JP2011013078A - Method for detecting non-spherical particle in particle dispersion liquid - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a non-spherical particle detection method capable of accurately detecting non-spherical particles existing in a particle dispersion liquid, and to provide an inspection method and a manufacturing method of particle dispersion liquids that uses the method.SOLUTION: The non-spherical particle detection method includes the measurement of the angle dependence on the scattering intensity distribution of a particle dispersion liquid by a dynamic light-scattering method. According to the detection method, it is possible to detect a trace amount of non-spherical particles present in an inorganic particle dispersion liquid, and further, the detection method is, preferably, useful for product analysis, inspection, quality management, manufacturing, etc.

Description

  The present invention relates to a method for detecting non-spherical particles mixed in a particle dispersion, a method for inspecting a particle dispersion using the method, and the like.

  Conventionally, spherical particles such as colloidal silica and polymer particles have been used for polishing magnetic disk substrates and semiconductor substrates. In recent years, in order to increase the recording density of magnetic disk drives, magnetic disk substrates have improved smoothness and flatness (surface roughness, waviness, There is a strong demand for reduction of edge sagging and defects (reduction of scratches, protrusions, pits, etc.).

  It is known that the smoothness / flatness and defects on the substrate surface are affected by the shape of the abrasive particles used. In the field of abrasives, the shape of abrasive particles has been evaluated by electron micrograph observation, dynamic light scattering, laser diffraction, and the like (Patent Documents 1 to 4).

JP 2003-133267 A JP 2006-80406 A JP 2007-137972 A JP 2002-141314 A

  However, the conventional particle shape evaluation method can grasp the average shape of the whole particle, but cannot detect a small amount of non-spherical particles present in a mixture with spherical particles. In addition, when the particle size of the particles is small, accurate measurement is not always possible, and it is difficult to strictly control the shape of the particles.

  Furthermore, in order to achieve a further increase in capacity of the magnetic disk substrate, stricter shape control of the abrasive particles is required than before, and the proportion of non-spherical particles present in the spherical particles is more accurately evaluated. There is a need for a measurement method that can be used.

  Therefore, the present invention provides a method for detecting non-spherical particles that can more accurately detect non-spherical particles present in the particle dispersion, and a method for inspecting particle dispersion using the method.

  The present invention relates to a method for detecting non-spherical particles in a particle dispersion, which includes measuring the angular dependence of the scattering intensity distribution of the particle dispersion by a dynamic light scattering method.

  As another aspect, the present invention relates to a particle dispersion inspection method and / or manufacturing method including a step of inspecting a particle dispersion using the method for detecting non-spherical particles of the present invention. The present invention further includes, as another aspect, a particle dispersion inspection process according to the inspection method of the present invention or a particle dispersion manufacturing process according to the manufacturing method of the present invention, wherein the particle dispersion is a polishing liquid composition. Further, the present invention relates to a method for manufacturing a substrate including a step of polishing a substrate to be polished using the polishing composition.

  According to the method for detecting non-spherical particles of the present invention, preferably, non-spherical particles present in a minute amount in a spherical particle dispersion, which has been difficult to detect conventionally, can be detected more accurately.

FIG. 1 shows the scattering intensity measured by the dynamic light scattering method when silica particles I (non-spherical: rhombus) or silica particles II (spherical: square) are added to a polishing liquid composition that is a colloidal silica dispersion. It is a graph which shows the change of the angle dependence ((DELTA) CV value) of distribution. FIG. 2 is a graph plotting lots used for the ΔCV value and the number of scratches on the substrate surface after polishing for each of silica A and silica B.

[Non-spherical particles]
In this specification, “non-spherical particles” refers to particles having a higher degree of deformity than “spherical particles”, and may include aggregated particles and aggregated particles of “spherical particles” that are primary particles. Further, in the present specification, “non-spherical particles” preferably refer to particles having a degree of irregularity greater than 1.55. Here, the degree of irregularity is a value obtained by dividing the average particle diameter measured by the dynamic light scattering method by the average particle diameter measured by the BET method. In general, when the particles are spherical, the value of the degree of irregularity is obtained. Converges to 1 and increases as the particle deviates from the true sphere (see, for example, JP-A-2007-137972).

Further, as one of methods for measuring whether the average shape of the particles in the dispersion is spherical (true spherical) or non-spherical, a diffusion coefficient (D = Γ / q ² ) as an index (see, for example, Japanese Patent Laid-Open No. 10-195152). Specifically, the smaller the angle dependency shown in the graph plotting Γ / q ² with respect to the scattering vector q ^{2, the} more the average shape of the particles in the dispersion is judged to be spherical, and the angle dependency is This is a method for determining that the average shape of the particles in the dispersion is non-spherical as the value increases.

  The inventors of the present invention pay attention to the fact that the quality of the substrate surface after polishing (such as the number of scratches) is greatly different even in a polishing composition produced as a particle dispersion having the same particle size and particle size distribution. It has been found that the cause is that a small amount of non-spherical particles of 200 nm or less that cannot be removed by the conventional coarse particle removal (precise filter filtration) step exists in the particle dispersion.

  The present invention provides a scattering intensity distribution measured by the dynamic light scattering method even for a sample that is recognized as a true spherical particle dispersion based on the angular dependence of the diffusion coefficient measured by the dynamic light scattering method. Focusing on the angular dependence of the sample, it is possible to detect non-spherical particles present in the sample, that is, the content of non-spherical particles in the particle dispersion and the scattering intensity measured by the dynamic light scattering method. Based on the finding that the angular dependence of the distribution correlates.

  Therefore, in one aspect, the present invention is a method for detecting non-spherical particles in a particle dispersion (hereinafter, also referred to as “the non-spherical particle detection method of the present invention”), which is based on a dynamic light scattering method. The present invention relates to a method for detecting non-spherical particles, including measuring the angular dependence of the scattering intensity distribution of the particle dispersion. According to the method for detecting non-spherical particles of the present invention, preferably, non-spherical particles present in a minute amount in a particle dispersion, which has been difficult to detect conventionally, can be detected more accurately.

  That is, the non-spherical particle detection method of the present invention is different from the conventional method of detecting and measuring the particle shape, particle size, etc., assuming that uniform particles are dispersed throughout the system. A non-spherical particle existing in a part of the particle can be found. The mechanism by which the non-spherical particles present in a minute amount in the particle dispersion, which has been difficult to detect by the non-spherical particle detection method of the present invention, can be detected more accurately is as follows. That is, in the dynamic light scattering method, when measuring a spherical particle dispersion solution of 200 nm or less in principle, the scattering intensity distribution can obtain a substantially constant result regardless of the detection angle, so the measurement result does not depend on the detection angle. . However, the scattering intensity distribution of dynamic light scattering of a spherical dispersion containing non-spherical particles varies greatly depending on the detection angle due to the presence of non-spherical particles, and the distribution of the scattering intensity distribution becomes broader at lower detection angles. . Therefore, the measurement result of the scattering intensity distribution of dynamic light scattering depends on the detection angle, and by measuring the angular dependence of the scattering intensity distribution of dynamic light scattering, the non-spherical particles present in the particle dispersion solution are detected. It can be detected. However, the present invention is not limited to this mechanism.

Scattering intensity distribution
In this specification, “scattering intensity distribution” means three particle size distributions of sub-micron or less particles obtained by dynamic light scattering (DLS) or quasielastic light scattering (QLS). It means the particle size distribution of scattering intensity among (scattering intensity, volume conversion, number conversion). Usually, the sub-micron particles have Brownian motion in a solvent, and the intensity of scattered light changes (fluctuates) with time when irradiated with laser light. For this fluctuation of scattered light intensity, for example, an autocorrelation function is obtained using a photon correlation method (JIS Z 8826), and a diffusion coefficient (D) indicating a Brownian motion velocity is calculated by cumulant method analysis. Furthermore, the average particle diameter (d: hydrodynamic diameter) can be obtained using the Einstein-Stokes equation. In addition to polydispersity index (PI) by cumulant method, particle size distribution analysis includes histogram method (Marquardt method), Laplace inverse transformation method (CONTIN method), non-negative least square method (NNLS method), etc. There is.

In the particle size distribution analysis of the dynamic light scattering method, the polydispersity index (PI) by the cumulant method is generally widely used. However, in the detection method of non-spherical particles of the present invention that enables detection of non-spherical particles slightly present in the particle dispersion, the particle size distribution by the histogram method (Marquardt method) or the Laplace inverse transformation method (CONTIN method). Obtain the average particle size (d50) and standard deviation from the analysis, calculate the CV value (Coefficient of variation: a value obtained by dividing the standard deviation by the average particle size and multiply by 100), and use its angular dependence (ΔCV) Is preferred.
(Reference document)
Text of the 12th Scattering Study Group (held on November 22, 2000) Scattering Basic Course "Dynamic Light Scattering Method" (Professor Mitsuhiro Shibayama, The University of Tokyo)
Text of the 20th Scattering Study Group (held on December 4, 2008) Measurement of size distribution of nanoparticles by dynamic light scattering (Doshisha University, Professor Yasushi Mori)

"Angle dependence of scattering intensity distribution"
In this specification, “angle dependency of the scattering intensity distribution of the particle dispersion” means scattering according to the scattering angle when the scattering intensity distribution of the particle dispersion is measured at different detection angles by the dynamic light scattering method. The magnitude of fluctuation in intensity distribution. For example, if the difference in the scattering intensity distribution between the detection angle of 30 ° and the detection angle of 90 ° is large, it can be said that the angle dependency of the scattering intensity distribution of the particle dispersion is large. Therefore, in the present invention, measurement of the angle dependence of the scattered intensity distribution includes obtaining a difference between measured values based on the scattered intensity distribution measured at two different detection angles. Examples of the measurement value include a standard deviation and a CV value based on a scattering intensity distribution obtained by measurement at one detection angle. In this specification, as described above, the CV value is obtained by dividing the standard deviation based on the scattering intensity distribution obtained by measurement at one detection angle by the average particle diameter based on the scattering intensity distribution at the detection angle. The value multiplied by 100. From the viewpoint of improving the accuracy of detection of non-spherical particles, the measured value is preferably a CV value. Therefore, as an index indicating the angle dependence of the scattered intensity distribution, from the viewpoint of improving the accuracy of detection of non-spherical particles, preferably, the difference in standard deviation based on the scattered intensity distribution measured at two different detection angles, and A difference (ΔCV value) between CV values based on scattering intensity distributions measured at two different detection angles, and more preferably a ΔCV value.

  As a combination of the two detection angles used in the measurement of the angle dependence of the scattering intensity distribution, a combination of forward scattering and side or back scattering is preferable from the viewpoint of improving the accuracy of detection of non-spherical particles. From the same viewpoint, the forward scattering detection angle is preferably 0 to 80 °, more preferably 0 to 60 °, still more preferably 10 to 50 °, and still more preferably 20 to 40 °. From the same viewpoint, the side or backscattering detection angle is preferably 80 to 180 °, more preferably 85 to 175 °.

"Particle dispersion"
In the present specification, the “particle dispersion” refers to a test sample in which the presence or absence of non-spherical particles is detected by the method for detecting non-spherical particles of the present invention. Further, in the present specification, the “particle dispersion liquid” is preferably an angle dependency of a diffusion coefficient by dynamic light scattering measurement, that is, an angle dependency shown in a graph plotting Γ / q ² with respect to a scattering vector q ² . Includes samples with small or no. A sample having such a small or no angular dependence of the diffusion coefficient by dynamic scattering measurement is conventionally regarded as a particle dispersion in which spherical particles are dispersed. One aspect of the present invention focuses on the detection of a small amount of non-spherical particles that may exist in a sample with little or no angular dependence of the diffusion coefficient by dynamic scattering measurement, focusing on the "angle dependence of the scattering intensity distribution". This is a method for detecting non-spherical particles.

[Detection method of non-spherical particles]
The method for detecting non-spherical particles of the present invention includes measuring the angle dependence of the scattering intensity distribution of the particle dispersion by a dynamic light scattering method. As described above, the value indicating the angle dependence of the scattering intensity distribution includes the difference in standard deviation based on the scattering intensity distribution and the CV value obtained by dividing the standard deviation by the average particle diameter and multiplying by 100. Difference (ΔCV value) and the like can be mentioned.

  As one embodiment of the method for detecting non-spherical particles of the present invention, the angle dependence of the scattering intensity distribution is previously determined by a control that does not contain non-spherical particles or a control that is considered to be substantially free of non-spherical particles from the polishing results. There is a detection method that includes measuring and determining whether or not the particle dispersion contains non-spherical particles as compared with the angular dependence of the scattering intensity distribution of the particle dispersion to be tested.

  As other embodiments of the method for detecting non-spherical particles of the present invention, a control that does not contain non-spherical particles or a control that is considered to be substantially free of non-spherical particles from the polishing results, and a control that contains known non-spherical particles Is used to create a calibration curve indicating the correlation between the angular dependence of the scattering intensity distribution and the content of non-spherical particles, and from the angular dependence of the scattering intensity distribution of the particle dispersion to be tested, The detection method including calculating | requiring content of a non-spherical particle | grain is mentioned. When creating the calibration curve, it is preferable to use a ΔCV value (difference in CV value) as a measured value indicating the angular dependence of the scattering intensity distribution. This is because the abundance of non-spherical particles in the particle dispersion and the ΔCV value show a stronger correlation, and more accurate detection of non-spherical particles becomes possible. However, the present invention is not construed as being limited to these embodiments.

  The non-spherical particle detection method of the present invention is preferably capable of more accurately detecting non-spherical particles present in the particle dispersion, so that the analysis, inspection, and quality control of the particle dispersion and / or the product containing the particle dispersion are performed. Useful for manufacturing. Accordingly, the present invention, in its aspect, relates to an analysis method, an inspection method, a quality control method, and / or a manufacturing method for a particle dispersion and / or a product containing the particle dispersion. Examples of the product containing the particle dispersion include polishing liquid compositions for semiconductor substrates and magnetic disk substrates, cosmetics, paints, and the like. Furthermore, this invention has the manufacturing process of the polishing liquid composition by the test | inspection process of the polishing liquid composition by the test | inspection method of this invention, or the manufacturing method of this invention, and uses the said polishing liquid composition in another aspect. The present invention relates to a substrate manufacturing method including a step of polishing a substrate to be polished.

  As a method for analyzing a particle dispersion and / or a product containing the particle dispersion, the presence / absence and / or content of the non-spherical particles in the product containing the particle dispersion and / or the particle dispersion is analyzed by the non-spherical particle detection method of the present invention. Embodiment which performs is mentioned. As a method for inspecting a particle dispersion and / or a product containing the particle dispersion, the presence or absence and / or content of the non-spherical particles in the particle dispersion and / or the product containing the particle dispersion are determined by the non-spherical particle detection method of the present invention. An embodiment to inspect is mentioned. As the quality control method of the particle dispersion and / or the product containing the particle dispersion, the presence or absence and / or content of the non-spherical particles in the product containing the particle dispersion and / or the particle dispersion by the non-spherical particle detection method of the present invention. An embodiment in which quality is managed by detecting the above is mentioned. As a method for producing a particle dispersion and / or a product containing the particle dispersion, the presence or absence of non-spherical particles in the product containing the particle dispersion and / or the particle dispersion is detected by the non-spherical particle detection method of the present invention. An embodiment including a process may be mentioned.

[particle]
Although it does not restrict | limit especially as a particle which exists in the particle dispersion liquid used for the detection method of the non-spherical particle | grains of this invention, For example, a particle | grain with an average particle diameter of the range of 1-200 nm may be sufficient. In conventional methods such as electron micrograph observation, dynamic light scattering method, and laser diffraction method, it has been difficult to detect non-spherical particles mixed in spherical particles at an average particle size in the above range. This is because microscopic observation is the observation of only a part of the solution, and the conventional dynamic light scattering method measures and calculates particles on the assumption that the entire solution system is uniform. In the method for detecting non-spherical particles of the present invention, as described above, the non-spherical particles present in the particle dispersion solution are detected by measuring the angular dependence of the scattering intensity distribution by the dynamic light scattering method. It is possible. In the present specification, the “average particle diameter” of particles means an average particle diameter based on a scattering intensity distribution measured at a detection angle of 90 degrees by a dynamic light scattering method, unless otherwise specified.

  The particles are preferably inorganic particles from the viewpoint of improving the accuracy of detection of non-spherical particles. Examples of the inorganic particles include particles of silica, alumina, zirconia, titania, magnesia, molybdenum oxide, tungsten oxide, zinc oxide, tin oxide, barium titanate, aluminum titanate, silicon carbide, silicon nitride, and the like. Other than these, inorganic oxides, inorganic nitrides, hard-positive ion crystals, covalent bonds, clay minerals, and the like may be used. From the viewpoint of application to an abrasive composition, silica and alumina are preferable as the inorganic particles, and colloidal silica synthesized by particle growth using water glass (silicic acid aqueous solution) or metal alkoxide as a raw material is more preferable.

[Polishing composition for magnetic disk substrate]
As one embodiment of the present invention, a method for detecting non-spherical particles in a polishing liquid composition for a magnetic disk substrate, an inspection method, a quality control method, a manufacturing method for the polishing liquid composition, and the polishing liquid composition are used. A substrate manufacturing method including a step of polishing a substrate to be polished will be described below.

  As an example of a polishing composition for a magnetic disk substrate, an average particle containing colloidal silica, water, an acid and an oxidant and based on a scattering intensity distribution measured by the dynamic light scattering method (detection angle 90 degrees) of the colloidal silica. A polishing composition having a diameter of 1 to 40 nm is used.

  As an example of applying the method for detecting non-spherical particles of the present invention to the polishing composition, a detection method using a CV value as a measurement value based on the scattering intensity distribution is given as follows. First, an average particle diameter and a standard deviation based on a scattering intensity distribution at a detection angle of 30 ° (forward scattering) were obtained using a dynamic light scattering method, and the standard deviation was divided by the average particle diameter and multiplied by 100. A CV value (CV30) is obtained. Next, a standard deviation based on a scattering intensity distribution at a detection angle of 90 ° (side scatter) is obtained using a dynamic light scattering method, and the standard deviation is divided by the average particle diameter and multiplied by 100. (CV90) is obtained. Then, a ΔCV value (ΔCV = CV30−CV90) which is a difference between CV30 and CV90 is obtained as a value indicating the angle dependency of the scattering intensity distribution of the polishing liquid composition. A higher ΔCV value indicates that more non-spherical silica particles are present in the polishing liquid composition. Moreover, it shows that there are few non-spherical silica particles in the said polishing liquid composition, so that the said (DELTA) CV value is low (closer to zero).

  In the polishing composition, from the viewpoint of reducing scratches on the substrate surface after polishing the substrate to be polished, the ΔCV value (ΔCV = CV30−CV90) is preferably 0 to 10%, more preferably 0 to 7%. More preferably, it is 0 to 5%.

  Therefore, the inspection method for the polishing composition of the present invention includes an inspection method including inspecting whether or not the ΔCV value of the polishing composition falls within the above range. Further, as a quality control method of the polishing composition of the present invention, whether or not the ΔCV value of the polishing composition is within the above range is inspected. Examples include a quality control method including adjusting the ΔCV value or stopping the use of the polishing composition. Furthermore, as a manufacturing method of the polishing liquid composition of this invention, the manufacturing method including the process of confirming that the said (DELTA) CV value of polishing liquid composition enters the said range is mentioned. The manufacturing method may further include the above-described ΔCV value adjusting step.

  A magnetic disk substrate, particularly a perpendicular magnetic recording type magnetic disk substrate, in which scratches of the substrate are reduced by using the polishing composition prepared by inspection, quality control and / or manufacturing using the method for detecting non-spherical particles of the present invention. The effect that can be manufactured is preferably produced.

  Although the mechanism of reducing scratches is not clear, 50-200 nm silica aggregates (non-spherical silica) produced by agglomeration of primary particles of an abrasive such as colloidal silica are the causative substances for the generation of scratches. It is estimated that scratches are reduced because there are few aggregates. However, the present invention is not limited to these estimation mechanisms.

  Hereinafter, preferred embodiments of the polishing composition for a magnetic disk substrate will be further described.

[Colloidal silica]
The colloidal silica used in the polishing composition may be obtained by a known production method or the like produced from an aqueous silicic acid solution. The usage form of the silica particles is preferably a slurry from the viewpoint of operability.

  From the viewpoint of reducing scratches, the average particle size based on the scattering intensity distribution measured by the colloidal silica dynamic light scattering method (detection angle 90 degrees) is preferably 1 to 40 nm, more preferably 5 to 37 nm, More preferably, it is 10-35 nm.

  The CV value (CV30 and CV90) of colloidal silica is preferably 1 to 35%, more preferably 5 to 34%, and still more preferably 10 to 33% from the viewpoint of reducing scratches.

  The content of colloidal silica in the polishing composition is preferably 0.5 to 20% by weight, more preferably 1 to 15% from the viewpoint of improving the polishing rate and further improving the flatness of the substrate surface. %, More preferably 3 to 13% by weight, even more preferably 4 to 10% by weight.

[water]
Water in the polishing composition is used as a medium, and examples thereof include distilled water, ion exchange water, and ultrapure water. From the viewpoint of the surface cleanliness of the substrate to be polished, ion exchange water and ultrapure water are preferable, and ultrapure water is more preferable. The content of water in the polishing composition is preferably 60 to 99.4% by weight, and more preferably 70 to 98.9% by weight. Moreover, you may mix | blend organic solvents, such as alcohol, in the range which does not inhibit the effect of this invention.

[acid]
The polishing liquid composition preferably contains an acid and / or a salt thereof. Acids and / or salts thereof include inorganic acids such as nitric acid, sulfuric acid, sulfurous acid, persulfuric acid, hydrochloric acid, perchloric acid, phosphoric acid, phosphonic acid, phosphinic acid, pyrophosphoric acid, tripolyphosphoric acid, amidosulfuric acid, and salts thereof, 2 -Aminoethylphosphonic acid, 1-hydroxyethylidene-1,1-diphosphonic acid, aminotri (methylenephosphonic acid), ethylenediaminetetra (methylenephosphonic acid), diethylenetriaminepenta (methylenephosphonic acid), ethane-1,1, -diphosphonic acid Ethane-1,1,2-triphosphonic acid, ethane-1-hydroxy-1,1-diphosphonic acid, ethane-1-hydroxy-1,1,2-triphosphonic acid, ethane-1,2-dicarboxy-1 , 2-diphosphonic acid, methanehydroxyphosphonic acid, 2-phosphonobutane-1,2-dicarboxylic acid, 1-phospho Organic phosphonic acids such as honobutane-2,3,4-tricarboxylic acid, α-methylphosphonosuccinic acid and salts thereof, aminocarboxylic acids such as glutamic acid, picolinic acid and aspartic acid and salts thereof, acetic acid, citric acid, tartaric acid, Examples thereof include carboxylic acids such as oxalic acid, nitroacetic acid, maleic acid, and oxaloacetic acid, and salts thereof. Among these, from the viewpoint of reducing scratches, inorganic acids, organic phosphonic acids, and salts thereof are preferable. Among inorganic acids and salts thereof, nitric acid, sulfuric acid, hydrochloric acid, perchloric acid and salts thereof are more preferable, and sulfuric acid is more preferable. Among organic phosphonic acids and salts thereof, 1-hydroxyethylidene-1,1-diphosphonic acid, aminotri (methylenephosphonic acid), ethylenediaminetetra (methylenephosphonic acid), diethylenetriaminepenta (methylenephosphonic acid) and their salts are more preferred. 1-hydroxyethylidene-1,1-diphosphonic acid and aminotri (methylenephosphonic acid) are more preferable. These acids and salts thereof may be used alone or in admixture of two or more.

  The counter ion in the case of using these acid salts is not particularly limited, and specific examples thereof include salts with metals, ammonium, alkylammonium and the like. Specific examples of the metal include metals belonging to the periodic table (long-period type) 1A, 1B, 2A, 2B, 3A, 3B, 4A, 6A, 7A, or Group 8. Among these, a salt with a metal belonging to Group 1A or ammonium is preferable from the viewpoint of reducing scratches.

  The content of the acid and its salt in the polishing composition is preferably 0.001 to 5% by weight, more preferably 0.01 to 4% by weight from the viewpoints of improving the polishing rate, surface roughness, and reducing scratches. %, More preferably 0.05 to 3% by weight, still more preferably 0.1 to 2.0% by weight.

[Oxidant]
It is preferable that the polishing composition contains an oxidizing agent. Preferable oxidizing agents include hydrogen peroxide, iron (III) nitrate, peracetic acid, ammonium peroxodisulfate, iron (III) sulfate, and iron (III) ammonium sulfate. As a more preferable oxidizing agent, hydrogen peroxide is mentioned from the viewpoint that metal ions do not adhere to the surface and are generally used and inexpensive. These oxidizing agents may be used alone or in admixture of two or more.

  The content of the oxidizing agent in the polishing composition is preferably 0.01 to 4% by weight, more preferably 0.05 to 2% from the viewpoints of improving the polishing rate and reducing the surface roughness, waviness and scratches. % By weight, more preferably 0.1 to 1% by weight.

[Other ingredients]
Other components can be blended in the polishing composition as necessary. Examples of other components include a thickener, a dispersant, a rust inhibitor, a basic substance, and a surfactant. 0-10 weight% is preferable and, as for content of these other arbitrary components in polishing liquid composition, 0-5 weight% is more preferable.

[PH of polishing composition]
The pH of the polishing composition is preferably from 0.5 to 4.0, more preferably from 0.8 to 3.0, and even more preferably from the viewpoint of improving the polishing rate and reducing the surface roughness. ~ 2.5.

[Method for preparing polishing liquid composition]
The polishing liquid composition can be prepared, for example, by mixing water, colloidal silica, and, if desired, an acid and / or salt thereof, an oxidizing agent, and other components by a known method. Under the present circumstances, colloidal silica may be mixed in the state of the concentrated slurry, and may be mixed after diluting with water etc. The content and concentration of each component in the polishing liquid composition are in the ranges described above, but as another aspect, the polishing liquid composition may be prepared as a concentrate.

  In another aspect, the present invention provides a method for preparing a polishing liquid composition for a magnetic disk substrate containing colloidal silica, wherein the angular dependence of the scattering intensity distribution of the polishing liquid composition is measured by a dynamic light scattering method. A method for preparing a polishing liquid composition for a magnetic disk substrate. More specifically, the average particle diameter measured at a detection angle of 90 ° in the dynamic light scattering method is 1 to 40 nm, and the standard deviation measured at the detection angle of 90 ° in the dynamic light scattering method is the average particle diameter. The CV value (CV90) divided by 100 is 1 to 35%, and the standard deviation measured at a detection angle of 30 ° in the dynamic light scattering method is divided by the average particle size and multiplied by 100. A method for preparing a polishing composition for a magnetic disk substrate, comprising selecting and / or confirming and using colloidal silica having a difference between CV value (CV30) and CV90 (ΔCV = CV30−CV90) of 0 to 10% Can provide. The polishing composition for a magnetic disk substrate using the colloidal silica can reduce scratches after polishing.

[Method of manufacturing magnetic disk substrate]
In another aspect, the present invention relates to a method for manufacturing a magnetic disk substrate, which includes measuring the angular dependence of the scattering intensity distribution of the polishing composition by a dynamic light scattering method. The method for producing a magnetic disk substrate of the present invention includes an inspection step by the polishing liquid composition inspection method of the present invention, a quality control step by the quality control method of the polishing liquid composition of the present invention, or a polishing liquid composition of the present invention. It is preferable to use a polishing composition for a magnetic disk substrate that includes a manufacturing process according to the above manufacturing method and that has a ΔCV value (ΔCV = CV30−CV90) falling within the above range. In addition, the method for manufacturing a magnetic disk substrate of the present invention preferably includes a step of polishing the substrate to be polished using the polishing liquid composition.

  According to the method for manufacturing a magnetic disk substrate of the present invention, it is possible to preferably provide a magnetic disk substrate in which scratches on the substrate after polishing are reduced. The manufacturing method of the present invention is particularly suitable for a method for manufacturing a magnetic disk substrate for perpendicular magnetic recording.

  As a specific example of the method for polishing a substrate to be polished using the polishing liquid composition, the polishing substrate is sandwiched between a surface plate to which a polishing pad such as a non-woven organic polymer polishing cloth is attached, and the polishing liquid composition There is a method of polishing the substrate to be polished by moving the surface plate or the substrate to be polished while supplying to the polishing machine.

  When the polishing process of the substrate to be polished is performed in multiple stages, the polishing process using the polishing liquid composition is preferably performed in the second and subsequent stages, and more preferably in the final polishing process. At that time, in order to avoid mixing of the polishing material and polishing liquid composition in the previous process, different polishing machines may be used, and in the case of using different polishing machines, polishing is performed for each polishing process. It is preferable to clean the substrate. The polishing machine is not particularly limited, and a known polishing machine for polishing a magnetic disk substrate can be used.

  In one embodiment of the method for manufacturing a magnetic disk substrate of the present invention, the average particle diameter measured at a detection angle of 90 ° in the dynamic light scattering method is 1 to 40 nm, and the detection angle of 90 ° in the dynamic light scattering method. The CV value (CV90) of the average particle diameter measured at 1 is 35 to 35%, and the standard deviation measured at a detection angle of 30 degrees in the dynamic light scattering method is divided by the average particle diameter and multiplied by 100. Further, it may include selecting and / or confirming and using a polishing liquid composition containing colloidal silica having a difference (ΔCV = CV30−CV90) between the CV value (CV30) and the CV90 of 0 to 10%.

[Polishing pad]
The polishing pad used in the present invention is not particularly limited, and a polishing pad of a suede type, a nonwoven fabric type, a polyurethane closed-cell foam type, or a two-layer type in which these are laminated can be used. From the viewpoint, a suede type polishing pad is preferable.

[Polishing load]
The polishing load in the polishing step using the polishing composition is preferably 5.9 to 20 kPa, more preferably 6.9 to 18 kPa, from the viewpoint of suppressing a decrease in the polishing rate and suppressing the generation of scratches. 5-16 kPa is more preferable. In the present invention, the polishing load refers to the pressure of the surface plate applied to the polishing surface of the substrate to be polished during polishing. The polishing load can be adjusted by applying air pressure or weight to at least one of the surface plate and the substrate to be polished.

[Supply of polishing liquid composition]
From the viewpoint of reducing scratches, the supply rate of the polishing composition in the polishing step using the polishing composition is preferably 0.05 to 15 mL / min, more preferably 0, per 1 cm ^{2 of the} substrate to be polished. 0.06 to 10 mL / min, more preferably 0.07 to 1 mL / min, even more preferably 0.08 to 0.5 mL / min, and even more preferably 0.12 to 0.5 mL / min.

  Examples of a method for supplying the polishing composition to the polishing machine include a method of continuously supplying the polishing composition using a pump or the like. When supplying the polishing composition to the polishing machine, in addition to the method of supplying one component containing all the components, considering the stability of the polishing composition, etc., it is divided into a plurality of compounding component liquids, Two or more liquids can be supplied. In the latter case, for example, the plurality of compounding component liquids are mixed in the supply pipe or on the substrate to be polished to obtain the polishing liquid composition of the present invention.

[Polished substrate]
Examples of the material of the substrate to be polished preferably used in the method of manufacturing a magnetic disk substrate include metals or semimetals such as silicon, aluminum, nickel, tungsten, copper, tantalum, and titanium, or alloys thereof, glass, and glass. Glassy substances such as glassy carbon and amorphous carbon, ceramic materials such as alumina, silicon dioxide, silicon nitride, tantalum nitride, and titanium carbide, and resins such as polyimide resin. Among these, a substrate to be polished containing a metal such as aluminum, nickel, tungsten, copper, or an alloy containing these metals as a main component is preferable. In particular, it is suitable for Ni-P plated aluminum alloy substrates and glass substrates such as crystallized glass, tempered glass, aluminosilicate glass, etc. Among them, Ni-P plated aluminum alloy substrates are suitable.

[Polishing method]
In another aspect, the present invention relates to a method for polishing a substrate to be polished, which comprises polishing the substrate to be polished while bringing the polishing composition described above into contact with a polishing pad. By using the polishing method of the present invention, the substrate to be polished can be polished without impairing the productivity, and both the surface roughness and the scratch are reduced, particularly the magnetic disk substrate of the perpendicular magnetic recording system. Is preferably provided. Examples of the substrate to be polished in the polishing method of the present invention include those used in the manufacture of magnetic disk substrates and magnetic recording medium substrates, as mentioned above. A substrate used for production is preferred. The specific polishing method and conditions can be as described above.

[Example 1]
A polishing liquid composition containing colloidal silica was prepared, nonspherical silica particles or spherical silica particles were further added to the polishing liquid composition, ΔCV value was measured, and the angle dependence of the scattering intensity distribution exerted by the added silica particles The effect on sex was confirmed. The measurement results are shown in Table 1 and FIG. The method for preparing the polishing composition and the method for measuring the ΔCV value are as follows.

[Method for preparing polishing liquid composition]
Colloidal silica, sulfuric acid (special grade, manufactured by Wako Pure Chemical Industries), HEDP (1-hydroxyethylidene-1,1-diphosphonic acid, manufactured by Thermos Japan), and hydrogen peroxide (concentration: 35% by weight, manufactured by Asahi Denka) Were added to ion-exchanged water, and these were mixed to prepare a polishing liquid composition. The resulting polishing composition was finally filtered through a 0.45 μm filter. The contents of sulfuric acid, HEDP, and hydrogen peroxide in the polishing composition were 0.4 wt%, 0.1 wt%, and 0.4 wt%, respectively.

In addition, colloidal silica is silica particles in an amount of 5.0% by weight in the polishing composition (average particle size (D _BET ) by BET adsorption method: 27 nm, average particle size (D _DLS ) by dynamic light scattering method) : 37 nm, degree of irregularity (D _DLS / D _BET ) = 1.37), 0.5, 1.0, 1.5, and 2.0 wt% of the following silica particles I, or In addition, 0.5% and 1.5% by weight of the following silica particles II were added to the silica particles.

[Mixed silica particles I and silica particles II]
Silica particles I:
D _BET = 18 nm, D _DLS = 80 nm, degree of irregularity (D _DLS / D _BET ) = 4.44
Silica particles II:
D _BET = 50 nm, D _DLS = 65 nm, irregularity (D _DLS / D _BET ) = 1.30

[Measuring method of average particle diameter, CV value, and ΔCV value of colloidal silica]
[Average particle size and CV value]
The polishing composition prepared as described above was measured with a dynamic light scattering apparatus DLS-6500 manufactured by Otsuka Electronics Co., Ltd., and particle size distribution analysis was performed by a histogram method (Marquardt method). As the average particle diameter of the colloidal silica at the detection angles of 30 ° and 90 °, the average particle diameter (d50) obtained by integrating 200 times at the detection angles of 30 ° and 90 °, respectively, was used. Further, the CV value was obtained by dividing the standard deviation of the scattering intensity distribution by the average particle diameter (d50) and multiplying by 100 to obtain CV30 and CV90, respectively.
[ΔCV value]
A value obtained by subtracting the CV value (CV90) of the colloidal silica particles at a detection angle of 90 ° from the CV value (CV30) of the colloidal silica particles at a detection angle of 30 °, which was measured according to the above measurement method, was obtained as a ΔCV value.
(Measurement conditions for DLS-6500)
Detection angle: 90 °
Sampling time: 4 (μm)
Correlation Channel: 256 (ch)
Correlation Method: TI
Sampling temperature: 26.0 (° C.)
Detection angle: 30 °
Sampling time: 10 (μm)
Correlation Channel: 1024 (ch)
Correlation Method: TI
Sampling temperature: 26.0 (° C.)

  Table 1-1 shows the results of average particle diameter, standard deviation, CV value, and ΔCV value at detection angles of 30 ° and 90 ° when silica particles I (non-spherical) are added, and Table 1-2 above. Shows the results when silica particles II (spherical) are added. Moreover, the graph of FIG. 1 shows the relationship between the addition amount of particle | grains in Table 1-1 and Table 1-2, and (DELTA) CV value. As shown in Table 1-1 above, the difference between the average particle diameters of the detection angles of 30 ° and 90 ° (that is, the angle dependency of the scattering intensity) did not correlate with the content of the silica particles I. However, as shown in FIG. 1, the ΔCV value (angle dependency of the scattering intensity distribution) showed a strong correlation with the content of the silica particles I (white diamonds). On the other hand, when silica particles II were added, no significant change was observed in the ΔCV value (white squares). That is, it was shown that the presence and content of non-spherical particles contained in colloidal silica can be detected by measuring the ΔCV value.

[Example 2]
About two types of colloidal silica (silica A and silica B), the polishing liquid composition was prepared for every lot. The substrate to be polished was polished using the prepared polishing liquid composition, and scratches on the substrate after polishing were evaluated. The evaluation results are shown in Table 2 below and FIG. The method for preparing the polishing composition, the polishing conditions (polishing method), and the evaluation method are as follows. Note that the method for measuring the ΔCV value is the same as in Example 1.

[Method for preparing polishing liquid composition]
Colloidal silica, sulfuric acid (special grade, manufactured by Wako Pure Chemical Industries), HEDP (1-hydroxyethylidene-1,1-diphosphonic acid, manufactured by Thermos Japan), and hydrogen peroxide (concentration: 35% by weight, manufactured by Asahi Denka) Were added to ion-exchanged water, and these were mixed to prepare a polishing liquid composition. The contents of silica particles, sulfuric acid, HEDP, and hydrogen peroxide in the polishing composition were 5.0% by weight, 0.4% by weight, 0.1% by weight, and 0.4% by weight, respectively. Finally, the blended polishing composition was filtered through a 0.45 μm filter for polishing evaluation.

In addition, as colloidal silica, six types of lots (Lot Nos: 1 to 6) of silica A (D _BET = 17 nm, D _DLS = 26 nm, D _DLS / D _BET = 1.52) and silica B (D _BET = 27 nm) 7 lots (Lot Nos: 1 to 7) of D _DLS = 37 nm and D _DLS / D _BET = 1.37) were used. The results for silica A are shown in Table 2-1 below and FIG. The result of silica B is shown in the following Table 2-2 and FIG.

[Polishing]
The following to-be-polished substrate was grind | polished on the grinding | polishing conditions shown below using the polishing liquid composition prepared as mentioned above. Next, scratches on the polished substrate were measured and evaluated based on the following conditions. The results are shown in Table 2 below. The data shown in Table 2 below is obtained by measuring four surfaces of each substrate to be polished for each polishing liquid composition, measuring both surfaces of each substrate to be polished, and averaging the data of four surfaces (front and back). did.

[Polished substrate]
As the substrate to be polished, a substrate obtained by rough polishing an aluminum alloy substrate plated with Ni-P in advance with a polishing composition containing an alumina abrasive was used. The substrate to be polished has a thickness of 1.27 mm, an outer diameter of 95 mm, an inner diameter of 25 mm, a center line average roughness Ra measured by AFM (Digital Instrument Nanoscope IIIa Multi Mode AFM), 1 nm, and a long wavelength. The amplitude of the undulation (wavelength 0.4 to 2 mm) was 2 nm, and the amplitude of the short wavelength undulation (wavelength 50 to 400 μm) was 2 nm.

[Polishing conditions]
Polishing tester: "Fast double-sided 9B polishing machine" manufactured by Speedfam
Polishing pad: Fujibo's suede type (thickness 0.9mm, average hole diameter 30μm)
Polishing liquid composition supply amount: 100 mL / min (supply rate per 1 cm ^{2 of} polishing substrate: 0.072 mL / min)
Lower platen rotation speed: 32.5 rpm
Polishing load: 7.9 kPa
Polishing time: 4 minutes

[Scratch measurement method]
Measuring instrument: OSA6100, manufactured by Candela Instruments
Evaluation: Four substrates were randomly selected from the substrates put into the polishing tester, and each substrate was irradiated with a laser at 10,000 rpm to measure scratches. The total number of scratches (both) on each of the four substrates was divided by 8 to calculate the number of scratches per substrate surface. The results are shown in Table 2 below.

  Table 2-1 above shows the average particle diameter, standard deviation, CV value, ΔCV value, and scratch results at detection angles of 30 ° and 90 ° when polishing liquid compositions using various lots of silica A were prepared. Table 2-2 shows the results when various lots of silica B were used. Moreover, the graph of FIG. 2 is the graph which plotted the lot in Table 2-1 and Table 2-2 about (DELTA) CV value and a scratch, respectively. As shown in Table 2 and FIG. 2, in both silica A and silica B, the ΔCV value is not constant between lots, and in the lot having a high ΔCV value, the number of scratches on the substrate surface after polishing increases, and ΔCV In the lot with a low value, the number of scratches on the substrate surface after polishing was reduced. Therefore, it was shown that the scratch reduction performance can be managed by managing the quality of the ΔCV value of the colloidal silica used.

  The present invention is useful in the field of analysis, inspection, quality control, or production of a product containing a particle dispersion.

Claims (9)

  A method for detecting non-spherical particles in a particle dispersion, comprising measuring the angular dependence of the scattering intensity distribution of the particle dispersion by a dynamic light scattering method.   The non-spherical particle detection according to claim 1, wherein the measurement of the angle dependence of the scattering intensity distribution includes obtaining a difference between measured values based on the scattering intensity distribution of the particle dispersion measured at two different detection angles. how to. The two different detection angles are a detection angle in forward scattering and a detection angle in side or back scattering,
The method for detecting non-spherical particles according to claim 2, wherein the measured value based on the scattered intensity distribution is at least one selected from the group consisting of a standard deviation based on the scattered intensity distribution and a CV value based on the scattered intensity distribution. .   The non-spherical particle according to any one of claims 1 to 3, wherein an average particle diameter of the particles in the particle dispersion based on a scattering intensity distribution measured at a detection angle of 90 ° by a dynamic light scattering method is 1 to 200 nm. How to detect.   The method for detecting non-spherical particles according to claim 1, wherein the particles in the particle dispersion are inorganic particles.   The method for detecting non-spherical particles according to claim 5, wherein the inorganic particles are colloidal silica.   The inspection method of an inorganic particle dispersion which has the process of test | inspecting an inorganic particle dispersion using the method in any one of Claim 1 to 6.   The manufacturing method of an inorganic particle dispersion which has the process of test | inspecting an inorganic particle dispersion using the method in any one of Claim 1 to 6. An inspection step of the inorganic particle dispersion by the inspection method according to claim 7, or a manufacturing step of the inorganic particle dispersion by the manufacturing method according to claim 8,
The inorganic particle dispersion is a polishing composition,
A method for producing a substrate, comprising a step of polishing a substrate to be polished using the polishing composition.

Images