WO1998014801A1

WO1998014801A1 - Strengthened optical glass filter

Info

Publication number: WO1998014801A1
Application number: PCT/US1997/016726
Authority: WO
Inventors: Suresh T. Gulati
Original assignee: Corning Incorporated
Priority date: 1996-09-30
Filing date: 1997-09-18
Publication date: 1998-04-09
Also published as: EP0929827A1; JP2001501743A; EP0929827A4; KR20000048738A

Abstract

A laminated, optical filter lens (10) having particular application in a high-fluence laser, and a method of forming such lens. The filter lens (10) comprises a core (12) and a cladding (14 and 16), the core (12) having an essentially pure, single, fused oxide composition and the cladding (14 and 16) having a fused oxide composition different from that of the core (12) and having a coefficient of thermal expansion lower than that of the core (12).

Description

STRENGTHENED OPTICAL GLASS FILTER

FIELD OF THE INVENTION

Laminated, optical glass filters, their use in articles such as lasers, and their method of production.

BACKGROUND OF THE INVENTION

It is well known that glass, in its pristine state, has a high mechanical strength.

However, even slight surface damage can greatly diminish this intrinsic strength. Further, surface cracks, not normally visible, can grow due to a combination of tensile stress and atmospheric effects, particularly moisture.

As a consequence, numerous physical and chemical methods of strengthening glass have been proposed. These include air tempering, surface ion-exchange and forming a crystalline layer on the glass surface. It has also been proposed to encase a high expansion glass with a lower expansion glass to yield high surface compressive stresses, and, consequently, high strength. However, attempts to strengthen high silica glasses, such as fused silica or a 96% silica glass, have met with problems, primarily because of the temperatures involved.

The problem of glass strength is aggravated in applications involving reduced pressure on one side of a glass article. Such a situation is encountered, for example, in a spatial filter lens for a high-fluence laser, such as are used in atom research programs. The term "spatial filter lens" refers to a lens through which a laser beam passes into an evacuated space. The term "fluence" refers to energy expressed in Joules/cm². A pressure differential exists across the filter lens due to the vacuum. This induces tensile stress on the vacuum side of the lens. This stress can be kept below a threshold level by special precautions. However, use of the laser itself induces damage in the glass lens.

The contribution of tensile stress, laser-induced damage, and an imperfect vacuum, can lead to slow growth of even minute, surface cracks. In turn, this leads to implosive failure of the filter if the stress intensity approaches a critical value for the glass.

This problem has been countered by increasing the thickness of the lens, or by resorting to a square lens design. Thickening of the lens has an adverse impact on optical performance. The square lens design limits design flexibility. There exists, then, a need for a more satisfactory method of producing a strengthened, optical filter lens. The number of optical glasses that are acceptable for production of such a lens are limited. Of these, fused silica is a preferred material because of its excellent transmissive properties over a wide range of wavelengths from 0.35-1.05 μm. The superior transmissive property is due in large measure to the essential absence of damaging inclusions in fused silica. This is a result of the material being produced by flame hydrolysis.

Fused silica, by its very nature, does not lend itself to the familiar glass strengthening practices of air tempering, surface modification by ion-exchange, or surface crystallization. Thus, there is a particular need for a glass strengthening procedure that is applicable to fused silica. It is a basic purpose of the present invention to meet this observed need. It is a further purpose to provide an optical filter lens having increased mechanical strength and/or less thickness. Another purpose is to provide such a lens having the transmissive characteristics of fused silica. A specific purpose is to provide a relatively thin, spatial filter lens for a high-fluence laser that can withstand the necessary pressure differential in service. It is also a purpose to provide a method of producing such an improved filter lens. SUMMARY OF THE INVENTION

The article of the invention is a laminated, optical glass filter lens comprising a core and a cladding, the core having an essentially pure, fused oxide composition and the cladding having a different, fused oxide composition, the cladding having a coefficient of thermal expansion (CTE) lower than that of the core glass.

The method aspect of the invention comprises depositing particles of a fused metal oxide to form an optical lens core and then depositing particles of a different fused metal oxide composition to form a cladding on the core, the cladding having a lower

CTE than the core, whereby the clad lens has compressively stressed surfaces.

PRIOR ART

Literature of possible relevance is supplied separately.

BRIEF DESCRIPTION OF THE DRAWING

FIGURES 1 and 2 are schematic drawings illustrating, respectively, a perspective view and a side view of an article in accordance with the invention.

DESCRIPTION OF THE INVENTION

As indicated above, fused silica is a preferred material for production of optical filter lenses, in particular, lenses for use in a high-fluence laser. Accordingly, the invention was developed for, and is described with respect to, an article of this material. However, it will be apparent that the basic principles involved apply equally to any fused metal oxide composition otherwise meeting the necessary qualifications, including binary metal oxide compositions. Fused silica has traditionally been produced by flame hydrolysis of silicon tetra chloride, the process being well known to those active in the art. Recently, it has been proposed to use silane derivatives, in particular, octamethylcyclotetrasiloxane, as a substitute precursor for silica. With either precursor, vapors are formed, hydrolyzed and converted to SiO₂ particles commonly referred to as "soot." The soot particles may be collected, pressed into a desired form, and then fired to form a solid body. Alternatively, the soot particles may be deposited on a mandrel to form a body, and later consolidated. This is a practice followed in optical fiber production. For present purposes, it is preferred to deposit the particles in a molten state. The invention is further described with reference to the accompanying drawing in which FIGURE 1 is a perspective view of a symmetrically laminated, optical glass article 10 illustrating the invention. Laminated article 10 has a flat core section 12 that has a first cladding layer 14 on one surface and a second cladding layer 16 on the opposite surface. Article 10 may, for example, be a spatial filter lens. It is shown as having a cylindrical shape. However, it may also have a square, other rectangular, circular or disc shape if desired.

Core 12 is a pure, fused-oxide glass, in this case, fused-silica with a CTE of about 5xlO^"7/° C. Cladding layers 14 and 16 are the same binary glass, here a Si0₂-TiO₂ glass. The properties of this binary glass depend on the TiO₂ content which may be up to about 10% by weight.

FIGURE 2 is a schematic side view of the laminated article of FIGURE 1. Such an article may be manufactured using the same flame hydrolysis process that is commonly used to produce a monolithic, SiO₂ article. The primary considerations in forming the laminated body are good control of thickness in the layers, and control of the

SiO₂:TiO weight fraction or ratio.

Known equipment and procedure for producing molten particles of silica may be employed in producing the present laminated article. It is necessary only to provide dual vapor streams to the flame. One stream will contain a TiO₂ precursor, such as TiCl₄, and the other stream will contain a Si0₂ precursor, such as SiCL». The contents of the streams will be adjusted to provide the ultimate desired TiO₂/SiO₂ weight ratio in the binary glass particles deposited to form the initial cladding layer.

Once a predetermined thickness of the TiO₂-doped SiO₂ layer 16 is deposited, the TiCl source is stopped. This permits depositing a layer 12 of pure SiO₂ to form the core of the laminated body. Then, the TiCl₄ source is reopened to deposit the upper cladding layer 14.

The cladding must have a lower CTE, preferably up to 10xl0^"7/° C. lower, than the core. For example, a SiO₂-TiO₂ cladding may have a CTE of close to 0 while the silica core has a CTE of about 5xl0^"7/° C. If desired, the TiO content in the cladding may have a gradient profile.

The resulting laminated body, upon cooling to room temperature, has significant compressive stresses induced in the cladding layer due to the expansion differential. These compressive stresses must be overcome before any tensile stresses can be introduced by bending. Furthermore, the thermal expansion of the cladding glass can be tailored by adjusting its composition to induce the desired level of surface compression.

Internal tension in the fused-silica body of the lens is necessary to balance surface compression. This is also controlled by limiting the expansion mismatch. This minimizes stored elastic energy, and the associated fragmentation pattern should the lens fracture. Finally, the cladding glass can be incorporated using the same flame hydrolysis process as that employed for manufacturing monolithic, fused-silica lenses.

Long term experience and testing have demonstrated that crack growth is sufficiently small at stress levels below 6.9 MPa (1,000 psi) that it can be neglected. Accordingly, that value has been taken as the allowable design stress for annealed bulk glass. It is commonly known as the threshold value.

In the case of high-fluence laser lenses, the fused silica lens can experience surface damage on the vacuum side if the tensile stress exceeds the threshold value. The consequences of lens failure can be very costly, particularly to ancillary equipment. Consequently, it has been found necessary to keep the failure probability several orders of magnitude lower than that of most glass products. This is done by reducing the allowable design stress to about 5.5 MPa (800 psi). Furthermore, the stored elastic energy at a peak flexure stress of 5.5 MPa (800 psi) is sufficiently low to contain damage in the remote case of lens failure.

Internal tension σ_c develops in the fused-silica core 12, and compressive stress σ_s develops in the SiO₂-TiO₂ cladding layers 14 and 16. The values may be calculated utilizing properties designated by the subscript _c for the core glass and the subscript _s for the cladding glass. The properties involved are thermal expansion coefficient (α), Young's modulus (E), Poisson's ratio (v), the glass setting temperature (T) and the laminae thickness (t). The Young's modulus and Poisson's ratio values for fused-silica and a SiO -TiO₂ glass with a low TiO₂ content approach each other. Therefore, the tensile and compressive stresses σ_c and σ_s, in the core and cladding, may be expressed as,

The thickness of a cladding layer (t_s) is much lower than the thickness of the silica core (t_c). The typical t₀:t_s ratio is about 20: 1. This cladding layer thickness has been found sufficient to contain surface damage within the compression layer.

Fused silica, the core material, has been found to have the following set of properties: Setting Temperature (T_c) = 990°C.

Young's modulus (E) = 10.6 x 10⁶

Poission's ratio (v) = 0.16

CTE at T_c = 3.5 x l0^"7/°C

The corresponding properties for a SiO₂-TiO₂ cladding glass at its setting point, as a first approximation, vary linearly with the weight fraction (f) of the titania. These properties may be expressed as follows:

CTE_S = CTE_c-62.5 (τ)

T_s = T_c-1500 τ E_s E_c(l-τ)

V_s

TABLE I τ σ_s σ_c ∑c (cmkg) ∑s (cmkg)

(wt. fraction) (psi) MPa (psi) MPa (in lb/in²) cm (in lb/in²) cm

0 010 -700 -4 8 70 0 48 0 0004 tc 0 00035tc 0 039 t_s 0 034t_s

0 015 -1140 -7 8 1 14 0 78 O OOlO t_c 0 00086t_c O lO ts 0 086t_s

0 030 -2140 -14 8 214 1 48 0 0035 t_c 0 v ,)302tc 0 37 t_s 0 3191,

0 045 -2970 -20 5 297 2 05 0 007 t_c 0 00604t_c 0 72 t_s 0 62 It,

0 060 -3840 -26 6 384 2 66 0 0117 tc O OlOltc 1 23 t_s 1 06t_s

0 075 -4520 -31 1 452 3 1 1 0 0162 t_c 0 0140t_c 1 73 t. 1 49t_s

0 090 -5320 -36 7 532 3 67 0 0224 tc 0 0193t_c 2 44 t_s 2 1 Its

TABLE I summarizes the residual stresses σ_c and σ_s and stored elastic energy ∑_c and ∑_s in core and cladding layers as a function of titania content A compressive stress of 4 8 MPa (700 psi) in the cladding layer, which can be obtained by doping with as little as one weight % of Ti0₂, will increase the allowable design stress (σ_d) from 5 5 MPa (800 psi) to 10 3 MPa (1500 psi) This nearly doubles the design stress for an annealed, fused-silica lens

Furthermore, the stored elastic energy in a fused-silica core ∑_c, which will contribute to fragmentation of a laminated lens, is negligibly small compared with that due to flexure of a lens under vacuum loading This is due to a low value of internal tension This low internal tension value is also desirable to avoid risk of fracture when a laminated lens is ground and polished to a required power

TABLE I also shows that the allowable design stress can be increased 6- or 7- fold while keeping the internal tension below 3 5 MPa (500 psi) The 7-fold increase in design stress, if permissible from a stored elastic energy point of view, will help reduce the lens thickness t by more than 60%, thereby improving the optical performance of a lens The increase in design stress is simply the surface compression σ_s whose magnitude depends on laminate geometry and properties. The latter, however, depend on the TiO₂ content. Thus,

Assuming a t_<A of 20, and substituting Eo and vo values, TABLE II summarizes σ_s, σd, t R and Σ/R values for TiO₂ contents of 0-3% by weight in a laminated article. σ_d denotes the safe design stress, and Σ/R the net stored energy due to bending and lamination, R denotes the radius in a circular lens. It denots one-half the side in a square lens, and one-half the short side in a rectangular lens.

TABLE π σ_s σ_d τ MPa MPa t/R Σ/R

0 0 5.5 0.1475 596 x 10^"6

0.005 -2.4 7.8 0.1236 973 x 10^"6

0.010 -4.7 10.2 0.1085 1355 x 10^"6

0.015 -7.0 12.5 0.0981 1740 x lO^-6

0.020 -9.4 14.7 0.0902 2140 x lO^"6

0.025 -11.4 16.9 0.08415 2546 x 10^"6

0.030 -13.7 19.1 0.07916 2967 x 10^"6

It will be noted from TABLE II that the thickness t of a laminated lens can be reduced significantly compared with that of a monolithic, silica lens while limiting the net tensile stress on the vacuum side to 6.9 MPa (800 psi). For example, TiO₂ levels as low as 1.5% can reduce the lens thickness by 33.5% with 43% lower stored energy than that in a monolithic lens stressed to 12.5 MPa (1810 psi). However, the net stored energy in the thin, laminated lens will be higher due to larger contributions from bending.

Claims

WE CLAIM:

1. A laminated, optical glass filter lens comprising a core and a cladding, the core having an essentially pure, single fused oxide composition, the cladding having a fused oxide composition that is different from that of the core and that has a coefficient of thermal expansion lower than the core.

2. A laminated lens in accordance with claim 1 wherein the core is fused silica.

3. A laminated lens in accordance with claim 1 wherein the cladding is a binary oxide glass consisting essentially of silica and up to 10% of a second oxide.

4. A laminated lens in accordance with claim 3 wherein the second oxide is titania.

5. A laminated lens in accordance with claim 1 wherein the thickness of the core is greater than that of the cladding.

6. A laminated lens in accordance with claim 5 wherein the ratio of core thickness to cladding thickness is about 20: 1.

7. A laminated lens in accordance with claim 1 wherein the design stress between the core and the cladding is not over 5.5 MPa (800 psi).

8. A laminated lens in accordance with claim 1 which is a spatial filter lens for a high-fluence laser.

9. A high-fluence laser comprising a laminated, spatial filter lens comprising a core and a cladding, the core having an essentially pure, single fused oxide composition, the cladding having a fused oxide composition that is different from that of the core and that has a coefficient of thermal expansion up to 10x10^" /° C. lower than that of the core.

10. A high-fluence laser in accordance with claim 9 wherein the laminated lens has a core of fused silica and a cladding of a SiO₂-TiO₂ glass in which the TiO₂ component is not over 10%.

11. A high-fluence laser in accordance with claim 9 wherein the laminated lens is subject to a pressure differential that generates a bending stress limited to not over 5.5 MPa (800 psi) by the laminated structure of the lens.

12 A high-fluence laser in accordance with claim 9 wherein the thickness of the core is greater than that of the cladding.

13. A high-fluence laser in accordance with claim 12 wherein the ratio of the core thickness to that of the cladding thickness is about 20: 1.

14. A method of minimizing the thickness of an optical filter lens which comprises cladding a core glass of essentially pure fused metal oxide with a thin layer of a binary oxide glass.

15. A method in accordance with claim 14 wherein the core glass is fused silica and the binary cladding glass is a SiO₂-TiO₂ glass containing up to 10% TiO₂.

16. A method in accordance with claim 13 which comprises forming molten particles of a SiO₂-TiO₂ composition, depositing the particles to form a cladding layer, forming molten particles of fused silica, depositing those molten particles of fused silica on the cladding layer to form a core layer, repeating the step of forming and depositing molten SiO₂-TiO₂ glass particles to form a second cladding layer on the surface of the core layer.