Glasses having non-frangible stress profiles

Glass without fragile stress distribution curve

GLASSES HAVING NON-FRANGIBLE STRESS PROFILES

This patent application claims the priority rights of U.S. Provisional Patent Application Serial No. 62/014372 filed on June 19, 2014 in accordance with the patent law. All citation methods are incorporated in this article.

This disclosure is about strengthened glass. More specifically, the present disclosure relates to strengthened glass that does not exhibit brittle behavior.

Chemically strengthened glass is widely used as protective glass for mobile devices, touch function displays, and the like. Generally speaking, non-fragile ion-exchange glass is better as the protective glass of the touch screen device, in order to reduce the risk of damage from small glass pieces due to the self-accelerating highly fragmented cracks, the highly fragmented cracks It is a characteristic of a highly fragile stress state. This state is often caused by the combination of excessive compressive stress and central tension in the sample. The recently revealed non-fragile standard based on the thickness-dependent maximum central tension (CT) is only applicable to relatively small thicknesses when the depth of compression (DOL) achieved by chemical strengthening is considerably smaller than the thickness of the sample. (Ie <0.8mm). For layer depths that are considerably greater than the thickness.

Provide glass that exhibits non-fragile behavior in areas where there may be a considerable amount of central tension without reaching friability. This area allows the compression depth to extend longer, where the defects that cause chipping are contained, even though the high central tension area in the sample does not make the glass fragile.

Provided is a strengthened glass that has a deep compression layer and does not exhibit brittle behavior (that is, the glass is non-fragile). The glass has a compression extending from the surface to at least about 0.08% of the total thickness of the glass The surface compression layer of the depth DOC and the compression stress CS and the physical central tension CT, of which CT-CS<img file="TWI697403B_D0001.tif" />350MPa.

Therefore, one aspect of the present disclosure is to provide a glass having a compression layer, a central region, and a thickness t. The compression layer extends from the surface of the glass to a compression depth DOC and is under the maximum compressive stress CS. The area has the largest physical central tension CT in the center of the glass, the central area extends from the center to the compression depth, and the thickness t is in the range from about 0.3 mm to about 1.0 mm, where the DOC<img file="TWI697403B_D0002.tif" />0.08. t and CT-CS<img file="TWI697403B_D0003.tif" />350MPa.

The second aspect of the present disclosure is to provide a glass having a compression layer, a central region, and a thickness t. The compression layer extends from the surface of the glass to a compression depth DOC and is under the maximum compressive stress CS. The central region There is a maximum physical central tension CT at the center of the glass, the central area extends outward from the center to the compression depth into the glass, and the thickness t is in the range from about 0.3 mm to about 1.0 mm. The compression depth DOC is greater than or equal to 0.08. t, and the glass has less than about 200J/m². The average elastic energy density in mm.

The third aspect of the present disclosure is to provide a glass comprising: a compression layer extending from the surface of the glass to a compression depth DOC, the compression surface layer has the maximum compressive stress CS; the center of the glass has the maximum physical central tension The central area of CT. The central area extends outward from the center of the glass to the compression depth. The glass has a thickness t in the range from about 0.3mm to about 1.0mm, where the DOC<img file="TWI697403B_D0004.tif" />0.08. t and CT-CS<img file="TWI697403B_D0005.tif" />350MPa. When 0.3mm<img file="TWI697403B_D0006.tif" />t<img file="TWI697403B_D0007.tif" />When 0.5mm, the physical central tension CT is greater than 0.681×(57-9.0×ln(<i>t</i>)+49.3×(ln(<i>t</i>))²). When 0.5mm<img file="TWI697403B_D0008.tif" />t<img file="TWI697403B_D0009.tif" />When 0.7mm, the physical central tension CT is greater than 0.728×(57-9.0×ln(<i>t</i>)+49.3×(ln(<i>t</i>))²). When 0.7mm&lt;t<img file="TWI697403B_D0010.tif" />At 1.0mm, the physical central tension CT is greater than<maths><img file="TWI697403B_D0011.tif" /></maths>

The fourth aspect of the present disclosure is to provide a glass comprising: a compression layer extending from the surface of the glass to a compression depth DOC, the compression surface layer has a maximum compressive stress CS; The center of the glass has a central area of the maximum physical central tension CT, and the central area extends from the center of the glass to the compression depth, wherein the glass has less than 200J/m². mm average elastic energy density; and thickness t in the range from about 0.3 mm to about 1.0 mm, where DOC<img file="TWI697403B_D0012.tif" />0.08. t. When 0.3mm<img file="TWI697403B_D0013.tif" />t<img file="TWI697403B_D0014.tif" />When 0.5mm, the physical central tension CT is greater than 0.681×(57-9.0×ln(<i>t</i>)+49.3×(ln(<i>t</i>))²). When 0.5mm<img file="TWI697403B_D0015.tif" />t<img file="TWI697403B_D0016.tif" />When 0.7mm, the physical central tension CT is greater than 0.728×(57-9.0×ln(<i>t</i>)+49.3×(ln(<i>t</i>))²), and when 0.7mm&lt;t<img file="TWI697403B_D0017.tif" />At 1.0mm, the physical central tension CT is greater than<maths><img file="TWI697403B_D0018.tif" /></maths>

These and other aspects, advantages, and salient features will become apparent from the following embodiments, drawings, and the scope of additional patent applications.

<p>100Glassware</p><p>110First surface</p><p>112Second Surface</p><p>120First compression zone</p><p>122Second compression zone</p><p>130Central area</p><p>d₁Compression Depth (DOC)</p><p>d₂Second Compression Depth (DOC)</p><p>tThickness</p>

Figure 1 is a schematic cross-sectional view of a chemically strengthened glass product; Figure 2 is an approximation of the erfc distribution curve characteristic of linear diffusion using CT_AAnd the calculated physical central tension CT (CT(erfc)) ratio chart; the third picture is the CT₁Shows the fragility limit CT diagram; the fourth diagram is the CT₃The graph showing the fragility limit CT; Fig. 5 is the refractive index distribution curve of transverse magnetic field (TM) and transverse electric field (TE) extracted by IWKB-based algorithm through Mirror coupling measurement; No. 6 The picture shows that at 440°C, it contains 50% NaNO by weight₃And 50% KNO₃The stress distribution curve of 0.4mm thick glass exchanged in the bath for 17.7 hours; Figure 7 is an example of the stress distribution curve extracted by the IWKB method; Figure 8 is the stress distribution of 0.55mm thick glass after double ion exchange Graph; Figure 9 is the TE and TM refractive index distribution curve of the dual ion exchange glass sample in Figure 8; Figure 10a is a photo showing strengthened glass products 1) showing fragile behavior when broken; and 2) showing non-fragile behavior when broken; and picture 10b showing non-fragile behavior when broken Photo of the strengthened glass sheet.

In the following description, in the several views shown in the drawings from beginning to end, the same reference signs indicate the same or corresponding parts. It should also be understood that unless otherwise specified, terms such as "top", "bottom", "outward", "inward" and the like are convenient terms and are not construed as restrictive terms. In addition, as long as a group is described as containing at least one of a group of elements and a combination of the above-mentioned elements, it is understood that the group can contain any number of these listed elements, or consist essentially of or consist of any number of these listed elements. These listed elements are composed either individually or in combination with each other. Similarly, when a group is described as being composed of at least one of a group of elements and a combination of the aforementioned elements, it is understood that the group can be composed of any number of these listed elements, whether individually Or combined with each other. Unless otherwise indicated, when a range of values is stated, the range of values includes both the upper and lower limits of the range and any range in between. As used herein, the indefinite article "a" and the corresponding definite article "the" mean "at least one" or "one or more" unless otherwise specified. It is also understood that the various features disclosed in this specification and the drawings can be used in any and all combinations.

The term "glassware" as used herein is used in its broadest sense to include any object made wholly or partly of glass. Unless otherwise specified, all ingredients are expressed in mole percent (mole%).

It is worth noting that the term "substantially" or "about" may be used herein to express the inherent uncertainty that may be attributed to any quantitative comparison, value, measurement, or other expression. These terms can also be used herein to indicate that the possible degree of difference between the expressed amount and the reference substance does not cause a change in the basic function of the subject substance in question. Therefore, glass that is "substantially free of MgO" It is a glass in which MgO is not actively or batchwise added to the glass, but can be present as an impurity in a very small amount.

With general reference to the drawings, especially Figure 1, it will be understood that these drawings are used for the purpose of describing specific embodiments and are not intended to limit the scope of the disclosure or the attached patent application. The drawings are not necessarily drawn to scale, and for the benefit of clarity and conciseness, certain features and certain views in the drawings may be illustrated on an exaggerated scale or in a schematic manner.

The terms "layer depth" and "DOL" used herein refer to the depth of the compressed layer measured by surface stress (FSM) measurement using a commercially available instrument such as FSM-6000.

The terms "compression depth" and "DOC" as used herein refer to the depth at which the stress in the glass changes from compression to tensile stress. In the DOC, the stress transitions from positive (compressive) stress to negative (tensile) stress and therefore has a zero value.

According to the convention commonly used in the technical field, compression is expressed as negative (&lt;0) stress, and tension is expressed as positive (&gt;0) stress. However, unless otherwise specified, this specification expresses the compressive stress CS as a positive or absolute value from beginning to end-that is, as described herein, CS=|CS|, and expresses the central tension or tensile stress as a negative value. Better visualize the compressive stress distribution curve described in this article.

Ion exchange is often used to chemically strengthen glass. In a specific example, the alkali metal cations in this cation source (such as a molten salt or "ion exchange" bath) are exchanged with smaller alkali metal cations in the glass to achieve compressive stress near the glass surface (CS ) Of the layer. For example, potassium ions from a cation source are often exchanged with sodium ions in the glass. The compression layer extends from the surface to a depth in the glass, and typically decreases from the maximum at the surface to 0 at the compression depth DOC.

In one embodiment, the strengthened glass described herein has a maximum compressive stress of at least about 150 MPa, and in some embodiments at least about 200 MPa. In certain embodiments, the compressive stress is less than about 250 MPa.

The cross-sectional schematic diagram of the flat ion exchange glass product is shown in Figure 1. The glass product 100 has a thickness<b><i>t</i></b>, The first surface 110, and the second surface 112. Although the embodiment illustrated in Figure 1 illustrates that the glass product 100 is a flat sheet or plate, the glass product may also have other configurations, such as a three-dimensional shape or a non-planar configuration. The glass article 100 has a first compression zone 120 that extends from the first surface 110 to a compression depth (DOC)<b>d</b>_<b>1</b>Enter the main body of the glass product 100. In the embodiment illustrated in Figure 1, the glass article 100 also has a second compression zone 122, which extends from the second surface 112 to a second compression depth (DOC)<b>d</b>_<b>2</b>. The glass product 100 also has a central area 130, which is from<b>d</b>_<b>1</b>Extend to<b>d</b>_<b>2</b>. The central area 130 is under tensile stress or physical central tension (CT), and the tensile stress or physical central tension (CT) balances or counteracts the compressive stresses of the areas 120 and 122. The depth of the first and second compression zones 120, 122<b>d</b>_<b>1</b>、<b>d</b>_<b>2</b>It can protect the glass product 100 from cracks caused by rapid impact on the first and second surfaces 110, 112 of the glass product 100, and at the same time, the compressive stress minimizes the cracks passing through the first and second compression regions 120, 122 depth<b>d</b>_<b>1</b>、<b>d</b>_<b>2</b>The possibility.

In some embodiments, the compression depth DOC is at least about 8% of the total thickness t of the glass article-ie DOC<img file="TWI697403B_D0019.tif" />0.8t- and in some embodiments, when the thickness t is greater than 0.75mm, the DOC<img file="TWI697403B_D0020.tif" />0.8t. In other embodiments, the compression depth DOC is at least about 9% of the thickness t (DOC<img file="TWI697403B_D0021.tif" />0.8t), and in some embodiments, when the thickness t is greater than 0.5mm, the DOC<img file="TWI697403B_D0022.tif" />0.9t.

The compressive stress CS and the layer depth DOL are measured using tools known in the art. Such tools include, but are not limited to, the surface stress test (FSM) carried out using commercially available instruments manufactured by Luceo Co., Ltd. (Tokyo, Japan) such as FSM-6000 or the like, and the title of "Chemical Strengthening Plate Standard Specification for Chemically Strengthened Flat Glass" ASTM 1422C-99 and ASTM 1279.19779 "Standard for non-destructive photoelastic measurement of edge and surface stress in annealed, thermally strengthened, and fully tempered flat glass Test method (Standard Test Method for Non-Destructive Photoelastic Measurement of Edge and Surface Stresses in Annealed, Heat-Strengthened, and Fully-Tempered The method of measuring compressive stress and layer depth described in "Flat Glass)", the content of the above method is fully incorporated herein by reference. The surface stress measurement relies on the accurate measurement of the stress optical coefficient (SOC), which is related to the birefringence of glass. The SOC is then measured by those methods known in the art, such as the fiber and four-point bending method. Both methods are described in the title "Standard Test Method for Measuring Glass Stress-Optical Coefficient" In the ASTM standard C770-98 (2008) of "Measurement of Glass Stress-Optical Coefficient", the content of the above method is incorporated herein by reference, as well as the bulk cylinder method.

In some embodiments, the relationship between CS and physical central tension CT can be approximated by the following expression: CT=(CS˙DOL)/(<b>t</b>-2DOL) (1), where t is the thickness of the glass product in micrometers (μm). In the various parts of the present disclosure, the unit of central tension CT and compressive stress CS herein is megapascals (MPa), the unit of thickness t is either micrometer (μm) or millimeter (mm), and the unit of layer depth DOL is micron. (μm).

For strengthened glass products in which the compressive stress layer extends to deeper depths in the glass, FSM technology may encounter contrast issues that affect the observed DOL value. At deeper DOL values, the contrast between TE and TM spectra may be insufficient, making the calculation of the difference between TE and TM spectra-and the determination of DOL-more difficult. In addition, FSM technology cannot measure the compressive stress distribution curve (that is, the change in compressive stress as a function of depth in the glass). In addition, FSM technology cannot measure the depth of the layer resulting from the ion exchange of certain elements (such as lithium).

The following techniques have been developed to produce more accurate measurements of the depth of compression (DOC) and compressive stress distribution curves of strengthened glass products.

In Rostislav V. Roussev et al. filed an application on May 3, 2012, entitled "Systems And Methods for Measuring the Stress Distribution Curve of Ion Exchange Glass" Measuring the Stress Profile of Ion-Exchanged Glass)" and claiming the priority of the U.S. Provisional Patent Application No. 61/489,800 filed on May 25, 2011 and having the same title, U.S. Patent Application No. 13/463,322 No. (hereinafter referred to as "Roussev I"), reveals two methods for capturing detailed and accurate stress distribution curves (stress as a function of depth) for tempered or chemically strengthened glass. The combined optical mode spectrum of TM and TE polarizations is collected by the coupling technology, and all are used to obtain detailed and accurate TM and TE refractive index distribution curves<i>n</i>_TM(z) and<i>n</i>_TE(z). The content of the above-mentioned application is fully incorporated herein by reference.

In one embodiment, the detailed refractive index profile is obtained from the mode spectrum using the Inverse Wenzell-Kramers-Brillouin (IWKB) method.

In another embodiment, the detailed refractive index distribution curve is calculated by adapting the measured mode spectrum to a numerically calculated spectrum in the form of a predefined function describing the shape of the refractive index distribution curve and obtaining the function from the best fit The parameters of the form are obtained. The detailed stress distribution curve S(z) is calculated from the difference between the obtained TM and TE refractive index distribution curves using the known stress-optical coefficient (SOC) value: S(z)=[n_TM(z)-n_TE(z)]/SOC (2).

Due to the small SOC value, the birefringence n at any depth z_TM(z)-n_TE(z) is the refractive index n_TM(z) and n_TE(z) A small portion of either (usually on the 1% level). Obtaining a stress distribution curve that is not significantly distorted due to noise in the measured mode spectrum requires the measurement of the effective refractive index of the mode to have an accuracy of 0.00001RIU. Although there are noise and/or poor contrast in the collected TE and TM mode spectra or mode spectra images, the method disclosed by Roussev I further includes a technique that is applied to raw data to ensure that the measured mode refractive index has such high accuracy. This technique includes noise averaging, filtering, and curve adaptation to find the extreme position corresponding to the pattern with sub-pixel resolution.

Similarly, Rostislav V. Roussev and others filed an application on September 23, 2013 with the title "Systems and Methods for Measuring Birefringence in Glass and Glass Ceramics (Systems and Methods for Measuring Birefringence in Glass and Glass). -Ceramics)" and advocated U.S. Patent Application No. 14/033,954 (hereinafter referred to as "Roussev II") with priority of U.S. Provisional Patent Application Serial No. 61/706,891 filed on September 28, 2012 and with the same title is disclosed for An apparatus and method for optically measuring birefringence on the surface of glass and glass ceramics (including opaque glass and glass ceramics). Unlike Roussev I, which identifies individual mode spectra, the method disclosed by Roussev II relies on careful analysis of the angular intensity distribution of the TM and TE light reflected from the interface of the sample in the measurement framework. The content of the above-mentioned application is fully incorporated in this article by reference.

Therefore, the correct contrast angle of the reflected light intensity distribution is far more important than in the traditional coupling stress measurement that only seeks individual mode positions. To this end, the methods disclosed by Roussev 1 and Roussev II include techniques for normalizing intensity spectra, including normalization to a reference image or signal, correction of detector nonlinearity, averaging multiple images to reduce image noise and speckle, and application Digital filtering to further smooth the intensity angular spectrum. In addition, one method includes forming a contrast signal that is otherwise standardized to correct for the fundamental difference in shape between the TM and TE signals. The above method relies on realizing two almost identical signals, and measuring the mutual displacement with sub-pixel resolution by comparing the part of the signal containing the steepest area. Using coefficients determined by the design of the device, including the geometry and refractive index of the lens, the focal length of the lens, and the pixel pitch on the sensor, the birefringence will be proportional to the mutual displacement. The stress is determined by multiplying the measured birefringence by the known stress-optical coefficient.

In another disclosed method, the derivative of the TM and TE signals is determined after applying some combination of the aforementioned signal adjustment techniques. The position of the maximum derivative of the TM and TE signals is obtained using the sub-pixel resolution, and the birefringence is proportional to the distance between the two maximum values, and the coefficients were previously determined by device parameters.

Associated with the requirement for correct intensity extraction, the device contains several enhancements, such as the use of a light scattering surface (static diffuser) close to or on the incident surface to improve the angular uniformity of the illumination, When the light source is coherent or partially coherent, use a moving diffuser to reduce speckles, and use light-absorbing coatings on part of the input and output surfaces of the scallop and the side of the scallop to reduce A spurious background that tends to distort the intensity signal. In addition, the device may include an infrared light source to enable measurement of opaque materials.

In addition, Roussev II reveals the wavelength range and attenuation coefficient of the studied sample, where the measurement is enhanced by the described method and equipment. The range is determined by α_sλ&lt;250πσ_sDefined, where α_sIs the optical attenuation coefficient at the measured wavelength λ, and σ_sIn order to use the expected stress value measured by the precision measurement that is typically required in the actual application. This wide range allows to obtain practical importance measurements at wavelengths where the large optical attenuation makes the pre-existing measurement methods unsuitable. For example, Roussev II revealed that the stress-induced birefringence of an opaque white glass-ceramic was successfully measured at a wavelength of 1550 nm with an attenuation greater than about 30 dB/mm.

Although it has been noted that FSM technology has some problems with deeper DOL values, FSM is still a useful traditional technology that can be used when it understands that there may be up to +/-20% error range at deeper DOL values. The terms "layer depth" and "DOL" used in this article refer to the DOL value calculated using FSM technology, while the terms "compression depth" and "DOC" refer to the compression layer measured by the method described in Roussev I &amp; II depth.

As mentioned above, glass products can be chemically strengthened by ion exchange. In this process, ions on or near the glass surface are replaced (or exchanged) with larger ions with the same valence or oxidation state. In those embodiments where the glass article contains alkali metal aluminosilicate glass, consists essentially of alkali metal aluminosilicate glass, or consists of alkali metal aluminosilicate glass, the amount of ions in the glass surface layer is larger Ions are monovalent alkali metal cations, such as Li+ (when present in glass), Na⁺, K⁺, Rb⁺, And Cs⁺. Alternatively, the monovalent cations in the surface layer may be replaced by monovalent cations other than alkali metal cations, such as Ag⁺Or the like.

The ion exchange process is usually performed by immersing the glass product in a molten salt bath containing larger ions that will exchange with smaller ions in the glass. Those with ordinary knowledge in the technical field will understand that the parameters used in the ion exchange process include, but are not limited to, the composition and temperature of the bath, the immersion time, the number of times the glass is immersed in the salt bath (or multiple baths), and various Use and addition of salt bath Steps such as annealing, washing, and the like are usually determined by the composition of the glass and the desired layer depth from the strengthening operation and the compressive stress of the glass. For example, the ion exchange of alkali metal-containing glass can be achieved by immersing in at least one molten bath containing salts such as, but not limited to, nitrates, sulfates, and chlorides of larger alkali metal ions. The temperature of the molten salt bath is generally in the range from about 380°C to as high as about 450°C, and the immersion time ranges from about 15 minutes to as long as about 40 hours. However, temperatures and immersion times different from those described above can also be used.

In addition, a non-limiting example of an ion exchange process in which glass is immersed in a variety of ion exchange baths with washing and/or annealing steps between the immersions is described in Douglas C. Allan et al. on October 22, 2013, with the title It is "Glass with Compressive Surface for Consumer Applications" and claims the priority of U.S. Provisional Patent Application No. 61/079,995 filed on July 11, 2008 In Patent No. 8,561,429, the glass is strengthened by multiple continuous ion exchange treatments immersed in salt baths of different concentrations; and Christopher M. Lee et al. were certified on November 20, 2012 with the title "Used Dual-stage ion exchange for chemical strengthening of glass (Dual Stage Ion Exchange for Chemical Strengthening of Glass)" and in U.S. Patent No. 8,312,739 claiming the priority of U.S. Provisional Patent Application No. 61/084,398 filed on July 29, 2008, in which glass is diluted by the use of effluent ions. Ion exchange is performed in one bath, and then immersed in a second bath with a lower effluent ion concentration than the first bath for strengthening. The contents of U.S. Patent Nos. 8,561,429 and 8,312,739 are fully incorporated herein by reference.

Compressive stress is generated by chemically strengthening the glass product, for example, by the ion exchange process described above, in which a plurality of second metal ions are used to exchange a plurality of first metal ions in the outer region of the glass product, so that the outer region contains The plurality of second metal ions. Each first metal ion has a first ion radius, and each second alkali metal ion has a second ion half path. The second ion radius is larger than the first ion radius, and the presence of the larger second alkali metal ion in the outer region generates compressive stress in the outer region.

At least one of the first metal ion and the second metal ion is an alkali metal ion. The first ion may be lithium, sodium, potassium, and rubidium ions. The second metal ion may be one of sodium, potassium, rubidium, and cesium, provided that the ion radius of the second alkali metal ion is greater than that of the first alkali metal ion.

This article describes chemically strengthened glass, such as Corning Gorilla® glass, which is used as protective glass for mobile electronic devices and touch-sensitive displays. In particular, the development of chemically strengthened glass focuses on the stress distribution curve with a large compression layer depth, which helps to reduce the possibility of blasting or breaking the fragile glass when the device is dropped on a hard, rough surface. sex. Due to self-accelerating highly fragmented cracks, such cracks eject glass fragments with a large amount of kinetic energy. The highly fragmented cracks are due to the characteristics of a highly fragile stress state generated by the combination of excessive compressive stress and central tension in the glass.

The fragile behavior is characterized by at least one of the following: the strengthened glass product (such as a plate or sheet) is broken into multiple small pieces (such as<img file="TWI697403B_D0023.tif" />1mm); the number of fragments formed on the glass product per unit area; multiple cracks are branched from the initial crack in the glass product; at least one fragment (for example, about 5cm or about 2 inches) is violently ejected at a specified distance from the original position ); and any combination of the aforementioned crushing (size and density), cracking and ejection behavior. The terms "fragile behavior" and "fragility" as used herein refer to those violent or high-energy fragmentation modes that strengthen glass products without any external restraints (such as coatings, adhesive layers, or the like). Although coatings, adhesive layers, and the like can be used in combination with the strengthened glass products described herein, these external restraints are not used to determine the fragility or fragile behavior of the glass product.

Examples of fragile behavior and non-fragile behavior when the strengthened glass product is hit by a sharp indenter point are shown in Figure 10a and Figure 10b. The point impact test used to determine fragile behavior includes A device delivered to the surface of a glass article that uses just enough force to release the internally stored energy present in the strengthened glass article. That is, the point impact force is sufficient to generate at least one new crack on the surface of the strengthened glass sheet, and make the crack extend through the compressive stress CS region (ie, layer depth) into the region under the central tension CT. The impact energy required to generate or form cracks in a strengthened glass sheet depends on the compressive stress CS and layer depth DOL of the product, and thus depends on the conditions under which the glass sheet is strengthened (that is, the conditions used to strengthen the glass by ion exchange). In addition, each ion exchange glass plate illustrated in Figures 10a and 10b is brought into contact with a sharp dart indenter (such as a SiC indenter) sufficient to extend the crack into the inner area of the glass plate. The system is under tensile stress. The force applied to the glass plate is just enough to reach the start of the inner area, allowing the energy to drive the crack from the tensile stress of the inner area, rather than the force of the dart hitting the outer surface. The degree of ejection can be determined, for example, by placing the glass sample in the center of the grid, hitting the sample, and measuring the ejection distance of each piece using the grid.

Refer to figure 10a, glass plate<b>a</b>Can be classified as fragile. In particular, the glass plate<b>a</b>Fragmented into a number of ejected small pieces, and showed a high degree of rupture branched from the initial crack, thereby producing these small pieces. About 50% of the fragments are less than 1mm in size, and it is estimated that about 8 to 10 cracks branch off from the initial crack. The glass sheet was also ejected about 5 cm from the original glass plate a, as shown in Figure 10a. Glassware exhibiting any of the three criteria described above (i.e., multi-crack branching, ejection, and extreme chipping) are classified as fragile. For example, if the glass only exhibits excessive branching, but does not exhibit the above-mentioned spouting or extreme fragmentation, the glass is still characterized as fragile.

glass plate<b>b</b>、<b>c</b>(Figure 10b) and<b>d</b>(Picture 10a) is classified as not fragile. In each of these samples, the glass sheet was broken into a small number of large pieces. For example, glass plate<b>b</b>(Picture 10b) Broken into two large pieces with no crack branches; glass plate<b>c</b>(Figure 10b) Broken into four pieces, and two cracks branch from the initial crack; and the glass plate<b>d</b>(Figure 10a) Break into four pieces, and two cracks branch from the initial crack. Based on the absence of ejected fragments (that is, there is no glass that is forced to eject more than 2 inches from the original position) Piece), no size<img file="TWI697403B_D0024.tif" />1mm fragments are visible to the naked eye, and the smallest number of crack branches are observed, the sample<b>b</b>、<b>c</b>,and<b>d</b>Is classified as not fragile or generally not fragile.

Based on the foregoing, the fragility index (Table 1) can be constructed to quantify the degree of fragile or non-fragile behavior of glass, glass ceramic, and/or ceramic products when hit by another object. Index numbers ranging from 1 for non-fragile behavior to 5 for highly fragile behavior have been assigned to describe different levels of fragility or non-fragility. Using the index, the fragility can be characterized by many parameters: 1) the percentage of the total number of fragments with a diameter (ie the largest size) less than 1mm ("fragment size" in Table 1); 2) per unit area (in this example, cm²) The number of fragments formed in the sample ("fragment density" in Table 1); 3) the number of cracks branched from the initial crack formed during impact ("crack branch" in Table 1); and 4) The percentage of the total number of fragments ejected from the original position exceeding about 5 cm (or about 2 inches) ("Ejection" in Table 1).

<tables><img file="TWI697403B_D0025.tif" /></tables>

If the glass product meets at least one criterion related to a specific index value, the fragility index is assigned to the product. Alternatively, if the glass product meets the criteria between two specific fragility levels, a fragility index range (for example, a fragility index of 2-3) can be assigned to the product. Glass products can be assigned the highest fragility index value, as determined from the various standards listed in Table 1. In many cases, it is impossible to determine each of the standard values listed in Table 1 (for example, the fragmentation density or the percentage of fragments that are ejected from the original position by more than 5 cm). Therefore, different standards are regarded as individual, alternative measures of fragility behavior and fragility index, so that glass products falling within a standard level will be designated as relative measures. The corresponding fragility and fragility index. If the fragility index based on any of the four standards listed in Table 1 is 3 or greater, the glass product is classified as fragile.

Apply the aforementioned fragility index to the sample shown in Figure 10a and Figure 10b, the glass plate<b>a</b>Fragmented into multiple ejected small pieces, and showed a large degree of rupture that branched from the initial crack to produce small pieces. About 50% of the fragments are less than 1mm in size, and it is estimated that there are about 8 to 10 cracks branching from the initial crack. Based on the standards listed in Table 1, the glass plate<b>a</b>It has a fragility index between about 4-5 and is classified as having a medium to high fragility.

Glass products with a fragility index of less than 3 (low fragility) can be considered non-fragile or substantially non-fragile. glass plate<b>b</b>、<b>c</b>,and<b>d</b>Each lacks fragments with a diameter of less than 1 mm, multiple branches from the initial crack formed upon impact, and fragments that are ejected from the original position by more than 5 cm. glass plate<b>b</b>、<b>c</b>,and<b>d</b>It is not fragile and thus has a fragility index of 1 (not fragile).

As previously discussed, the glass plate a (showing fragile behavior) and the glass plate observed in Figure 10a and Figure 10b<b>b</b>、<b>c</b>,and<b>d</b>The difference in behavior (showing non-fragile behavior) can be attributed to the difference in central tension CT between the test samples. The possibility of this fragile behavior is a consideration in the design of various glass products, such as protective panels or windows for portable or mobile electronic devices, and displays for information terminal (IT) devices. Portable or mobile electronic devices such as mobile phones, entertainment devices, and the like, and information terminal (IT) devices such as notebook computers. In addition, the compression layer depth DOL and the maximum compressive stress CS that can be designed or provided to the glass article are both limited by this fragile behavior.

Therefore, in some embodiments, the strengthened glass article described herein exhibits a fragility index of less than 3 when subjected to a point impact sufficient to break the strengthened glass article. In other embodiments, non-fragile strengthened glass products can achieve a fragility index of less than 2 or less than 1.

The recently revealed non-fragile standard based on the thickness-dependent maximum physical center tension CT is only when the depth of chemical strengthening (DOL) is greater than the thickness of the sample.<b>t</b>In a very small state (i.e. DOL&lt;0.1<b>t</b>) Only suitable for relatively small thickness (ie &lt;0.8mm). As described in this article, when DOL contains a larger proportion of the overall thickness<b>t</b>At this time, it is possible to have glass whose central tension is considerably larger than previously disclosed, but does not reach the limit of fragility. This additional non-fragile area allows the compression depth to be extended further without making the glass brittle, despite the development of high central tension within the sample. The increased depth of the compression layer allows defects that cause deeper fragmentation to be contained.

In one aspect of the present invention, the upper limit of the sum of CS and CT that allows the DOL to increase indefinitely without reaching the fragility limit is disclosed, including the central tension CT which is increased to a much higher limit than the recently known fragility limit of the CT Circumstances, this situation was disclosed in the recent patent application entitled "Strengthened Glass Articles and Methods of Making" issued by Kristen Barefoot et al. on April 9, 2013, US Patent No. 8,415,013 (Hereinafter referred to as "Barefoot I").

In one aspect, the upper limit of the sum of CS and CT is the same as K in the sample⁺The upper limit of the maximum spatial variation of the concentration is related. The spatial change is achieved by using K in the glass substrate⁺Single step ion exchange Na⁺Obtained, where Na⁺Or Na⁺And K⁺It is the only alkali metal ion in glass.

In another aspect, an additional fragility criterion based on the total stored elastic energy is introduced to allow the DOL to account for the thickness of the sample<b>t</b>Predict fragility stress conditions for a considerable part of the situation. In one embodiment, DOL&gt;0.1<b>t</b>, And in other embodiments DOL&gt;0.15<b>t</b>. Under these conditions, a single-step or two-step ion exchange is used to obtain fragility conditions controlled by the stress profile. In addition, the total stored elastic energy standard allows the correct control of the fragility of the stress profile obtained by simultaneous or multi-step ion exchange involving the relative diffusion of more than two ions.

The total elastic energy standard allows rapid, non-destructive quality control of fragility based on stress measurements, such as the coupling of single and double ion exchange compressive stress distributions with large layer depths.

Barefoot I describes the fragility limit of glass with a thickness of less than about 0.75mm. It was found that the extrapolation of the linear relationship that was found for a larger thickness and known earlier would underestimate the upper limit of the non-fragile design space. "Non-linear critical central tension CT₁"Is given by the empirical formula.

CT₁(MPa)<img file="TWI697403B_D0026.tif" />-38.7(MPa/mm)×ln(t)(mm)+48.2(MPa) (3), where t is the thickness of the sample. The CT measurement value compared with the above formula is calculated by the following formula_A(CS,DOL,t)=(CS×DOL)/(t-2DOL)(4).

The subscript "A" is added to CT to indicate that the above approximation used to find CT has been accepted in the field of chemically strengthened glass and is widely used in process and quality control. According to Barefoot et al., the fragility limit CT₁The range is from 48.2MPa with a thickness of 1mm substrate to 94.8MPa with a thickness of 0.3mm.

In the U.S. Provisional Patent Application No. 61/657,279 (hereinafter referred to as " Barefoot II'') reveals a higher non-linear fragility CT_ALimit. For the thickness range of 0.1mm-0.75mm, the fragility limit CT₃Formulated as a function of thickness CT₃(MPa)=57(MPa)-9.0(MPa/mm). ln(t)(mm)+49.3(MPa/mm). ln²(t)(mm) (5).

Will be between 0.3mm and 1mm (in CT₃In the case of 0.3mm to 0.75mm), the non-linear fragility limit of several thicknesses in the range of CT₁And CT₃Summarized in Table 2. Therefore, according to Barefoot I and Barefoot II, for thickness less than 0.75mm, CT_AGreater than CT₃Of glass poses an unacceptable risk of fragility (&gt;5%). Similarly, for thickness above 0.75mm, CT_AGreater than CT₁Of glass presents an unacceptable risk of fragility (&gt;5%).

For the glass used in Barefoot I and Barefoot II with a thickness in the range of 0.3mm to 0.5mm, when the DOL range is about 0.085<b>t</b>And 0.126<b>t</b>Between, in nominal pure KNO₃During the ion exchange period, fragility began to be observed. As can be seen from Figure 2, the DOL/<b>t</b>Between, CT_AFor physical CT (denoted as CT in the following formula_<i>phys</i>) Ranges from about 1.373 to about 1.469, flat Both are about 1.421. Therefore, CT/CT_AThe ratio ranges from about 0.681 to about 0.728, with an average value of about 0.704. Therefore, for 0.3-0.5mm (CT₃) Thickness, corresponding to the prior art CT_AThe physical CT limit of the limit is<maths><img file="TWI697403B_D0027.tif" /></maths>

For thicknesses ranging from about 0.5mm to about 0.75mm, the fragile DOL/<b>t</b>The ratio is in the range of 0.064-0.085, where CT_A/CT_<i>phys</i>The ratio is about 1.332 to about 1.374. Therefore, CT_<i>phys</i>/CT_AThe ratio ranges from about 0.728 to about 0.751, and the limit of fragility expressed in physical central tension can be compared with the CT of Barefoot II₃The relationship between the limits is defined as<maths><img file="TWI697403B_D0028.tif" /></maths>

For samples with a thickness greater than 0.75mm and not greater than 1.0mm, the relevant CT limit described in Barefoot I is CT₁. Fragile DOL/ in the Barefoot I instance<b>t</b>Ratio is usually in the range of 0.048 to 0.060, while CT_A/CT_<i>phys</i>The ratio ranges from about 1.302 to about 1.324, CT_A/CT_<i>phys</i>The inverse range of is from 0.755 to 0.768.

Therefore, for the thickness range 0.75mm&lt;<b>t</b><img file="TWI697403B_D0029.tif" />For 1.0mm, the fragility limit of physical CT can be derived from the empirical fragility limit of Barefoot II:<maths><img file="TWI697403B_D0030.tif" /></maths>

<tables><img file="TWI697403B_D0031.tif" /></tables><tables><img file="TWI697403B_D0032.tif" /></tables>

As described in this article, if the maximum CS and DOL in the glass are quite different from those described by Barefoot et al., for a glass with the same overall composition and thickness, it may be at a quite different CT_AThe value begins to appear fragile. Contains about 16 mol% Na₂O and almost no K₂O aluminosilicate glass, when it is at 390°C, it basically contains pure KNO₃When the ions in the bath are exchanged to a layer depth of about 36μm measured by FSM-6000 surface stress measurement, the 0.4mm thick glass substrate becomes fragile. The compressive stress generated by the same surface stress meter during the measurement is about 920MPa, while CT_AIt is about 101MPa. However, when it contains 37wt% NaNO at 440°C₃And 63wt% KNO₃At 11.7 hours of ion exchange in the bath, the same type of glass did not exhibit brittle behavior. Under these ion exchange conditions, the glass developed a CS of 301 MPa, a DOL of 114.7 μm measured by FSM-6000, and a CT of 202 MPa._A, Almost a CT with a thickness of 0.4mm₃The fragility limit (106.6MPa, Table 2) is twice as large. In another example, it was found that the same type of glass was fragile after 13.7 hours of ion exchange in the same bath and at the same temperature, and produced a compressive stress of 279 MPa and a layer depth of 120.6 μm DOL and CT._A=212MPa. Compared with the pure bath case where the layer depth is only 9% of the thickness, these experiments show how the CT determined by the formula used by Barefoot et al. can have a DOL of 30% of the thickness (0.3<b>t</b>) Is twice as large as the fragility limit.

In related experiments, a sample with a thickness of 0.50mm contained 37% NaNO by weight at 440°C.₃And 63% KNO₃After 15.3 hours of ion exchange in the ion exchange bath, it exhibits non-fragile behavior. The ion exchange sample has a CS of 304MPa, a DOL of 120.8μm, and a CT of 142MPa_A, Barefoot II CT with glass thicker than 0.5mm₃The limit of 86.9MPa (Table 2) is quite high.

In addition, it contains 45wt% NaNO at 440°C₃And 55% KNO₃No fragility was observed on the samples with ion exchange time in the bath for more than 25 hours, and the DOL of the ion exchange samples exceeded 150um. In one example, after ion exchange at 440°C for 21 hours, a 0.4mm thick sample obtained a CS of 213MPa, a DOL of at least 149.3μm, and a CT of at least 314MPa._A. In another example, after ion exchange at 440°C for 25.25 hours, a 0.5mm thick substrate yielded a CS of 221MPa, a DOL of at least 147μm, and a CT of at least 172MPa._A. After ion exchange at 440°C for 25.25 hours, a substrate with a thickness of 0.6mm yielded a CS of 254MPa, a DOL of at least 148μm, and a CT of at least 124MPa._A, Which is better than the 74.5MPa CT observed on 0.6mm thick glass₃Quite a lot. After the substrate with a thickness of 0.8 mm is ion exchanged under the same conditions, a CS of 272 MPa, a DOL of at least 144 μm, and a CT of at least 76 MPa are obtained_A. This is better than in the same thickness and CT₃56.8MPa CT observed on₁The value is quite large. A value of 59.3 MPa is observed at a thickness of 0.75 mm. A 1.0mm thick substrate has a CS of 278MPa (this is better than the 48.2MPa CT obtained with the same thickness)₁The value is quite large), DOL of at least 142μm, and CT of at least 55MPa_A。

Containing 50wt% NaNO at 440°C₃And 50wt% KNO₃After ion exchange in the bath for more than 30 hours, no fragility of the 0.4mm thick substrate was observed, and a layer depth exceeding 170um was achieved. Ion exchange in the same bath for 14 hours and 20 minutes resulted in a compressive stress of 235 MPa, a DOL of at least 111 μm, and a CT of at least 150 MPa._A. At 440°C in 50wt% NaNO₃/50wt% KNO₃After 16.7 hours of ion exchange in the bath, a compressive stress of 227 MPa and a DOL of at least 131 μm were measured, and the CT_AIt is at least 215MPa. For 17.7 to 20.7 hours, 25 hours, And 30 hours of ion exchange time, FSM-6000 cannot estimate DOL and CT_A, But DOL will be greater than 131μm, and CT_AWill be greater than about 215 MPa.

Due to the limited ability to measure DOL when DOL exceeds 100μm, especially when DOL exceeds about 130μm, the FSM-6000 instrument cannot estimate the depth of the slice and the CT of the deepest distribution curve._A. When the DOL is greater than about 100μm-and especially when the DOL is greater than 130μm-due to the limited ability of the instrument to resolve the dark lines of the mode spectrum (when the DOL is very large, the dark lines of the mode spectrum become very dense), FSM-6000 usually underestimates DOL.

In related experiments, samples with the same glass and larger thicknesses of 0.5, 0.6, 0.8, and 1.0 mm contained 50wt% NaNO at 440°C.₃And 50wt% KNO₃The in-bath ion exchange totals 26 hours and 43 hours. All samples are not fragile. Because the layer depth of these samples exceeds 150μm, it is impossible to measure DOL and CT on the FSM-6000 instrument_A。

In the above example, the DOL measured by FSM-6000 exceeds 0.1<b>t</b>, And the fragile CT was observed for the first time_AThe value is significantly higher than the CT determined by the empirical formula of Barefoot I and Barefoot II₁Fragility value.

As shown by the above example, when DOL&gt;0.1<b>t</b>Time, because the allowable physical CT exceeds the previously prescribed CT₁And CT₃Fragility limit, so the combination of relatively high CS and large DOL can be used to obtain stronger glass.

The actual physical central tension in the midplane of the sample is usually different from the approximate value CT_A, Because CT can be easily calculated based on known thickness and CS and DOL usually reported by FSM-6000_A, CT_AHas been widely adopted. Assuming that the relevant refractive index profile is a linearly truncated profile, FSM-6000 estimates DOL from the measured number of guided optical modes in the ion exchange layer. However, in practice, the refractive index profile is different from the linearly truncated profile, especially at the deep end of the profile.

In many cases, the distribution curve can be closely approximated by the complementary error function (erfc). This is usually the effective diffusion coefficient (interdiffusion coefficient) of ion exchange in the concentration distribution of the diffusing agent. The case where the variation of the concentration range spanned by the line is relatively small. This kind of glass is used in the glass described by Barefoot I and Barefoot II.⁺Exchange Na⁺Circumstance, Barefoot I and Barefoot II reveal the CT observed on those glass₁And CT₃Fragility limit. K⁺The central tension CT of the erfc-shaped distribution of the concentration can be proportional to the local K by considering the local change of a specific volume⁺The concentration is calculated by applying the required balance force. The application of the required balance force requires that the spatial integral of the stress in the compression region of the substrate is equal to and opposite in sign to the stress integral between the tension regions.

CT to be approximated_AThe ratio of the calculated real physical CT (CT(erfc)) to the characteristic of the linear diffusion erfc distribution curve as a function of the ratio of layer depth DOL to layer thickness is shown in Figure 2, where DOL is the same as calculated by FSM-6000 erfc-shaped refractive index profile, and FSM-6000 regards DOL as a linear truncated profile.

Suppose the concentration distribution curve of chemically enhanced ions follows the functional form of linear diffusion:<maths><img file="TWI697403B_D0033.tif" /></maths>Where x₀Is the effective penetration depth. x₀The following equation is related to the DOL measured by FSM<i>DOL</i>=1.3825<i>x</i>₀ (10)。

Then determine CT from CS from force balance:<maths><img file="TWI697403B_D0034.tif" /></maths>

The ratio of physical CT to CS depends on DOL in the following way:<maths><img file="TWI697403B_D0035.tif" /></maths>

On the other hand, CT_AThe FSM formula is<maths><img file="TWI697403B_D0036.tif" /></maths>Therefore, the traditional approximation CT_AThe ratio of physical CT to linear diffusion (erfc distribution curve) is:<maths><img file="TWI697403B_D0037.tif" /></maths>

Take CT_AFragility limit indicated by CT₁And the corresponding physical CT limit values are from CT₁Calculate, and assume that DOL is 0.03, 0.04, and 0.05mm generally measured by the surface stress meter FSM-6000. For thickness ranges&gt; 0.3mm, and the glass composition described by US Barefoot et al. is nominally pure KNO₃Ion exchange is performed in the medium, the CS is between about 700 and 900 MPa, and the DOL is greater than about 0.03 mm at the beginning of fragility. Just CT_AFor example, according to the prior art CT₁The area above the curve is fragile. This means that in terms of physical CT, for DOL=0.03mm, it means CT_erfcThe entire area above the continuous line is considered fragile by the prior art.

The boundary line that separates the fragile and non-breakable glass areas in the two-dimensional space of thickness and CT is shown in Fig. 3. Figure 3 includes CT based on Barefoot II_ADefine the separation line ((a) of Figure 3), and show three physical CTs represented and calculated for having the same CT as Barefoot I_AThe other lines of the erfc-shaped distribution curve. These lines are calculated for the different DOL measured by FSM-6000, and indicate that the glass revealed by Barefoot II is nominally pure KNO₃The typical DOL range of fragility appears after the medium ion exchange. In these lines, the CT represented by physical CT is drawn_AThe maximum CT limit of the curve corresponds to the minimum DOL (0.03mm; line b in Figure 3).

For thicknesses less than 0.75mm, the higher CT limit is described below. For thickness greater than 0.75mm, the space above the curve is expressed in CT_AOr the fragile glass condition represented by physical CT depends on the curve.

Pure KNO in nominal₃After medium ion exchange, based on CT₃The indicated standard shows the boundary line separating the fragile and non-fragile glass areas in the two-dimensional space of thickness and CT as shown in Fig. 4. Figure 4 includes CT_AThe defined separation line (Figure 3(a)) and three other erfc-shaped distribution curves represented by physical CT. These distribution curves have the same CT as line a_AAnd is calculated for FSM-6000 measurement Different DOL. These distribution curves represent the typical DOL range where fragility occurs in the glass disclosed by Barefoot II. Among the lines illustrated in Fig. 4, the CT represented by physical CT is drawn_AThe highest physical CT limit of the curve corresponds to the smallest DOL.

Due to pure KNO₃The fragility of ion exchange bath and thickness t&gt;0.3mm usually appears when DOL&gt;0.03mm, so the entire area above the curve (curve b) corresponding to DOL=0.03mm is based on the fragility area of Barefoot et al.

The specific glass composition used to illustrate the embodiments of the present disclosure is described by Timothy M. Gross filed an application on November 15, 2012, entitled "Ion Exchangeable Glass with High Crack Initiation Threshold" Crack Initiation Threshold)" U.S. Patent Application No. 13/678,013, and Timothy M. Gross filed on November 15, 2012, titled "Ion Exchangeable Glass with High Crack Initiation Threshold" High Crack Initiation Threshold) in the US Patent Application No. 13/677,805, both of the above two patents claim the priority of the US Provisional Patent Application No. 61/560,434 filed on November 16, 2011. This glass contains Na₂O is the main alkali metal oxide, and because K is not completely removed from the starting material₂O, there is negligible amount of K in the substrate₂O. In this case, using K⁺Ion exchange Na⁺Generally nonlinear diffusion occurs in the process, which has a low K⁺The interdiffusion coefficient is low in the concentration region, and in the K⁺Concentration accounted for K⁺And Na⁺The interdiffusion coefficients are considerably higher in those regions where the majority of the total concentration (&gt;25%). In this case, the shape of the erfc function does accurately represent the shape of the refractive index and the stress distribution curve, and a detailed nonlinear diffusion model is required to accurately describe the distribution curve and the relationship between the distribution curve and the ion exchange conditions. The use of IWKB-based algorithms to extract detailed stress distribution curves is described in the application by Rostislav V. Roussev et al. on May 3, 2012, entitled "System and Method for Measuring Stress Distribution Curves of Ion Exchange Glass ( Systems And Methods for Measuring the Stress Profile of Ion-Exchanged Glass)" and claimed that the U.S. provisional patent application with the same title was filed on May 25, 2011. In U.S. Patent Application No. 13/463,322 (hereinafter referred to as "Roussev I") with priority of application No. 61/489,800, the contents of the above patents are incorporated herein by reference in order to determine the stress Distribution curve. An example of the non-erfc-extracted refractive index profile curve of the actual substrate is shown in Figure 5. Figure 5 is the turning point of the optical mode captured by the IWKB-based algorithm through the coupling measurement to the last captured optical mode. The transverse magnetic field (TM) and transverse electric field (TE) refractive index profiles. The glass substrate contains 50% NaNO by weight at 440°C₃And 50% KNO₃Ion exchanged 0.4mm thick glass in the bath for 17.7 hours. The glass substrate composition is described in U.S. Patent Application No. 13/678,013. The shape of the refractive index profile is quite different from the erfc shape. Figure 6 shows 50% NaNO by weight at 440°C₃And 50% KNO₃The stress distribution curve of 0.4mm thick glass with ion exchange in the bath for 17.7 hours. The composition of the glass sample is described in US Patent Application No. 13/678,013. The stress distribution curve has a compressive stress of 219 MPa on the surface (depth=0 μm), a compression depth DOC of 78 μm, and a central tension CT of 86 MPa. For glass with a thickness of 0.4mm, this physical CT is considerably higher than the 62MPa physical CT limit taught by Barefoot et al. The difference CT-CS is about 86-(-219)=305MPa.

Integrating along the depth dimension, the elastic energy per unit area in the compression zone is estimated to be about 13.4J/m², And in the tension area is about 15.7J/m². Therefore, the total elastic energy is about 29.1J/m². Considering the thickness of 0.4mm, the total elastic energy per unit thickness is 72.8J/(m². mm).

By applying a force balance condition between the stress depth integral in the compression zone and the stress depth integral in the tension zone, the exact value of the actual physical central tension CT can be determined. In general, this physical CT should correspond to a physical CT based on erfc, and the physical CT based on erfc can be calculated under the condition of basically linear diffusion mentioned earlier. The stress distribution curve found by the IWKB method is usually limited by the smaller depth of the deepest turning point of the TM and TE optical modes in the waveguide region. When the DOL is very large, the stress distribution curve at depths close to these maximum depths will sometimes be disturbed by obvious noise. Therefore, a parabola is used to approximate the shape of the stress distribution curve between the compression depths DOC, and the The distribution curve has a stress at a larger depth where the depth of chemical penetration becomes substantially flat, which is substantially equal to the central tension from the depth to the center of the substrate. An example of the stress distribution curve extracted by the IWKB method is shown in Fig. 7. The solid line (line a) in Figure 7 represents a quadratic approximation used to imitate the shape of the distribution curve of the tension zone, and is used to accurately estimate the stress integral of the tension zone. The variable part of the stress distribution curve in the tension zone is represented by a parabola (dotted line (b) in Figure 7) extending between the compression depth (DOC), and the depth is equal to 1.15. DOL. For the specific glass described above, the depth to which the distribution curve becomes flat is about 1.15. DOL, where the DOL of the same ion exchange glass is measured by the FSM-6000 instrument. In those cases where the stress distribution curve can be captured with very low noise, the deepest part of the stress distribution curve has a stress close to the central tension, which can be found by the aforementioned force balance method between tension and compression. The force balance condition represents the fact that in the absence of external force, the shape of the sample remains unchanged over time.

For the specific example illustrated in Figures 4 and 5, the difference CT-CS is about 305 MPa, where the tensile stress is positive and the compressive stress is negative by traditional physical conventions. The thickness of 0.4mm glass contains 50wt% NaNO₃And 50wt% KNO₃In the process of ion exchange in the bath, fragility has never appeared, even when the ion exchange time exceeds 30 hours and the stress distribution curve (at a level where the measured signal is approximately equal to the noise level) is close to very high from both sides of the substrate. The same goes for being close to the center.

During and after the ion exchange without stress relaxation, at the maximum diffusing agent (K⁺) The concentration difference between the concentration and the minimum diffusing agent concentration in the center CT-CS=|CT|+|CS| is directly related to the composition of the ion exchange bath and the ion exchange temperature. This difference still depends to a large extent on the ion exchange time, until finally the distribution curve from the two ends of the substrate meets in the middle, and a measurable diffusing agent (K⁺Or K₂O) The concentration increases. At this point the concentration difference between the maximum concentration and the minimum concentration decreases, and the difference CT-CS thus begins to decrease beyond this point, even in the absence of stress relaxation. At temperatures below 450°C and in NaNO₃+KNO₃NaNO in ion exchange bath₃In the ion-exchange salt mixture composition with a fraction of &gt;30wt%, the stress relaxation is relatively small. In addition, with the ion The exchange time and FSM DOL (due to small stress relaxation) increase, the difference CT-CS decreases very slowly, and can be approximated as a constant. Therefore, what has been discovered is that for CT-CS<img file="TWI697403B_D0038.tif" />For 305MPa, even if the physical CT is substantially greater than any CT corresponding to the previously disclosed CT limit, the ion exchange substrate will not become fragile, and in fact, as long as the above inequality is observed, the substrate can never become fragile. This situation applies to all substrate thicknesses greater than or equal to the thickness of 0.4 mm described herein.

In addition, for the temperature at 440 °C containing 45wt% NaNO₃And 55wt% KNO₃The mixture was ion-exchanged for about 42 hours and had a substrate with poor CT-CS depending on the thickness ranging from about 311 MPa to about 324 MPa, and no fragility was observed.

In one aspect, regardless of the depth of the layer, a physical CT, DOL&gt;0.1 that is greater than the previously known fragility CT limit (curve a in Figure 3)<b>t</b>, And strengthened glass with CT-CS less than or equal to about 350 MPa (and in some embodiments less than or equal to about 340 MPa) does not exhibit brittle behavior. The difference of short ion exchange time CS-CT (i.e. 10μm<img file="TWI697403B_D0039.tif" />DOL_short<img file="TWI697403B_D0040.tif" />40μm) indicates that there is a moderate amount of stress relaxation.

In another aspect, instead of realizing the CT required to quickly extend the defect through the stretched central area of the glass, but when the DOL is large (usually DOL&gt;0.1<b>t</b>) By storing the amount of elastic energy to obtain a state that can limit the occurrence of fragility. Specifically, it can be obtained when DOL is greater than about 0.15<b>t</b>When the CT may exceed the previously disclosed fragile CT limit. If the amount of stored elastic energy in the compression and tension zone is not sufficient to form a large, new free surface during crack extension and bifurcation, fragility is prevented.

The elastic energy stored by the stress distribution curve is calculated according to the following equation<maths><img file="TWI697403B_D0041.tif" /></maths>in<i>v</i>Is the Poisson's ratio (0.22 for the exemplary glass composition described above), E is the Young's modulus (approximately 68 GPa for our example glass 5318), and σ is the stress.

The elastic energy of glass per unit area in each compression zone (area on each main outer surface of the substrate) is:<maths><img file="TWI697403B_D0042.tif" /></maths>

The elastic energy in the tension region from the compression depth to the center of the glass substrate is:<maths><img file="TWI697403B_D0043.tif" /></maths>

The total elastic energy stored in the substrate is twice the sum of the elastic energy of a single compression and tension zone, multiplied by 2 to calculate the two compression zones and half of the central tension zone appearing in the chemically strengthened substrate. The units of the different variables in the above equation are as follows: For stress: [<i>σ</i>]=<i>MPa</i>≡10⁶<i>N</i>/<i>m</i>² (18); For depth: [<i>x</i>]=<i>μm</i>=10^-6<i>m</i> (19); Elastic energy per unit area of substrate:<img file="TWI697403B_D0044.tif" />;as well as Elastic energy per unit substrate area per unit thickness: J/m². mm.

Use the quadratic approximation of the variable part of the stress in the tension region, and the chemical depth d_chem=1.15. DOL_FSM, Applying the force balance condition to produce the following specific formula for determining the physical CT of the specific glass composition and distribution curve under consideration:<maths><img file="TWI697403B_D0045.tif" /></maths>This is found by integrating the stress between the compression regions of the distribution curve, such as by the IWKB-based algorithm<maths><img file="TWI697403B_D0046.tif" /></maths>

The energy in the compression region is directly found by the square of the integral stress through the previously described definition of energy in the compression region (Equation 7). In the specific case of the variable part of the distribution curve in the quadratic approximation of the tension zone, the effective tension zone energy is expressed as:<maths><img file="TWI697403B_D0047.tif" /></maths>

Table 3 summarizes the results obtained for glasses with a thickness ranging from 0.4 to 1.0 mm. The glass contains about 45wt% NaNO at 440°C₃And about 55wt% KNO₃Ion exchange at different times in the bath. In these examples, the DOL measured by FSM-6000 ranges from about 0.14t to about 0.39<b>t</b>. As mentioned previously, the CT-CS difference ranges from about 311 MPa to at least 324 MPa. Depending on the thickness and DOL, the CS ranges from about 222 MPa to about 270 MPa. It was found that all the glass samples listed in Table 3 were not fragile. For all thicknesses, physical CT exceeds the fragility limit of physical CT corresponding to the previous technical limit, and CT_ACT that exceeds the prior art_ALimit.

<tables><img file="TWI697403B_D0048.tif" /></tables><tables><img file="TWI697403B_D0049.tif" /></tables>

Contain about 45wt% NaNO at 440°C₃And about 55wt% KNO₃After 21 hours of ion exchange in the bath, the sample with a thickness of 0.4mm showed a compression depth of 81.6μm, a physical CT of at least 102.7MPa, and 15.1J/m in the compression zone.²The storage elastic energy, and at least 8.9J/m in half of the tension area²The storage elastic energy. The total elastic energy is at least 48J/m², When normalized to thickness, the total elastic energy is at least 120J/m². mm. In this embodiment, for a thickness of 0.4mm, it is found that DOL is greater than about 0.1<b>t</b>The new non-fragile area. The physical CT is greater than the value of about 63MPa consistent with the previous disclosure. For the 0.4mm sample thickness of Barefoot I and Barefoot II, the physical CT is greater than 76MPa, which is the same as the CT_A=CT₃=106.6 unanimous.

In another example, for a thickness of 0.4mm, it contains about 50wt% NaNO at 440°C₃And 50wt% KNO₃The non-breakable glass produced by ion exchange in the bath for 26.5 hours has a CS of about 191 MPa, a CT of at least 94 MPa, and a DOC of about 85 microns.

In another example, at 440° C. containing about 50wt% NaNO₃And 50wt% KNO₃The 0.4mm thick sample with ion exchange in the bath for 26.5 hours is not fragile and has a CS of about 191 MPa, a CT of at least 94 MPa, and a DOC of about 85 μm. The physical CT is considerably higher than the physical value of 76MPa. For the same thickness, the physical CT corresponds to the previously revealed value CT_A=CT₃=106.6MPa.

Will contain about 40wt% NaNO at 440°C₃And 60wt% KNO₃After ion exchange in the bath, DOL&gt;0.1<b>t</b>And examples of non-breakable and breakable glass of various thicknesses are summarized in Table 4. The 42.6-hour ion exchange example has an FSM-type DOL that is substantially higher than 150 μm, and because it is difficult to resolve these closely spaced modes, some higher-order modes may not be detected. Therefore, the calculated value of DOL, physical CT, tension energy, and total elastic energy are lower bound estimates. The non-breakable example exhibited a CT-CS value as high as 334 MPa. In the three examples that are not fragile, for a thickness of 0.6mm (CT&lt;52MPa), 0.8mm (CT&lt;44.3MPa), and 1.0mm (CT&lt;38MPa), the physical CT generally exceeds the corresponding CT limit previously reported.

<tables><img file="TWI697403B_D0050.tif" /></tables>

In another example listed in Table 4, it was found that 40wt% NaNO₃And 60wt% KNO₃The 0.4mm thick glass sample with ion exchange in the bath for 21.5 hours is fragile, with about 56.2J/m²When the thickness is standardized, the total storage elastic energy is equivalent to about 140.4J/m²mm. Therefore, the newly discovered non-fragile area is characterized as having a thickness of less than 56.246.6 J/mm for a glass sample thickness of 0.4 mm²The storage elastic energy, and for all thicknesses, especially for thickness greater than or equal to 0.4mm, the elastic energy that is standardized to thickness is less than 140.4J/m²mm.

In another example listed in Table 3, obtained at 440 °C in 45wt% NaNO₃/55wt% KNO₃The 0.5mm thick glass sample that has been ion-exchanged in the ion exchange bath for 25.25 hours is a non-breakable glass with 9.6MPa CT. For a sample thickness of 0.5 mm and a DOL of 0.04 mm, this CT is significantly greater than the value reported by Barefoot I by about 56 MPa. CT of the sample listed here_AIt is estimated to be 183MPa, much larger than 86.9MPa CT₃(0.5mm). The DOC of the sample is as high as 91.6μm, and the energy in the compression zone is 18.7J/m², And the energy in the half-tension zone is at least 8.7J/m². The total storage elastic energy is at least 54.8J/m², When the thickness is standardized, the total storage elastic energy is at least 109.7J/m². mm. The CT-CS difference is about 316MPa.

Listed in Table 3 at 440°C, containing about 45% NaNO₃And 55% KNO₃The 0.6mm thick sample that was ion-exchanged in the bath for 25.25 hours was found not to be brittle. The ion-exchanged sample has a CS of about 248 MPa, a DOL of about 153 μm, a DOC of 98.6 μm, and a physical CT of at least 65.6 MPa. The latter is equivalent to the limit of about 51 MPa reported by Barefoot et al. in terms of physical CT and DOL of about 40 μm. many. CT_AIt is estimated to be 130MPa, which is higher than the 75.5MPa CT previously reported₃Quite a lot. The elastic energy is estimated to be 21.5J/m in the compression zone², And in the tension area is about 8.1J/m². The total elastic energy is about 59.4J/m², And the elastic energy per unit area and unit thickness is about 98.9J/m²mm.

The 0.6mm thick samples listed in Table 4 contain about 40wt% NaNO at 440°C₃And about 60wt% KNO₃The in-bath ion exchange for 25.7 hours was found to be not brittle. The ion-exchanged sample has a CS of about 255 MPa, a DOL of close to 150 μm, a DOC of 100 μm, and a physical CT of about 70.2 MPa even higher, which is considerably higher than the previously reported value of about 56 MPa. Similarly, the non-fragile sample exhibited a CT of 129.8MPa_A, Which is lower than the previously reported fragility limit CT_A=CT₃(0.6mm)=74.5MPa is quite large. The elastic energy in the compression zone is about 24.2J/m², And at least 39.4J/m in the tension half area². The total elastic energy is estimated to be at least 67.3J/m², And the elastic energy per unit area and unit thickness is at least 112J/m²mm.

The sample with a thickness of 0.8mm (Table 3) was stored at 440°C and contained 45wt% NaNO₃And about 55wt% KNO₃The in-bath ion exchange for 25.25 hours was found to be non-fragile and had a CS of about 268 MPa, a DOL of about 153 microns, a DOC of about 107 μm, and a physical CT of about 49 MPa. For a thickness of 0.8mm, the physical CT ratio corresponds to CT_A=CT₁The physical CT of 43.5MPa The fragility limit is higher. The elastic energy in the compression zone is 26.7J/m², And the tension half area has 7.2J/m²The elastic performance. The total elastic energy is about 67.7J/m², When the thickness is standardized, it is about 84.6J/m²mm.

Another sample listed in Table 4 with the same thickness of 0.8mm contains 40wt% NaNO at 440°C₃And about 60wt% KNO₃After 25.5 hours of ion exchange in the bath, it showed non-fragile behavior. The sample has a CS of about 281 MPa, a DOL of about 146 μm, a DOC of about 109 μm, and a physical CT of about 45 MPa. The latter is much larger than the prior art limit expressed in physical CT (43.5 MPa) with a thickness of 0.8 mm. The elastic energy in the compression zone is about 30.2J/m², And the tension half area is about 10.6J/m², Resulting in a total of about 77.1J/m². Elastic energy density, that is, the elastic energy per unit area and unit thickness is about 96.4J/m²mm. The difference CT-CS of this non-breakable glass is at least about 334 MPa.

Four examples of deep ion exchange on a 1mm thick substrate are also listed in Table 3. Ion exchange is at 440°C, containing about 45wt% NaNO₃And about 55wt% KNO₃In the bath. The ion exchange time was 25.25, 30, 36, and 42 hours, and the resulting physical CT values were estimated to be 39.3 MPa, 42.5 MPa, at least 44.9 MPa, and 48.4 MPa, respectively. The value may be slightly underestimated, especially the 36-hour ion exchange, because the DOL exceeds 160 μm, which poses a challenge to the accurate analysis of higher-order modes. CT_AThe value ranges from about 58.6 to about 76.2MPa, and they are all significantly higher than the CT limit of the prior art₁=48.2MPa. DOL ranges from about 143 μm to more than 170 μm, while DOC ranges from about 115 μm to about 136 μm. The range of the difference CT-CS is from about 313 MPa to about 325 MPa. The total storage elastic energy ranges from about 73.4J/m²To at least about 81.7J/m², The average energy density is 81.7J/(m². mm).

The samples with a thickness of 1.0mm in Table 4 were stored at 440°C containing about 40wt% NaNO₃And about 60wt% KNO₃Ion exchange in the bath for 42.6 hours. The resulting strengthened glass is not brittle, and has a CS of about 272 MPa and a physical CT of at least about 52.8 MPa, which is considerably larger than the estimated limit of 37 MPa for the physical CT fragility of 1 mm thick glass with a DOL of about 50 μm. CT of non-fragile samples_AIs about 80.2MPa, which is less than the limit of Barefoot I fragility CT_A=CT₁(1mm)=48.2MPa is quite high. DOL is estimated to be about 185μm or more, DOC is about 139μm, and the elastic energy in the compression zone is about 36.6J/m², In the half area of tension greater than about 10.4J/m². The total elastic energy is at least 49.9J/m²mm, and mean that the average elastic energy density is at least 49.9J/m²mm.

Examples show that when DOL occupies a considerable part of the thickness of the glass, the CT value for fragility will change with DOL, depending on the total elastic energy stored. In the case of dual ion exchange glasses with moderately compressed deep regions and highly compressed shallow regions (where the stress varies strongly with depth), the total elastic energy becomes even more important (Figures 7 and 8) . The samples depicted in Figures 7 and 8 are double ion exchanged 0.55mm thick glass. The first ion exchange step involves 40wt% NaNO at 450°C₃/60wt% KNO₃Soak in the molten mixture for 7.75 hours. The first ion exchange step produces a deep, slowly changing portion A of the stress distribution curve. In the second step, the glass contains about 99.5wt% KNO at 390°C₃And 0.5wt% NaNO₃The ions in the bath are exchanged for 12 minutes, resulting in shallow and steep area B of the stress distribution curve. A sample with this stress profile is unlikely to be fragile, but any significant or even slight additional ion exchange to increase the depth of the first or second zone will produce fragile glass. IWKB analysis shows CS of about 891MPa, DOC of about 70.6 microns, and physical CT of about 61MPa, which are similar to the corresponding CT_A=CT₃(0.55mm) Fragility limit of physical CT limit. The elastic energy is about 44.7J/m in the compression zone², And about 7.8MJ/m in the tension half area². The total elastic energy is about 105J/m², Which means that the average energy density is about 191J/m²mm. Which is greater than about 0.12<b>t</b>The highest average elastic energy density observed in non-fragile samples with a large chemical penetration depth, and CT_AThan CT₃The fragility limit of the prior art is quite large.

As described in this article, when DOL accounts for a considerable amount of glass thickness (ie<img file="TWI697403B_D0051.tif" />10%), the CT value of fragility can change with DOL, depending on the total elastic energy stored. When glass is strengthened by a two-step (or double) ion exchange process, the total elastic energy plays an even more important role. The glass has a deep region with moderate compression and a shallow surface region with high compression. The stress in the shallow surface region changes with depth. Very fast (Figure 8). Figure 8 is the stress distribution curve of 0.55mm thick glass subjected to double ion exchange. The first step involves 40wt% NaNO at 450°C₃And 60wt% KNO₃Melt mixing The ion exchange in the compound was 7.75 hours. The first step produces the deep and gradually changing part (A) of the stress distribution curve. In the second step, the glass contains about 99.5wt% KNO at 390°C₃And 0.5wt% NaNO₃The ions in the bath are exchanged for 12 minutes, resulting in a shallow and steep area of the stress distribution curve (B).

The sample with the stress profile shown in Figure 8 was found to be not brittle, but any significant additional ion exchange to increase the depth of the first or second zone will produce brittle glass. The IWKB analysis of the glass shows a CS of about 891MPa, a DOC of about 70.6μm, and a physical CT of about 61MPa. The latter is considerably larger than the fragility limit of the physical CT estimated based on the previous criteria of strengthened glass with a thickness of 0.55mm and DOL40μm .

The elastic energy of the sample shown in Figure 6 is about 44.7J/m in the compression zone², And about 9.5MJ/m in the area of tension². The total elastic energy is about 54.1J/m², Which means about 98.4J/m². The average energy density in mm. This is the highest average elastic energy density observed in non-breakable samples. It is estimated that the maximum average elastic energy density of non-breakable glass with thickness ranging from 0.4mm to 1mm is about 98J/m²mm and 116.5J/m²The latter value is the lowest value observed to be fragile for 0.4mm thick glass with large DOL.

In some embodiments, the elastic energy density is less than about 200 J/m². mm. In other embodiments, the elastic energy density is less than about 140 J/m². mm, and in yet other embodiments, the elastic energy is less than about 120J/m². mm.

Figure 9 shows the TE and TM refractive index profiles of the sample whose stress profile is shown in Figure 8. For using K⁺Ion exchange Na⁺In other words, the refractive index increases due to ion exchange, and the refractive index profile is a monotonic function of depth, which makes it convenient to use IWKB analysis to capture and evaluate the stress profile. The refractive index profile in Figure 9 shows that, in addition to the approximate surface compressive stress, DOL and FSM-6000 will significantly underestimate the depth of chemical penetration in the deep region, and will not provide relevant steepness in the case of the dual ion exchange (DIOX) profile. Direct information in shallow areas. This is because the widely used DOL reported by FSM-6000 assumes that the refractive index profile can have a single fixed A single linear segment of slope and single penetration depth is a good representation of what is calculated. Widely used CT calculated based on DOL and surface CS obtained using FSM-6000_AIt is often 2 to 3 times larger than the physical CT of the DIOX distribution curve, so it is not convenient to predict fragility. It should be clear that the analysis disclosed in this disclosure in terms of physical CT and storage elastic energy is far more effective than CT-based analysis._AThe standard has a wider range of applications.

In addition, ion exchange that does not increase the refractive index can be used in some cases (for example, in Li-rich₂O on the glass substrate with Na⁺Exchange Li⁺In the process) to obtain a stress distribution curve with a large compression depth. Although the traditional DOL based on the guided optical modulus measurement is not provided in these cases, the compressed depth DOC is still a physical quantity that can be measured and expressed by various optical rotation techniques and represents the depth of chemical strengthening. As can be seen from Table 1 and Table 2, for all examples of non-fragile glass whose physical CT exceeds the fragility limit of the prior art, the DOC is greater than 0.1<b>t</b>, Usually more than 0.12<b>t</b>, And most often exceed 0.15<b>t</b>。

When using salt composition and allowable in 10μm<img file="TWI697403B_D0052.tif" />DOL_short<img file="TWI697403B_D0053.tif" />Realize CT-CS at 40μm<img file="TWI697403B_D0054.tif" />At a temperature of 350MPa, regardless of the DOL, the non-fragile standard based on poor CT-CS can be equivalently restated as the non-fragile area with CT-CS&lt;330MPa. This allows the DOC to increase indefinitely without the risk of fragility. Similarly, the fragility standard for storing elastic energy should be less than about 233J/m². mm, and in some embodiments less than about 197J/m². mm can be applied to all kinds of DOC&gt;0.1<b>t</b>Glass, including Na may be useful⁺Ion exchange Li⁺, And also use Na⁺And K⁺Ion exchange Li⁺Rich in Li₂O glass. In this case, the standard 10μm<img file="TWI697403B_D0055.tif" />DOL_short<img file="TWI697403B_D0056.tif" />40μm can be standard 10μm<img file="TWI697403B_D0057.tif" />DOL_short<img file="TWI697403B_D0058.tif" />40μm instead, because DOL can not be defined in terms of FSM-6000.

Figure 9 is the TE and TM refractive index distribution curves of the double ion exchange 0.55mm thick glass sample shown in Figure 8. For using K⁺Ion exchange Na⁺, The refractive index decreases as a result of ion exchange. The refractive index profile is a monotonic function of depth, which makes it convenient to use IWKB analysis to capture and evaluate the stress profile. The refractive index profile of Figure 9 shows that in the double ion exchange In the case of the (DIOX) distribution curve, the DOL estimated by FSM-6000 will significantly underestimate the chemical penetration depth in the deep compression zone, and will not provide direct information about the steep and shallow areas, and can only roughly estimate the surface compressive stress. This is because the widely used DOL reported by FSM-6000 is calculated on the assumption that the refractive index profile is well represented by a single linear section with a single fixed slope and a single penetration depth. Widely used CT calculated based on DOL and surface CS_AIt is often 2 to 3 times larger than the physical CT of the DIOX distribution curve, so it is not convenient to predict fragility. Therefore, the analysis described in this disclosure in terms of physical CT and storage elastic energy is far more effective than CT-based analysis._AA wider range of applications of the prior technical standards. Layer depth DOL used in this DIOX example_FSMIs 75μm, while CT_AIs about 167MPa, which is lower than the prior art limit CT_A=CT₃(0.55)=80MPa is more than twice as large.

In some cases, ion exchange can be used (e.g. in Li-rich₂Na on the glass of O⁺Exchange Li⁺In the process of) obtaining a stress distribution curve with a large compression depth DOC without causing an increase. In these cases, DOL based on guided optical modulus measurement is not provided. However, the compression depth DOC is a physical quantity representing the depth of chemical strengthening, which can be measured by various optical rotation and refractive near field (RNF) techniques. As can be seen in Table 3 and Table 4, for all examples of non-breakable glass whose physical CT exceeds the fragility limit of the prior art, for a smaller thickness, the DOC is greater than 0.09<b>t</b>, Usually more than 0.12<b>t</b>, And most often exceed 0.15<b>t</b>(<b>t</b>Is the thickness).

Use salt composition and at 10μm<img file="TWI697403B_D0059.tif" />DOL_short<img file="TWI697403B_D0060.tif" />The temperature of 40μm can allow CT-CS value up to 350MPa. Regardless of the DOL, the non-fragile standard based on poor CT-CS can be equivalently restated as the non-fragile area with CT-CS&lt;330MPa, thus allowing unlimited DOC Increase without the risk of fragility. Similarly, the fragility standard for storing elastic energy should be &lt;233J/m². mm, and in some embodiments less than about 197J/m². mm can be applied to all kinds of DOC&gt;0.1<b>t</b>Glass, including Na may be useful⁺Ion exchange Li⁺, And also use Na⁺And K⁺Ion exchange Li⁺Rich in Li₂O glass. In this case, the standard 10μm<img file="TWI697403B_D0061.tif" />DOL_short<img file="TWI697403B_D0062.tif" />40μm can be standard 10μm<img file="TWI697403B_D0063.tif" />DOC_short<img file="TWI697403B_D0064.tif" />40μm instead, because DOL can not be defined by FSM-6000 data.

In another embodiment, a fragility criterion in the form of standardized total energy is provided. The normalized total energy is defined as:<maths><img file="TWI697403B_D0065.tif" /></maths>

In many of the above examples, when DOL&gt;0.1<b>t</b>At that time, especially when the thickness is 0.4mm, the fragility predicted based on the fixed CT limit starts to become inaccurate. In these cases, the total normalized energy provides a better prediction of fragile behavior. Although the total normalized energy value will vary with the mechanical parameters of the glass substrate (i.e. Poissons ratio)<i>v</i>And Young's modulus E) change, but it is reasonable to assume that these values fall in a relatively small range.

Therefore, in one embodiment, the central tension CT is higher than the limit CT₃(For thickness less than or equal to 0.75mm) or higher than the limit CT₁(For thickness greater than 0.75mm) ion-exchange glass products have a thickness of less than or equal to 37.5 x 10 per unit thickness³MPa²The total normalized elastic energy of μm. For a thickness of 0.4mm, CT_ASubstrates greater than 106.6MPa should be stored less than or equal to 15 x 10⁶MPa²Standardized elastic performance of μm.

Depending on the glass composition and the mechanical properties of the glass, the limit of the total standardized energy can be changed. However, these values fill the range of most glasses of interest and include practical limitations to avoid fragility.

In another embodiment, for a 0.4mm thick substrate, the total normalized energy is less than 7.5 x 10⁶MPa²μm. For other thicknesses, the standardized storage elastic energy per unit thickness is less than about 19 x 10³MPa²μm.

The glass articles described herein may include any glass chemically strengthened by ion exchange or consist of any glass chemically strengthened by ion exchange. In some embodiments, the glass is alkali metal aluminosilicate glass.

In one embodiment, the alkali metal aluminosilicate glass comprises or consists essentially of: at least one of aluminum oxide and boron oxide, and one of alkali metal oxides and alkaline earth metal oxides At least one of them, of which -15 mol%<img file="TWI697403B_D0066.tif" />(R₂O+R<i>'</i>O-Al₂O₃-ZrO₂)-B₂O₃<img file="TWI697403B_D0067.tif" />4 mol%, where R is one of Li, Na, K, Rb and Cs, and R<i>'</i>It is at least one of Mg, Ca, Sr, and Ba. In some embodiments, the alkali metal aluminosilicate glass comprises or consists essentially of: from about 62 mol% to about 70 mol% SiO₂; From 0 mol% to about 18 mol% Al₂O₃; From 0 mol% to about 10 mol% of B₂O₃; Li from 0 mol% to about 15 mol%₂O; from 0 mol% to about 20 mol% Na₂O; K from 0 mol% to about 18 mol%₂O; from 0 mol% to about 17 mol% of MgO; from 0 mol% to about 18 mol% of CaO; and from 0 mol% to about 5 mol% of ZrO₂. In some embodiments, the glass includes aluminum oxide and boron oxide and at least one alkali metal oxide, wherein -15 mol%<img file="TWI697403B_D0068.tif" />(R₂O+R<i>'</i>O-Al₂O₃-ZrO₂)-B₂O₃<img file="TWI697403B_D0069.tif" />4 mol%, where R is at least one of Li, Na, K, Rb and Cs, and R<i>'</i>Is at least one of Mg, Ca, Sr and Ba; of which 10<img file="TWI697403B_D0070.tif" />Al₂O₃+B₂O₃+ZrO₂<img file="TWI697403B_D0071.tif" />30 and 14<img file="TWI697403B_D0072.tif" />R₂O+R'O<img file="TWI697403B_D0073.tif" />25; Wherein the silicate glass contains or consists essentially of the following: 62-70 mol% of SiO₂; 0-18 mole% Al₂O₃; 0-10 mole% of B₂O₃;0-15 mole% Li₂O; 6-14 mole% Na₂O; 0-18 mole% K₂O; 0-17 mol% MgO; 0-18 mol% CaO; and 0-5 mol% ZrO₂. The glass is described in the U.S. patent application titled "Glasses Having Improved Toughness And Scratch Resistance" filed on November 25, 2008 by Matthew J. Dejneka et al. No. 12/277,573, and Matthew J. Dejneka et al. filed on August 17, 2012, entitled "Glasses Having Improved Toughness And Scratch Resistance" U.S. Patent No. 8,652,978 In No. 6, the above two patent cases all claim the priority of U.S. Provisional Patent Application No. 61/004,677 filed on November 29, 2008. The contents of all the above patent cases are fully incorporated into this article by reference.

In another embodiment, the alkali metal aluminosilicate glass comprises or consists essentially of: from about 60 mol% to about 70 mol% SiO₂; From 6 mol% to about 14 mol% Al₂O₃; From 0 mol% to about 15 mol% of B₂O₃; Li from 0 mol% to about 15 mol%₂O; from 0 mole% to Approximately 20 mole% Na₂O; K from 0 mol% to about 10 mol%₂O; from 0 mol% to about 8 mol% of MgO; from 0 mol% to about 10 mol% of CaO; from 0 mol% to about 5 mol% of ZrO₂; SnO from 0 mol% to about 1 mol%₂; From 0 mol% to about 1 mol% CeO₂; Less than about 50ppm As₂O₃; And less than about 50ppm of Sb₂O₃; Of which 12 mole%<img file="TWI697403B_D0074.tif" />Li₂O+Na₂O+K₂O<img file="TWI697403B_D0075.tif" />20 mol% and 0 mol%<img file="TWI697403B_D0076.tif" />MgO+CaO<img file="TWI697403B_D0077.tif" />10 mol%. In some embodiments, the alkali metal aluminosilicate glass comprises or consists essentially of: 60-70 mol% SiO₂; 6-14 mole% Al₂O₃; 0-3 mole% of B₂O₃; 0-1 mole% Li₂O; 8-18 mole% Na₂O; 0-5 mole% of K₂O; 0-2.5 mol% of CaO; greater than 0 to 3 mol% of ZrO₂; 0-1 mole% of SnO₂; And 0-1 mole% CeO₂, Of which 12 mole%&lt;Li₂O+Na₂O+K₂O<img file="TWI697403B_D0078.tif" />20mol%, and the silicate glass contains less than 50ppm As₂O₃. In some embodiments, the alkali metal aluminosilicate glass comprises or consists essentially of: 60-72 mol% SiO₂; 6-14 mole% Al₂O₃; 0-3 mole% of B₂O₃; 0-1 mole% Li₂O; 0-20 mole% Na₂O; 0-10 mole% K₂O; 0-2.5 mol% CaO; 0-5 mol% ZrO₂; 0-1 mole% of SnO₂; And 0-1 mole% CeO₂, Of which 12 mole%<img file="TWI697403B_D0079.tif" />Li₂O+Na₂O+K₂O<img file="TWI697403B_D0080.tif" />20mol%, and the silicate glass contains less than 50ppm As₂O₃And less than 50ppm of Sb₂O₃. The glass is described in US Patent No. 8,158,543 entitled "Fining Agents for Silicate Glasses" filed by Sinue Gomez et al. on February 25, 2009; Sinue Gomez et al. United States Patent No. 8,431,502 entitled "Silicate Glasses Having Low Seed Concentration" filed on June 13, 2012; and Sinue Gomez et al. on June 2013 In U.S. Patent No. 8,623,776, filed on 19th, entitled "Silicate Glasses Having Low Seed Concentration", all of the above-mentioned patents claim to be filed on February 26, 2008 Priority of the applied U.S. Provisional Patent Application No. 61/067,130. The contents of all the above patent cases are fully incorporated into this article by reference.

In another embodiment, the alkali metal aluminosilicate glass contains SiO₂And Na₂O, where the glass is at temperature T_35kpIt has a viscosity of 35 kpoise, where zircon decomposes to form ZrO₂And SiO₂Temperature T_{break down}Greater than T_35kp. In some embodiments, the alkali metal aluminosilicate glass comprises or consists essentially of: from about 61 mol% to about 75 mol% SiO₂; From about 7 mol% to about 15 mol% Al₂O₃; From 0 mol% to about 12 mol% of B₂O₃; From about 9 mol% to about 21 mol% Na₂O; K from 0 mol% to about 4 mol%₂O; from 0 mol% to about 7 mol% of MgO; and 0 mol% to about 3 mol% of CaO. The glass was described in an application filed by Matthew J. Dejneka et al. on August 10, 2010, entitled "Zircon Compatible Glasses for Down Draw" and claimed on August 29, 2009 The priority of the U.S. Provisional Patent Application No. 61/235,762 filed on Japan is in the U.S. Patent Application No. 12/856,840. The contents of the above patent cases are fully incorporated into this article by reference.

In another embodiment, the alkali metal aluminosilicate glass contains at least 50 mol% SiO₂And at least one modifier selected from the group consisting of alkali metal oxides and alkaline earth metal oxides, wherein [(Al₂O₃(Mol%)+B₂O₃(Mole%))/(Σ alkali metal modifier (mole%))]&gt;1. In some embodiments, the alkali metal aluminosilicate glass comprises or consists essentially of: from 50 mol% to about 72 mol% SiO₂; From about 9 mol% to about 17 mol% Al₂O₃; From about 2 mol% to about 12 mol% of B₂O₃; From about 8 mol% to about 16 mol% Na₂O; and K from 0 mol% to about 4 mol%₂O. In some embodiments, the glass comprises or consists essentially of: at least 58 mol% SiO₂; At least 8 mole% Na₂O; from 5.5 to 12 mole% of B₂O₃; And Al₂O₃; Where [(Al₂O₃(Mol%)+B₂O₃(Mole%))/(Σ alkali metal modifier (mole%)))&gt;1,Al₂O₃(Mol%)&gt;B₂O₃(Mol%),0.9&lt;R₂O/Al₂O₃&lt;1.3. The glass is described in an application filed by Kristen L. Barefoot et al. on August 18, 2010, under the title "Crack And Scratch Resistant Glass and Enclosures Made Therefrom" The United States Patent No. 8,586,492, Kristen L. Barefoot et al. filed on November 18, 2013, entitled Crack And Scratch Resistant Glass and Enclosures Made Therefrom" US Patent Application No. 14/082,847, both of the above two patents are claimed to be filed on August 21, 2009 Priority of the applied U.S. Provisional Patent Application No. 61/235,767. The contents of all the above patent cases are fully incorporated into this article by reference.

In another embodiment, the alkali metal aluminosilicate glass contains SiO₂, Al₂O₃, P₂O₅, And at least one alkali metal oxide (R₂O), of which 0.75<img file="TWI697403B_D0081.tif" />[(P₂O₅(Mol%)+R₂O(mol%))/M₂O₃(Mol%))<img file="TWI697403B_D0082.tif" />1.2, where M₂O₃=Al₂O₃+B₂O₃. In some embodiments, the alkali metal aluminosilicate glass comprises or consists essentially of: from about 40 mol% to about 70 mol% SiO₂; From 0 mol% to about 28 mol% of B₂O₃; From 0 mol% to about 28 mol% Al₂O₃; From about 1 mol% to about 14 mol% of P₂O₅; And from about 12 mol% to about 16 mol% R₂O; And, in some embodiments, from about 40 mol% to about 64 mol% SiO₂; From 0 mol% to about 8 mol% of B₂O₃; From about 16 mol% to about 28 mol% Al₂O₃; From about 2 mol% to about 12 mol% of P₂O₅; And from about 12 mol% to about 16 mol% R₂O. The glass is described in an application filed by Dana C. Bookbinder et al. on November 28, 2011, entitled ``Ion Exchangeable Glass with Deep Compressive Layer and High Damage Threshold (Ion Exchangeable Glass with Deep Compressive Layer and High Damage Threshold). )" and claiming the priority of U.S. Provisional Patent Application No. 61/417,941 filed on November 30, 2010 in U.S. Patent Application No. 13/305,271. The contents of all the above patent cases are fully incorporated into this article by reference.

In yet another embodiment, the alkali metal aluminosilicate glass contains at least about 50 mol% SiO₂And at least about 11 mol% Na₂O, and the compressive stress is at least about 900 MPa. In some embodiments, the glass further contains Al₂O₃And B₂O₃, K₂At least one of O, MgO and ZnO, of which -340+27.1. Al₂O₃-28.7. B₂O₃+15.6. Na₂O-61.4. K₂O+8.1. (MgO+ZnO)<img file="TWI697403B_D0083.tif" />0 mole%. In a specific embodiment, the glass comprises or consists essentially of: from about 7 mol% to about 26 mol% Al₂O₃; From 0 mol% to about 9 mol% of B₂O₃; From about 11 mol% to about 25 Mole% Na₂O; K from 0 mol% to about 2.5 mol%₂O; from 0 mol% to about 8.5 mol% of MgO; and from 0 mol% to about 1.5 mol% of CaO. The glass was described in an application filed by Matthew J. Dejneka et al. on June 26, 2012, entitled "Ion Exchangeable Glass with High Compressive Stress" and claimed in July 2011. The priority of the U.S. Provisional Patent Application No. 61/503,734 filed on January 1st is in the U.S. Patent Application No. 13/533,298. The contents of all the above patent cases are fully incorporated into this article by reference.

In other embodiments, the alkali metal aluminosilicate glass is ion-exchangeable and contains: at least about 50 mol% SiO₂; At least about 10 mole% R₂O, where R₂O contains Na₂O; Al₂O₃; And B₂O₃, Where B₂O₃-(R₂O-Al₂O₃)<img file="TWI697403B_D0084.tif" />3 mole%. In some embodiments, the glass includes: at least about 50 mol% SiO₂; At least about 10 mole% R₂O, where R₂O contains Na₂O; Al₂O₃, Where Al₂O₃(Mol%)&lt;R₂O (mol%); and 3-4.5 mol% of B₂O₃, Where B₂O₃(Mol%)-(R₂O(mol%)-Al₂O₃(Mol%))<img file="TWI697403B_D0085.tif" />3 mole%. In certain embodiments, the glass comprises or consists essentially of: at least about 50 mole% SiO₂; From about 9 mol% to about 22 mol% Al₂O₃; From about 3 mol% to about 10 mol% of B₂O₃; From about 9 mol% to about 20 mol% Na₂O; K from 0 mol% to about 5 mol%₂O; at least about 0.1 mol% of MgO, ZnO, or a combination of MgO and ZnO, where 0<img file="TWI697403B_D0086.tif" />MgO<img file="TWI697403B_D0087.tif" />6 and 0<img file="TWI697403B_D0088.tif" />ZnO6 mol%; and optionally, at least one of CaO, BaO, and SrO, of which 0 mol%<img file="TWI697403B_D0089.tif" />CaO+SrO+BaO<img file="TWI697403B_D0090.tif" />2 mole%. In some embodiments, the glass has a Vickers crack initiation threshold of at least about 10 kgf after ion exchange. Such glass is described in an application filed by Matthew J. Dejneka et al. on May 28, 2013, entitled "Zircon Compatible, Ion Exchangeable Glass with High Damage Resistance". In the US Patent Application No. 14/197,658, the application No. 14/197,658 was filed by Matthew J. Dejneka et al. on May 28, 2013 with the title "High resistance to damage Zircon Compatible, Ion Exchangeable Glass with High Damage Resistance" is a continuation of the U.S. Patent Application No. 13/903,433. Both of the above two patents claim that they were filed on May 31, 2012. Priority of Provisional Patent Application No. 61/653,489. The contents of these applications are fully incorporated herein by reference.

In some embodiments, the glass includes: at least about 50 mol% SiO₂; At least about 10 mole% R₂O, where R₂O contains Na₂O; Al₂O₃, Of which -0.5 mol%<img file="TWI697403B_D0091.tif" />Al₂O₃(Mol%)-R₂O (mol%)<img file="TWI697403B_D0092.tif" />2 mole%; and B₂O₃, And where B₂O₃(Mol%)-(R₂O(mol%)-Al₂O₃(Mol%))<img file="TWI697403B_D0093.tif" />4.5 mol%. In other embodiments, the glass has a zircon decomposition temperature equal to the temperature at which the viscosity of the glass is greater than about 40 kPoise, and the glass includes: at least about 50 mol% SiO₂; At least about 10 mole% R₂O, where R₂O contains Na₂O; Al₂O₃; And B₂O₃, Where B₂O₃(Mol%)-(R₂O(mol%)-Al₂O₃(Mol%))<img file="TWI697403B_D0094.tif" />4.5 mol%. In still other embodiments, the glass is ion-exchanged, has a Vickers crack initiation threshold of at least about 30 kgf, and includes: at least about 50 mol% of SiO₂; At least about 10 mole% R₂O, where R₂O contains Na₂O; Al₂O₃, Of which -0.5 mol%<img file="TWI697403B_D0095.tif" />Al₂O₃(Mol%)-R₂O (mol%)<img file="TWI697403B_D0096.tif" />2 mole%; and B₂O₃, Where B₂O₃(Mol%)-(R₂O(mol%)-Al₂O₃(Mol%))<img file="TWI697403B_D0097.tif" />4.5 mol%. Such glass is described in a U.S. patent application entitled "Ion Exchangeable Glass with High Damage Resistance" filed by Matthew J. Dejneka et al. on May 28, 2013 In No. 903,398, Application No. 903,398 claims the priority of U.S. Provisional Patent Application No. 61/653,485 filed on May 31, 2012. The contents of these applications are fully incorporated herein by reference.

In certain embodiments, the alkali metal aluminosilicate glass contains at least about 4 mol% P₂O₅, Where (M₂O₃(Mol%)/R_xO(mol%))&lt;1, where M₂O₃=Al₂O₃+B₂O₃, And the R_xO is the sum of monovalent and divalent cation oxides present in alkali metal aluminosilicate glass. In some embodiments, the monovalent and divalent cationic oxides are selected from Li₂O, Na₂O, K₂O, Rb₂O, Cs₂A group consisting of O, MgO, CaO, SrO, BaO and ZnO. In some embodiments, the glass contains 0 mol% B₂O₃. In some embodiments, the glass is ion-exchanged to a layer depth of at least about 10 μm and contains at least about 4 mol% P₂O₅, Where 0.6&lt;[M₂O₃(Mol%)/R_xO(mole%)]&lt;1.4; or 1.3&lt;[(P₂O₅+R₂O)/M₂O₃]<img file="TWI697403B_D0098.tif" />2.3; where M₂O₃=Al₂O₃+B₂O₃, R_xO is the sum of monovalent and divalent cation oxides in alkali metal aluminosilicate glass, and R₂O is the sum of the divalent cation oxides in the alkali metal aluminosilicate glass. The glass is described in the US patent application titled "Ion Exchangeable Glass with High Crack Initiation Threshold" filed by Timothy M. Gross on November 15, 2012. No. 13/678,013 and Timothy M. Gross filed on November 15, 2012, titled "Ion Exchangeable Glass with High Crack Initiation Threshold", the U.S. Patent Application No. In No. 13/677,805, the above two applications both claim the priority of U.S. Provisional Patent Application No. 61/560,434 filed on November 16, 2011. The contents of these applications are fully incorporated herein by reference.

In other embodiments, the alkali metal aluminosilicate glass includes: from about 50 mol% to about 72 mol% SiO₂; From about 12 mol% to about 22 mol% Al₂O₃; Up to about 15 mol% of B₂O₃; Up to about 1 mol% of P₂O₅; From about 11 mol% to about 21 mol% Na₂O; up to about 5 mol% of K₂O; up to about 4 mol% MgO; up to about 5 mol% ZnO; and up to about 2 mol% CaO. In some embodiments, the glass comprises: from about 55 mol% to about 62 mol% SiO₂; From about 16 mol% to about 20 mol% Al₂O₃; From about 4 mol% to about 10 mol% of B₂O₃; From about 14 mol% to about 18 mol% Na₂O; K from about 0.2 mol% to about 4 mol%₂O; up to about 0.5 mol% of MgO; up to about 0.5 mol% of ZnO; and up to about 0.5 mol% of CaO, wherein the glass is substantially free of P₂O₅. In some embodiments, Na₂O+K₂O-Al₂O₃<img file="TWI697403B_D0099.tif" />2.0 mol%, and in some embodiments, Na₂O+K₂O-Al₂O₃<img file="TWI697403B_D0100.tif" />0.5 mol%. In some embodiments, B₂O₃- (Na₂O+K₂O-Al₂O₃)&gt;4mol%, and in some embodiments, B₂O₃-(Na₂O+K₂O-Al₂O₃)&gt;1mol%. In some embodiments, 24 mol%<img file="TWI697403B_D0101.tif" />RAlO₄<img file="TWI697403B_D0102.tif" />45 mol%, and in other embodiments, 28 mol%<img file="TWI697403B_D0103.tif" />RAlO₄<img file="TWI697403B_D0104.tif" />45 mol%, where R is at least one of Na, K, and Ag. The glass is described in a U.S. patent application entitled "Fast Ion Exchangeable Glasses with High Indentation Threshold" filed on November 26, 2013 by Matthew J. Dejneka et al. In Case No. 61/909,049, the content of the above-mentioned application is incorporated into this article by reference.

In some embodiments, the glasses described herein are substantially free of arsenic, antimony, barium, strontium, bismuth, lithium, and at least one of the foregoing compounds. In other embodiments, the glass may include up to about 5 mol% Li₂O, and in some embodiments, the glass may include up to about 10 mol% Li₂O.

In some embodiments, the glasses described herein can resist defects introduced by sharp or sudden impacts after undergoing ion exchange. Therefore, these ion exchange glasses exhibit a Vickers crack initiation critical value of at least about 10 kilogram force (kgf). In certain embodiments, these glasses exhibit a Vickers crack initiation threshold of at least about 20 kgf, and in some embodiments, these glasses exhibit a Vickers crack initiation threshold of at least about 30 kgf.

In some embodiments, the glass described herein can be drawn by a process known in the art, such as slit drawing, fusion drawing, redrawing, and the like, and has a liquid content of at least 130 kpo Phase line viscosity. In addition to those listed above, various other ion exchange alkali metal aluminosilicate glass compositions can also be used.

Although typical embodiments have been described for illustrative purposes, the foregoing description should not be considered as a limitation on the scope of the present disclosure or the scope of the appended patent application. Therefore, persons with ordinary knowledge in the technical field can make various modifications, adaptations and substitutions without departing from the spirit and scope of the present disclosure or the scope of the appended patent application.