WO2022254677A1

WO2022254677A1 - Processing method and processing apparatus for optical device

Info

Publication number: WO2022254677A1
Application number: PCT/JP2021/021268
Authority: WO
Inventors: 雅太田; 慶太山口; 賢哉鈴木; 摂森脇
Original assignee: 日本電信電話株式会社
Priority date: 2021-06-03
Filing date: 2021-06-03
Publication date: 2022-12-08
Also published as: JPWO2022254677A1

Abstract

Provided is a processing method for an optical device, the method allowing the thickness of a film in a minute region on a wafer surface to be accurately processed. The processing method for an optical device in which a film with a desired film thickness is processed by scanning along an arbitrary coordinate axis on a wafer includes; a step for calculating, on the basis of distribution of the desired film thickness, scanning information that is processing information for each scanning when processing is performed along a scanning axis; a step for converting, on the basis of the scanning information, a control signal for controlling a processing unit; and a step for processing, on the basis of the control signal, the wafer to have the distribution of the desired film thickness by scanning along the scanning axis by the processing unit.

Description

Optical device processing method and processing apparatus

The present invention relates to an optical device processing method and processing apparatus, and more particularly, to a light beam capable of processing a clad film and a core film formed on a wafer into a shape that makes the optical characteristics of the optical device uniform. The present invention relates to a device processing method and processing apparatus.

With the further increase in capacity of wavelength multiplexing optical communication, research and development of optical circuits such as optical wavelength multiplexing/demultiplexing circuits and optical switch circuits are being actively carried out. In many cases, these optical circuit components consist of embedded optical waveguides formed on a wafer, and include a plurality of optical signal paths with different optical path lengths and multiplexing/demultiplexing elements. Using the interference of light waves propagating in an optical circuit, functions such as multiplexing/demultiplexing and switching of light of multiple wavelengths are realized.

The interference characteristics of light waves depend on the difference in optical path length between optical signal paths. It is mainly determined by the height, width, etc. Therefore, the optical characteristics of an optical circuit fabricated on a wafer fluctuate depending on the wafer in-plane distribution of the optical constant of the optical waveguide material and the dimensional variation of the waveguide structure. In particular, the height of the waveguide core is determined by the film thickness during core deposition, and variations in optical characteristics of the fabricated optical device depend on the accuracy of the deposition process.

Conventionally, as a method for arbitrarily adjusting the distribution of the core film thickness formed by the flame deposition method or the like, for example, the manufacturing method described in Patent Document 1 has been known. When the second thin film is formed on the first thin film to form the waveguide core, the first thin film is processed into a coarse-dense pattern according to a predetermined film thickness distribution to form the second thin film. Low viscosity in the process to form the core. In order to accurately form the core film thickness, it is necessary to process the pattern shape of the first thin film formed on the substrate with high accuracy. However, conventionally, it has been difficult to process the shape of a very small area with a diameter of less than 1 mm on the surface of a wafer with high accuracy.

JP-A-2003-4966

An object of the present invention is to provide an optical device processing method and processing apparatus capable of microfabrication of the core film and the clad film in a minute area of the wafer surface, particularly the film thickness, with high accuracy.

In order to achieve such an object, one embodiment of the present invention is an optical device processing method for processing a film of a desired thickness by scanning along an arbitrary coordinate axis on a wafer, comprising: A step of calculating scan information, which is processing information for each scan when performing processing along the scan axis, based on the distribution of the desired film thickness; and a control of controlling a processing unit based on the scan information. and processing the wafer to the desired film thickness distribution by scanning the processing unit along the scan axis based on the control signal. do.

FIG. 1 is a diagram showing a schematic configuration of a processing apparatus according to one embodiment of the present invention; FIG. 2 is a diagram showing a first example of the processing unit of the processing device according to the present embodiment; FIG. 3 is a diagram showing a second example of the processing unit of the processing device according to the present embodiment; FIG. 4 is a diagram for explaining a wafer processing method in a processing unit; FIG. 5 is a diagram for explaining a method of calculating scan information; FIG. 6 is a diagram showing the scan speed distribution of the calculated scan information; FIG. 7 is a diagram showing the processing result of the wafer in the processing unit; FIG. 8 is a diagram showing a processing example of the Mach-Zehnder interferometer according to the first embodiment of the present invention; FIG. 9 is a diagram showing a processing example of an arrayed waveguide grating according to Example 2 of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

FIG. 1 shows a schematic configuration of a processing device according to one embodiment of the present invention. The processing apparatus 10 includes an analysis unit 11 that analyzes the waveguide distribution of an optical circuit fabricated on a wafer, a processing unit 12 that processes the wafer, and a controller that controls the processing unit 12 based on the analysis results of the analysis unit 11. a portion 13;

When mask information or waveguide distribution information for processing a wafer is input as an input signal, the analysis unit 11 analyzes this information and determines the scan axis. The scan axis is set in the direction in which the ratio of the optical axis direction is the largest in the waveguide formed in the optical circuit. Specifically, for each waveguide of an optical circuit fabricated on a wafer, the waveguide length is calculated for each predetermined optical axis direction. The optical axis direction is, for example, divided into 18 directions within a range of angles of 5 degrees from the direction parallel to the orientation flat of the wafer to the direction perpendicular to it. Next, the length of each waveguide is summed up for each section, and the direction of the longest waveguide length is taken as the scan axis.

The analysis unit 11 obtains processing profile information derived in advance, and calculates scan information including scan speed distribution and pitch information. The processing profile information is processing information obtained from the structure of an optical circuit to be fabricated on a wafer. is the shape information of That is, a desired height distribution of the waveguide cores formed according to the function of the optical circuit is defined. Further, the machining profile information includes characteristic information regarding machining characteristics in the machining unit 12 . For example, in the case of a microwave plasma etching processing unit, processing characteristics such as how much a film on a wafer can be etched are defined according to the scan speed and microwave power. The scan information is processing information for each scan when processing is performed along the determined scan axis in the processing unit 12 that processes the wafer. The shape information of the processing profile information is scanned along the scan axis, and the film thickness distribution of the core film is calculated for each scan. In order to obtain the calculated film thickness, the distribution of the scan speed for each scan, the pitch, etc. are defined according to the characteristic information of the processing profile information. For example, in the case of a microwave plasma etching method, if the height of the waveguide core is to be lowered (the core film is made thinner), the scanning speed of the capillary, which will be described later, is decreased, and the height of the waveguide core is increased. If you want to thicken the core film, increase the scan speed. Also, the pitch is narrowed where the shape change of the core is large, and the pitch is widened where the shape change is small.

The control unit 13 sends an instruction signal to the analysis unit 11 , inputs the scan information from the analysis unit 11 , converts it into a control signal for controlling the processing unit 12 , and outputs it to the processing unit 12 . For example, in the case of a microwave plasma etching processing unit, the control signal is for supply of raw material gas, control of the position of a capillary described later, control of a stage on which a wafer is placed, and control of microwave (RF signal) output. is a signal of

The processing unit 12 processes the wafer based on the input control signal. The wafer will be processed based on the scan information along the determined scan axis. The processing unit 12 is equipped with a sensor that scans the wafer surface, and a detected sensor signal is fed back to the control unit 13 .

Feedback control in the control unit 13 will be described. The control unit 13 calculates an adjustment amount (feedback amount) to be added to the control signal described above from the sensor signal from the processing unit 12 . A feedback amount is calculated by defining a feedback function and a feedback coefficient for each signal output from each sensor. For example, when the feedback signals from the two types of sensors are s ₁ and s ₂ respectively, and the feedback function is s ₁ as a quadratic function and s ₂ as a linear function, the feedback amount is predetermined by the coefficients A to D. be
f( _s1 , _s2 )= _As12 + ^Bs1 + _Cs2 ₊ D
is required as

FIG. 2 shows a first example of the processing section of the processing device according to this embodiment. The microwave plasma etching type processing unit 20 applies microwaves (RF signals) to a capillary that discharges a raw material gas to generate plasma, and irradiates the wafer with the plasma to etch the wafer. A wafer 26 is fixed on a stage 22 provided in a vacuum chamber 21 of a processing section 20 , and a plasma source gas is irradiated from a capillary 24 at the tip of a nozzle 23 . A microwave (RF) power supply 27 is connected between the nozzle 23 and an RF electrode 25 on the stage 22, and the material gas discharged from the capillary 24 is turned into plasma.

The raw material gas is supplied from the gas chambers 28a, b to the vacuum chamber 21 and the nozzle 23 via MFCs (Mass Flow Controllers) 29a, b, MFMs (Mass Flow Meters) 34a, b, and valves 30a, b. 31 is discharged. The type of raw material gas is fluorine-based, hydrogen-based or chlorine-based. Further, these raw material gases may be mixed with a combustion-supporting gas or an inert gas and supplied to the vacuum chamber.

The control unit 13 controls the RF power supply 27 that outputs microwaves (RF signals), the MFC 29 that controls the gas flow rate, and the valve 30, controls the nozzle 23 and the stage 22, and moves the wafer 26 along the scan axis. A film having a desired film thickness is processed by scanning along the . Further, the control unit 13 receives signals from the RF power meter 32, the vacuum gauge 33, and the MFM 34 as sensors, calculates the feedback amount as described above, and sends control signals to the RF power supply 27, MFC 29, and valve 30. Output to perform feedback control.

The processing unit 20 is a processing device in which the raw material gas is discharged from the capillary 24 of the nozzle 23, but is a suction type processing device that sucks the raw material gas from the nozzle and locally applies microwaves (RF signals). may be

FIG. 3 shows a second example of the processing section of the processing device according to this embodiment. The ion milling processing unit 40 irradiates the wafer with an Ar (argon) ion beam to etch the wafer. A wafer 46 is fixed to a wafer holder 45 on a stage 42 provided in a vacuum chamber 41 of a processing section 40 and is irradiated with an Ar ion beam from an ion gun 43 . The raw material gas is supplied from the gas chamber 48 to the vacuum chamber 41 via the MFC 49 and the valve 50 and discharged from the exhaust port 51 . Ar or Ga (gallium) or the like is mainly used as the ion beam species.

The control unit 13 controls the ion gas flow rate, controls the stage 42 on which the wafer 46 is mounted, and the ion gun 43, and scans the wafer 46 along the scan axis to process a film of a desired thickness. . The control unit 13 also receives a signal from a sensor (not shown), calculates the amount of feedback as described above, and performs feedback control.

A method of processing a wafer in the processing unit will be described with reference to FIG. FIG. 4(a) shows an optical circuit formed on a wafer. A plurality of optical circuits 26 a are formed on one wafer 26 . As described above, the scan axis is determined by analyzing the waveguide included in this optical circuit 26a. FIG. 4(b) shows how processing is performed for each scan along the x-axis direction, where the direction parallel to the orientation flat is the x-axis and the direction perpendicular to the orientation flat is the y-axis. For example, in the processing unit 20 of the microwave plasma etching method, a plasma material is discharged from a capillary 24 at the tip of a nozzle 23 according to scan information.

As described above, the scan axis is set in the direction in which the ratio of the optical axis direction of the waveguide is the largest, but depending on the function of the optical circuit, the scan axis may be the direction in which the ratio is the smallest. FIG. 4(c) shows how processing is performed for each scan along the y-axis direction. In addition, when the scan axis is not parallel to the orientation flat, as shown in FIGS. You may

A method of calculating scan information will be described with reference to FIG. FIG. 5(a) shows how the wafer 26 shown in FIG. 4(a) is processed for each scan along the x-axis direction as shown in FIG. 4(b). In one scan 26b, when the machining distribution in the direction (y-axis) perpendicular to the scan axis (x-axis) direction can be approximated by a Gaussian distribution, w is the machining width and A is the peak value of the machining depth. The processing distribution is

is required as FIG. 5B shows the obtained machining distribution, showing the machining depth in the y-axis direction from the central axis of the capillary 24 . Similarly, the next one scan 26c is processed in the -x-axis direction at intervals of pitch p. The peak value A and the processing width w are, for example, proportional to the scan speed and the power of the microwave in the case of a processing portion of the microwave plasma etching method. stipulated. Specifically, when the scan speed value is small, the residence time of the nozzle on the wafer is long, and the processing is deeper. Further, the larger the microwave power, the wider the area processed by the plasma, and the wider the processing width of the processing region.

In this way, reciprocating scanning is continued along the scanning axis. FIG. 5C shows the target processing amount in the y-axis direction at the position of x=0 when the center of the wafer 26 is set to x=0 and y=0. distribution. The peak value A of the machining depth is calculated so that the superposition of machining distributions for one scan becomes the target machining distribution shown in FIG. Derive the scan speed distribution. For the derivation, the method of least squares, regression analysis, etc. are used.

FIG. 6 shows the scan speed distribution of the calculated scan information. When the center of the wafer 26 is set to x=0 and y=0, it is the scan speed at arbitrary positions a, b, and c on the x-axis. One black dot indicates the scan speed in one scan, and indicates the distribution of the scan speed in the y-axis direction at arbitrary positions a, b, and c on the x-axis.

FIG. 7 shows the processing result of the wafer in the processing section. This is an example in which a core film is formed on a substrate using a processing unit 20 of a microwave plasma etching method. The thick dashed line is the target machining amount in the y-axis direction at the position of x=0 when the center of the wafer 26 is x=0 and y=0. In this example, the core film thickness distribution is formed in the range of 2.0 to 2.5 μm. A thin line indicates the amount of processing for each scan, which is the amount of processing when scanning is performed at a pitch p=1 mm in the y-axis direction.

The circular mark is the processing result obtained by superimposing the processing distribution of one scan. It can be seen that the film thickness distribution of the target processing amount is processed with high accuracy.

FIG. 8 shows a processing example of the Mach-Zehnder interferometer according to Example 1 of the present invention. A Mach-Zehnder interferometer (MZI) 60 shown in FIG. 8A includes a first arm 63 and a first arm 63 having a path length longer than the first arm 63 by ΔL between two

directional couplers

61 and 62 . 2 arms 64 are connected. When the horizontal direction in FIG. 8A is taken as the x-axis, the length of the waveguide in the x-axis direction is the longest, so this direction is taken as the scan axis.

As described above, when multiple scans are sequentially superimposed along the scan axis, minute processing errors remain in the area where two different adjacent scans overlap. If the scanning axis is set in the direction of the optical axis of the waveguide, it is possible to suppress an increase in insertion loss and phase error caused by unevenness in the effective refractive index caused by this processing error.

FIG. 8(b) shows the processing result when the target processing amount is 10 μm. The film thickness variation in the y-axis direction is suppressed to ±0.1 μm or less. Since the waveguide width of the MZI 60 is on the micro-order or nano-order, it is made sufficiently narrower than the scan pitch, and the scan axis is set in the direction in which the ratio of the optical axis direction of the waveguide is the largest. can be minimized.

According to Example 1, the height of the waveguide core (film thickness of the core film) is in the range of 100 nm to 20 μm, and the width of the waveguide core is in the range of 100 nm to 100 μm. Circuit waveguides can be made.

FIG. 9 shows a processing example of an arrayed waveguide grating according to Example 2 of the present invention. Arrayed waveguide grating (AWG) 70 shown in FIG. are connected by a plurality of arrayed waveguides 75 having different path lengths by ΔL. The AWG 70 is used for optical wavelength filters and the like.

FIG. 9(b) is a cross-sectional view taken along line BB' of FIG. 9(a), in which a plurality of cores 75a to 75n of an arrayed waveguide 75 are embedded in a clad layer 77 formed on a substrate 76. ing. A central wavelength λ ₀ , which is a typical optical characteristic of the AWG 70, is obtained by the following formula.

Here, _narray is the equivalent refractive index of the arrayed waveguide 75, .DELTA.L is the path length difference between the arrayed waveguides, and m is the diffraction order. Also, the distribution of n _array has the following relationship.

Here, n _core is the core refractive index, n _clad is the clad refractive index, w is the waveguide width, and T(x, y) is the core film thickness distribution.

From the above equation, by making the core film thickness distribution T(x, y) between the arrayed waveguides 75 uniform, it is possible to suppress variations in the characteristics of the optical device of the AWG 70 . That is, when a plurality of waveguides with different path lengths are distributed over a wide area like the AWG 70, it is better to make the film thickness uniform in the array direction rather than in the optical axis direction of the waveguides. better characteristics. Therefore, in order to ensure the uniformity of the equivalent refractive index between waveguides, the scan axis is set in the y-axis direction where the ratio of the waveguides in the optical axis direction is the smallest.

Claims

An optical device processing method for processing a film of a desired film thickness by scanning along an arbitrary coordinate axis on a wafer, comprising:
a step of calculating scan information, which is processing information for each scan when performing processing along the scan axis, based on the distribution of the desired film thickness;
a step of converting the scan information into a control signal for controlling the processing unit;
and processing the wafer into the desired film thickness distribution by scanning the processing unit along the scan axis based on the control signal.
2. The optical device according to claim 1, wherein the scan axis is set in a direction having the largest or smallest ratio of optical axis directions in a waveguide of an optical circuit formed on the wafer. processing method.
The processing unit uses a microwave plasma etching method, the scan information includes a scan speed distribution and a pitch, and the control signal includes the output of a microwave power supply, gas flow rate, and plasma source gas irradiation. 3. The method of processing an optical device according to claim 1, further comprising controlling a nozzle that holds the wafer and a stage on which the wafer is fixed.
The processing unit uses an ion milling method, the scan information includes a scan speed distribution and a pitch, and the control signal controls an ion gas flow rate, an ion gun, and a stage on which the wafer is fixed. 3. The method for processing an optical device according to claim 1 or 2.
An optical device processing apparatus for processing a film of a desired film thickness by scanning along an arbitrary coordinate axis on a wafer,
an analysis unit that calculates scan information, which is processing information for each scan when performing processing along the scan axis, based on the distribution of the desired film thickness;
a processing unit that processes the wafer into the desired film thickness distribution by scanning along the scan axis with a control signal converted based on the scan information. processing equipment.
6. The optical device according to claim 5, wherein the scan axis is set in a direction having the largest or smallest ratio of optical axis directions in the waveguide of the optical circuit formed on the wafer. processing equipment.
The processing unit uses a microwave plasma etching method, the scan information includes a scan speed distribution and a pitch, and the control signal includes the output of a microwave power supply, gas flow rate, and plasma source gas irradiation. 7. The optical device processing apparatus according to claim 5 or 6, wherein the nozzle for processing and the stage on which the wafer is fixed are controlled.
The processing unit uses an ion milling method, the scan information includes a scan speed distribution and a pitch, and the control signal controls an ion gas flow rate, an ion gun, and a stage on which the wafer is fixed. 7. The optical device processing apparatus according to claim 5 or 6.