US20180169826A1

US20180169826A1 - Polishing extremely thin silica sheets and polished sheets

Info

Publication number: US20180169826A1
Application number: US15/846,778
Authority: US
Inventors: Jonas Bankaitis; Karan Mehrotra
Original assignee: Corning Inc
Current assignee: Corning Inc
Priority date: 2016-12-21
Filing date: 2017-12-19
Publication date: 2018-06-21

Abstract

A method of manufacturing a sheet of fused silica includes polishing a sheet of fused silica having a thickness of less than 500 μm and a major face surface area of at least 6π square inches. The polishing is performed by removing less than 100 micrometers depth of material of a major surface of the sheet, such as by bonding the sheet to a substrate, polishing a first major side of the sheet, debonding the sheet from the substrate, flipping the sheet, bonding the flipped sheet to the substrate or a new substrate, and polishing a second major side of the sheet. Prior to polishing, the sheet has a peak-to-valley waviness of at least 1 micrometer, but after polishing has peak-to-valley waviness less than 500 nanometers.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

This application claims the benefit of priority of U.S. Provisional Application Ser. No. 62/446,670 filed on Jan. 16, 2017 which claims the benefit of priority of U.S. Provisional Application Ser. No. 62/437,404 filed on Dec. 21, 2016 the contents of which are relied upon and incorporated herein by reference in their entirety as if fully set forth below.

BACKGROUND

Aspects of the technology disclosed herein relate to methods and equipment for polishing extremely thin, fragile sheets of fused silica, such as having thicknesses of less than 500 μm for example, that have large surfaces areas, such as at least 6π in²for example.
Thin fused silica sheets may be useful as semiconductor substrates, or for other purposes. Today, fused silica is typically made in large boules, and then sliced into sheets. These sheets may then be lapped and polished to achieve high quality surfaces and planarization. However, limitations on conventional polishing techniques constrain the ability to go thinner while maintaining wide surface areas, high surface quality, and planarity, as now explained.
The boule cutting process is composed of a plurality of steps and is generally complex and expensive. The boule cutting process typically employs cerium oxide chemical-mechanical polishing (CMP) almost exclusively. That is, in the boule cutting process, conventional finishing of semiconductor grade glass wafers is generally conducted via conventional abrasive lapping to remove wire saw damage and terminated with cerium oxide polishing to effect removal of corresponding lapping damage and establish the finished surface. Cerium oxide polishing is a form of CMP, for example, and is characterized by a corresponding planarization efficiency. Planarization efficiency (PE) in CMP is defined as the degree to which protruding regions of a workpiece are polished or reduced while recessed regions are protected.
The boule cutting process may be limited in several important aspects, particularly where finished wafer thickness capability and material removal budget are concerned. In the case of the former, the boule cutting process is generally limited to production of highly polished glass wafers of no less than 0.5 mm in final thickness, especially for specifications requiring high surface quality and large area with planarity. There are several reasons for this limitation.
It is Applicant's understanding, a main reason the boule cutting process and other processes are incapable of producing very thin, highly polished glass wafers is rooted in a planarization/surface quality conundrum that accompanies flat substrate finishing. For example, in processing sequences such as that employed in the boule cutting process, finished wafer planarity is established in the primary lapping step. Planetary double side lapping is highly effective in accomplishing the following processing objectives simultaneously removing some damage incurred during initial boule slicing, reducing sliced wafer thickness to a prescribed proximity to the finished product thickness target to the degree possible without fracturing due to thinness and fragility of the sheet, and establishing a degree of finished wafer base planarity. In accomplishing the above, however, the lapping process may impose damage commensurate with contact of substrate with a thin film of lapping compound dynamically pressed between incompressible steel (martensitic) lapping plates. A challenge for the remainder of the wafering process, therefore, is to establish the high quality surface finish on the substrate with minimal degradation to planarity established during lapping.
Chemical-mechanical polishing (CMP) is widely regarded as a method of choice for generating requisite surface quality on glass or semiconductor substrate material surfaces planarized in the primary, and possibly secondary and even tertiary, lapping process steps. Furthermore, CMP carried out using the same planetary polishing technology as that used in lapping (double side polishing) may be preferred because the degree of planarization achieved in lapping is only minimally degraded. However, this is where a limitation of CMP via double-side polishing (DSP) becomes apparent: very thin wafers, such as those less than 0.5 mm are difficult to handle as they are not sufficiently mechanically rigid and are often not mechanically strong enough to withstand forces imposed upon them during free (unrestrained) planetary polish processing, because as the finished wafer thickness decreases so must the DSP carrier thickness decrease to compensate—conventional lapping and polishing processes employ polishing equipment (double side lappers/polishers) which require carriers of thicknesses less than that of the wafers they must house/restrain during polishing. The carrier thickness must be thinner than the substrate by a minimum of the intended material removal. As the finished wafer thickness shrinks so must the carrier thickness, thereby creating a need for very thin carriers that are sufficiently mechanically rigid to withstand torsional forces imposed upon them by the driving pin rings on planetary DSP equipment and such carriers may not exist in commercially available forms.
Applicants invented a process to produce unique sheets of fused silica that are extremely thin, such as less than 0.5 mm, where the sheets have a relatively large surface area, such as at least 6π in²(English units are used because substrate discs are often identified by diameter in inches, however the present substrates may be non-round in some embodiments), and where the sheets are highly planar and have polished surface quality.

SUMMARY

Applicants have discovered a way to produced polished, extremely thin sheets of fused silica that may be useful as semiconductor substrates or for other applications. Instead of starting from a boule and cutting, lapping, and polishing down the silica sheets, Applicants start with extremely thin silica sheets made by a thin-soot sheet deposition and sintering process, where the sintered silica is subsequently heat pressed to establish planarity (low-warp), and then only a small amount of material (removal to a depth on the order of microns, instead of hundreds of microns) on top and/or bottom of the sheet is removed. This unique process allows the thin sheets to have polished surfaces without losing planar attributes of the sheets.
Aspects of the present disclosure relate generally to a method of manufacturing a sheet of fused silica, such as at least 99.99 wt % silicon dioxide. The method includes steps of polishing a sheet of fused silica having a thickness of less than 500 μm by removing less than 100 micrometers depth of material of a major surface of the sheet, such as less than 50 micrometers. The sheet of fused silica may then have a thickness of less than 300 micrometers, a major face surface area of at least 6π square inches, a total thickness variation of less than 10 micrometers excluding edge rolloff, a peak-to-valley waviness of less than 500 nanometers, and a root mean square roughness over a square millimeter of major surface of less than 100 nanometers. Prior to polishing, the sheet may have peak-to-valley waviness of greater than 3 micrometers. The sheet may be made by depositing silica soot to form a sheet of silica soot and then sintering the sheet of silica soot to form the sheet of fused silica, prior to the polishing step.
In some such embodiments, the polishing step of the method further includes a step of bonding the sheet to a substrate and then polishing the major surface of the sheet, and may be performed without simultaneously polishing a second major surface of the sheet. In some embodiments, the polishing step further includes debonding the sheet from the substrate, flipping the sheet, bonding the flipped sheet to the substrate or a new substrate, and polishing the second major surface of the sheet. The method may further include a step of flattening the sheet of fused silica prior to the polishing step.
Other aspects of the present disclosure relate generally to a method of manufacturing a sheet of fused silica, which includes a step of shaping a sheet of fused silica having a thickness of less than 500 micrometers, a major face surface area of at least 6π square inches, and a peak-to-valley waviness of at least 1 micrometer. The shaping may be flattening of the sheet so that the sheet has a warp of less than 50 micrometers, or the shaping may include imparting curvature. The method may further include a step of polishing the sheet, after the shaping step. The polishing may, in part, be performed by bonding the sheet to a substrate, polishing a first major side of the sheet, debonding the sheet from the substrate, flipping the sheet, bonding the flipped sheet to the substrate or a new substrate, and polishing a second major side of the sheet. After completing the method, the sheet of fused silica may have a total thickness variation of less than 10 micrometers excluding edge rolloff and peak-to-valley waviness of less than 500 nanometers.
In some such embodiments, the shaping step may include heating the sheet to a temperature below melting temperature of the fused silica. In some embodiments, the shaping step may include applying a load, such as compression, tension, etc., to the sheet to displace at least a portion of the sheet. In some embodiments, the shaping step may include, after the heating step, a step of cooling the sheet under the load such that displacement is at least in part set. After the polishing step, the warp may still be less than 50 micrometers. According to an exemplary embodiment, polishing and shaping does not fracture the sheet.
Still other aspects of the present disclosure relate generally to a sheet of fused silica that has a thickness of less than 300 micrometers, a major face surface area of at least 6π square inches, a total thickness variation of less than 10 micrometers excluding edge rolloff, a peak-to-valley waviness of less than 500 nanometers, and a root mean square roughness over a square millimeter of major surface of less than 100 nanometers. In some embodiments, warp of the sheet is less than 50 micrometers. In some embodiments, the fused silica is amorphous and is at least 99.99 wt % silicon dioxide. In some embodiments, the peak-to-valley waviness is less than 100 nanometers, the total thickness variation is less than 5 micrometers, and/or the major face surface area is at least 8π square inches.
Additional features and advantages are set forth in the Detailed Description that follows, and in part will be readily apparent to those skilled in the art from the description or recognized by practicing the embodiments as described in the written description and claims hereof, as well as the appended drawings. It is to be understood that both the foregoing general description and the following Detailed Description are merely exemplary, and are intended to provide an overview or framework to understand the nature and character of the claims.

BRIEF DESCRIPTION OF THE FIGURES

The accompanying Figures are included to provide a further understanding, and are incorporated in and constitute a part of this specification. The drawings illustrate one or more embodiments, and together with the Detailed Description serve to explain principles and operations of the various embodiments. As such, the disclosure will become more fully understood from the following Detailed Description, taken in conjunction with the accompanying Figures, in which:

FIG. 1 is perspective view of a conceptual rendering of a manufacturing line for producing thin sheets of fused silica.

FIG. 2 is a side view of a warped sheet of fused silica, as may be produced with the manufacturing line of FIG. 1.

FIG. 3A-B are digital images of a fused silica sheet before and after using processes disclosed herein to flatten the sheet, according to an exemplary embodiment.

FIG. 4A is a perspective view of a three-dimensional surface profile of a sheet of fused silica showing peaks and valleys resulting from the manufacturing line of FIG. 1.

FIG. 4B is a plot of a two-dimensional surface profile of the surface of FIG. 4A.

FIG. 5 is a flow diagram of a process for polishing the thin, flattened sheets, such as the sheet of FIG. 3B.

FIG. 6A is a top view of a three-dimensional surface profile after polishing via the process of FIG. 5.

FIG. 6B is a plot of two-dimensional surface profiles of the surface of FIG. 6A.

FIG. 7 is a perspective view of a thin, polished sheet of fused silica, according to an exemplary embodiment.

DETAILED DESCRIPTION

Before turning to the following Detailed Description and Figures, which illustrate exemplary embodiments in detail, it should be understood that the present inventive technology is not limited to the details or methodology set forth in the Detailed Description or illustrated in the Figures. For example, as will be understood by those of ordinary skill in the art, features and attributes associated with embodiments shown in one of the Figures or described in the text relating to one of the embodiments may well be applied to other embodiments shown in another of the Figures or described elsewhere in the text.
Referring generally to the figures, various embodiments of a sintered silica glass sheet/material as well as related systems and methods are shown. In various embodiments, the system and method disclosed herein utilizes one or more glass soot generating device (e.g., a flame hydrolysis burner) that is directed or aimed to deliver a stream of glass soot particles on to a soot deposition device or surface forming a glass soot sheet.
The soot sheet may then be sintered using a laser or other heat source (e.g., high temperature furnace, tunnel kiln) forming a silica glass sheet. In general, the laser beam is directed onto the soot sheet such that the soot densifies forming a fully sintered or partially sintered silica glass sheet. In various embodiments, the configuration and/or operation of the glass soot generating device, the soot deposition surface and/or the sintering laser are configured to form a sintered glass sheet having very high surface smoothness as compared to some sintered silica glass sheets formed from sintered silica soot.
Further, the configuration and/or operation of the glass soot generating device, the soot deposition surface and/or the sintering laser are configured to form a sintered glass sheet having very low levels of certain contaminants (e.g., sodium (Na), surface hydroxyl groups, etc.) commonly found in silica materials formed using some other methods. Applicant has found that by using the laser sintering process and system discussed herein, sintered silica glass sheets can be provided with a high surface smoothness and low contaminant content.
In yet additional embodiments, sintered silica glass sheets discussed herein also include a high level of flatness (e.g., a low degree of warp, as discussed below), even in very thin sheets. In various embodiments, the high level of flatness is achieved through a flattening process in which a sintered silica glass sheet having a relatively high degree of warp is heated and then is flattened through application of a force, such as force provided between upper and lower silica plates.
In some embodiments, the flattened sheet of sintered silica glass may then be polished using a unique polishing process that deviates from standard wafer polishing by polishing each side of the sheet separately, using substrate-supported mechanical polishing to only a very shallow depth. The polishing approach allows the flatness and low warp to be maintained while micro-scale waviness on surfaces of the sheet are reduced to nano-scale or eliminated.
Referring to FIG. 1, a system and method for forming a high purity, high smoothness silica glass sheet is shown according to an exemplary embodiment. As shown in FIG. 1, system 10 includes a soot deposition device, shown as deposition drum 12, having an outer deposition surface 14. System 10 includes a soot generating device, shown as soot burner 16 (e.g., a flame hydrolysis burner), that directs a stream of glass soot particles 18 onto deposition surface 14 forming glass soot sheet 20.
As shown in FIG. 1, drum 12 rotates in the clockwise direction such that soot sheet 20 is advanced off of drum 12 in a processing direction indicated by the arrow 22 and advanced past sintering laser 24. In some embodiments, soot sheet 20 is in tension (e.g., axial tension) in the direction of arrow 22. In specific embodiments, soot sheet 20 is only in tension (e.g., axial tension) in the direction of arrow 22 such that tension is not applied widthwise across soot sheet 20. However, in at least some other embodiments, soot sheet 20 is tensioned in the widthwise direction. In some embodiments, tensioning in different directions is selected to control the bow or warp of the sintered soot sheet.
As will be explained in more detail below, sintering laser 24 generates a laser beam 26 toward soot sheet 20, and the energy from laser beam 26 sinters glass soot sheet into a partially or fully sintered glass sheet 28. As will be understood, the energy from sintering laser beam 26 causes the densification of glass soot sheet 20 into a partially or fully sintered glass sheet 28. Specifically, laser sintering of silica soot sheet 20 uses laser 24 to rapidly heat soot particles to temperatures above the soot melting point, and as a result of reflow of molten soot particles a fully dense, thin silica glass sheet 28 is formed.
In various embodiments, soot sheet 20 has a starting density between 0.2 g/cc to 0.8 g/cc, and silica glass sheet 28 is a fully sintered silica glass sheet having a density of about 2.2 g/cc (e.g., 2.2 g/cc plus or minus 1%). As will be explained in more detail below, in some embodiments, silica glass sheet 28 is a fully sintered silica glass sheet including voids or bubbles such that the density of the sheet is less than 2.2 g/cc. In various other embodiments, soot sheet 20 has a starting density between 0.2 g/cc to 0.8 g/cc, and silica glass sheet 28 is a partially sintered silica glass sheet having a density between 0.2 g/cc and 2.2 g/cc.
In various embodiments, sintered glass sheet 28 has length and width between 1 mm and 10 m, and in specific embodiments, at least one of the length and width of sintered glass sheet 28 is greater than 18 inches. It is believed that in various embodiments, system 10 allows for formation of sintered glass sheet 28 having length and/or width dimensions greater than the maximum dimensions of silica structures formed by other methods (e.g., silica boules which are typically limited to less than 18 inches in maximum dimension).
System 10 is configured to generate a soot sheet 20 having a smooth surface topology which translates into glass sheet 28 also having a smooth surface topology. In various embodiments, soot burner 16 is positioned a substantial distance from and/or at an angle relative to drum 12 such that soot streams 18 form a soot sheet 20 having a smooth upper surface. This positioning results in mixing of soot streams 18 prior to deposition onto surface 14. In specific embodiments, the outlet nozzles of soot burner 16 are positioned between 1 inch and 12 inches, specifically 1 inch to 4 inches, and more specifically about 2.25 inches, from deposition surface 14, and/or are positioned at a 30-45 degree angle relative to soot deposition surface 14. In specific embodiments, soot stream 18 can be directed to split above and below drum 12 with exhaust, and in other embodiments, soot stream 18 is directed only to one side of drum 12. In addition, the velocity of soot streams 18 leaving burner 16 may be relatively low facilitating even mixing of soot streams 18 prior to deposition onto surface 14. Further, burner 16 may include a plurality of outlet nozzles, and burner 16 may have a large number of small sized outlet nozzles acting to facilitate even mixing of soot streams 18 prior to deposition onto surface 14.
In addition, burner 16 may be configured to better mixing of constituents and soot within channels inside the burners such as via a venturi nozzle and flow guides that generate intermixing and eddies. In some embodiments, these structures may be formed via 3D printing.
In various embodiments, laser 24 is configured to further facilitate the formation of glass sheet 28 having smooth surfaces. For example in various embodiments, sintering laser 24 is configured to direct laser beam 26 toward soot sheet 20 forming a sintering zone 36. In the embodiment shown, sintering zone 36 extends the entire width of soot sheet 20. As will be discussed in more detail below, laser 24 may be configured to control laser beam 26 to form sintering zone 36 in various ways that results in a glass sheet 28 having smooth surfaces. In various embodiments, laser 24 is configured to generate a laser beam having an energy density that sinters soot sheet 20 at a rate that forms smooth surfaces.
In various embodiments, laser 24 generates a laser beam having an average energy density between 0.001 J/mm²and 10 J/mm², specifically 0.01 J/mm²and 10 J/mm², and more specifically between 0.03 J/mm2 and 3 J/mm2 during sintering. In some embodiments, laser 24 may be suited for sintering particularly thin soot sheets (e.g., less than 1000 μm, less than 500 μm, less than 200 μm, 100 μm, 50 μm, etc. thick), and in such embodiments, laser 24 generates a laser beam having an average energy density between 0.001 J/mm²and 0.01 J/mm².
In other embodiments, system 10 is configured such that relative movement between soot sheet 20 and laser 24 occurs at a speed that facilitates formation of glass sheet 28 with smooth surfaces. In general, the relative speed in the direction of arrow 22 is between 0.1 mm/s and 10 m/s. In various embodiments, the relative speed in the direction of arrow 22 is between 0.1 mm/s and 100 mm/s, specifically between 0.5 mm/s and 5 mm/s, and more specifically between 0.5 mm/s and 2 mm/s. In various embodiments, system 10 is a high speed sintering system having a relative speed in the direction of arrow 22 between 1 m/s and 10 m/s.
As shown in FIG. 1, in one embodiment, laser 24 utilizes dynamic beam shaping to form sintering zone 36. In this embodiment, laser beam 26 is rapidly scanned over soot sheet 20 generally in the direction of arrows 38. The rapid scanning of laser beam 26 emulates a line-shaped laser beam generally in the shape of sintering zone 36. In a specific embodiment, laser 24 utilizes a two-dimensional galvo scanner to scan laser beam 26 forming sintering zone 36. Using a two-dimensional galvo scanner, laser beam 26 can be rastered across the entire width of soot sheet 20 or across a specific subarea of soot sheet 20. In some embodiments, laser beam 26 is rastered as soot sheet 20 is translated in the direction of arrow 22. During the sintering process the rastering speed may vary depending on the desired sintering characteristics and surface features. In addition, the rastering pattern of laser beam 26 may be linear, sinusoidal, uni-directional, bidirectional, zig-zag, etc., in order to produce sheets with designed and selected flatness, density or other attributes.
In such embodiments, laser 24 may use galvo, polygonal, piezoelectric scanners and optical laser beam deflectors such as AODs (acousto-optical deflectors) to scan laser beam 26 to form sintering zone 36. In various embodiments, relative movement between soot sheet 20 and laser beam 26 may be accomplished via directing laser beam 26 with or without moving laser 24.
In a specific embodiment using a dynamic laser beam shaping to form sintering zone 36, a CO₂laser beam was scanned bi-directionally at a speed of 1500 mm/s. The CO₂laser beam has a Gaussian intensity profile with 1/e2 diameter of 4 mm. The step size of the bi-directional scan was 0.06 mm. At settings of scan length of 55 mm and a laser power of 200 W, a soot sheet 20 of roughly 400 μm in thickness was sintered into a silica glass sheet 28 of ˜100 μm thickness. The effective sintering speed was ˜1.6 mm/s, and the sintering energy density was 0.65 J/mm². In other embodiments, as discussed below, the sintering laser is a CO laser.
In various embodiments, laser 24 can be a laser at a wavelength or pulse width such that there is enough absorption by the soot particles to cause sintering. The absorption can be linear or nonlinear. In a specific embodiment, laser 24 is a CO₂laser. In another embodiment, laser 24 may be a CO laser with a wavelength of around 5 μm. In such embodiments, a CO laser 24 can penetrate deeper into soot sheet 20, and thus a CO laser 24 may be used to sinter thicker soot sheets 20. In various embodiments, the penetration depth of a CO₂laser 24 in silica soot sheet 20 is less than 10 μm, while the penetration depth of the CO laser is ˜100 μm. In some embodiments, soot sheet 20 may be pre-heated from the backside and/or front side, for example, using a resistive heater, an IR lamp, etc., to further increase the depth of sintering formed via laser 24.
In some embodiments, system 10 is configured to maintain a constant sintering temperature during the laser sintering process. This can be achieved by adding temperature sensors along the sintering line. The temperature sensor data can be used to control the laser power in order to maintain constant sintering temperature. For example, a series of germanium or silicon detectors can be installed along the sintering line. The detector signals are read by a controller. The controller can process the signals and use the info to control the laser output power accordingly.
In contrast to some silica glass formation processes (e.g., boule formation processes), system 10 is configured to produce silica glass (e.g., fused and/or amorphous) sheet 28 having very high purity levels with very low thicknesses. In various embodiments, silica glass sheet 28 has a thicknesses (i.e., the dimension perpendicular to the major and minor surfaces) of less than 500 μm, of less than 300 μm, of less 200 μm and of less than 150 μm. Further, in various embodiments, silica glass sheet 28 is least 99.9 mole % silica (e.g. fused silica, amorphous silicon dioxide), and specifically at least 99.99 mole % silica. In other contemplated embodiments, boule formation processes may be used in conjunction with some of the inventive teachings of the present disclosure, such as polishing techniques disclosed herein.
In addition, silica glass sheet 28 is formed having very low levels of contaminant elements common in silica glass formed by other methods. In specific embodiments, silica glass sheet 28 has a total sodium (Na) content of less than 50 ppm. In various embodiments, the sodium content of silica glass sheet 28 is substantially consistent throughout sheet 28 such that the total sodium content is less than 50 ppm at all depths within silica glass sheet 28. This low total sodium content and the even sodium distribution is in contrast to some silica structures (e.g., silica boules) which have higher overall sodium content that varies at different depths within the boule. In various embodiments, it is believed that the low sodium content discussed herein provides glass sheet 28 with optical loss reduction, index of refraction uniformity and chemical purity/non-reactivity as compared to other silica materials with higher sodium content. In other contemplated embodiments, sheets of silica have higher sodium concentrations and still benefit from inventive teachings disclosed herein, such as those related to flattening and/or polishing.
In other embodiments, silica glass sheet 28 has a low level of hydroxyl (OH) concentration. In various embodiments, the OH concentration can be controlled to impact the viscosity, refractive properties, and other properties of silica glass sheet 28. In various embodiments, silica glass sheet 28 has a beta OH concentration of less than 0.2 abs/mm (e.g., less than 200 ppm OH), and more specifically of less than 0.12 abs/mm (120 ppm OH). In various embodiments, silica glass sheet 28 has a particularly low concentration of OH, and in such embodiments, beta OH is less than 0.02 abs/mm and more specifically is less than 0.002 abs/mm. In some embodiments, the OH concentration of silica glass sheet 28 formed using laser sintering system 10 is less than the OH concentration of silica material formed using some other formation methods (e.g., plasma sintering, flame sintering and/or sintering process that dry using chlorine prior to sintering). In contrast to some silica materials that utilize a surface treatment with a material such as hydrofluoric acid, silica glass sheet 28 has a low surface halogen concentration and a low surface OH concentration. In other contemplated embodiments, sheets of silica have higher OH concentrations and still benefit from inventive teachings disclosed herein, such as those related to flattening and/or polishing.
In various embodiments, sintered glass sheet 28 has opposing first and second major surfaces, at least one of which has a high level of smoothness. In various embodiments, the roughness (Ra) of at least one of the first major surface and the second major surface of sintered glass sheet 28, such as after polishing as disclosed herein, is less than 50 nanometers, such as less than 10 nm, between 0.025 nm and 1 nm, specifically between 0.1 nm and 1 nm and specifically between 0.025 nm and 0.5 nm, over at least one 0.023 mm²area, such as over 0.05 mm², such as over 0.5 mm², such as over 1 mm². In particular embodiments, the roughness (Ra) of at least one of the first major surface and the second major surface of sintered glass sheet 28 is particularly low such that the roughness is between 0.025 nm and 0.2 nm over at least one 0.023 mm²area.
Ra is determined using a Zygo optical profile measurement as shown in FIGS. 4A-B and 6A-B, and specifically determined using the Zygo with a 130 μm×180 μm spot size. In some embodiments, the roughness (Ra) of at least one of the first major surface and the second major surface of sintered glass sheet 28 is between 0.12 nm and 0.25 nm as measured using AFM over a 2 μm line scan. In specific embodiments, sintered glass sheet 28 has a low roughness level on a small scale measurement, and a larger roughness level with a larger scale measure. In various embodiments, the roughness (Ra) of at least one of the first major surface and the second major surface of sintered glass sheet 28 is between 0.025 nm and 1 nm over at least one 0.023 mm2 area, and an Ra of between 1 μm and 2 μm using a profilometer and a scan length of 5 mm.
While the major surfaces of sintered glass sheet 28 are smooth, the surfaces do have a surface topology prior to polishing, as disclosed herein, including a series of raised and recessed features or peaks and valleys, as shown in FIGS. 4A-B. In the embodiments discussed herein, the raised and recessed features are relatively small contributing to the low surface roughness. In various embodiments, each raised feature has a maximum peak height that is between 0.1 μm and 10 μm, and specifically between 1 μm and 2 μm, relative to the average or baseline height of the topology as measured using a profilometer and a scan length of 5 mm. In specific embodiments, the topology of one or more surface of glass sheet 28 is such that the average and/or maximum vertical distance between the bottom of a recessed feature (e.g., a valley) and the top of a raised feature (e.g., a peak) is greater than 1 micrometer before polishing, as disclosed herein, and less than 0.5 micrometer after polishing, as measured by a Zygo optical profile measurement.
In some embodiments, silica glass sheet 28 may have bulk curvature or warp such that the opposing major surfaces of silica glass sheet 28 deviate somewhat from a planar configuration. In some embodiments, one of the major surfaces of silica glass sheet 28 has concave shape extending across the width of sheet 28 such that the center of one of the major surfaces of sheet 28 is positioned lower than the lateral edges of sheet 28. In various embodiments, the warp of sheet 28 is between 0.5 mm and 8 mm, such as before flattening as disclosed herein, as measured within an area of 3750 mm². In an example, the warp of a sample of sheet 28 was measured and taken from the Werth gauge on sheet 28 having dimensions 50 mm×75 mm. In another embodiment, the warp of sheet 28 is less than 20 μm across a 150 mm×150 mm square area. Alternatively, as discussed in more detail below, in various embodiments sheet 28 may be sintered in a manner to reduce warp and/or may be flattened following sintering to reduce warp.
In various embodiments, silica glass sheet 28 has two major surfaces, the upper surface formed from the portion of soot sheet 20 facing soot burner 16, and the lower surface formed from the portion of soot sheet 20 which is in contact with drum 12. In various embodiments, either the upper surface or the lower surface or both of silica glass sheet 28 may have any of the characteristics discussed herein. In specific embodiments, upper surface of silica glass sheet 28 may have the surface characteristics discussed herein, and the lower surface has surface configuration, topology, roughness, surface chemistry, etc. that is different from the upper surface resulting from the contact with drum 12. In a specific embodiment, the lower surface of silica sheet has a roughness (Ra) that is greater than that of the upper surface, and the Ra of the lower surface of silica glass sheet 28 may be between 0 and 1 μm. In another embodiment, lower surface of silica sheet 28 has a roughness (Ra) that is less than that of the upper surface, and in such embodiments, cleaning of the soot deposition surface (e.g., surface 14 of drum 12) following removal of the soot sheet may result in the high level of smoothness of the lower surface of silica sheet 28.
Referring to FIG. 1, system 10 includes a cutting laser 30 that generates a cutting laser beam 32 that cuts a subsection 34 of sintered glass from glass sheet 28. In addition to cutting subsection 34 from glass sheet 28, cutting laser 30 is configured to form an edge structure surrounding and defining the outer perimeter of cut subsection 34. In various embodiments, the edge structure is a thickened or bulblike section of melted silica material that may act to strengthen the cut subsection 34.
In various embodiments, cutting laser 30 is a focused CO₂laser beam. In one exemplary embodiment, a CO₂laser beam with a focal length of about 860 mm is focused down to 500 μm in diameter. At a laser power of 200 W, the average power density at the focus is 1020 W/mm². At this power density, laser ablation occurs, and a 100 μm thick silica sheet was cut at a speed of 70 mm/s. The peak energy density during the laser ablation process is 11 J/mm². In contrast to prior laser cutting contemplated by Applicant, it was found that this high powered, energy dense laser created the strengthening edge profile discussed below.
In yet further embodiments, system 10 is configured to produce sintered silica glass sheets, such as sheet 28, or cut glass subsections, such as subsection 34, having a very low degree of warp (e.g., a high degree of flatness). In various embodiments, the highly flat, sintered glass sheets discussed herein also include any combination of the silica sheet features (e.g., roughness, purity, chemical compositions, surface characteristics, strengthening edge shape, fictive temperature characteristics, etc.) discussed herein. As discussed in detail herein, highly flattened silica sheets may be produced via a post-sintering flattening process, alone or in combination, with control of various sintering process parameters that increase or result in sheet flatness. In various embodiments, high levels of flatness may provide various advantages in various applications, such as increasing uniformity in the deposition, growth, alignment, fixturing, machining and or stacking of multiple silica sheets 28. In particular, improved flatness may increase repeatable alignment of parts incorporating sheets 28, in various assembly operations.
Referring to FIGS. 3A-3B, a system 210 and method for post-sinter flattening of a sintered glass sheet 212A (before) 212B (after), is shown and described. Flattening system 210 is shown, according to one embodiment. Flattening system includes a lower plate 214 or support, shown as setter plate, a top plate 216, and a heating system, such as induction heater 218. In general, sintered silica glass sheet 212A is placed on setter plate 214. As can be seen in FIG. 3A, glass sheet 212A has a relatively high degree of warp, and in particular, glass sheet 212A has an arched shape in which the central region is spaced a distance above (in the orientation of FIG. 3A) of the outer perimeter of glass sheet 212A. Thus, prior to heating, outer perimeter of glass sheet 212A is in contact with upper surface of setter plate 214, and central region of glass sheet 212A is spaced from upper surface of setter plate 214. As shown in FIG. 3B, after glass sheet 212A is placed onto setter plate 214, top plate 216 is placed on top of sheet 212A such that lower surface of top plate 216 is in contact with the upper surface of glass sheet 212A.
With glass sheet 212A between plates 214 and 216, induction heater 218 heats glass sheet 212A. As glass sheet 212A is heated, the weight of top plate 216 acts as a force acting downward on glass sheet 212A. Through the application of heat by induction heater 218 and of the force applied by top plate 216, glass sheet 212A is flattened forming flattened glass sheet 212B as shown in FIG. 3B. In particular embodiments, glass sheet 212B is heated to above its glass transition temperature but below melting such that it may be flattened under the weight of plate 216. In specific embodiments, system 210 reduces the degree of warp present in sheet 212A to produce flattened sheet 212B while maintaining the various other properties of sheet 212B discussed herein, such as surface roughness, surface features, purity, etc. In the exemplary embodiment show, glass sheet 212B is a 50 mm by 50 mm sintered glass sheet, and plates 214, 216 have a thickness of 1 mm.
In various embodiments, plates 214, 216 are formed from a silica material, and in particular are formed from a highly pure silica material, such as high purity fused silica. By contacting sheet 212A with high purity silica plates during flattening, the high silica purity of glass sheet 212A can be maintained by preventing glass sheet 212A from absorbing contaminants from plates 214, 216. However, Applicant discovered that glass sheet 212A and silica plates 214, 216 tend to bond together during heating if the temperature is too high or if surfaces of plates 214, 216, respectively, are too smooth. Accordingly, Applicant identified that silica plates 214, 216 having surfaces each having surface roughness (Ra) greater than 500 nm, specifically greater than 600 nm and more specifically greater than 700 nm allows glass sheet 212B to easily release from plates 214, 216 following flattening. Applicants found that plates 214, 216 having the surface roughness did not bond to glass sheet 212B during flattening.
Applicant has also identified that induction heater 218 may be controlled to facilitate release of plates 214, 216 from glass sheet 212B following flattening. In particular, without being bound by a particular theory, Applicant believes that if system 210 is heated too high for too long, the roughness of surfaces 214, 216 will decrease which in turn increases the degree of bonding between plates 214, 216 and glass sheet 212A. In various embodiments, Applicant has found that in order to maintain the roughness of sheet 214, 216 surfaces, induction heater 218 is controlled to maintain the maximum temperature of plates 214, 216 below 1800 degrees C., and specifically between 1300 and 1800 degrees C. As will be understood, because the degree to which surfaces of the plates 214, 216 lose their roughness during heating varies based on the amount of time spent at a particular temperature, the maximum allowable temperature is inversely related to the amount of time plates 214, 216 are exposed to heating during the flattening operation.
It should be understood that while FIG. 3B shows an induction based heating system as part of flattening system 210, system 210 may include any of a variety of heating systems that can reach glass transition temperatures, such as resistive, gas, microwave, laser, etc., heating systems. However it is believed that induction based heating systems are particularly suitable options allowing for fast cycle time and flexibility in part shape and size.
In various embodiments, flattening system 210 may be further configured to provide desirable heat distribution and/or to maintain high silica purity of glass sheet 212A. For example, system 210 may include a susceptor below setter plate 214. As will be generally understood, graphite susceptor is a block of resistive material capable of absorbing electromagnetic energy from induction heater 218 which in turn heats susceptor, and the heat from susceptor is conducted to glass sheet 212A. In other embodiments, the susceptor may be formed from a metal material, and in other embodiments, continuous process furnaces designed for high throughput part flattening may be used.
Further, system 210 may include an enclosure for controlling the atmosphere that glass sheet 212A is exposed to during heating and flattening. By controlling the atmosphere during heating, the purity of glass sheet 212A can be maintained or controlled by controlling the degree to which impurities are imparted to glass sheet 212A during flattening. In various embodiments, enclosure may be filled with an inert or non-reactive atmosphere (a nitrogen atmosphere, noble gas atmosphere, etc.) during flattening. In other embodiments, a vacuum may be drawn within enclosure during flattening. In particular embodiments, removing the atmospheric air and/or providing inert gas around the graphite susceptor acts to remove O₂and/or moisture from system 210 during flattening. It is believed that O₂may allow the graphite susceptor to burn, and H₂O may cause plates 214, 216 to stick more easily to glass sheet 212A during flattening. Thus, operation of system 210 may be improved by the removal of O₂and/or H₂O from the atmosphere within the enclosure.
It should be understood that while the specific embodiment of flattening system 210 shown utilizes a top plate 216 to provide the flattening force onto glass sheet 212A, the flattening force may be applied to glass sheet 212A in other ways. For example, in one embodiment, the flattening force applied to glass sheet 212A is the gravitational force acting on sheet 212A, and in such an embodiment, glass sheet 212A is flattened under its own weight. In other embodiments, gas pressure or gas jets are directed onto glass sheet 212A pressing the glass sheet onto setter plate 214. In yet another embodiment, a vacuum may be applied to the lower surface of glass sheet 212A pulling glass sheet 212A downward, and in a specific embodiment, setter plate 214 includes a plurality of apertures allowing the pulling vacuum to be evenly distributed across a portion or across all of the surface.
In yet other embodiments, the flattening process may be utilized to further alter glass sheet 212A. In a particular embodiment, sheet 212B may include a shape, pattern, etc., which is imprinted or embossed onto the lower and/or upper surfaces of glass sheet 212B during shaping, such as flattening.
In general, as used herein, warp refers to the shape of glass sheet 110 at a macroscopic or sheet-wide scale. FIG. 2 provides an illustration of how warp is determined, measured or calculated, according to an exemplary embodiment. As shown in FIG. 2, line C is the least squares focal plane defined along an article (e.g., sheet 110, sheet 28, etc.) at a cross-sectional position through the article. In at least some embodiments, when warp is determined using the definition shown in FIG. 2, the sheet is in a free or unweighted/unclamped state. As shown, point B is the lowest point of the sheet, and point A is the highest point of the sheet. In this definition of warp/flatness, warp is the maximum distance between the highest point (A) and lowest point (B) from the least squares focal plane (C). In this embodiment, warp measurements are positive, and warp is determined by measuring the displacement from the least squares focal plane across the entire sheet or across an entire defined subsection (rather than simply measuring at a particular set of points, such as at the center point).
In various embodiments, warp of flattened glass sheet 110 (i.e. representative of flattened sheets disclosed herein) is less than 1 mm, less than 500 μm, less than 50 μm, or less than 10 μm. In particular embodiments, these warp measurements are maximum warp as measured over an entire area of sheet 110 utilizing the definition shown in FIG. 2. In particular embodiments, these warp measurement are the maximum warp as measured over at least one section of the glass sheet having an area of 50 mm by 50 mm or alternatively having an area greater than 2500 mm²utilizing the definition shown in FIG. 2. As noted above, sheet 28 may have levels of warp greater than 1 mm, and thus, flattening system 100 is able to achieve high levels of flattening relative to the initial levels of warp. In various embodiments, warp of flattened glass sheet 110 is less than 50% of the warp of sheet 28, less than 10% of the warp of sheet 28, and even less than 1% of the warp of sheet 28. In some such embodiments, the surface roughness Ra of the major surfaces of sheet 28 and sheet 110 remains with the same range both before and after flattening. In some such embodiments, the purity of sheet 28 and sheet 110 remains within the same range both before and after flattening.
In various embodiments, flattened glass sheet 110 includes the low levels of warp discussed herein in combination with any of the other glass sheet properties discussed herein. In a particular embodiment, flattened glass sheet 110 includes the low warp measurements discussed herein and one or both of the major surfaces of flattened glass sheet 110 has a total indicator run-out (TIR) measurement of less than 50 μm, a microwaviness measurement (Wa) of less than 0.5 μm, and/or a microwaviness measurement (Wt) of less than 20 μm. Referring to FIG. 3A, measurements of warp and microwaviness of flattened sheet 110 are shown according to exemplary embodiments. In one exemplary embodiment, glass sheet 28 has an area of 50 mm by 50 mm, warp of between 1 and 1 mm, and TIR of more than 10 mm. Following flattening using the process described above, TIR was reduced to below 50 μm (specifically 36.7 μm).
In various embodiments, instead of or in addition to utilizing post-sintering flattening system 210, various aspects of system 10 may be controlled to increase flatness of sintered glass sheet 28 produced by system 10. In some such embodiments, glass sheet 28 may have any of the low warp characteristics of sheet 110 discussed above without the need to be processed through flattening system 210.
In some embodiments, sintered silica glass sheet 28 consists of at least 99.9% by weight, and specifically at least 99.99% by weight, of a material of the composition of (SiO₂)_1-x-y. M′_xM″_y, where either or both of M′ and M″ is an element (e.g., a metal) dopant, or substitution, which may be in an oxide form, or combination thereof, or is omitted, and where the sum of x plus y is less than 1, such as less than 0.5, or where x and y are 0.4 or less, such as 0.1 or less, such as 0.05 or less, such as 0.025 or less, and in some such embodiments greater than 1×10⁻⁶for either or both of M′ and M″. In certain embodiments, sintered silica glass sheet 28 is crystalline, and in some embodiments, sintered silica glass sheet 28 is amorphous.
In various embodiments, sintered silica glass sheet 28 is a strong and flexible substrate which may allow a device made with sheet 28 to be flexible. In various embodiments, sintered silica glass sheet 28 is bendable such that the thin sheet bends to a radius of curvature of at least 500 mm without fracture when at room temperature of 25° C. In specific embodiments, sintered silica glass sheet 28 is bendable such that the thin sheet bends to a radius of curvature of at least 300 mm without fracture when at room temperature of 25° C., and more specifically to a radius of curvature of at least 150 mm without fracture when at room temperature of 25° C. Bending of sintered silica glass sheet 28 may also help with roll-to-roll applications, such as processing across rollers in automated manufacturing equipment.
In various embodiments, sintered silica glass sheet 28 is a transparent or translucent sheet of silica glass. In one embodiment, sintered silica glass sheet 28 has a transmittance (e.g., transmittance in the visual spectrum, transmittance of light having a wavelength between 300 and 2000 nm) greater than 90% and more specifically greater 93%. In various embodiments, the sintered silica glass sheets discussed herein have a softening point temperature greater than 700° C. and more specifically greater than 1100° C. In various embodiments, the sintered silica glass sheets discussed herein have a low coefficient of thermal expansion less than 10×10⁻⁷/° C. in the temperature range of about 50 to 300° C.
While other sintering devices may be used to achieve some embodiments, Applicants have discovered advantages with laser sintering in the particular ways disclosed herein. For example, Applicants found that laser sintering may not radiate heat that damages surrounding equipment which may be concerns with sintering via induction heating and resistance heating. Applicants found that laser sintering has good control of temperature and repeatability of temperature and may not bow or otherwise warp sheet 28, which may be a concern with some other sintering methods. In comparison to such other processes, laser sintering may provide the required heat directly and only to the portion of the soot sheet needing to be sintered. Laser sintering may not send significant amounts of contaminates and gases to the sintering zone, which may upset manufacturing of the thin sheets. Further, laser sintering is also scalable in size or for speed increases.
In various embodiments, the silica soot sheets disclosed herein are formed by a system that utilizes one or more glass soot generating device (e.g., a flame hydrolysis burner) that is directed or aimed to deliver a stream of glass soot particles on to a soot deposition surface. As noted above, the silica sheets discussed herein may include one or more dopant. In the example of a flame hydrolysis burner, doping can take place in situ during the flame hydrolysis process by introducing dopant precursors into the flame. In a further example, such as in the case of a plasma-heated soot sprayer, soot particles sprayed from the sprayer can be pre-doped or, alternatively, the sprayed soot particles can be subjected to a dopant-containing plasma atmosphere such that the soot particles are doped in the plasma. In a still further example, dopants can be incorporated into a soot sheet prior to or during sintering of the soot sheet. Example dopants include elements from Groups IA, IB, IIA, IIB, IIIA, IIIB, IVA, IVB, VA, VB and the rare earth series of the Periodic Table of Elements. In various embodiments, the silica soot particles may be doped with a variety of materials, including germania, titania, alumina, phosphorous, rare earth elements, metals and fluorine.
Following the flattening steps disclosed above, the sheet 28 may be relatively flat, having low warp as disclosed herein, but may have surface waviness on a microscale, as shown in FIGS. 4A and 4B. The waviness may be in the form of recurring peaks and valleys that are periodic and directional, generally all extending in the same direction. The peaks and valleys may be an artifact of the above described process for manufacturing thin silica sheets, such as due to laser sintering and/or soot deposition through discrete nozzles.
Referring now to FIG. 5, in at least one embodiment, a thin substrate, such as less than 300 micrometers in thickness, of surface area equivalent to a 150 mm diameter circle is highly polished to form a high quality thin glass wafer. To achieve this polished quality, the thin wafer (see process of FIG. 5) is temporarily bonded to a clean, highly polished silicon wafer to produce a mechanically rigid material stack. The material stack bearing the thin wafer is polished by bringing the silica wafer into contact with a mechanically rigid (low compressibility, high hardness) polishing pad affixed to a platen of a planetary polisher. In some such embodiments, the polishing pad is first dynamically impregnated with nano-diamond slurry particles via extended contact with a high concentration (50 cts/liter) water-based nano-diamond slurry. According to an exemplary embodiment, the mechanically planarized thin glass wafer (still in stack form) is then submitted for conventional cerium oxide (CMP) polishing to achieve a high degree of surface quality suitable for bonding to a compatible substrate.
According to an exemplary embodiment, following debonding of the mechanically rigid material stack. The silica glass wafer is mechanically cleaned to remove all debris detrimental to temporary bonding. A layer of a polymeric bonding agent is applied to the B-side of the glass wafer, the glass wafer is then brought into contact with a clean, polished silicon wafer substrate in a vacuum resulting in a bonded silicon to silica glass wafer stack. The bonded silicon wafer stack is then loaded into a simple polishing fixture in preparation for polishing.
According to an exemplary embodiment, the polishing system may include Logitech benchtop planetary polisher equipped with a flat steel polishing platen, a mechanically rigid (low compressibility, high hardness) polishing pad affixed to the platen, such as a Nitta-Haas MH14 polishing pad, silicon wafer then affixed to a steel puck of the same diameter (e.g., 100 mm diameter) in preparation for use as a pad break-in substrate.
According to an exemplary embodiment, the polishing pad was then hydrated as follows. A slurry delivery line was positioned above the center of the polishing pad. The polishing platen was set in motion to a rotational speed of 50 rpm. A flow of potable water of 100 ml/min was then established using the peristaltic pump. A mechanically stiff hand brush was then pressed against the polishing pad with moderate force to impregnate the pad with water and hydrate it, where hydration was determined visually. The machine was stopped and the steel puck bearing the silicon wafer was loaded onto the polishing platen such that the silicon surface was put in contact with the hydrated polishing pad. A flow (e.g. 10 ml/min) of concentrated (50 cts/liter) nano-diamond slurry (particle size ranging from 30 nm to 25 μm) was then established in place of the potable water and the platen slowly set in motion ramping in 5 rpm increments until a speed of about 50 rpm was established. The above condition was maintained for a period of at least 10 minutes or until it was clear that the nano-diamond slurry particles had thoroughly impregnated the polishing pad surface. Next, the platen was stopped and the silicon break-in puck unloaded, however, the nanodiamond slurry pump was allowed to continue depositing new slurry onto the polishing pad.
In some such embodiments, the bonded silicon wafer stack was placed on the polishing platen such that the silica was put into contact with the polishing pad. The platen speed was slowly ramped up in increments of 5 rpm while increasing the polishing downforce (done by carefully placing calibrated weight steel pucks onto the back of the material stack) until independent substrate rotation is established. In such embodiments, these conditions can be described by the following process parameter settings: platen rotation speed of about 70 rpm, polishing downforce of about 1 psi, nanodiamond slurry volumetric flow rate of about 6 ml/min, inner sweep of 0%, outer sweep of about 58%. The silica surface under the above conditions was polished for about 5 minutes. Upon completion of the above polishing cycle, the platen and the slurry flow were shut down and then the material stack was removed, rinsing off the silica surface and pad surface with clean water. This pad cleansing (mechanical brush under high water flow) was repeated until the pad was free of nanodiamond impregnation. Alternatively, an additional pad may be prepared.
In some such embodiments, the polishing the silica surface was finished using the method described above for 5 minutes using cerium oxide slurry; in this embodiment finish polishing conditions were as follows: platen rotation speed of 70 rpm, polishing downforce of 1 psi, cerium oxide slurry included Super Cerite 416, cerium oxide slurry volumetric flow rate of 10 ml/min, inner sweep of 0%, and outer sweep of 58%. FIG. 6A-6B show a surface of a silica sheet after such polishing. As can be seen, the peaks and valleys are substantially reduced relative to those in FIG. 4A-4B.
In another contemplated embodiment, a thin substrate as disclosed herein of 150 mm diameter circular wafer form factor of high purity fused silica was highly polished using double side polishing technology to form a high quality thin glass wafer. The thin carrier issue was overcome by temporarily bonding the thin substrates to each side of a rigid carrier wafer thereby producing a thick material stack that may then be treated as a thick wafer in a double side polishing. The double side polishing pads may be impregnated with diamond Nanoslurry in a manner analogous to the single side polishing (flip polishing) method of the first embodiment above.
In still other contemplated embodiments, a thin substrate as disclosed herein of 150 mm diameter circular wafer form factor of high purity fused silica was differentially polished employing the Twyman Effect to mechanically ease stresses resident in the glass during its formation (consolidation). One face of the silica substrate may be coarsely polished while the other is polished in a manner resulting in a fine surface, thereby enabling shape relaxation via the Twyman Effect.
In another contemplated embodiment a material stack comprising a standard thickness (˜725 μm) silicon wafer substrate and ultrathin (≤50 μm) silica wafer were permanently bonded (thermal oxide bonded) together.
Still referring to FIGS. 1, 3A-B, and 5, aspects of the present disclosure relate generally to a method of manufacturing a sheet of fused silica, such as at least 99.99 wt % silicon dioxide. The method includes steps of polishing a sheet of fused silica having a thickness of less than 500 μm by removing less than 100 micrometers depth of material of a major surface of the sheet, such as less than 50 micrometers. The sheet of fused silica may then have a thickness of less than 300 micrometers, a major face surface area of at least 6π square inches, a total thickness variation of less than 10 micrometers excluding edge rolloff, a peak-to-valley waviness of less than 500 nanometers, and a root mean square roughness over a square millimeter of major surface of less than 100 nanometers. Prior to polishing, the sheet may have peak-to-valley waviness of greater than 3 micrometers. The sheet may be made by depositing silica soot to form a sheet of silica soot and then sintering the sheet of silica soot to form the sheet of fused silica, prior to the polishing step.
In some such embodiments, the polishing step of the method further includes a step of bonding the sheet to a substrate and then polishing the major surface of the sheet, and may be performed without simultaneously polishing a second major surface of the sheet. In some embodiments, the polishing step further includes debonding the sheet from the substrate, flipping the sheet, bonding the flipped sheet to the substrate or a new substrate, and polishing the second major surface of the sheet. The method may further include a step of flattening the sheet of fused silica prior to the polishing step.
Other aspects of the present disclosure relate generally to a method of manufacturing a sheet of fused silica, which includes a step of shaping a sheet of fused silica having a thickness of less than 500 micrometers, a major face surface area of at least 6π square inches, and a peak-to-valley waviness of at least 1 micrometer. The shaping may be flattening of the sheet so that the sheet has a warp of less than 50 micrometers, or the shaping may include imparting curvature. The method may further include a step of polishing the sheet, after the shaping step. The polishing may, in part, be performed by bonding the sheet to a substrate, polishing a first major side of the sheet, debonding the sheet from the substrate, flipping the sheet, bonding the flipped sheet to the substrate or a new substrate, and polishing a second major side of the sheet. After completing the method, the sheet of fused silica may have a total thickness variation of less than 10 micrometers excluding edge rolloff and peak-to-valley waviness of less than 500 nanometers.
In some such embodiments, the shaping step may include heating the sheet to a temperature below melting temperature of the fused silica. In some embodiments, the shaping step may include applying a load, such as compression, tension, etc., to the sheet to displace at least a portion of the sheet. In some embodiments, the shaping step may include, after the heating step, a step of cooling the sheet under the load such that displacement is at least in part set. After the polishing step, the warp may still be less than 50 micrometers. According to an exemplary embodiment, polishing and shaping does not fracture the sheet.
Referring now to FIG. 7, other aspects of the present disclosure relate generally to a sheet 810 of fused silica that has a thickness T of less than 300 micrometers, a major face 814 surface area of at least 6π square inches (in this case length L times width W), a total thickness variation of less than 10 micrometers excluding edge 812 rolloff, a peak-to-valley waviness of less than 500 nanometers, and a root mean square roughness over a square millimeter of major surface of less than 100 nanometers. In some embodiments, warp of the sheet is less than 50 micrometers. In some embodiments, the fused silica is amorphous and is at least 99.99 wt % silicon dioxide. In some embodiments, the peak-to-valley waviness is less than 100 nanometers, the total thickness variation is less than 5 micrometers, and/or the major face surface area is at least 8π square inches.
Advantages of technology disclosed herein include high planarization efficiency by the flattening, enabling removal of defects with minimal material removal. Using this method it has been demonstrated in samples polished to thicknesses ˜100 μm that waviness can be reduced to as low as <100 nm measured as the distance from highest max peak and lowest adjoining valley from a plane across 1/10 of the longest part dimension. Using polishing methods disclosed herein, the average surface roughness has been demonstrated to be less than 50 nm.
As indicated in multiple instances above, peak-to-valley waviness for embodiments of the sheets disclosed herein may be less than 500 nanometers. For example, peak-to-valley waviness for such embodiments may be less than 100 nanometers, such as 10 nanometers, and/or at least 10 nanometers.
The construction and arrangements of the methods, systems, and equipment, as shown in the various exemplary embodiments, are illustrative only. Although only a few embodiments have been described in detail in this disclosure, many modifications are possible (e.g., variations in sizes, dimensions, structures, shapes, and proportions of the various elements, values of parameters, mounting arrangements, use of materials, colors, orientations) without materially departing from the novel teachings and advantages of the subject matter described herein. Some elements shown as integrally formed may be constructed of multiple parts or elements, the position of elements may be reversed or otherwise varied, and the nature or number of discrete elements or positions may be altered or varied. The order or sequence of any process, logical algorithm, or method steps may be varied or re-sequenced according to alternative embodiments. Other substitutions, modifications, changes and omissions may also be made in the design, operating conditions and arrangement of the various exemplary embodiments without departing from the scope of the present inventive technology.

Claims

What is claimed is:

1. A method of manufacturing a sheet of fused silica, comprising steps of:

polishing a sheet of fused silica having a thickness of less than 500 μm by removing less than 100 micrometers depth of material of a major surface of the sheet;

wherein, following the above steps, the sheet of fused silica has:

a thickness of less than 300 micrometers,

a major face surface area of at least 6π square inches,

a total thickness variation of less than 10 micrometers excluding edge rolloff,

a peak-to-valley waviness of less than 500 nanometers, and

a root mean square roughness over a square millimeter of major surface of less than 100 nanometers.

2. The method of claim 1, wherein the polishing step further comprises a step of bonding the sheet to a substrate and then polishing the major surface of the sheet.

3. The method of claim 2, wherein the polishing of the major surface of the sheet is performed without simultaneously polishing a second major surface of the sheet.

4. The method of claim 3, wherein the polishing step further comprises debonding the sheet from the substrate, flipping the sheet, bonding the flipped sheet to the substrate or a new substrate, and polishing the second major surface of the sheet.

5. The method of claim 1, further comprising a step of flattening the sheet of fused silica prior to the polishing step.

6. The method of claim 1, wherein the fused silica is amorphous and is at least 99.99 wt % silicon dioxide.

7. The method of claim 1, wherein the polishing step is performed by removing less than 50 micrometers depth of material of the major surface of the sheet.

8. The method of claim 1, wherein the sheet of fused silica, prior to the polishing step, has peak-to-valley waviness of greater than 3 micrometers.

9. The method of claim 1, further comprising steps of depositing silica soot to form a sheet of silica soot and then sintering the sheet of silica soot to form the sheet of fused silica, prior to the polishing step.

10. A method of manufacturing a sheet of fused silica, comprising steps of:

shaping a sheet of fused silica having a thickness of less than 500 micrometers, a major face surface area of at least 6π square inches, and a peak-to-valley waviness of at least 1 micrometer; and

polishing the sheet, after the shaping step, by bonding the sheet to a substrate, polishing a first major side of the sheet, debonding the sheet from the substrate, flipping the sheet, bonding the flipped sheet to the substrate or a new substrate, and polishing a second major side of the sheet,

wherein, following the above steps, the sheet of fused silica has a total thickness variation of less than 10 micrometers excluding edge rolloff and peak-to-valley waviness of less than 500 nanometers.

11. The method of claim 10, wherein the shaping step includes a step of heating the sheet to a temperature below melting temperature of the fused silica.

12. The method of claim 11, wherein the shaping step includes a step of applying a load to the sheet to displace at least a portion of the sheet.

13. The method of claim 12, wherein the shaping step includes, after the heating step, a step of cooling the sheet under the load such that displacement is at least in part set.

14. The method of claim 13, wherein the shaping step is more specifically a flattening step that flattens the sheet such that warp of the sheet is less than 50 micrometers.

15. The method of claim 14, wherein, after the polishing step, the warp is still less than 50 micrometers.

16. A sheet of fused silica comprising a thickness of less than 300 micrometers, a major face surface area of at least 6π square inches, a total thickness variation of less than 10 micrometers excluding edge rolloff, a peak-to-valley waviness of less than 500 nanometers, and a root mean square roughness over a square millimeter of major surface of less than 100 nanometers.

17. The sheet of claim 16, wherein warp of the sheet is less than 50 micrometers.

18. The sheet of claim 16, wherein the fused silica is amorphous and is at least 99.99 wt % silicon dioxide.

19. The sheet of claim 16, wherein peak-to-valley waviness is less than 100 nanometers.

20. The sheet of claim 16, wherein the total thickness variation is less than 5 micrometers.

21. The sheet of claim 16, wherein the major face surface area is at least 8π square inches.