JP2015224150A - Method for producing alkali-free glass - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an alkali-free glass which has a high strain point, a high Young's modulus and excellent solubility during producing and can be easily subjected to float molding.SOLUTION: There is provided an alkali-free glass which has a Young's modulus of 87 GPa or more, a strain point of 680°C or more and an average thermal expansion coefficient at 50 to 350°C of 30×10to 47×10/°C, comprises, by mass% on the basis of oxide, 55 to 69 of SiO, 17 to 27 of AlO, 0 to 3 of BO, 0 to 20 of MgO, 2 to 20 of CaO, 0 to 2 of SrO, 0 to 2 of BaO and 0.01 to 1 of SnOand satisfies [SiO]×6.7+[AlO]+[BO]×4.4-458≤0, in which SiO+AlO+MgO+CaO is 95 or more and MgO+CaO+SrO+BaO is 12 to 23.

Description

  The present invention is a non-alkali glass which is suitable for display substrate glass and photomask substrate glass used in the production of various flat panel displays (FPD) and which can be float-molded substantially without containing an alkali metal oxide. And a manufacturing method thereof.

Conventionally, various display substrate glasses, particularly those in which a metal or oxide thin film is formed on the surface, have been required to have the following characteristics as shown in Patent Document 1, for example.
(1) When an alkali metal oxide is contained, alkali metal ions diffuse into the thin film and deteriorate the film characteristics, so that the alkali metal ions are not substantially contained.
(2) When exposed to a high temperature in the thin film forming process, the strain point is high so that the deformation (thermal shrinkage) associated with glass deformation and glass structural stabilization can be minimized.
(3) Sufficient chemical durability against various chemicals used for semiconductor formation. In particular, buffered hydrofluoric acid (BHF: mixture of hydrofluoric acid and ammonium fluoride) for etching SiO _x and SiN _x , and chemicals containing hydrochloric acid used for etching ITO, various acids used for etching metal electrodes (Nitric acid, sulfuric acid, etc.) Resistant to alkali of resist stripping solution.
(4) There are no defects (bubbles, striae, inclusions, pits, scratches, etc.) inside and on the surface.

In addition to the above requirements, in recent years, there are the following situations.
(5) The weight reduction of the display is required, and the glass itself is desired to have a low density glass.
(6) A reduction in the weight of the display is required, and a reduction in the thickness of the substrate glass is desired.

(7) In addition to the conventional amorphous silicon (a-Si) type liquid crystal display, a polycrystalline silicon (p-Si) type liquid crystal display having a slightly higher heat treatment temperature has been produced (a-Si). : About 350 ° C. → p-Si: 350 to 550 ° C.).
(8) In order to increase the temperature raising / lowering rate of the heat treatment for manufacturing the liquid crystal display to increase the productivity and the thermal shock resistance, a glass having a small linear expansion coefficient is required.

On the other hand, dry etching has progressed, and the demand for BHF resistance has become weaker. As the conventional glass, a glass containing 6 to 10 mol% of B ₂ O ₃ has been often used in order to improve the BHF resistance. However, B ₂ O ₃ tends to lower the strain point. Examples of non-alkali glass not containing B ₂ O ₃ or having a low content are as follows.

Patent Document 2 discloses a glass containing 0 to 3% by weight of B ₂ O ₃ , but the strain point of Examples is 690 ° C. or lower.

Patent Document 3 discloses a glass containing 0 to 5 mol% of B ₂ O ₃ , but the average thermal expansion coefficient at 50 to 300 ° C. exceeds 50 × 10 ⁻⁷ / ° C.

JP 2001-348247 A JP-A-4-325435 Japanese Patent Laid-Open No. 5-232458

  As the size of the FPD increases, there is a concern that deformation due to its own weight deflection occurs in the manufacturing process and yield decreases. Further, in order to sufficiently ensure the practical strength of a large FPD, it is useful to improve the fracture toughness of the substrate glass.

As a result of intensive studies, the inventors of the present application have found that alkali-free glass suitable for such applications tends to form a foam layer (hereinafter referred to as a foam layer) on the surface of the molten glass at the time of melting, and in particular SnO ₂ is used as a fining agent. It was found that it was remarkable when used. When the foam layer is present, bubbles in the glass cannot be sufficiently removed, and it becomes difficult to satisfy the quality requirement (4). In addition, when melting the glass raw material in the melting kiln, there is a case where combustion by a burner is used as a heat source, but when a foam layer is generated on the surface layer of the molten glass, heat is not efficiently transmitted to the lower surface of the molten glass, A large amount of time is required to dissolve the glass raw material. In addition, the heat reflected by the foam layer heats the upper furnace material of the melting kiln more than necessary, causing deterioration of the furnace material.

In addition, when bubbles with large diameter (hereinafter referred to as large bubbles) occur near the surface layer of the molten glass during melting, heat is not efficiently transferred to the glass around the large bubbles, and the temperature in the glass becomes inhomogeneous. This makes it difficult to satisfy the quality requirement (4). In addition, various gas components contained in the large bubbles diffuse into the glass around the large bubbles, causing a difference in composition between the large bubbles and the other glass, causing striae.
The large bubble in this specification refers to what is judged as a large bubble in the evaluation method of the large bubble in the Example mentioned later.
In addition, an above-described large bubble is what arises when the bubbles in an above-mentioned foam layer united, what is produced when bubbles adjoin when a bubble floats out from molten glass, and a bubble rises. This may be caused by the bubbles themselves expanding. It is considered that both the above-described foam layer and the above-described large foam are due to the presence of many bubbles in the molten glass.

An object of the present invention is to provide an alkali-free glass that has a high Young's modulus, a high strain point, hardly generates a foam layer or large bubbles even when SnO ₂ is used as a fining agent, and is easy to float. .

The present invention has a Young's modulus of 87 GPa or more, a strain point of 680 ° C. or more, an average coefficient of thermal expansion at 50 to 350 ° C. of 30 × 10 ^{−7 to} 47 × 10 ⁻⁷ / ° C., and an oxide. Reference mass% display
SiO ₂ 55~69,
Al ₂ O ₃ 17-27,
B ₂ O ₃ 0-3,
MgO 0-20,
CaO 2-20,
SrO 0-2,
BaO 0-2,
SnO ₂ 0.01 to 1,
SiO ₂ + Al ₂ O ₃ + MgO + CaO is 95 or more, MgO + CaO + SrO + BaO is 12 to 23, and [SiO ₂ ] × 6.7 + [Al ₂ O ₃ ] + [B ₂ O ₃ ] × 4.4 An alkali-free glass satisfying −458 ≦ 0 is provided.

  The alkali-free glass of the present invention is suitable as various display substrate glasses and photomask substrate glasses, but can also be used as a magnetic disk glass substrate. However, considering the demand for larger and thinner glass plates for various display substrate glasses and photomask substrate glasses, it has a high Young's modulus, so various display substrate glasses and photomask substrate glasses It is effective as

Although the alkali-free glass of the present invention contains SnO ₂ that acts as a fining agent, the generation of a foam layer when the glass raw material is dissolved is suppressed. Therefore, when using combustion by a burner as a heat source at the time of melt | dissolving a glass raw material with a melting kiln, there is no possibility of requiring a lot of time for melt | dissolution of a glass raw material. Moreover, there is no possibility that the upper furnace material of the melting kiln will deteriorate due to the heat reflected by the foam layer.

Further, when a two-stage melting process described later is used for melting the glass raw material, the time required for the foam layer to disappear between the melting process 1 and the melting process 2 is shortened. Thereby, the time required for clarification of molten glass is shortened. Moreover, the foam which remains in the molten glass conveyed downstream from a melting kiln decreases.
These effects are also exhibited when using energization heating as a heat source when melting the glass raw material in the melting kiln.

FIG. 1 is a graph showing the relationship between the retention time and the number of remaining bubbles for Examples 1, 2, and 10.

Next, the composition range of each component will be described. If SiO ₂ exceeds 69% (mass%, the same unless otherwise specified), the Young's modulus becomes low. In addition, the viscosity is increased, and there is a risk that bubbles may be mixed in due to an increase in melting temperature or bubbles that cannot be completely removed during clarification. Further, devitrification of mullite easily occurs, and the devitrification temperature T _L increases. If it is less than 55%, the thermal expansion coefficient will increase. Preferably it is 56-68%, More preferably, it is 57-67%, Most preferably, it is 58-65.

Al ₂ O ₃ suppresses the phase separation of the glass, lowers the thermal expansion coefficient, and raises the glass transition point Tg, but if it is less than 17%, this effect will not appear. In addition, the Young's modulus decreases, and the amount of heat shrinkage increases. Since Al ₂ O ₃ works as a network former like SiO _2, if it exceeds 27%, the viscosity increases, so that the melting temperature may increase and bubbles may be mixed. Further, devitrification such as mullite, anorthite, and spinel is likely to occur, and the devitrification temperature _TL may be increased. Preferably it is 17 to 26%, more preferably 18 to 25%, particularly preferably 18 to 24%.

B ₂ O ₃ is not essential, but can be contained up to 3% in order to improve the melting reactivity of the glass and to lower the devitrification temperature _TL . However, if it is too much, the strain point is lowered and the Young's modulus is also lowered. In addition, the environmental load increases. Therefore, it is preferably 2% or less, more preferably 1% or less, still more preferably 0.5% or less, and particularly preferably not substantially contained. “Substantially not contained” means not containing any inevitable impurities (hereinafter the same).

Although MgO is not essential, it can be contained in alkaline earths because it does not increase the expansion and does not excessively lower the strain point, improves solubility and increases the Young's modulus. However, if it exceeds 20%, the amount of heat shrinkage will increase. Further, devitrification such as cordierite and diopsite is likely to occur, and the devitrification temperature _TL increases. Preferably it is 2-19%, More preferably, it is 3-17%, More preferably, it is 3-15%, Most preferably, it is 5-12%.

  Since CaO improves solubility and contains together with MgO, the occurrence of devitrification can be suppressed, so it is necessary to contain 2% or more. However, if it exceeds 20%, the thermal expansion coefficient becomes large. It also causes an increase in the amount of heat shrinkage. Preferably it is 2-18%, More preferably, it is 3-16%, More preferably, it is 4-14%, Most preferably, it is 5-12%.

SrO is not essential, but can be contained up to 2% in order to improve the solubility without increasing the devitrification temperature T _{L of the} glass. However, if the amount is too large, the thermal expansion coefficient increases. In addition, when SrCO ₃ is used as the SrO raw material, the temperature at which carbon dioxide gas is released by thermal decomposition becomes high, which causes a foam layer or large bubbles to be generated when the glass raw material is melted. Accordingly, the content is preferably 2% or less, more preferably 1% or less, and particularly preferably substantially not contained.

BaO is not essential, but can be contained up to 2% in order to improve the solubility of the glass. However, if the amount is too large, the thermal expansion coefficient increases. The density also increases greatly. Further, when BaCO ₃ is used as a BaO raw material, the temperature at which carbon dioxide gas is released due to thermal decomposition becomes high, which causes a foam layer or large bubbles to be generated when the glass raw material is melted. Therefore, the content of BaO is preferably 2% or less, more preferably 1% or less, and particularly preferably not substantially contained.

In addition to the above-mentioned components, the glass of the present invention contains other oxides such as ZrO ₂ and ZnO in a total amount within a range not departing from the object of the present invention in order to adjust various mechanical properties, meltability and formability. Up to 3%. 1% or less is preferable, 0.5% or less is more preferable, 0.3% or less is more preferable, 0.2% or less is further more preferable, and it is particularly preferable that it is not substantially contained.

In the alkali-free glass of the present invention, SiO ₂ , Al ₂ O ₃ , MgO, and CaO are 95% or more in total amount (SiO ₂ + Al ₂ O ₃ + MgO + CaO). If it is less than 95%, it becomes impossible to achieve both a high strain point and a high Young's modulus. Preferably it is 97% or more, Most preferably, it is 99% or more.

When the total amount of MgO, CaO, SrO, BaO is less than 12% (MgO + CaO + SrO + BaO), the temperature T _{4 at} which the glass viscosity becomes 10 ⁴ dPa · s increases, and the float bath casing structure is formed during float forming. Otherwise, the life of the heater may be shortened extremely. 12.5% or more is preferable, and 13% or more is more preferable. If it exceeds 23%, there is a risk that the thermal expansion coefficient cannot be reduced. It is preferably 22.5% or less, and more preferably 22% or less.

The inventors of the present application have examined in detail the relationship between glass composition and physical properties of non-alkali glass having a SiO ₂ + Al ₂ O ₃ + MgO + CaO content of 95% or more and an Al ₂ O ₃ content of 17% or more. As a result, the present inventors have found that in order to suppress the generation of a foam layer when glass raw materials are dissolved while satisfying the required physical properties of the display substrate, each component needs to have a specific blending ratio shown below. That is, in addition to the above, by satisfying [SiO ₂ ] × 6.7 + [Al ₂ O ₃ ] + [B ₂ O ₃ ] × 4.4-458 ≦ 0, the strain point of the glass to be produced and Young's While maintaining the rate high, the formation of the foam layer can be suppressed when the glass raw material is melted. Preferably, [SiO ₂ ] × 6.7 + [Al ₂ O ₃ ] + [B ₂ O ₃ ] × 4.4-458 ≦ −5, and more preferably [SiO ₂ ] × 6.7 + [Al ₂ O ₃ ] + [B ₂ O ₃ ] × 4.4-458 ≦ −10. More preferably _{[SiO 2] × 6.7 + [} Al 2 O 3] + [B 2 O 3] × 4.4-458 ≦ -15, more preferably _{[SiO 2] × 6.7 + [} Al 2 O ₃ ] + [B ₂ O ₃ ] × 4.4-458 ≦ −20.

Alkali-free glass of the present invention contains 0.01% to 1% of the tin compound in terms of SnO _2. In the present specification, when described as SnO ₂ content refers to the content of the tin compound of the terms of SnO _2.
Tin compounds typified by SnO ₂ generate O ₂ gas in the glass melt.
In the glass melt, it is reduced from SnO ₂ to SnO at a temperature of 1450 ° C. or more, generating O ₂ gas and acting to grow bubbles greatly. During the production of the alkali-free glass of the present invention, the glass raw material is heated to 1500 to 1800 ° C. and melted, so that bubbles in the glass melt are more effectively increased. The tin compound in the raw material is prepared so as to be contained at 0.01% or more in terms of SnO _{2 with} respect to 100% of the total amount of the glass mother composition. When the SnO ₂ content is less than 0.01%, the clarification action during melting of the glass raw material is lowered. Preferably it is 0.05% or more, More preferably, it is 0.1% or more. If the SnO ₂ content exceeds 1%, the glass may be colored or devitrified. The content of the tin compound in the alkali-free glass is preferably 1.0% or less, more preferably 0.7% or less, still more preferably 0.5% in terms of SnO _{2 with} respect to 100% of the total amount of the glass mother composition. In the following, it is more preferably 0.4% or less, particularly preferably 0.3% or less.

Fe ₂ O ₃ contributes to infrared absorption in a temperature range of 1400 ° C. to 1800 ° C., which is a typical temperature at which the alkali-free glass of the present invention melts. Thereby, the efficiency at the time of melt | dissolving a glass raw material improves, and stability of the manufacturing process of an alkali free glass increases. Therefore, the alkali-free glass of the present invention desirably contains an iron compound in an amount of 0.005% or more in terms of Fe ₂ O ₃ . Meanwhile alkali-free glass of the present invention contains a tin compound which acts as a fining agent, the iron compound is present in the glass composition, the release of O ₂ gas by tin is inhibited by valence change of the iron. Therefore, the iron compound content is preferably 0.08% or less in terms of Fe ₂ O ₃ . More preferably, it is 0.06% or less, still more preferably 0.04 or less, and particularly preferably 0.02 or less.

In the alkali-free glass of the present invention, when the glass raw material is melted, the silica sand as the SiO ₂ raw material is melted at a lower temperature, and the unmelted silica sand does not remain undissolved in the glass melt.
If unmelted silica sand remains undissolved in the glass melt, the unmelted silica sand is taken into the bubbles generated in the glass melt, so that the clarification action during melting is reduced.
Also, if unmelted silica sand remains undissolved in the glass melt, the unmelted silica sand taken in by the bubbles gathers near the surface of the glass melt, so that the surface layer of the glass melt and parts other than the surface layer A difference in the composition ratio of SiO ₂ occurs between the glass and the homogeneity of the glass, and the flatness also decreases.

In order to improve the solubility, clarity and formability of the glass, the total amount of SO ₃ , F and Cl in the glass raw material is 0.5% or less, preferably 0.3% or less, more preferably 0.1% or less. Can be contained.

The glass of the present invention does not contain an alkali metal oxide in excess of the impurity level (ie substantially) in order not to cause deterioration of the characteristics of the metal or oxide thin film provided on the glass surface during panel production. The impurity level in this case is 2000 ppm or less on a mass basis. In order to facilitate recycling of the glass, it is preferable that PbO, As ₂ O ₃ and Sb ₂ O ₃ are not substantially contained.

The alkali-free glass of the present invention is preferably produced by the following method. First, the raw materials of each component normally used are blended so as to become target components. This is put into a melting furnace and melted by a melting step described later to obtain molten glass. A glass sheet can be obtained by forming this glass melt into a predetermined plate thickness by a float method, and cutting it after slow cooling.
Here, with respect to the glass melt before shaping | molding by the float glass process, it can degas using a vacuum degassing apparatus other defoaming apparatus as needed.

The melting step is a two-stage process including: a melting step 1 in which a raw material is charged into a melting furnace and heated to make the raw material into molten glass; and a melting step 2 in which the molten glass is further heated to defoam bubbles in the glass. It is preferable to employ the dissolution step. This is because oxygen due to SnO ₂ reduction reaction is generated all at once, and an alkali-free glass with fewer bubbles is obtained. Moreover, since the foam layer generated at the time of melting the glass raw material disappears between the melting step 1 and the melting step 2, the time required for refining the molten glass is shortened.

Then, the temperature of the raw material in the melting step 1 becomes molten glass (hereinafter, referred to as initial temperature) to a temperature of the molten glass in the melting step 2 (hereinafter, referred to reach temperature), at once by oxygen reduction reaction of SnO ₂ From the viewpoint of generating an alkali-free glass with fewer bubbles, it is preferably 50 ° C. or higher, more preferably 70 ° C. or higher, and still more preferably 90 ° C. or higher.

The initial temperature described above is 1400 to 1550 ° C. and preferably 1430 to 1530 ° C. from the viewpoint of effectively dissolving the raw material but still suppressing the generation of oxygen due to the reduction reaction of SnO _2. .
Further, when the ratio of Sn valence at the initial temperature (Sn-redox) is determined by, for example, a well-known wet analysis method by oxidation-reduction titration or Mossbauer spectroscopy, in the dissolution step 1, oxygen due to the reduction reaction of SnO ₂ From the standpoint of still suppressing the generation of Sn ²⁺ / (Sn ⁴⁺ + Sn ²⁺ ) in the alkali-free glass, the value of the ratio is preferably 0.25 or less, More preferably, it is 0.15 or less.
Here, the measurement of Sn-redox by Mossbauer spectroscopy can be performed by the method described in JP-A-2008-150228.

The ultimate temperature in the melting step 2 is to raise Sn-redox to generate oxygen at once by the reduction reaction of SnO ₂ , further lower the viscosity η to increase the bubble rising speed, and obtain an alkali-free glass with fewer bubbles. From the viewpoint of being obtained, it is 1500-1800 degreeC, and it is preferable that it is 1550-1800 degreeC.
The Sn-redox at the ultimate temperature is preferably 0.15 or more, more preferably 0.20 or more, and even more preferably 0.25 or more.
Further, the Sn-redox at the ultimate temperature is preferably 0.05 or more higher than the Sn-redox at the initial temperature, and more preferably 0.1 or more.

  Since the alkali-free glass of the present invention has a Young's modulus of 87 GPa or more, the fracture toughness is improved, and it is suitable for various display substrate glasses and photomask substrate glasses that require a larger or thinner glass plate. is there. 89 GPa or more is more preferable, and 91 GPa or more is more preferable.

In addition, since the alkali-free glass of the present invention has a specific elasticity (Young's modulus / density) of 33 GPa · cm ³ / g or more, the self-weight deflection is reduced. For this reason, there is little deformation resulting from its own weight deflection in the manufacturing process, and it is suitable for various display substrate glasses and photomask substrate glasses that require a larger or thinner glass plate. 34 GPa · cm ³ / g or more is more preferable, and 35 GPa · cm ³ / g or more is more preferable.

Since the alkali-free glass of the present invention has a strain point of 680 ° C. or higher, thermal shrinkage during panel production can be suppressed. Further, a laser annealing method can be applied as a method for manufacturing the p-Si TFT. 690 ° C. or higher is preferable, and 700 ° C. or higher is more preferable.
Since the alkali-free glass of the present invention has a strain point of 680 ° C. or higher, the glass fictive temperature tends to increase in the production process (for example, a thickness of 0.7 mm or less, preferably 0.5 mm or less, more preferably Suitable for organic EL display substrates or illumination substrates having a thickness of 0.3 mm or less, or thin display substrates or illumination substrates having a thickness of 0.3 mm or less, preferably 0.1 mm or less.
When forming a sheet glass having a plate thickness of 0.7 mm or less, further 0.5 mm or less, further 0.3 mm or less, and further 0.1 mm or less, the drawing speed at the time of forming tends to increase. Increases, and the amount of heat shrinkage of the glass tends to increase. In this case, heat shrinkage can be suppressed when the glass has a high strain point.

  The alkali-free glass of the present invention has a glass transition point Tg of preferably 730 ° C. or higher, preferably 740 ° C. or higher, and more preferably 750 ° C. or higher for the same reason as the strain point.

The alkali-free glass of the present invention has an average coefficient of thermal expansion at 50 to 350 ° C. of 30 × 10 ^{−7 to} 47 × 10 ⁻⁷ / ° C., has high thermal shock resistance, and has high productivity during panel production. it can. In the alkali-free glass of the present invention, the average thermal expansion coefficient at 50 to 350 ° C. is preferably 35 × 10 ⁻⁷ or more. The average thermal expansion coefficient at 50 to 350 ° C. is preferably 46 × 10 ⁻⁷ / ° C. or less, more preferably 45 × 10 ⁻⁷ / ° C. or less, and further preferably 44 × 10 ⁻⁷ / ° C. or less.

In the alkali-free glass of the present invention, the temperature T _{2 at} which the viscosity becomes 10 ² poise (dPa · s) is preferably 1760 ° C. or less, more preferably 1720 ° C. or less. If temperature T ₂ is within the above range, it is relatively easy dissolution of the glass raw material.

In the alkali-free glass of the present invention, the temperature T _{4 at} which the viscosity becomes 10 ⁴ poise (dPa · s) is preferably 1350 ° C. or less, more preferably 1330 ° C. or less, 1320 ° C. or less, and further 1310 ° C. In the following, it is particularly preferably 1300 ° C. or lower, and molding by the float process is possible.

In addition, the alkali-free glass of the present invention preferably has a devitrification temperature T _L of 1320 ° C. or less because molding by the float method becomes easy. More preferably, it is 1310 degrees C or less, More preferably, it is 1300 degrees C or less, Most preferably, it is 1290 degrees C or less. Further, the temperature T ₄ which is a measure of the float formability (glass viscosity η is 10 ⁴ poise (dPa · s) and consisting Temperature unit: ° C.) the difference between the liquidus temperature T _L (T ₄ -T _L) is Preferably they are -50 degreeC or more, -30 degreeC or more, Furthermore, 0 degreeC or more, More preferably, it is 10 degreeC or more, More preferably, it is 20 degreeC or more.
In this specification, the devitrification temperature is obtained by putting crushed glass particles in a platinum dish and performing heat treatment for 17 hours in an electric furnace controlled at a constant temperature. It is an average value of the maximum temperature at which crystals are deposited inside and the minimum temperature at which crystals are not deposited.

The alkali-free glass of the present invention preferably has a small amount of shrinkage during heat treatment. In manufacturing a liquid crystal panel, the heat treatment process differs between the array side and the color filter side. For this reason, particularly in a high-definition panel, when the thermal shrinkage rate of glass is large, there is a problem in that dot displacement occurs during fitting.
The thermal shrinkage rate can be evaluated by the following procedure.
First, after melting the target glass at 1500 ° C. to 1800 ° C., the molten glass is poured out, cooled into a plate shape, and then cooled. The obtained plate glass is polished to obtain a glass plate of 100 mm × 20 mm × 1 mm.
Next, the obtained glass plate is heated to the glass transition point Tg + 70 ° C., held at this temperature for 1 minute, and then cooled to room temperature at a temperature drop rate of 40 ° C./min. Thereafter, two indentations are made on the surface of the glass plate in the long side direction at intervals A (A = 90 mm) to obtain a sample before processing. In the present invention, the indentation is applied using a Vickers indenter, for example, a load of 50 g and an indentation time of 10 seconds.
Next, the pre-treatment sample is heated to 600 ° C. at a heating rate of 100 ° C./hour, held at 600 ° C. for 1 hour, and then cooled to room temperature at a cooling rate of 100 ° C./hour to obtain treated sample 1.
And the distance B between impressions of the sample 1 after a process is measured.
The thermal contraction rate is calculated from A and B thus obtained using the following formula.
Thermal contraction rate [ppm] = (A−B) / A × 10 ⁶
In the above evaluation method, the thermal shrinkage rate is preferably 70 ppm or less, more preferably 60 ppm or less, and even more preferably 50 ppm or less.

  Since the alkali-free glass of the present invention has relatively low solubility, it is preferable to use the following as a raw material for each component.

(Silicon source (SiO ₂ raw material))
Silica sand can be used as a raw material for SiO ₂ , but the ratio of particles having a median particle diameter D ₅₀ of 20 μm to 300 μm, a particle diameter of 2 μm or less is 0.3% by volume or less, and a particle diameter of 400 μm or more. Using silica sand with a volume of 2.5 vol% or less can suppress the agglomeration of the silica sand and melt it, facilitating the melting of the silica sand, less bubbles, high uniformity and flatness. Alkali glass is preferable because it is obtained.

The “particle size” in this specification is the equivalent sphere diameter of silica sand (meaning the primary particle size in the present invention), and specifically, in the particle size distribution of the powder measured by the laser diffraction / scattering method. Refers to particle size.
Further, in this specification, “median particle diameter D ₅₀ ” means that the volume frequency of particles larger than a certain particle diameter is 50% of that of the whole powder in the particle size distribution of the powder measured by the laser diffraction method. The particle diameter occupied. In other words, it refers to the particle diameter when the cumulative frequency is 50% in the particle size distribution of the powder measured by the laser diffraction method.
Further, “the ratio of particles having a particle diameter of 2 μm or less” and “the ratio of particles having a particle diameter of 400 μm or more” in this specification are measured by measuring the particle size distribution by a laser diffraction / scattering method, for example.

Further, if the median particle size D ₅₀ of silica sand is 300 μm or less, it is more preferable because melting of the silica sand becomes easier.

(Alkaline earth metal source)
An alkaline earth metal compound can be used as the alkaline earth metal source. Specific examples of the alkaline earth metal compound include carbonates such as MgCO ₃ , CaCO ₃ , BaCO ₃ , SrCO ₃ , (Mg, Ca) CO ₃ (dolomite), MgO, CaO, BaO, SrO and the like. Examples thereof include oxides and hydroxides such as Mg (OH) ₂ , Ca (OH) ₂ , Ba (OH) ₂ , and Sr (OH) ₂ . However, when BaCO ₃ or SrCO ₃ is used as the alkaline earth metal compound, carbon dioxide gas is released due to thermal decomposition even at a high temperature of 1000 ° C. or more, which causes a foam layer or large bubbles to be generated during the dissolution of the raw material. Therefore, it is desirable to use a hydroxide as a raw material for Ba and Sr.

(Tin source (Sn raw material))
Tin compounds are Sn oxides, sulfates, chlorides, fluorides, etc., but SnO ₂ is particularly preferred because it significantly increases bubbles. If the particle size of SnO ₂ is too large, SnO ₂ particles may not be completely dissolved in the glass material, so the average particle size (D ₅₀ ) of SnO ₂ is 200 μm or less, preferably 150 μm or less, more preferably 100 μm. The following. Further, if the particle size of SnO ₂ is too small, the SnO ₂ may agglomerate in the glass melt and remain undissolved, so it is preferably 5 μm or more, more preferably 10 μm or more.

  The raw materials were prepared so as to become the glasses of Tables 1 and 2, and melted at a temperature of 1650 ° C. using a platinum crucible. In melting, the mixture was stirred using a platinum stirrer to homogenize the glass. Next, the molten glass was poured out, formed into a plate shape, and then slowly cooled.

In Tables 1 and 2, the glass composition (unit: mass%) and strain point (unit: ° C) of Examples 1 to 11 (measured by the fiber method described in JIS R3103), glass transition point Tg (unit: ° C), 50 Average thermal expansion coefficient (unit: x 10 ^-7 / ° C.), specific gravity (unit: g / cm ³ ) (measured by Archimedes method), Young's modulus (unit: GPa) (measured by ultrasonic method) , Specific elastic modulus (unit: GPa · cm ³ / g), high temperature viscosity value, temperature T ₂ that is a measure of solubility (temperature at which glass viscosity η becomes 10 ² poise, unit: ° C.), and float moldability And T ₄ (temperature at which the glass viscosity η becomes 10 ⁴ poise, unit: ° C.) (measured with a rotational viscometer), devitrification temperature T _L (unit: ° C.) (measured by the above method) ), T ₄ -T _L (unit: ° C.), heat shrinkage (unit: p m) indicating the (measured by the above method). Here, Examples 1 to 6 are Examples, and Examples 7 to 11 are Comparative Examples. Table 1 shows the median particle size D ₅₀ , the proportion of particles having a particle size of 2 μm or less, and the proportion of particles having a particle size of 400 μm or more as the particle size of silica sand in the raw material used at this time. In addition, Tables 1 and 2 also show the foam layer thickness, the number of large bubbles, the number of initial bubbles, and the foam attenuation coefficient evaluated by the following procedure.
In Tables 1 and 2, the values shown in parentheses are calculated values.

[Method for evaluating foam layer thickness]
The raw materials were prepared so that 250 g of the glass of Examples 1 to 11 was obtained, melted at a temperature of 1500 ° C. for 1 hour using a platinum crucible having a diameter of 50 mm to 90 mm, and then held at a temperature of glass transition point Tg + 30 ° C. for 60 minutes. Thereafter, the glass was gradually cooled to room temperature (25 ° C.) at a rate of 1 ° C./min to obtain a glass for evaluation. Using this evaluation glass, the thickness of the foam layer on the upper surface of the glass was measured. The foam layer thickness was measured by the following procedure. First, the crucible central part of the glass for evaluation was cut into a diameter of 40 mm with a core drill, and a 30 mm portion from the upper surface of the glass was cut out as a cylindrical glass. A glass plate sample having a thickness of 1 mm was cut out from the cylindrical glass in parallel to the central axis including the central axis. Further, both surfaces of the cut surface when cut out were subjected to optical polishing (mirror polishing finish) to obtain a sample 1 for evaluation. The evaluation sample 1 was used for evaluation observation with a stereomicroscope in a direction perpendicular to the optical polishing surface. The observation area is every 0.5 mm such as a portion corresponding to 0 to 0.5 mm, a portion corresponding to 0.5 to 1.0 mm from the upper surface of the glass of the crucible (the surface where the glass was in contact with air during melting). In each observation area, the area where (the total projected area of bubbles on the observation surface) / (area of the observation area) is 0.8 or more is the foam layer, and the area that is not determined as the foam layer from the upper surface of the crucible glass The distance until it appeared was taken as the foam layer thickness.
In order to efficiently melt the glass raw material in the melting furnace and prevent deterioration of the upper furnace material of the melting furnace due to the heat reflected by the foam layer, the foam layer thickness is preferably 1.5 mm or less. It is more preferably 0 mm or less, and even more preferably 0.5 mm or less.
[Evaluation method for large bubbles]
Using the sample 1 for evaluation described above, in the region having a depth of 20 mm from the upper surface of the glass (large bubble confirmation region), a bubble having a major axis of 2 mm or more in the manner of evaluation observation described in the method for evaluating the foam layer thickness. It was a large bubble. The number of large bubbles in the large bubble confirmation region described above was measured as the number of large bubbles. The large bubbles described above are preferably absent.
[Evaluation method of initial bubble number and bubble damping coefficient]
The same raw material batch as in Examples 1 to 11 was melted at a temperature of 1500 ° C. for 1 hour using a platinum crucible, held at a temperature of 1600 ° C. for a predetermined time, and then held at 840 ° C. for 60 minutes, and then 1 ° C./min. The glass was gradually cooled to room temperature (25 ° C.) at a speed to obtain a glass for evaluation. Using this evaluation glass, processing similar to that of the above-described evaluation sample 1 was performed to obtain an evaluation sample 2. In the sample 2 for evaluation, the optical polishing surface was observed with a stereomicroscope for a portion corresponding to between 10 and 20 mm from the upper surface of the glass of the crucible, the number of bubbles having a diameter of 50 μm or more was measured, and the value was measured on the glass plate. Dividing by the volume, the number obtained was taken as the number of bubbles. The regression equation is obtained by setting the retention time at a temperature of 1600 ° C. to x, the number of bubbles to y, and y = A × exp (−B) x, and A is the initial bubble number (the retention time at the temperature of 1600 ° C. described above is 0 B) is the foam attenuation coefficient.
The initial number of bubbles is preferably 10,000 or less, more preferably 9000 or less, and even more preferably 8000 or less. The foam attenuation coefficient is preferably 0.030 or more, and more preferably 0.035 or more.

As is apparent from Tables 1 and ₂ , Examples 1 to 6, 10, and 11 in which SiO ₂ + Al ₂ O ₃ + MgO + CaO is 95% or more and Al ₂ O ₃ content is 17% or more are SiO ₂ Compared with Examples 7 to 9 where + Al ₂ O ₃ + MgO + CaO does not satisfy the above range, it has a high strain point of 680 ° C. or higher, a high Young's modulus of 87 GPa or more, and it is confirmed that there are no large bubbles. It was. Further, among Examples 1 to 6, 10, and 11 in which SiO ₂ + Al ₂ O ₃ + MgO + CaO is 95% or more and the content of Al ₂ O ₃ is 17% or more, [SiO ₂ ] × 6.7 + [ In Examples 1 to 6 that satisfy Al ₂ O ₃ ] + [B ₂ O ₃ ] × 4.4 to 458 ≦ 0, the foam layer thickness is reduced as compared with Examples 10 and 11 that do not satisfy the above range. It was confirmed that the solubility was improved.

Table 3 shows the number of initial bubbles and the bubble attenuation coefficient for Examples 1, 2 and 10. FIG. 1 is a graph showing the relationship between the retention time at the temperature of 1600 ° C. and the number of bubbles in Examples 1, 2 and 10. As is clear from this, in Examples 1 and 2, it was confirmed that the number of bubbles in the glass was well reduced and the clarity was improved with respect to the holding time at the temperature of 1600 ° C. compared to Example 10. It was done.

Table 4 shows the glass composition (unit: mass%) and physical properties of Examples 12 and 13 in which Fe ₂ O ₃ was changed. Examples 12 and 13 are examples. Table 4 shows the median particle size D ₅₀ , the proportion of particles having a particle size of 2 μm or less, and the proportion of particles having a particle size of 400 μm or more as the particle size of silica sand in the raw material used at this time. Table 4 also shows the foam layer thickness, the number of large bubbles, the initial number of bubbles, and the foam attenuation coefficient when the raw material is melted. Various characteristics were evaluated in the same tests as in Examples 1-11.
As is apparent from Table 4, Example 12 in which Fe ₂ O ₃ is 0.08% or less has a higher bubble attenuation coefficient than Example 13 in which Fe ₂ O ₃ exceeds 0.08%. It was confirmed that the clarity was further improved.

  The alkali-free glass of the present invention is suitable as various display substrate glasses and photomask substrate glasses, but can also be used as a magnetic disk glass substrate. However, as the substrate glass for various displays and the substrate glass for the photomask, considering that the glass plate needs to be enlarged or thinned, it has a high Young's modulus, and when exposed to high temperatures in the thin film formation process, Considering that it is required to minimize the dimensional change associated with glass deformation and glass structural stabilization, the amount of heat shrinkage is small, which is effective as substrate glass for various displays and substrate glass for photomasks.

Claims (6)

The Young's modulus is 87 GPa or more, the strain point is 680 ° C. or more, the average thermal expansion coefficient at 50 to 350 ° C. is 30 × 10 ^{−7 to} 47 × 10 ⁻⁷ / ° C., and the mass based on oxide In% display
SiO ₂ 55~69,
Al ₂ O ₃ 17-27,
B ₂ O ₃ 0-3,
MgO 0-20,
CaO 2-20,
SrO 0-2,
BaO 0-2,
SnO ₂ 0.01 to 1,
SiO ₂ + Al ₂ O ₃ + MgO + CaO is 95 or more, MgO + CaO + SrO + BaO is 12 to 23, and [SiO ₂ ] × 6.7 + [Al ₂ O ₃ ] + [B ₂ O ₃ ] × 4.4 Alkali-free glass satisfying −458 ≦ 0. The alkali-free glass according to claim 1, wherein the content of Fe ₂ O ₃ is 0.08% or less in terms of mass% based on oxide. The alkali-free glass according to claim 1 or 2, wherein the specific elasticity is 33 GPa · cm ³ / g or more.   The alkali-free glass according to claim 1, wherein the glass is melted at 1500 ° C. for 1 hour in a platinum crucible, held at a temperature of glass transition point Tg + 30 ° C. for 60 minutes, and then brought to room temperature at a rate of 1 ° C./minute. An alkali-free glass having a foam layer thickness of 1.5 mm or less from the upper surface of the glass when slowly cooled.   5. The alkali-free glass according to claim 1, wherein the glass is melted at 1500 ° C. for 1 hour in a platinum crucible, held at a temperature of glass transition point Tg + 30 ° C. for 60 minutes, and then brought to room temperature at a rate of 1 ° C./minute. An alkali-free glass in which bubbles having a major axis of 2 mm or more do not exist in a region 20 mm from the upper surface of the glass when slowly cooled. As the silicon source of the SiO ₂ raw material, the median particle size D ₅₀ is 20 μm to 300 μm, the proportion of particles having a particle size of 2 μm or less is 0.3% by volume or less, and the proportion of particles having a particle size of 400 μm or more is 2.5% by volume or less. The manufacturing method of the alkali free glass in any one of Claims 1-5 using the silica sand of this.