desired variation, thereby creating a plurality of ports of different wavelengths.

The ion gun 126 is directed obliquely to the substrate holder 113, its output being indicated by the dashed line 127, and its input being in communication with the reactive gas mixture 128 via line 130. The ion gun was positioned such that its ion and mixed gas output covered the entire substrate holder 113, in this example the top of the ion gun was 20 inches from the planetary turntable. The main function of the ion gun is twofold. The first function is to modify and improve film properties in a manner conceptually similar to U.S. Pat. No. 4,793,908 to Scott et al. The second function may be more important, its role being to remain lowThe background pressure of the reaction gas. The reaction gas is ionized by an ion gun and directed toward the substrate. The momentum of the reactant gas then directs it only toward the substrate (rather than toward the magnetron where it arcs and slows it down). The small amount of gas that diffuses towards the magnetron does not have a significant effect on its operation. The pressure of the reaction gas is generally 2X 10^-5torr-1.5X 10^-4In the range of torr, preferably in the range of 3X 10^-5torr-9X 10^-4In torr.

A suitable hot cathode pressure gauge 131 may also be connected to the vacuum chamber 111 to measure the pressure within the vacuum chamber. The vacuum chamber is also provided with a flap 132 which is pivotable about a handle 133 to block the output of the magnetron arrangement 112 as indicated by the dotted line 134. The handle 133 is connected to the platform 135 and a means for swinging the handle (not shown) in any suitable manner. The baffle is used in a pre-sputter source to remove contaminants from the target that condense on the surface of the target during idle times between deposition of a layer onto the substrate and deposition of a layer.

As shown in fig. 9, the magnetron apparatus 112 includes a target holder 136 having a cavity 137 bounded by a wall 138 and a target material 140. Centrally located within the cavity is a conventional magnet block 141 which is water cooled by a circulating flow of water into and out of the cavity 136 through

conduits

142 and 143. The metal target material 140 clamped by the fixture is also cooled with water. Line 144, which is slightly spaced from fixture 136 and sealed with insulator 145, is connected to the source of working gas 116 (fig. 8) through conduit 117, allowing the gas to flow completely around the top of the fixture and over the metal target material 140. The opening 145 of the line 144 is sized substantially the same as the size of the metal target material to emit the working gas of the sputtered target material as indicated by line 134. Magnetrons are commercially available from Material Sciences of Boulder, colorado, and are typically 6-8 inches in diameter with strong magnet strength.

It would also represent a significant advance in the art of the present invention when the present invention achieves the production of high quality thin film coatings by magnetron sputtering without the limitations of IBS or other known techniques.

The above dimensions and pressures of the present invention also show a great difference between the present invention and the prior art, and the relevant data are: a throw of 16 inches, a diameter of the planetary turntable of 15 inches, a diameter of the substrate of 15 inches or less, a distance of 20 inches from the top of the ion gun to the planetary turntable, and a very low pressure of the reaction gas of 2X 10^-5torr-1.5X 10^-4In the torr range, while the very low pressure of the inert gas is 5X 10^-5torr-2X 10^-4The range of torr.

The throughput of the preferred embodiment of the present invention in manufacturing laser quality mirrors was then compared to the throughput of a typical IBS system:

	the invention	IBS
	the invention	IBS	Coating rate	2-5 angstroms/second	2-1 angstrom/sec
Area of substrate	800-²	50-100 inches²	Coating rate	2-5 angstroms/second	2-1 angstrom/sec
Area of substrate	800-²	50-100 inches²	(5 Total area of turntable)

As can be seen from the above data, the throughput of the present invention is 20-120 times faster than that of the general IBS system. Coating throughput is a function of coating rate and substrate area.

In addition, the process of the present invention is easily scaled up to larger equipment sizes. All of the above dimensions are easily increased at least by a factor of 2 to enable low-loss coating of optical substrates having a diameter of 30 inches or more with a laser having good uniformity. Scaling up is a simple linear problem. Larger systems use larger magnetrons and more working gas (e.g., argon). A corresponding increase in vacuum pump is required to accommodate the larger vacuum chamber and increased gas flow in the process.

It is therefore apparent that the present invention is capable of producing, for example, laser quality mirrors that are many times larger in diameter than the known sizes produced by current IBS systems.

The throw of the preferred embodiment of the invention is 16 inches long, preferably 20 inches or more, and the chamber pressure is low, which allows two or more materials to be deposited simultaneously to form a high quality optical film composed of a mixture of materials. As an example of multiple sources, FIG. 10 shows two sources, a magnetron apparatus 112 and a magnetron apparatus 112a, within a vacuum chamber. (for ease of description, subscripts are used for additional sources and all other reference numbers are the same as in FIG. 8.)

By controlling the power level of each source, which is effective to control the deposition rate, coatings having selected refractive indices can be formed as mixtures of two or more material compositions. The mixture may be uniform throughout the coating to form a film having a selected refractive index, or non-uniform, such that the composition of the coating, and thus the refractive index, varies throughout the film. One common form of non-uniform film is known as a "rugate" filter, whose refractive index varies in a sinusoidal manner, which has the effect of forming a narrow notch reflector.

In order to keep such multi-source systems at low pressure, the pumping speed must be increased approximately twice for two simultaneously deposited sources, or N times for N sources. With the benefit of the present invention, it will be a simple operation for those skilled in the art to increase the pumping speed, which typically includes increasing the size of the pump, or adding more pumps to the vacuum chamber. In practice, however, since the rates provided by the sources are additive, it is not necessary to drive two simultaneously operating sources with the level used by a single source in order to maintain the coating rate, and thus the source size can be reduced and less gas used.

Another device that may be used in the present invention is the arc reduction electronic device (arc reduction electronic device) sold under the trademark SPARC-LE by Advanced Energy of Boulder, Corona. In FIG. 8, the SPARC-LE is shown connected to the magnetron apparatus 112 by a conductive line 147, the SPARC-LE having its own DC power supply 148. Similarly, as shown in fig. 9, the SPARC-LE is connected to two

magnetron devices

112 and 112 a. This device facilitates arc extinction, but it is not necessary in the method and apparatus of the present invention.

As can be seen from the above description, the magnetron system operates at a very low pressure. The chamber pressure of the inert gas will be a function of the magnetron pressure. Most importantly in the present invention, as shown in FIG. 9, is the region 150(A + O) of lower total pressure₂) Is generally at a much lower pressure than the high argon pressure region 152.

The pressure in the vacuum chamber can be simulated using the well-known pressure-flow equation (see the 1993 Leybold product & technical reference manual, pages 18-5):

P_{vacuum chamber}=Flow_Ar/C_P

P_Magnetron=Flow_Ar/C_M+P_{Vacuum chamber}

Wherein,

P_{vacuum chamber}Is the pressure in the vacuum chamber;

Flow_Aris the flow rate of argon flowing into the vacuum chamber through the magnetron;

C_Pis the conductance of the high vacuum pump (vacuum chamber pumping speed);

P_magnetronIs a magnetronPressure of (2);

C_Mis the conductance due to the holding of the gas by the magnetron (holding efficiency of the magnetron).

Instead of the items, the chamber pressure can be written as:

P_{vacuum chamber}＝P_Magnetron/(C_P/C_M+1)。

This is an important relationship because it indicates that the pressure in the vacuum chamber depends on the pumping speed (C) of the vacuum chamber_P). It also means that if the pumping speed of the vacuum chamber is low, the pressure in the vacuum chamber will be approximately equal to the pressure in the magnetron. This type of system with a low pumping speed is known from the prior art, in which a throttle valve arrangement is placed in front of the pump, thereby reducing the pumping speed. See, thin film processing, pp 156, published in 1978 by Vossent and kern, new york academic press. However, if the pumping speed in the vacuum chamber is high, as taught by the present invention, the chamber pressure will be lower than the magnetron pressure.

Using the above equations, the chamber pressure can be determined for substantially any new chamber for which the pumping speed is known, as shown in figures 11 and 12. As the shown figures clearly demonstrate, any suitable desired pressure can be obtained by increasing the pumping speed of the vacuum chamber. If the pressure of the inert working gas in the magnetron is reduced, which is possible with some types of magnetrons, the overall pressure curve will be correspondingly reduced. This can be seen by comparing the two pressure curves of fig. 11 and 12. For the pressure curve of FIG. 11, the magnetron pressure is 0.7 microns, the conductance of the magnetron device (C)_M) It was 3000 l/sec. Whereas for the pressure curve of fig. 12, the magnetron pressure is 0.4 microns, the conductance of the magnetron device (C)_M) It was 3000 l/sec. The pumping speeds shown on the abscissa can be achieved by, for example, operating a commonly used 20-inch diffusion pump at 17500 l/sec, and a 32-inch diffusion pump at 32000 l/sec.

Figures 13 and 14 show another preferred embodiment which uses a physical mask between the source and the substrate to control and customize the thickness of the filter coating. In other respects, it can be seen that the apparatus of fig. 13 and 14 corresponds to the embodiment described above. Specifically, the physical shield 150 is placed between the magnetron apparatus 112 and the substrate surface 115. The degree of shielding varies with distance from the center of the substrate such that at a certain radial distance the resulting filter is tuned to a first specific wavelength, and at a second radial distance from the center the filter is tuned to another wavelength. The screen may be fixed or may be movable, for example, spinning or rotating. For example, in general, the screen may be rotated about an axis common to the substrate surface, although other rotation means may be used and will be apparent to those skilled in the art after understanding the benefits of the present invention. Generally, it is desirable to place the substrate as close to the substrate surface 115 as possible, preferably less than 0.5 inches, more preferably 0.25 inches, but most preferably about 0.125 inches, for example. Preferably, the wobble or oscillation of any rotation or other movement of the membrane is zero or close to zero, preferably less than 0.001 inches. In general, the looser the tolerance specification or the wider the bandwidth of the filter, the greater the degree of swing the filter can withstand. The screen is preferably rotated or spun at high speed, preferably several hundred revolutions per layer, for example, about 50-100 revolutions per minute.

From the foregoing discussion, it will be apparent that various additions and modifications may be made to the optical multiplexer as described herein without departing from the true scope and spirit of the present invention. It is intended that all such variations and additions be covered by the following claims.

Claims

1. An optical multiplexer comprising an optical element having an optical port through which collimated light of multiple wavelengths can be transmitted; a continuously variable thickness interference filter extending over a multiport surface of said optical element and defining a plurality of ports spaced from one another along said multiport surface, said continuously variable thickness interference filter transmitting different wavelength subranges of said multiple wavelength collimated light and reflecting other wavelengths at each of said plurality of ports; and means for transmitting the optical ladder from one of the multiple ports to another port along a multi-drop transmission path.

2. The optical multiplexing device in accordance with claim 1 wherein the interference filter is continuously variable.

3. The optical multiplexing device in accordance with claim 1 wherein the interference filter is linearly continuously variable.

4. The optical multiplexing device in accordance with claim 1 wherein the means for cascading light comprises a reflective coating on the second surface of the optical block.

5. The optical multiplexing device in accordance with claim 4 wherein the second surface of the optical block is spaced from and substantially parallel to the multiport surface.

6. The optical multiplexing device in accordance with claim 4 wherein the reflective coating is continuous over the second surface at least coextensive with the multi-point transmission path.

7. The optical multiplexing device in accordance with claim 6 wherein the reflective coating is a broadband highly reflective film coating that reflects substantially uniformly all of the sub-ranges of the multiple wavelength collimated light.

8. The optical multiplexing device in accordance with claim 6 wherein the reflective coating forms a plurality of additional ports spaced from one another along the second surface, the reflective coating transmitting a different wavelength subrange of the multiple wavelength collimated light and reflecting other wavelengths at each of the plurality of additional ports.

9. The optical multiplexing device in accordance with claim 4 wherein the reflective coating comprises a plurality of discrete reflective film elements spaced from one another along the second surface.

10. The optical multiplexing device in accordance with claim 4 wherein the means for cascading light further comprises means for directing multiple wavelength collimated light into the optical block through the optical port at an angle between 4 ° and 15 ° relative to the multiport surface.

11. The optical multiplexing device in accordance with claim 1 wherein each of the plurality of ports has an associated lens means for focusing collimated light transmitted by one of the plurality of ports.

12. The optical multiplexing device in accordance with claim 11 wherein the lens means comprises a GRIN lens in communication with an optical fiber.

13. The optical multiplexing device in accordance with claim 1 wherein the optical block comprises a solid block of material substantially transparent to the multiple wavelength collimated light and selected from the group consisting of glass and fused silica, the continuously variable thickness interference filter being located on an outside surface of the solid block.

14. The optical multiplexing device in accordance with claim 1 wherein the optical block comprises a closed cavity.

15. The optical multiplexing device in accordance with claim 1 wherein the optical block is substantially linear and the optical ports are located on a front surface of the optical block opposite and parallel to the multiport surface of the optical block.

16. The optical multiplexing device in accordance with claim 15 wherein (a) the means for cascading light comprises a reflective film coating on the front surface that does not extend over the optical ports; (b) at least 8 of said plurality of ports, each port being transparent to a discrete wavelength subrange separated from the wavelength subrange of an adjacent port of said plurality of ports by about 2 nanometers; (c) collimated light passes through the optical port at an angle of about 6-10 ° relative to the plane of the front surface; and (d) the plurality of ports are linearly spaced from one another along the multi-port surface.

17. The optical multiplexing device in accordance with claim 16 wherein the reflective film on the front surface of the optical block is a broadband highly reflective film coating.

18. The optical multiplexing device in accordance with claim 15 wherein the means for cascading light comprises a reflective film on the front surface that does not extend over the optical ports, the reflective film coating being a second continuously variable thickness interference filter extending over the front surface of the optical block forming a plurality of additional ports, the second interference filter transmitting a different sub-range of wavelengths at each of the plurality of additional ports but reflecting other wavelengths of the multiple wavelength collimated light.

19. The optical multiplexing device in accordance with claim 18 wherein there are at least 4 of the plurality of ports and at least 4 of the plurality of additional ports.

20. The optical multiplexing device in accordance with claim 1 wherein the continuously variable thickness interference filter forms an all-dielectric narrow bandpass filter at each of the plurality of ports.

21. The optical multiplexing device in accordance with claim 1 wherein the continuously variable thickness interference filter is a multi-cavity interference filter.

22. The optical multiplexing device in accordance with claim 21 wherein the continuously variable thickness interference filter comprises a thin film stack forming at least three interference cavities.

23. The optical multiplexing device in accordance with claim 1 wherein the multi-cavity interference filter of continuously variable thickness comprises a stack of thin films of alternating niobium pentoxide and silicon dioxide thin films.

24. A method for manufacturing an optical multiplexer comprising an optical element, said optical element having:

an optical port transparent to multiple wavelengths of collimated light;

a continuously variable thickness interference filter extending over a multiport surface of said optical element and defining a plurality of ports spaced from one another along said multiport surface, said continuously variable thickness interference filter transmitting a different wavelength subrange of said multiple wavelength collimated light at each of said plurality of ports but reflecting other wavelengths; and

means for transmitting the optical ladder from one of the multiple ports to another port along a multi-drop transmission path.

The method comprises the following steps:

fixing said optical element in a vacuum chamber having magnetron means and source means for sputtering particles, said multiport surface of said optical element facing said source means at a longer range from the source means;

operating the magnetron apparatus to sputter particles from the source apparatus, coating the multiport surface, including directing an inert gas at a closed pressure in the vicinity of the source apparatus;

rapidly drawing and evacuating inert gas from said vacuum chamber with a high-speed high-vacuum pump; and is

Directing ionized reactive gas toward the multiport surface for reactive coating to obtain the interference filter as a low loss optical coating.

25. The method for fabricating an optical multiplexing device in accordance with claim 24 wherein the longer range between the source and the multiport surface is at least 16 inches.

26. The method for fabricating an optical multiplexing device in accordance with claim 24 wherein the inert gas pressure in the vacuum chamber is maintained at less than 2.0 x 10^-4Sum of torr greater than 5X 10^-5In the range of torr.

27. The method for fabricating an optical multiplexing device in accordance with claim 24 wherein the masking means is secured in the vacuum chamber between the multiport surface and the source means.

28. The method for fabricating an optical multiplexing device in accordance with claim 27 wherein the masking means is substantially closer to the multiport surface than the source means.

29. The method for fabricating an optical multiplexing device in accordance with claim 28 wherein the masking means comprises a flat masking member that is less than 0.5 inches from the multiport surface.

30. The method for fabricating an optical multiplexing device in accordance with claim 29 further comprising the step of rotating the flat masking member while operating the magnetron apparatus.

31. The method for fabricating an optical multiplexing device in accordance with claim 30 wherein the optical element and the flat masking member rotate in opposite rotational directions.

32. The method for fabricating an optical multiplexing device in accordance with claim 24 wherein the optical block is mounted in the vacuum chamber at a position laterally offset from the magnetron means.

33. The method for fabricating an optical multiplexing device in accordance with claim 24 wherein the magnetron means comprises a magnetron and a shield partially obscuring the magnetron while permitting sputtered particles from the source means to diffuse to impinge on the multiport surface while preventing diffusion of inert gas from the source means.

34. The method for fabricating an optical multiplexing device in accordance with claim 24 further comprising the step of rotating the multiport surface relative to the vacuum chamber.