Glass cloth, prepreg, and printed circuit boards

Abstract

To provide a low dielectric glass cloth from which a prepreg excellent in insulation reliability can be produced, and a prepreg and a printed wiring board made of the low dielectric glass cloth.SOLUTION: A glass cloth including a glass yarn made of a plurality of glass filaments as warp and weft has a weight loss factor obtained as a product of a weight loss ratio derived from the glass component in heat treatment at 380°C for 2 hours and an average radius of the glass monofilament in the following equation (1) of 0.38 or more and 0.9 or less. The glass cloth has a whiteness degree of 95 or more. Weight loss factor=The weight loss ratio (%)×Average radius of the glass filament (μm) ... (1)SELECTED DRAWING: None

Description

[Technical Field]

[0001] This invention relates to glass cloth, prepreg, and printed circuit board. [Background technology]

[0002] With the recent development of the information and communication society, data communication and / or signal processing have become high-capacity and high-speed, leading to a significant reduction in the dielectric constant of printed circuit boards used in electronic devices. Therefore, many low-dielectric glass cloths have been proposed for use in the glass cloths that make up printed circuit boards.

[0003] For example, the low dielectric constant glass cloth disclosed in Patent Document 1 achieves this low dielectric constant by incorporating a large amount of B2O3 into the glass composition compared to conventionally used E glass cloth, while simultaneously adjusting the amount of other components such as SiO2. [Prior art documents] [Patent Documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2007-262632 [Overview of the Initiative] [Problems that the invention aims to solve]

[0005] Increasing the B2O3 content in glass yarn to reduce the dielectric constant of glass cloth lowers the elastic modulus of the glass yarn, making it more prone to breakage during the manufacturing process. Therefore, as described in Patent Document 1, glass fiber bundles are coated with a sizing agent during spinning and warping, and after weaving, a process called heat cleaning is performed to remove the sizing agent, which is an organic substance attached to the glass fiber bundles.

[0006] Conventional heat cleaning methods, as described in Patent Document 1, include, for example, batch heat cleaning methods at 350-500°C and heat cleaning methods that involve continuously passing the glass cloth through a heating furnace at high temperatures of 550-700°C. However, it has been found that when such heat cleaning methods are applied to low-dielectric glass cloth, the resulting glass cloth may contain prepregs with inferior insulation reliability.

[0007] The present invention has been made in view of the above-mentioned problems, and aims to provide a low-dielectric glass cloth that can provide a prepreg with excellent insulation reliability, as well as a prepreg and printed wiring board using the low-dielectric glass cloth. [Means for solving the problem]

[0008] As a result of diligent research to solve the above problems, the inventors of the present invention have found that the above problems can be solved by setting the whiteness to a predetermined level in glass yarn that has a predetermined weight reduction tendency, and have completed the present invention.

[0009] In other words, the present invention is as follows. [1] A glass cloth composed of glass threads made of multiple glass filaments as warp and weft threads, In the following formula (1), the weight loss coefficient, which is calculated as the product of the weight loss rate derived from the glass component during a heat treatment at 380°C for 2 hours and the average radius of the glass monofilament, is between 0.38 and 0.9. Weight reduction coefficient = Weight reduction rate (%) × Average radius of the glass filament (μm) ... (1) The whiteness is 95 or higher. Glass cloth. [2] The aforementioned glass cloth, The Si content is 40-60% by mass in terms of SiO2. The B content is 15-30% by mass, calculated as B2O3. The glass cloth according to [1]. [3] The Fe content is 0.001 to 0.10% by mass in terms of Fe2O2. The glass cloth according to [1] or [2]. [4] Having a dielectric constant of 5.0 or less at a frequency of 1 GHz. The glass cloth according to any one of [1] to [3]. [5] The glass cloth according to any one of [1] to [4], and a matrix resin impregnated in the glass cloth, having a prepreg. [6] Comprising the glass cloth according to any one of [1] to [4]. a printed wiring board. [Advantages of the Invention]

[0010] According to the present invention, it is possible to provide a low-dielectric glass cloth capable of providing a prepreg excellent in insulation reliability, and a prepreg and a printed wiring board using the low-dielectric glass cloth. [Embodiments for Carrying Out the Invention]

[0011] Hereinafter, embodiments of the present invention (hereinafter referred to as "the present embodiment") will be described in detail, but the present invention is not limited thereto, and various modifications are possible without departing from the gist thereof.

[0012] [Glass Cloth] The glass cloth of the present embodiment is a glass cloth composed of glass yarns made of a plurality of glass filaments as warp and weft, and in the following formula (1), the weight reduction coefficient obtained as the product of the weight reduction ratio derived from the glass component in the heat treatment at 380 ° C. for 2 hours and the average radius of the glass filaments is 0.38 or more and 0.90 or less, and the whiteness is 95 or more. Weight reduction coefficient = weight reduction ratio (%) × average radius of glass filaments (μm) ··· (1)

[0013] It has been found that when using low-dielectric glass cloth obtained through conventional heat cleaning, printed circuit boards with poor insulation reliability can be obtained. The greater the low dielectric effect of the glass cloth, the worse the insulation reliability tends to be. The reason for this is not clear, but it is thought that one of the causes is that the sizing agent, which should be removed by heating during heat cleaning, produces flame-retardant reaction products that remain on the glass cloth without being removed. In other words, it is inferred that the components constituting the low-dielectric glass cloth exert some kind of catalytic effect, causing the sizing agent to be modified into a flame-retardant component, making it difficult to remove. This situation was not apparent in conventional E-glass cloth, etc. It is thought that the decrease in insulation reliability when the amount of residue is similar tends to be more pronounced in low-dielectric glass cloth compared to E-glass cloth, etc. Furthermore, since the components constituting the glass of low-dielectric glass cloth tend to volatilize easily at high temperatures, there is also the limitation that it is difficult to perform sufficiently high-temperature treatment during heat cleaning. Therefore, in this embodiment, the glass cloth is specified to have a predetermined weight reduction tendency.

[0014] Furthermore, in the present embodiment, the whiteness of the glass cloth is also adjusted. The glass cloth of the present embodiment is obtained through a desizing process that removes the sizing agent attached to the glass filaments of the glass cloth, as will be described later. In this desizing process, the sizing agent attached to the glass filaments is removed by heat-treating the glass cloth. However, in the case of a glass cloth that is difficult to heat-treat at a high temperature as described above, the removal of the sizing agent may be insufficient and may remain on the glass cloth. When examining this residue, it was found that there is little residue of the sizing agent on the surface of the glass cloth, and even if there is residue, there are many whiteish ash-like substances that are powdery and easy to physically remove. On the other hand, in the intertwined parts of the glass filaments such as the weave (the intersection points of the warp and weft), it has been found that there are many thick and dark-colored substances that are difficult to remove. This is presumably because the sizing agent is difficult to remove in the intertwined parts of the glass filaments, and viscous substances (flame-retardant reaction products) are generated as the heat treatment progresses. Further investigation revealed that such viscous substances have relatively high hygroscopicity, which improves the hygroscopicity of the glass cloth and reduces the insulation reliability. In particular, it has been found that in the case where the weight loss tendency is a predetermined value, the part where the components have volatilized and escaped from the glass filaments is prone to moisture absorption, and when this viscous substance adheres to it, it causes a further reduction in insulation reliability.

[0015] In the present embodiment, whiteness is used as an index indicating the amount of such viscous substances, and in a glass cloth that satisfies a predetermined weight loss tendency, it is defined that the whiteness is not less than a predetermined value. Thereby, it is possible to provide a glass cloth that can provide a printed wiring board having excellent insulation reliability. Hereinafter, the configuration of the present embodiment will be described in more detail.

[0016] (Weight loss coefficient) The weight loss coefficient (hereinafter, also simply referred to as the "weight loss coefficient") obtained as the product of the weight loss ratio derived from the glass component and the average radius of the glass filaments when the glass cloth is heat-treated at 380°C for 2 hours is 0.38 or more and 0.90 or less, preferably 0.42 or more and 0.85 or less, and more preferably 0.46 or more and 0.80 or less.

[0017] "Weight loss percentage due to glass components" means that the weight loss percentage after heat treatment at 380°C for 2 hours is due to the loss of glass components due to volatilization during the heat treatment. As described later, in this embodiment, if surface treatment agents such as silane coupling agents or a large amount of organic impurities are attached to the glass cloth, the weight loss percentage is calculated after removing the surface treatment agents such as silane coupling agents and organic impurities that are physically adsorbed with a good solvent such as alcohol or acetone. Therefore, the weight loss percentage of the glass cloth after heat treatment, after removing such adhering components that decompose at 380°C, is the weight loss percentage due to glass components.

[0018] Furthermore, it was confirmed that this weight reduction rate depends on the filament diameter of the glass fiber. The weight reduction rate differs depending on the glass filament diameter, with smaller filament diameters resulting in greater weight reduction. On the other hand, the product of the weight reduction rate and the filament radius is approximately constant regardless of the filament diameter. Therefore, in this embodiment, the weight reduction coefficient is normalized by the filament diameter.

[0019] A weight loss coefficient of 0.38 or higher means that the material is susceptible to the effects of hygroscopic viscous substances, which can easily lead to a decrease in the insulation reliability of the resulting prepreg. In other words, when the amount of residual viscous substance is similar, the decrease in insulation reliability tends to be more pronounced in low-dielectric glass cloths compared to E-glass cloths with a smaller weight loss coefficient. In contrast, in this embodiment, whiteness is used as an indicator of the amount of residual viscous substance, and by defining and combining the weight loss tendency and whiteness within specific ranges, the insulation reliability of the resulting printed circuit board can be improved. Furthermore, because the weight loss coefficient is 0.90 or lower, the hygroscopicity of the glass itself does not become too high, thus suppressing a significant decrease in insulation reliability.

[0020] The weight loss percentage can be measured using the following procedure. First, the glass cloth is dried in a dryer at 105°C ± 5°C for 60 minutes, then transferred to a desiccator and allowed to cool to room temperature. After cooling, the weight of the glass cloth is measured to a unit of 0.1 mg or less (glass cloth weight a). Next, the glass cloth is heated at 380°C for 2 hours, then transferred to a desiccator and allowed to cool to room temperature. After cooling, the weight of the glass cloth is measured to a unit of 0.1 mg or less (glass cloth weight b after heat treatment). Then, the weight lost due to the heat treatment is determined, and the weight loss percentage (%) is calculated using the following formula (2). Weight reduction rate (%) = (ab) / a×100 ···(2)

[0021] The weight reduction rate obtained as described above is preferably 0.05 to 0.7%, more preferably 0.1 to 0.5%, and even more preferably 0.12 to 0.4%. A weight reduction rate of 0.05% or more is likely to cause a decrease in insulation reliability. However, by adjusting the whiteness described later, in this embodiment, the decrease in insulation reliability can be suppressed, and a glass cloth with a lower dielectric constant can also be obtained due to the composition of the glass cloth. Furthermore, a weight reduction rate of 0.7% or less can suppress a significant decrease in insulation reliability.

[0022] Next, the average diameter of the glass filaments of the glass thread constituting the glass cloth is measured in accordance with JIS R3420, and the average filament diameter is determined by taking half of the said filament diameter. In this embodiment, when simply referred to as glass filament, it means glass monofilament. The average radius of the glass filament used in calculating the weight loss coefficient is the average radius before heat treatment. The average radius of the glass filament thus obtained is preferably 1.25 to 4.5 μm, more preferably 1.5 to 3.75 μm, and even more preferably 1.75 to 2.7 μm.

[0023] Furthermore, the glass cloth used in the weight loss rate measurement method described above can be pre-treated as appropriate. For example, glass cloth pulled from an intermediate roll after de-adhesive treatment (heat cleaning) can be used directly in the weight loss rate measurement method described above because no adhering material is attached to the glass filaments.

[0024] On the other hand, when determining the weight loss ratio for glass cloth coated with a surface treatment agent such as a silane coupling agent, the surface treatment agent such as the silane coupling agent that has been physically adsorbed can be removed beforehand using a suitable solvent such as alcohol or acetone, and then the weight loss coefficient can be determined using the method described above.

[0025] Furthermore, "physically adsorbed silane coupling agents" refer to silane coupling agents that are attached to glass filaments and not silane coupling agents that are chemically bonded to the glass filaments. In contrast, silane coupling agents that are chemically bonded to glass filaments are called "chemically adsorbed silane coupling agents."

[0026] Furthermore, if the glass cloth contains organic impurities (such as starch-based sizing agents applied during the glass yarn manufacturing process, or combustion residues from the heat cleaning process of the sizing agent), the weight loss coefficient can be determined using the method described above after removing the organic impurities adhering to the glass cloth by a washing and removal operation using alcohols, acetone, etc.

[0027] The above cleaning process removes physically adsorbed silane coupling agents and organic impurities, but does not remove chemically adsorbed silane coupling agents. However, even if heated at 380°C for 2 hours, the chemically adsorbed silane coupling agents will not decompose, or if some decompose, it will not exceed the margin of error. Therefore, in measuring the weight loss rate in this embodiment, it is not necessary to remove the chemically adsorbed silane coupling agents in the pretreatment.

[0028] Furthermore, to simplify the decision of whether or not to perform pretreatment, glass cloth that has been uniformly washed in advance with a suitable solvent such as alcohol or acetone may be used for measuring the weight loss rate. This allows for the measurement of the weight loss rate to be performed under the same conditions, whether the glass cloth is pulled from an intermediate roll after de-adhesion treatment (heat cleaning) or whether it has physically adsorbed silane coupling agents or organic impurities attached to it.

[0029] Alternatively, when determining the weight loss coefficient mentioned above, the amount of surface treatment agent and organic impurities before and after heating can be quantified, and the weight loss due to the surface treatment agent can be subtracted from the obtained weight loss to determine the weight loss coefficient derived from the glass components. As a method for determining the weight loss due to the surface treatment agent, known methods such as the quantitative determination method for silane coupling agents described in Japanese Patent Publication No. 6472082 may be used.

[0030] The weight loss coefficient can be adjusted by increasing or decreasing relatively volatile components in the composition of the glass cloth, such as the B content, and can also be adjusted by increasing or decreasing other components from a similar perspective.

[0031] Furthermore, the weight reduction coefficient can also be adjusted by increasing or decreasing the opportunities for the glass surface to be exposed to a high-temperature atmosphere through adjustments such as adjusting the void filling rate of glass in the glass cloth (weaving density and thickness), adjusting the degree of unraveling of the monofilaments constituting the glass yarn bundle through processes such as fiber opening, and adjusting the monofilament diameter of the glass yarn used. In other words, the weight loss coefficient is not determined solely by the composition of the glass cloth.

[0032] (whiteness) The whiteness is 95 or higher, preferably 95.5 or higher, more preferably 96 or higher, and even more preferably 97 or higher. While there is no particular upper limit to the whiteness, the upper limit reflects the color tone of the glass itself and is approximately 99.5. A whiteness of 95 or higher further improves insulation reliability. The whiteness can be measured in accordance with JIS2000 L1916, the method for measuring the chromaticity of textile products.

[0033] Furthermore, whiteness can be improved by reducing the amount of sizing agent (also called adhesive), increasing the heat treatment temperature, changing the composition of the sizing agent, or changing the composition of the glass cloth.

[0034] (composition) The composition of the glass cloth in this embodiment will be described below. Note that the composition of the glass cloth is synonymous with the composition of the glass yarn that constitutes the glass cloth. In the composition of the glass cloth in this embodiment, the Fe content is preferably 0.001% by mass or more and 0.10% by mass or less, more preferably 0.001% by mass or more and 0.08% by mass or less, and even more preferably 0.01% by mass or more and 0.05% by mass or less, on an Fe2O3 basis. By having an Fe content of 0.10% by mass or less, the insulation reliability of the resulting prepreg can be further improved. This is presumed to be because reducing the Fe content, which is considered to have relatively general catalytic activity, to below a certain range suppresses the modification of the sizing agent into a flame-retardant component. The lower limit of the Fe content is not particularly limited, but it is 0.001% by mass or more. The Fe content can be adjusted according to the amount of raw materials used in glass filament production, purification and removal from raw materials by magnetic filters, etc., and additional additions.

[0035] The Si content of the glass cloth is preferably 40-60% by mass, more preferably 45-55% by mass, even more preferably 47-53% by mass, and 48-52% by mass, in terms of SiO2. Si is a component that forms the skeletal structure of the glass thread, and a Si content of 40% by mass or more improves the strength of the glass thread, and tends to further suppress the breakage of the glass cloth in the glass cloth manufacturing process and in subsequent processes such as the production of prepregs using the glass cloth. Also, a Si content of 40% by mass or more tends to further lower the dielectric constant of the glass cloth. On the other hand, a Si content of 60% by mass or less lowers the viscosity during melting in the glass filament manufacturing process, resulting in glass fibers with a more homogeneous glass composition, and as a result the glass cloth becomes less prone to breakage. The Si content can be adjusted according to the amount of raw materials used in the production of the glass filament.

[0036] The B content of the glass cloth is preferably 15-30% by mass, more preferably 17-28% by mass, even more preferably 20-27% by mass, even more preferably 21-25% by mass, and even more preferably 21.5-24% by mass, based on B2O3. A B content of 15% by mass or more tends to lower the dielectric constant. Furthermore, a B content of 30% by mass or less tends to improve moisture resistance and further improve insulation reliability. The B content can be adjusted according to the amount of raw materials used in the production of the glass filament. If the B content may fluctuate during the production of the glass filament, the amount of raw materials used can be adjusted in advance to account for this.

[0037] The F content of the glass cloth is preferably 0.001 to 0.1% by mass, more preferably 0.001 to 0.07% by mass, and even more preferably 0.001 to 0.05% by mass. Here, the F content is a value obtained in terms of F2. When the F content is within the above range, the breakage of the glass cloth tends to be further suppressed. The reasons why the breakage of the glass cloth is suppressed by adjusting the F content are not limited, but can be considered as follows: F reduces the viscosity of the molten glass in the glass manufacturing process. Therefore, when the F content is within a predetermined range, when metal components such as Fe are incorporated into the glass structure, they are uniformly dispersed without localization, and a uniform glass can be formed. It is thought that the effect of suppressing the modification of the sizing agent into a flame-retardant component is more effectively expressed when Fe is uniformly dispersed. The F content can be adjusted according to the amount of raw materials used in the production of the glass filament.

[0038] Furthermore, the glass filament may have other compositions besides those mentioned above. These other compositions are not particularly limited, but examples include Al, Ca, Mg, P, Na, K, Ti, and Zn.

[0039] The Al content of the glass cloth is preferably 10-20% by mass, more preferably 12-18% by mass, and even more preferably 14-17% by mass, based on Al2O3. Having an Al content within this range tends to improve electrical properties and strength. The Al content can be adjusted according to the amount of raw material used in the production of the glass filament.

[0040] The Ca content of the glass cloth is preferably 1.0 to 6.0% by mass, more preferably 2.0 to 5.0% by mass, and more preferably 2.5 to 4.0% by mass, in terms of CaO. A Ca content of 1.0% by mass or more tends to reduce the viscosity during melting in the glass filament manufacturing process, resulting in glass fibers with a more homogeneous glass composition. Furthermore, a Ca content of 6.0% by mass or less tends to improve the dielectric constant. The Ca content can be adjusted according to the amount of raw materials used in the glass filament production.

[0041] The Mg content of the glass cloth is preferably 5.0% by mass or less, more preferably 3.0% by mass or less, even more preferably 0.01 to 1.0% by mass or less, even more preferably 0.05 to 0.5% by mass or less, and even more preferably 0.05 to 0.3% by mass or less, in terms of MgO. A Mg content of 5.0% by mass or less tends to reduce the likelihood of breakage when the glass cloth passes through squeeze rolls or nip rolls while wet during processes such as fiber opening and surface treatment during glass cloth manufacturing. Furthermore, phase separation during glass filament manufacturing is suppressed, further improving the moisture resistance of the resulting glass filament. As a result, the resulting printed circuit board is less susceptible to the effects of high-humidity environments, and the environmental dependence of the dielectric constant can be reduced. The Mg content can be adjusted according to the amount of raw materials used in glass filament production.

[0042] The phosphorus (P) content of the glass cloth is preferably 1.0 to 7.0% by mass, more preferably 2.0 to 6.0% by mass, even more preferably 3.0 to 6.0% by mass, and even more preferably 3.0 to 5.5% by mass, based on P2O5. A P content of 1.0% by mass or more tends to further reduce the dielectric constant. Furthermore, a P content of 7.0% by mass or less tends to reduce the likelihood of breakage when the glass cloth passes through squeeze rolls or nip rolls while wet during processes such as fiber opening and surface treatment during glass cloth manufacturing. In addition, phase separation during glass filament manufacturing is suppressed, and the moisture resistance of the resulting glass filament is further improved. As a result, the resulting printed circuit board is less affected by high humidity environments, and the environmental dependence of the dielectric constant can be reduced. The P content can be adjusted according to the amount of raw materials used in glass filament production.

[0043] The above-mentioned content can be measured by ICP emission spectrometry. Specifically, the Si and B content can be obtained by pressurizing and decomposing a weighed glass cloth sample with sodium hydroxide, then dissolving it in dilute nitric acid and filtering it. The undissolved portion is then dissolved with sodium carbonate, and the mixture is combined with the filtrate to a fixed volume. The resulting sample is then measured by ICP emission spectrometry.

[0044] The Fe, Al, Ca, Ma, and P content are determined by heating and decomposing a weighed glass cloth sample with perchloric acid, nitric acid, hydrochloric acid, and hydrogen fluoride, then heating and dissolving it with dilute aqua regia and filtering it. The filtrate is then brought to a fixed volume. Undissolved components are then heated and decomposed with sulfuric acid, nitric acid, hydrochloric acid, and hydrogen fluoride to achieve a fixed solution, and the resulting sample is measured by ICP emission spectrometry. A Hitachi High-Tech Science PS3520VDD II ICP emission spectrometer can be used.

[0045] Furthermore, the fluoride content is determined by burning a weighed glass cloth sample in a tubular electric furnace, and then absorbing the resulting gas into an absorption solution. This solution is then analyzed by ion chromatography to determine the fluoride ions (F). -The amount of the substance in the sample can be determined by measuring its concentration. The combustion device can be an automated sample combustion device (AQF-2100S) manufactured by Mitsubishi Chemical Analytec, and the measuring device can be an ion chromatograph ICS-1500 manufactured by Thermo Fisher Scientific.

[0046] The elastic modulus of the glass cloth is preferably 50 to 70 GPa, more preferably 50 to 63 GPa, even more preferably 53 to 63 GPa, and particularly preferably 53 to 60 GPa. The lower the elastic modulus of the glass cloth, the more easily it breaks. Therefore, an elastic modulus of 50 GPa or higher tends to reduce the likelihood of breakage when the glass cloth passes through squeeze rolls or nip rolls while wet during manufacturing processes such as fiber opening and surface treatment. Furthermore, in subsequent processes such as prepreg manufacturing, breakage tends to be reduced when the glass cloth is passed through slits to control the amount of resin impregnation into the glass cloth. In addition, an elastic modulus of 70 GPa or lower tends to result in a relatively lower dielectric constant. The elastic modulus can be measured by the method described in the examples. The elastic modulus can also be adjusted by the composition of the glass yarn.

[0047] The dielectric constant of the glass cloth in this embodiment is preferably 5.0 or less, more preferably 4.7 or less, even more preferably 4.5 or less, and particularly preferably 4.0 or less at a frequency of 1 GHz. In this embodiment, unless otherwise specified, the term dielectric constant refers to the dielectric constant at a frequency of 1 GHz.

[0048] (composition) Glass yarn is obtained by bundling multiple glass filaments together and twisting them as needed, while glass cloth is obtained by weaving the above glass yarn as warp and weft threads. Glass yarn is classified as multifilament, and glass filaments are classified as monofilament.

[0049] The average diameter of the glass filaments constituting the warp and weft threads is preferably 2.5 to 9 μm, more preferably 3.0 to 7.5 μm, and even more preferably 3.5 to 5.4 μm, respectively. Having an average diameter of glass filaments within this range tends to improve processability when processing the resulting substrate with a mechanical drill, carbon dioxide laser, or UV-YAG laser. This allows for the realization of thin, high-density mounted printed circuit boards. In particular, when the average diameter is 5.4 μm or less, the weight loss during the same heating time increases, which tends to increase the hygroscopicity of the glass cloth itself. Furthermore, the increased surface area per unit volume of glass filaments leads to greater adhesion of hygroscopic viscous substances derived from sizing agents, further enhancing the hygroscopicity of the glass cloth. Therefore, the improved insulation reliability effect of this embodiment becomes even more important. Additionally, an average diameter of 2.5 μm or more tends to reduce the likelihood of breakage when the glass cloth passes through squeeze rolls or nip rolls while wet during glass cloth manufacturing processes such as fiber opening and surface treatment. Furthermore, in subsequent processes such as prepreg manufacturing, when passing the glass cloth through slits to control the amount of resin impregnation into the glass cloth, breakage tends to be less likely to occur.

[0050] The density of the warp and weft threads constituting the glass cloth is preferably 30 to 120 threads / 25 mm, more preferably 40 to 110 threads / 25 mm, and even more preferably 50 to 100 threads / 25 mm.

[0051] The thickness of the glass cloth is preferably 8 to 100 μm, more preferably 10 to 70 μm, and even more preferably 12 to 50 μm. A glass cloth thickness within this range tends to yield a thin yet relatively strong glass cloth. The weight loss coefficient depends on the area of ​​the glass cloth and the diameter of the filaments constituting the glass cloth, and is therefore less dependent on the thickness. This tendency is maintained at least within the above thickness range.

[0052] The weight (basis weight) of the glass cloth is preferably 8 to 250 g / m². 2 More preferably 8-100 g / m² 2 And more preferably 8-50 g / m 2 The g / m² g / m² is particularly preferably 8-35 g / m². 2 That is the case.

[0053] The weaving structure of the glass cloth is not particularly limited, but examples include plain weave, twill weave, satin weave, and twill weave. Among these, the plain weave structure is more preferred.

[0054] (Surface treatment) The glass cloth may be surface-treated with a surface treatment agent. The surface treatment agent is not particularly limited, but examples include silane coupling agents, and may be used in combination with water, organic solvents, acids, dyes, pigments, surfactants, etc., as needed.

[0055] The silane coupling agent is not particularly limited, but examples include the compound shown in formula (1). X(R) 3-n SiY n ...(1) (In formula (1), X is an organic functional group having at least one amino group and one unsaturated double bond group, Y is an alkoxy group, n is an integer between 1 and 3, and R is a group selected independently from the group consisting of a methyl group, an ethyl group, and a phenyl group.)

[0056] X is preferably an organic functional group having at least three amino groups and unsaturated double bond groups, and more preferably an organic functional group having at least four amino groups and unsaturated double bond groups.

[0057] Any form of the alkoxy group described above can be used, but from the viewpoint of stabilizing the treatment on glass cloth, an alkoxy group with 5 or fewer carbon atoms is preferred.

[0058] Specifically, silane coupling agents include N-β-(N-vinylbenzylaminoethyl)-γ-aminopropyltrimethoxysilane and its hydrochloride, N-β-(N-vinylbenzylaminoethyl)-γ-aminopropylmethyldimethoxysilane and its hydrochloride, N-β-(N-di(vinylbenzyl)aminoethyl)-γ-aminopropyltrimethoxysilane and its hydrochloride, N-β-(N-di(vinylbenzyl)aminoethyl)-N-γ-(N-vinylbenzyl)-γ-aminopropyltrimethoxysilane and its hydrochloride, N-β Examples include known individual elements or mixtures thereof such as -(N-benzylaminoethyl)-γ-aminopropyltrimethoxysilane and its hydrochloride, N-β-(N-benzylaminoethyl)-γ-aminopropyltriethoxysilane and its hydrochloride, γ-(2-aminoethyl)aminopropyltrimethoxysilane, γ-(2-aminoethyl)aminopropyltriethoxysilane, aminopropyltrimethoxysilane, vinyltrimethoxysilane, methacryloxypropyltrimethoxysilane, and acryloxypropyltrimethoxysilane.

[0059] The molecular weight of the silane coupling agent is preferably 100 to 600, more preferably 150 to 500, and even more preferably 200 to 450. Among these, it is preferable to use two or more silane coupling agents with different molecular weights. By treating the surface of the glass yarn with two or more silane coupling agents with different molecular weights, the density of the surface treatment agent on the surface of the glass cloth increases, and the reactivity with the matrix resin tends to improve further.

[0060] [Method for manufacturing glass cloth] The method for manufacturing glass cloth in this embodiment is not particularly limited, but examples include a weaving step of weaving glass yarn to obtain glass cloth, a fiber opening step of opening the glass yarn of the glass cloth, and a desizing step of removing sizing agents attached to the glass yarn of the glass cloth. Furthermore, a surface treatment step using a silane coupling agent may be included as needed.

[0061] The weaving method is not particularly limited as long as it involves folding the warp and weft threads to achieve a predetermined weave structure. Similarly, there are no particular limitations on the fiber opening method, but examples include methods using spray water (high-pressure water opening), a vibro-washer, ultrasonic water, or a mangle.

[0062] Furthermore, there are no particular limitations on the desizing method, but one example is a method of removing the sizing agent by heating. The sizing agent is used in the weaving process, etc., to protect the glass yarn from breaking. There are no particular limitations on such sizing agents, but examples include starch-based binders and polyvinyl alcohol-based binders. The temperature for removing the sizing agent by heating is preferably 300 to 500°C, more preferably 330 to 450°C, and even more preferably 350 to 430°C, from the viewpoint of sufficiently removing the sizing agent while maintaining the breaking strength.

[0063] Furthermore, surface treatment methods include contacting the glass cloth with a surface treatment agent containing a silane coupling agent and drying it. Contact of the glass cloth with the surface treatment agent can be achieved by immersing the glass cloth in the surface treatment agent, or by applying the surface treatment agent to the glass cloth using a roll coater, die coater, or gravure coater. There are no particular limitations on the drying method of the surface treatment agent, but examples include hot air drying or drying using electromagnetic waves.

[0064] [Prepreg] The prepreg of this embodiment comprises the glass cloth and a matrix resin composition impregnated into the glass cloth. The prepreg having the glass cloth has improved insulation reliability and a high yield of the final product. Furthermore, it can provide a printed circuit board that has excellent dielectric properties and excellent moisture resistance, resulting in less influence from the operating environment, especially in high humidity environments, and thus less fluctuation in dielectric constant.

[0065] The prepreg of this embodiment can be manufactured according to conventional methods. For example, it can be manufactured by impregnating the glass cloth of this embodiment with a varnish in which a matrix resin such as epoxy resin is diluted with an organic solvent, then evaporating the organic solvent in a drying oven to cure the thermosetting resin to the B stage (semi-cured state).

[0066] Examples of matrix resin compositions include, in addition to the epoxy resins mentioned above, thermosetting resins such as bismaleimide resins, cyanate ester resins, unsaturated polyester resins, polyimide resins, BT resins, and functionalized polyphenylene ether resins; thermoplastic resins such as polyphenylene ether resins, polyetherimide resins, all-aromatic polyester liquid crystal polymers (LCPs), polybutadiene, and fluororesins; and mixed resins thereof. From the viewpoint of improving dielectric properties, heat resistance, solvent resistance, and press moldability, a resin obtained by modifying a thermoplastic resin with a thermosetting resin may be used as the matrix resin composition.

[0067] Furthermore, the matrix resin composition may contain inorganic fillers such as silica and aluminum hydroxide; flame retardants such as bromine-based, phosphorus-based, or metal hydroxide-based agents; other silane coupling agents; heat stabilizers; antistatic agents; ultraviolet absorbers; pigments; colorants; lubricants, etc.

[0068] [Printed wiring board] The printed circuit board of this embodiment includes the glass cloth described above. The printed circuit board of this embodiment has improved insulation reliability and a high yield of the final product. Furthermore, because it has excellent dielectric properties and excellent moisture resistance, it can also have the effect of being less affected by the operating environment, especially in high humidity environments, where the dielectric constant fluctuates less. [Examples]

[0069] The present invention will be described more specifically below using examples and comparative examples. The present invention is not limited in any way by the following examples.

[0070] [Physical properties of glass cloth] The physical properties of the glass cloth, specifically the thickness of the glass cloth, the diameter and number of filaments constituting the warp and weft threads, and the weave density of the warp and weft threads were measured in accordance with JIS R3420.

[0071] [Weight reduction coefficient] The weight loss coefficient was measured using the following procedure. First, the glass cloth pulled from the intermediate roll was placed in a dryer at 105°C ± 5°C and dried for 60 minutes. After that, the glass cloth was transferred to a desiccator and allowed to cool to room temperature. After cooling, the weight of the glass cloth was measured to the nearest 0.1 mg (weight of glass cloth a). Next, the glass cloth was heated at 380°C for 2 hours, and then transferred to a desiccator and allowed to cool to room temperature. After cooling, the weight of the glass cloth was measured to the nearest 0.1 mg (weight of glass cloth after heat treatment b). The weight reduction due to the heat treatment was then determined, and the weight reduction percentage (%) was calculated using the following formula (2). Weight reduction rate (%) = (ab) / a×100 ···(2)

[0072] Next, the diameter of the monofilament was measured in accordance with Method B of JIS R3420, and half of that value was taken as the radius of the monofilament. Although Method B of JIS R3420 specifies measuring the diameter of 25 filament cross-sections at random, the diameter of all monofilaments constituting the glass fiber (multifilament) was measured, and the average value was used to determine the filament diameter. The weight loss coefficient was calculated using the weight loss percentage (%) and the average radius of the glass filament (μm) from the following formula (1). Weight loss coefficient = Weight loss rate (%) × Average radius of glass filament (μm) ... (1)

[0073] [Composition of glass fiber] The composition of the glass cloth was determined by ICP emission spectrometry. Specifically, the Si and B content were obtained by pressurizing and decomposing the weighed glass cloth sample with sodium hydroxide, then dissolving it in dilute nitric acid and filtering it. Undissolved components were dissolved with sodium carbonate, and the mixture was combined with the filtrate to a fixed volume. The resulting sample was then measured by ICP emission spectrometry.

[0074] Furthermore, the Fe, Al, Ca, Ma, and P content were determined by heating and decomposing the weighed glass cloth samples with perchloric acid, nitric acid, hydrochloric acid, and hydrogen fluoride, then heating and dissolving them with dilute aqua regia and filtering. The filtrate was then brought to a constant volume. Undissolved components were heated and decomposed with sulfuric acid, nitric acid, hydrochloric acid, and hydrogen fluoride to achieve a constant solution, and the resulting samples were measured by ICP emission spectrometry. A Hitachi High-Tech Science Corporation PS3520VDD II ICP emission spectrometer was used.

[0075] Furthermore, the fluoride content was determined by burning a weighed glass cloth sample in a tubular electric furnace, and then absorbing the resulting gas into an absorption solution. This solution was then analyzed by ion chromatography to determine the fluoride ion content (F). - The concentration of the substance in the sample was determined by measuring its concentration. The combustion device used was an automated sample combustion device (AQF-2100S) manufactured by Mitsubishi Chemical Analytec, and the measurement device was an ion chromatograph ICS-1500 manufactured by Thermo Fisher Scientific.

[0076] [Measurement of whiteness] Whiteness was measured using a spectrophotometer (CM-2600d, manufactured by Konica Minolta) in accordance with JIS2000 L1916, the method for measuring the self-chromaticity of textile products. Whiteness was measured with eight layers of glass cloth folded together, and the average of five measurements was used.

[0077] [Evaluation of insulation reliability] <Method for manufacturing circuit boards> The glass cloths obtained in the examples and comparative examples were impregnated with an epoxy resin varnish (a mixture of 40 parts by mass of a low-brominated bisphenol A type epoxy resin (manufactured by Mitsubishi Chemical Corporation), 10 parts by mass of an O-cresol type novolak epoxy resin (manufactured by Mitsubishi Chemical Corporation), 50 parts by mass of dimethylformamide, 1 part by mass of dicyandiamide, and 0.1 part by mass of 2-ethyl-4-methylimidazole), dried at 160 °C for 2 minutes, and prepregs with a resin content of 78% by mass were obtained for Examples 1 to 3, Comparative Examples 1 to 6, and Reference Example 1, a prepreg with a resin content of 73% by mass was obtained for Example 4, and a prepreg with a resin content of 58% by mass was obtained for Example 5. These prepregs were stacked in a predetermined number, and furthermore, copper foils with a thickness of 12 μm were stacked on the top and bottom, and heated and pressed at 175 °C and 40 kg / cm 2 for 60 minutes to obtain a substrate with a thickness of about 0.4 mm. A wiring pattern with through holes at an interval of 0.15 mm was formed on the copper foils on both sides of the obtained substrate to obtain a test substrate for insulation reliability evaluation.

[0078] <Evaluation method for insulation reliability of substrate> A voltage of 10 V was applied to the obtained test substrate in an atmosphere of a temperature of 120 °C and a humidity of 85% RH, and the change in the resistance value was measured. The case where the resistance became less than 1 MΩ within 500 hours after the start of the test was counted as insulation failure.

[0079] 100 test substrates were prepared, and insulation reliability was measured for 30 prepregs arbitrarily selected from them. Based on the measurement results, the insulation reliability was evaluated according to the following evaluation criteria. A: There was no test substrate showing insulation failure. B: There was one prepreg showing insulation failure. C: There were two prepregs showing insulation failure. D: There were three or more prepregs showing insulation failure.

[0080] 〔Examples 1 to 3〕 Using an air jet chamber, glass yarn (average filament diameter: 4.0 μm, number of filaments: 50) with the composition shown in Table 1 was woven to obtain a glass cloth with a warp and weft density of 95 threads / 25 mm and a thickness of 14 μm. Next, a heat cleaning treatment was performed by heating at 400°C for 24 hours, and a fiber opening process was carried out using a high-pressure water spray to obtain a glass cloth intermediate with a width of 1280 mm and a length of 2000 m. Subsequently, a surface treatment was performed using a silane coupling agent to produce glass cloth.

[0081] [Example 4] A roll of glass cloth was obtained in the same manner as in Example 2, except that a low-dielectric glass yarn consisting of 100 filaments with an average filament diameter of 5.0 μm was woven, and the warp thread density was 65 threads / 25 mm and the weft thread density was 67 threads / 25 mm. The thickness of the obtained glass cloth was 30 μm, and its composition is as shown in Table 1.

[0082] [Example 5] A roll of glass cloth was obtained in the same manner as in Example 2, except that a low-dielectric glass yarn consisting of 200 filaments with an average filament diameter of 7.0 μm was woven, and the warp and weft thread density was set to 52.5 threads / 25 mm. The thickness of the obtained glass cloth was 92 μm, and its composition is as shown in Table 1.

[0083] [Example 6] Except for attempting to wash away any remaining sizing agent residue after heat cleaning by rinsing the glass cloth with water after heat cleaning, a roll of glass cloth was obtained in the same manner as in Comparative Example 5. The composition of the obtained glass cloth is shown in Table 1.

[0084] [Comparative Examples 1, 3, 6] A roll of glass cloth was obtained in the same manner as in Example 1, except that the composition of the glass fibers was different. The composition of the obtained glass cloth is shown in Table 1.

[0085] [Comparative Example 2] A roll of glass cloth was obtained in the same manner as in Comparative Example 1, except that the heat cleaning treatment time was set to 48 hours. The composition of the obtained glass cloth is shown in Table 1.

[0086] [Comparative Example 4] A roll of glass cloth was obtained in the same manner as in Comparative Example 3, except that the heat cleaning treatment time was set to 48 hours. The composition of the obtained glass cloth is shown in Table 1.

[0087] [Comparative Example 5] A roll of glass cloth was obtained in the same manner as in Comparative Example 3, except that the glass cloth was washed with water before the heat cleaning process to attempt to reduce the sizing agent before the heat cleaning process. The composition of the obtained glass cloth was as shown in Table 1.

[0088] [Reference example 1] A roll of glass cloth was obtained in the same manner as in Example 1, except that the composition of the glass yarn was changed to a so-called E-glass composition. The composition of the obtained glass cloth is shown in Table 1.

[0089] [Table 1]

[0090] In all cases, substrates with high insulating reliability were obtained using the glass cloths described in the examples, and the occurrence rate of substrates with poor insulating reliability was low. Furthermore, although the glass cloth in Reference Example 1 had lower whiteness, the occurrence rate of substrates with poor insulating reliability was low, suggesting that the problem of reduced insulating reliability is less likely to occur with conventional E-glass and is a problem specific to low-dielectric glass cloths.

[0091] On the other hand, the glass cloths in Comparative Examples 1 and 3 were found to have low whiteness and poor insulation reliability. This is presumed to be due to the generation of flame-retardant viscous substances.

[0092] Furthermore, in Comparative Examples 2 and 4, it can be seen that while doubling the heat cleaning time gradually improves whiteness, the degree of improvement is not significant.

[0093] Furthermore, in Comparative Example 5, although an attempt was made to clean the sizing agent, etc., before the heat cleaning treatment, no significant difference was observed compared to Comparative Example 3. In Comparative Example 6, it was found that glass cloth, which has a high weight loss coefficient and is prone to volatilization, exhibits inferior insulation reliability. [Industrial applicability]

[0094] The present invention has industrial applicability as a low-dielectric glass cloth used in prepregs and the like.

Claims

1. A glass cloth composed of glass threads made of multiple glass filaments as warp and weft threads, In the following formula (1), the weight loss coefficient, which is calculated as the product of the weight loss rate derived from the glass component during a heat treatment at 380°C for 2 hours and the average radius of the glass filament, is 0.38 or more and 0.9 or less. Weight reduction coefficient = Weight reduction rate (%) × Average radius of the glass filament (μm) ... (1) The whiteness is 95 or higher. It does not contain white pigment. The Si content of the aforementioned glass cloth is SiO 2 In terms of conversion, it is 40-60% by mass, and the B content is B 2 O 3 In terms of conversion, this is 15-30% by mass. Having a dielectric constant of 5.0 or less at a frequency of 1 GHz, Glass cloth.

2. The Fe content of the glass cloth is Fe 2 This is 0.001 to 0.10 mass percent in terms of O3. The glass cloth according to claim 1.

3. The glass cloth according to claim 1 or 2, The glass cloth is impregnated with a matrix resin, Prepreg.

4. A glass cloth comprising the glass cloth described in claim 1 or 2, Printed circuit board.

Images