JP2001041952A - Method for evaluating cleanliness of metallic material - Google Patents

Abstract

PROBLEM TO BE SOLVED: To quickly, highly accurately and highly reliably evaluate cleanliness by obtaining the maximum diameter of an intervening substance from requirement time for reduction of the interposed substance and a calibration curve, and calculating the estimated maximum diameter according a specific formula. SOLUTION: An interposed substance, in each of (n) sample pieces collected from a metallic material to be analyzed, is reduced. A maximum diameter aj(j=1 or n) of the intervening substance is calculated for each sample piece from a calibration curve, in which reduction requirement times TE and maximum diameters of intervening substances are made corresponding to each other. Calibration curve is formed through acid dissolving and extracting the intervening substance in the metallic material, observing the substance, under a microscope to obtain the size of the substance. A recursion time T is obtained by T=(V+V0)/V0, where V0 is an inspection reference volume of one sample piece and V is an estimate volume. Moreover, a standardization variable ymax is obtained with ymax=-1n[-1n (T-1)/T}], where (n) is the number of inspection times. An estimate maximum diameter amax of the intervening substance is calculated from amax=t×ymax+u, where (t) is regression coefficient and (u) is a constant from linear regression expression of a maximum diameter aj and the standardization variable yj of the intervening substance, and cleanliness of the metallic material to be analyzed is evaluated.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for evaluating the cleanliness of a metal material. More specifically, the present invention relates to a method for performing a highly reliable cleanness evaluation on a large amount of metal materials from the relationship between the time required to reduce oxide-based inclusions in the metal material and the maximum diameter of the oxide-based inclusions. .

[0002]

2. Description of the Related Art With the recent improvement in metallurgical technology, the cleanliness of metallic materials such as steel has been greatly improved, and non-metallic inclusions in medium to large metallic materials exceeding 20 microns have been further reduced. The size is also smaller. Under such circumstances, it is extremely difficult to detect large inclusions that occur accidentally or with extremely low probability. However, for example, Al ₂ O ₃ , MgO · Al ₂ O ₃ , (Ca, M
g) Inclusions such as O.Al ₂ O ₃ tend to cause fatigue fracture in bearing steel and the like, and are still a problem.

At present, as a standard inspection method for checking the cleanliness of a metal material, a method using an optical microscope is standard. But,
In this method, the area to be inspected is as small as about 1000 mm ^2, and it cannot be practically used as a method for evaluating the cleanliness of a metal material having high cleanliness as described above (JIS G 05
55, ASTM E45, ASTMA295, DIN50602, ISO4967, etc.).

Further, a method of extracting inclusions from a metal material by acid dissolution and evaluating the particle size of the inclusions with a microscope has been proposed.
A method has been proposed in which a metal material is melted by a melting method and inclusions that have floated are observed with a microscope (Japanese Patent Application Laid-Open No. 9-1 / 1991).
25199, JP-A-9-125200), the inclusions may be dissolved in the acid, or the inclusions themselves may be melted or agglomerated, and the dissolution of the acid may take a long time, resulting in poor processing speed. Also, it was difficult to cope with the mass production process of products.

[0005] Under such circumstances, it has been desired to develop a new technology capable of actually evaluating and guaranteeing the cleanliness of a metal material.

[0006]

SUMMARY OF THE INVENTION Accordingly, the present invention provides a highly reliable metal material having a high degree of cleanliness in response to recent improvements in metallurgical technology and a great improvement in the cleanliness of metal materials such as steel. It does not provide an evaluation method for It is another object of the present invention to provide a method for quickly evaluating the cleanliness of a metal material that is compatible with such a metal material mass-production process.

[0007]

Means for Solving the Problems The present inventor has set forth the above object.
For the purpose of solving ₀
= 100mm^TwoAnd the number of samples n = 30 to 60 are collected,
Extreme value from the maximum inclusion diameter that appeared in each sample
Statistical procedures are used to determine the presence of inclusions in the metallic material being analyzed.
We have been studying a method to predict the large diameter.
The reliability of predicting large inclusions was low. Therefore,
In the method using a microscope, the evaluation of cleanliness of metal materials
It was not satisfactory.

The present inventors have further studied and, in a method for analyzing the amount of oxygen in a metal, have repeatedly studied the extraction waveform of carbon monoxide gas and the particle size of oxide-based inclusions in the metal. As a result, they have found that there is a correlation between the time required to reduce the oxide-based inclusions and the maximum particle size of the oxide-based inclusions, and have completed the present invention.

The present invention provides means for solving the above-mentioned problems, the gist of which is as set forth in the appended claims.

That is, the present invention is a method for evaluating the cleanliness of a metal material by obtaining the maximum diameter of an oxide-based inclusion, wherein n sample pieces are sampled from a metal material to be analyzed and each sample is sampled. The time T _E required for reducing the inclusions in the piece and reducing
They were respectively measured for each sample piece, a calibration curve and the maximum diameter of the inclusions T _E corresponding, maximum diameter a _j of n the inclusions for each specimen of the (j = 1, n) And calculating the estimated maximum diameter a _max of the inclusions in the metal material to be analyzed by the following equations (1) and (1 ′) to evaluate the cleanliness of the metal material to be analyzed. This is a method for evaluating the cleanliness of metal materials. [Equation 1] The maximum diameter a _j (j = 1, 2) of oxide-based inclusions
n) and a linear regression equation of the standardization variable y _j (j = 1, n) a = ty + u (1) where n = the number of inspections Standardization variable y _j = −ln [−ln ｛j / (n + 1)]] (j = 1,
n) t = regression coefficient u = constant [Formula 1 '] Formula for calculating the estimated maximum diameter a _max of oxide-based inclusions in the metal to be analyzed (regression formula) a _max = t × y _max + u... (1 ′) where V _o = inspection reference volume (mm ³ ) V = volume to perform prediction (mm ³ ) T (recursion period) = (V + V _o ) / V _o y _max (normalization Variable) =-ln [-ln ｛(T-1) / T]]

The present invention also relates to a method for evaluating the cleanliness of a metal by determining the maximum diameter of an oxide-based inclusion, wherein n pieces of a sample are collected from a metal material to be analyzed, and reducing the inclusions in, seek time T _E necessary for the reduction were measured for each sample piece _{T Ej (j = 1, n} ), the estimated maximum T _Emax by the following formula (2) and (2 ') calculates, from the calibration curve and the maximum diameter of the inclusions T _E corresponding, estimated maximum diameter a of the inclusions of the analyzed metal material
This is a method for evaluating the cleanliness of a metal material, wherein _max is calculated to evaluate the cleanliness of the metal material to be analyzed. [Equation 2] Time T _Ej (j required for reduction of oxide-based inclusions
= 1, n) and the reference variables _{y j (j = 1, n} ) of the linear regression equation _{T E = ty + u ··········· (} 1) However, n = number of tests standard variables y _j = −ln [−ln ｛j / (n + 1)]] (j = 1,
n) t = regression coefficient u = constant [Equation 2 '] Calculation equation (regression equation) of estimated maximum _TEmax of oxide-based inclusions in the metal to be analyzed _TEEmax = t * _ymax + u ... (1 ′) where V _o = inspection reference volume (mm ³ ) V = predicted volume (mm ³ ) T (recursion period) = (V + V _o ) / V _o y _max (standardization variable ) = − Ln [−ln ｛(T−1) / T]]

The present invention also provides a method for heating and melting a sample under an inert gas atmosphere and reacting the sample with a carbon source to generate a carbon monoxide gas. measured every, the measurements of each sample piece T _E
And a method for evaluating the cleanliness of said metal material.

[0013]

BEST MODE FOR CARRYING OUT THE INVENTION First, as a result of various studies, the present inventor accidentally, under a situation where non-metallic inclusions in a metal material exceeding 20 microns are further reduced and the size is reduced, Alternatively, it has been concluded that it is extremely difficult to detect large inclusions that occur with a very low probability by a method based on microscopic observation. It was considered that such large inclusions did not always appear on the test surface, but rather were often hidden and not observed.

For this reason, it has been considered that it is practically impossible to evaluate and assure the quality of the cleanliness of the metal material using a method based on microscopic observation.

Therefore, the present inventors have made various studies and found that
They found that the gas extraction waveform obtained by oxygen analysis of metallic materials could be used.

By applying this oxygen analysis method, the present inventors have succeeded in achieving the effect that the maximum diameter of inclusions can be inspected 1,000 to tens of thousands times larger than the conventional one. Hereinafter, embodiments of the present invention will be specifically described.

In the present invention, n sample pieces to be subjected to oxygen analysis are collected from the metal material to be analyzed. The n pieces may be collected from any part of the metal material to be analyzed, but a predetermined inspection part is set so that the metal material to be analyzed can be appropriately evaluated according to the purpose of the evaluation. However, it is preferable to collect from the inspection site.
For example, if a continuously cast steel slab is to be analyzed, the inspection site is set at the top, middle, and bottom of the continuously cast steel slab as shown in FIG. Thus, it can be set at a site where large non-metallic inclusions are likely to occur. Further, it is preferable to collect a plurality (for example, three) of sample pieces having the same shape from the set inspection site. For example, in this manner, all parts of the metal material to be analyzed can be inspected on average. Also,
In the case of using a continuously cast steel billet as the analysis target metal material, set the inspection site at all the top, middle, and bottom sites and collect samples to inspect the sites corresponding to the initial, middle, and end stages of casting. It will also be.

[0018] As the number n of the sample pieces increases statistically, the reliability is improved.
The number is 60, particularly preferably 30 to 40. Within this range, the test can be performed quickly without a great burden on the work, and statistically reliable estimation data of the maximum diameter of the inclusion can be obtained.

The weight of a sample piece collected from a metal to be analyzed and analyzed by an oxygen analysis method depends on the conditions of an analytical instrument for performing oxygen analysis, but is preferably 0.5 to 3 g, particularly preferably 0.1 to 3 g. 8 to 1 gram. This range is preferable in that reliable data can be obtained and the test can be performed quickly.

Here, in order to compare the analytical weight according to the method of the present invention with the number of visual fields measured by optical microscopy, the weight is converted into the visual field as follows.

The number of inclusions of high clean steel is small. For this reason, in order to accurately measure the particle size of inclusions by microscopic observation, it is generally necessary to observe at a magnification of about 400 times. 03
If it is 61 mm ² , about 100 visual fields are required. More specifically, for example, the surface of a 5 × 5 × 5 mm ³ (approximately 0.98 gram) metal sample piece is observed at a magnification of 400 × 5 × 5 mm ² , and when observation is completed, 10 μm is polished. If the observation was repeated in the same manner, 1
The number of fields required to finish observing one specimen is (5 × 5 ×
5) / (0.01 × 0.0361) = 3.5 × 10 ⁵ Therefore, the amount of information obtained by the oxygen analysis method is 3.5 × 10 5 in comparison with the conventional microscopy (100 visual fields).
³ times. Assuming that the number of sample pieces is n = 30 to 60, the amount of information per charge is 1-2 × 10 ⁵ times larger than that of the conventional microscopy. Therefore, the inspection was performed 100,000 times as compared with the conventional optical microscope.

[0022] In the evaluation method of the present invention is to reduce the oxide-based inclusions of the sample piece which had been collected from a analyte metallic material, determining the time T _E necessary for the reduction. Specifically, for example, a sample piece is heated and melted in an inert gas atmosphere and reacted with a carbon source to generate carbon monoxide gas (hereinafter sometimes referred to as “CO gas”), thereby generating CO gas. by measuring the time to the end for each specimen from the measured value by a T _E of the specimen, it is possible to determine the time T _E necessary for the reduction. That is, the same method as the oxygen analysis method for obtaining the amount of oxygen from the amount of extracted CO gas can be used. More specifically, for example, a so-called inert gas carrier melting / infrared absorption method (Japanese Patent Laid-Open No. 11-30613) ).

Inert gas transport melting-In the infrared absorption method,
In an extraction furnace in an inert gas atmosphere, a metal sample piece is heated and melted, and reacted with a carbon source in the extraction furnace to generate CO gas.The CO gas is transported by the inert gas and CO is absorbed by an infrared absorption method. Measure the gas. As a carbon source, for example,
When a carbon source is contained in a graphite crucible, a graphite powder, a graphite capsule, or a metal sample, it can be used.
When graphite powder or graphite capsules are used, they are mixed with a sample piece and heated. When a graphite crucible is used, the sample piece is put into a graphite crucible and heated and melted.

FIG. 1 illustrates an oxygen extraction curve when a steel sample is dropped into a graphite crucible in an inert gas atmosphere, heated and melted, and CO gas is extracted from the molten bath. Since it is obtained by converting from the extraction amount of CO gas, CO may be directly displayed on the vertical axis.) When the sample piece in the graphite crucible is melted, gas generation starts at the point S, and then the gas is generated. The amount of gas generated increases to point M, and then the amount of gas generated decreases to point E.

[0025] Here, the present inventors have found that the time required from point S to point E, i.e. made various investigations about the determinants of time T _E necessary for the reduction of oxide inclusions in the specimen in. As a result, factors that determine the T _E, it has become clear that there is an analysis device factors and material factors for analysis.
The former analysis material factors, that is, oxide inclusions, include: 1) particle size of oxide inclusions, 2) amount of oxide inclusions, and 3) composition (type) of oxide inclusions. And the latter factors of the analyzer are 1) heating pattern,
2) There is a gas flow rate and the like. In this, as a result of various investigations for factors on determination of T _E, in cause of the particle size of the oxide diameter inclusions found newly that is closely related to T _E.

[0026] That is, as illustrated in the graph of FIG. 3, it was found that the maximum particle diameter of the oxide diameter inclusions as T _E increases there is a correlation shown a tendency to increase.
The present invention resides in utilizing this new finding. The graph shown in FIG. 3 is a standard curve, it is a T _E obtained by oxygen analysis to determine the maximum diameter of the oxide inclusions in the sample pieces in. Calibration curve between the maximum diameter of the T _E and inclusions, the inclusions in the metal material is extracted with an acid dissolution, or by scraped, it can be prepared by, for example, obtaining the size was observed with a microscope. It is desirable to create a calibration curve every time a factor such as the type of metal material, the pattern of temperature rise of the analyzer, the gas flow rate, or the like is different.

The measured T _E every n each sample piece, maximum diameter a _j of n inclusions sample piece from the calibration curve of the maximum diameter of the inclusions T _E prepared in advance ( j = 1, n)
Is required.

Next, from the maximum diameter a _j (j = 1, n) of the inclusion for each sample piece thus determined, the above equation (1) is obtained.
The estimated maximum diameter a _max of the oxide-based inclusions in the metal material to be analyzed is calculated according to (1 ′). Equation (1) is a linear regression equation obtained from _aj (j = 1, n) previously obtained and the standardized variable _yi . Equation (1 ′) is used to determine the maximum diameter a _max of the oxide-based inclusions in the metal material to be analyzed. That is, the recursion period T is obtained from the volume (inspection reference volume) V _{0 of} one sample piece and the volume V for which prediction is performed,
Further, y _max is obtained, and a _max is calculated from Expression (1 ′).
The thus calculated a _max is the estimated maximum diameter of the oxide-based inclusions in the volume V of the metal material to be analyzed.

[0029] Another embodiment of the present invention, after obtaining the previously estimated maximum T _E, in terms of the estimated maximum size a _max from the estimated maximum T _E by a calibration curve of the maximum diameter of the inclusions T _E You can also ask. In this case, equations (2) and (2 ') are used. That is, after obtaining the T _E for each sample piece, the formula (2) and the estimated maximum value of T _E from (2 ') T
_Emax is calculated, and an estimated maximum diameter a _max of the inclusion can be obtained from a calibration curve in which T _E and the maximum diameter of the inclusion correspond. Other points such as the method of measuring T _E of the specimen can be performed in the same manner as embodiment described above.

The evaluation method of the present invention, a T _E obtained by measuring the oxygen analysis a diameter of the inclusions is associated, by using a more statistical method, part of the data of the analyzed metal material From this, the maximum diameter a _max of the oxide-based inclusions in the entire metal material to be analyzed was successfully estimated with high accuracy.

Although the evaluation method of the present invention can be widely used for metal materials, preferably, the evaluation method of the present invention is preferably used for Al alloys and F alloys.
e alloy, Mg alloy, Ti alloy, Cr alloy, Co alloy, N
i alloy, Cu alloy, Zn alloy, Ag alloy, Sn alloy, W
Alloys and the like, more preferably Al alloys and Fe alloys, and particularly preferably steel materials.

The evaluation method of the present invention can be suitably used, for example, for collecting molten steel at the time of continuous casting and evaluating cleanliness in a smelting charge.

Further, according to the evaluation method of the present invention, it is possible to estimate the maximum diameter of the oxide-based inclusions in the metal material. In particular, Al ₂ O ₃ , MgO.Al ₂ O ₃ , (Ca, M
g) It is suitable for estimating the maximum diameter of oxide inclusions such as O.Al ₂ O ₃ .

[0034]

Embodiments of the present invention will be described below in detail. (Example 1) 1. Analyte metal material and its treatment As a metal object to be analyzed, a round bar-shaped steel piece as shown in FIG. 2 of 160 t high carbon Cr bearing steel (for rod pipe) manufactured by a continuous casting method was used. As described above, the cleanliness was evaluated by the method of the present invention.

Inspection sites were set at the portions of the round bar-shaped steel slabs shown in FIG. 2 and test specimens of 5 × 5 × 5 mm were collected from each inspection site. A total of 30 analysis test pieces were cut out from the top part and the middle part, and four pieces each from the bottom part (that is, n = 30). Since the bottom part generally has a lower degree of cleanliness of the steel material than other parts, four sample pieces were collected from the bottom part more than the other parts.

Next, each test piece was dropped into a graphite crucible, heated and melted, and a gas was extracted from the molten bath and analyzed by an inert gas carrier melting / infrared absorption method. The T _E 1 th sample piece was measured as T _E1, was measured sequentially until T _E30.

Next, an inspection prepared based on microscopic observation was performed.
Quantitative curve (maximum diameter of inclusion and T_EEach sample piece
Diameter of inclusions, ie, maximum of oxide inclusions
Diameter a _j(J = 1, n) was determined.

2. Estimation of the maximum diameter of the oxide-based inclusions in the metal material to be analyzed From the maximum diameter a _j (j = 1, n) of the inclusions for each of the 30 test pieces obtained as described above, Thus, the estimated maximum diameter a _max of the inclusion was obtained.

First, the maximum diameter a of the inclusion is set in order from the minimum value.
A, in ascending order₁, A_Two, ... a _jDefined.

Here, 1, 2,.
The value obtained by calculating j twice by logarithm is the normalized variable y _i in the proviso to [Expression 1]. Table 1 summarizes j, a _j , and y _j .

[0041]

[Table 1]

In FIG. 4, the inclusion diameter is plotted in order from the smallest inclusion diameter (ie, a ₁ ) with the inclusion diameter on the abscissa and the normalized variable on the ordinate. A linear regression of the black circle is a straight line that rises to the right on the right side (the straight line is expressed by an equation, which is the equation (1) for the metal material to be analyzed in this example).

Here, in the oxygen analysis test, since the weight (or volume) of each test piece is constant, the standardized variable on the vertical axis represents the weight of the sample. In Table 1, y _i = −1.2
Since 337 is the reference weight (or volume) of one sample piece, (5 × 5 × 5) = 125 mm ³ (= V ₀ ) (approximately 0.9
8g), and y _i = −1.0082 corresponds to 2V ₀ in the reference volume for two sample pieces. Thus, the volume V
On the other hand, when the maximum diameter inclusion a _max included in the volume is predicted, it is sufficient to perform an inverse calculation from the value on the vertical axis corresponding to the volume V. This conversion formula is Expression (1 ′), and the value of the vertical axis corresponding to the volume V to be predicted by T (recursion period) in the proviso of [Expression 1 ′] may be obtained.

In the case of FIG. 4, the volume to be predicted is 620,000 mm.
^{3 The} maximum inclusion diameter indicated by the straight line rising to the right on the right side is 29.
2 μm. The volume of 620,000 mm ³ is approximately 500 g in terms of weight.

On the other hand, in the microscopic method (conventional method), the inspection reference area S ₀ = 100 mm ² , and the area S to be predicted S = 30000-
50,000 mm ² [provided T of expression 1 ′ (recursion period) =
300 to 500], and T = 500.
When the maximum inclusion diameter that appeared was estimated from FIG. 4, it was 16.1 μm.

3. Evaluation of cleanliness of metal material to be analyzed Cleanliness of metal material to be analyzed is estimated maximum inclusion diameter = a
_max, inspection reference volume V ₀ mm ^3, can be given as the volume Vmm ³ to make predictions. The above results are summarized as follows.

According to the method of the present invention, the evaluation of the cleanness of the round bar-shaped lump as the metal material to be analyzed is performed by estimating the maximum inclusion diameter a _max = 29.2 μm and the inspection reference volume V ₀ = 125 m.
m ³ , the volume V to be predicted was 6.2 × 10 ⁴ mm.

[0048] In contrast, the maximum inclusion diameter from about 1 inclusions findings by microscopy, the inspection reference area S ₀ = 100 mm ^2, it appears in the area S = 50,000 mm ² make predictions
It was 6.1 μm.

The following test was conducted to confirm the evaluation accuracy of these results. Small mechanical parts which were subjected to repeated stress and required to have fatigue strength were produced from the round bar-shaped mass used in this example, and subjected to a fatigue fracture test. When the diameter of the inclusions observed on the fracture surface of the fracture test piece was measured, it was 29.3.
μm.

Therefore, it was found that the value of the estimated maximum diameter of the present invention matched well with the actually measured value, and the prediction accuracy of the evaluation method of the present invention was extremely excellent.

(Example 2) Spring steel (JIS steel type SUP1)
0) was melted in an electric furnace for 150 tons. After degassing the RH, it was cast into a slab (plume) having a cross section of 380 × 490 mm by continuous casting. Then, it was subjected to bulk rolling to obtain a billet having a diameter of 167 mm and a weight of 2 t. This was rolled and processed into a φ5 valve spring. When this spring was used, it was broken during use. When the broken portion was examined, inclusions of 50 μm were found.

On the other hand, of the rolled material subjected to the spring, a sample piece was cut out from the remaining rolled material that had been stored without being subjected to spring processing, and was prepared in the same manner as in Example 1 described above. When the particle diameter of the oxide-based inclusions was evaluated by an analytical method, it was estimated that the maximum diameter of the oxide-based inclusions that could be present in about 2 t of the rolled material subjected to the spring working was 53 μm. Thus, it was found that the method was suitable as a method for evaluating the maximum diameter of oxide-based inclusions.

[0053]

According to the present invention, the cleanliness of a metal material can be quickly and accurately evaluated with high reliability.

The present invention also contributes to the recent evaluation of cleanliness and quality assurance of metal materials such as steel, and is a very useful invention that meets the needs of the industry.

[Brief description of the drawings]

FIG. 1 is an explanatory diagram showing an oxygen extraction curve by an oxygen analysis method.

FIG. 2 shows the setting of an inspection site in a metal material to be analyzed;
It is a figure which shows an example of the collection number of the sample piece from each test | inspection part.

3 is a diagram showing an example of a comparison calibration curve T _E obtained from the extraction curve of oxygen and the maximum diameter of the oxide inclusions.

FIG. 4 is a diagram showing a comparison between the optical microscopy (conventional method) and the method of the present invention in estimating the maximum diameter of oxide-based inclusions.

Continuation of the front page (72) Inventor Yoshiyuki Kato 3007 one character of Nakajima character in Shima, Himeji-shi, Hyogo F-term in Sanyo Special Steel Co., Ltd. 2G055 AA03 BA04 BA20 CA25 EA05 EA08 EA10 FA02

Claims (3)

[Claims] 1. A method for evaluating the cleanliness of a metal material by obtaining a maximum diameter of an oxide-based inclusion, wherein n sample pieces are collected from a metal material to be analyzed, and the inclusions were reduced, the time T _E necessary for the reduction were measured for each sample piece, a calibration curve and the maximum diameter of the inclusions T _E corresponding, in said n inclusions for each specimen The maximum diameter a _j (j = 1, n) is calculated, and the estimated maximum diameter a _max of the inclusions in the metal material to be analyzed is calculated by the following equations (1) and (1 ′), and the metal to be analyzed is calculated. A method for evaluating the cleanliness of a metal material, comprising evaluating the cleanliness of the material. [Equation 1] The maximum diameter a _j (j = 1, 2) of oxide-based inclusions
n) and a linear regression equation of the standardization variable y _j (j = 1, n) a = ty + u (1) where n = the number of inspections Standardization variable y _j = −ln [−ln ｛j / (n + 1)]] (j = 1,
n) t = regression coefficient u = constant [Equation 1 '] Formula for calculating the estimated maximum diameter a _max of oxide-based inclusions in the metal to be analyzed (regression formula) a _max = t × y _max + u... (1 ′) where V _o = inspection reference volume (mm ³ ) V = volume for prediction (mm ³ ) T (recursion period) = (V + V _o ) / V _o y _max (normalization Variable) =-ln [-ln ｛(T-1) / T]] 2. A method for evaluating the cleanliness of a metal by determining the maximum diameter of an oxide-based inclusion, comprising: collecting n pieces of a sample from a metal material to be analyzed; reducing things, determine the time T _E necessary for the reduction were measured for each sample piece _{T Ej (j = 1, n} ), calculates the estimated maximum T _Emax by the following formula (2) and (2 '), a calibration curve between T _E and the maximum diameter of the inclusions is corresponding, to assess the cleanliness of the analyzed metal material by calculating the estimated maximum size a _max of the inclusions of the analyzed metal material Characteristic metal material cleanliness evaluation method. [Equation 2] Time T _Ej (j required for reduction of oxide-based inclusions
= 1, n) and the reference variables _{y j (j = 1, n} ) of the linear regression equation _{T E = ty + u ··········· (} 1) However, n = number of tests standard variables y _j = −ln [−ln ｛j / (n + 1)]] (j = 1,
n) t = regression coefficient u = constant [Equation 2 '] Calculation equation (regression equation) of estimated maximum _TEmax of oxide-based inclusions in the metal to be analyzed _TEEmax = t * _ymax + u ... (1 ′) where V _o = inspection reference volume (mm ³ ) V = predicted volume (mm ³ ) T (recursion period) = (V + V _o ) / V _o y _max (standardization variable ) = − Ln [−ln ｛(T−1) / T]] 3. A sample piece is heated and melted in an inert gas atmosphere, reacted with a carbon source to generate carbon monoxide gas, and the time from generation to termination of carbon monoxide gas is measured for each sample piece. Te, characterized by the measured value T _E of each sample piece, cleanliness evaluation method for a metal material according to claim 1 or 2.