RU2131403C1 - Method of determining heat resistance of constructional ceramic materials - Google Patents

Abstract

FIELD: manufacture of building materials. SUBSTANCE: prismatic samples with lateral indent simulating crack and containing no inflicted defects on their tops are manufactured. Heat resistance of samples is determined from values of following relationships: (1-K $\genfrac{}{}{0ex}{}{\text{*}}{\text{1c}}$ /K_1c) x 100% and σ₁/σ₂, where K $\genfrac{}{}{0ex}{}{\text{*}}{\text{1c}}$ is critical coefficient of stress intensity after heat shock in a sample, K_1c average value of critical coefficient of stress intensity in samples before heat shock, σ₁ ultimate strength of a bent cut sample after heat shock, and σ₂ average value of ultimate strength of bent cut samples before heat shock. EFFECT: improved determination accuracy due to enabled quantitative evaluation of ceramic structure resistance to formation of heat cracks. 1 tbl

Description

 The invention relates to ceramic technology and can be used to correctly assess the heat resistance of structural ceramics operating under thermal stresses due to temperature gradient. Existing guest methods for determining heat resistance (GOST 7875-83) by the number of heating-cooling cycles until the first cracks, until complete failure, before losing 20% of the initial mass were developed for refractory products and allow only a qualitative assessment of heat resistance to be made without revealing it Relations with other material properties. The variety of existing design criteria for heat resistance / 1 / allows only to predict it with a certain reliability, given the thermophysical, strength, elastic, deformative and some other properties. The technique, based on direct measurement of the temperature gradient, leading to the destruction of the wall of the hollow cylinder / 2 / for a given heating mode, is very progressive, however, the influence of the shape factor and volume of the product (the so-called "scale factor") on heat resistance should be taken into account. That is, it can be argued that when the shape or volume of the test sample changes, the distribution of the temperature fields will also change and, as a result, the critical level of the resulting thermal stresses will change. Therefore, in practice, many researchers recommend to test directly the ceramic product for heat resistance in the mode of temperature differences that occur during high-temperature operation. This technique is very time-consuming and can not always be implemented. The closest to the claimed technical essence and the achieved effect is the method / 3 /, according to which experimental prismatic samples are made, a thermal shock is created by the heating-cooling cycle, and the amount of loss or residual strength of the samples during bending is determined with respect to its value before thermal shock. This method is the most affordable and used to assess the heat resistance of structural ceramics. However, due to the statistical nature of the strength of ceramics as a brittle material, there is a large scattering and, as a result, a significant error (10-30%) in measuring the value of the loss of strength after thermal shock. Therefore, for a statistically reliable assessment of heat resistance by this method requires a large number of measurements (10-30 samples). It is also known that a change in the volume of the test sample will significantly affect the heat resistance also due to the statistical aspect: the larger the volume of the sample, the greater the likelihood of a critical defect (or defects), from which the growth of thermal cracks will be initiated. Therefore, as a rule, samples of a smaller volume show greater heat resistance compared to samples of a larger volume (subject to the same geometric shape). The latter makes it difficult to reliable assessment of the heat resistance of large-sized products according to the test results of a series of prototypes. In addition, this method does not allow to quantify the resistance to initiation of thermal cracks provided by the structure of a ceramic material, as well as the sensitivity (insensitivity) of the ceramic structure to real defects (stress concentrators) that are formed as a result of thermal shock.

 The objective of the invention is to reduce the error in determining heat resistance while reducing the number of test samples tested, leveling the influence of the volume factor of the test sample on heat resistance, as well as to obtain a quantitative estimate of the resistance of the ceramic structure to the initiation of thermal cracks and its sensitivity to defects formed as a result of thermal shock.

To accomplish the task in a method for determining the heat resistance of structural ceramic materials, including the manufacture of experimental prismatic samples, creating a thermal shock by means of a heating-cooling cycle, and then determining the loss of physical and mechanical properties in relation to their values before thermal shock, the samples are made with a side notch simulating a crack and not containing induced defects at the apex, and without an incision, and heat resistance is determined by the values of the ratios
(1 - K * _1c / K _1c ) • 100%
and
σ ₁ / σ ₂ ,
where K _1c is the average value of the critical coefficient of stress intensity of the samples before thermal shock;
K * _1c is the critical stress intensity factor of the sample after thermal shock;
σ ₁ - tensile strength in bending of the notched sample after thermal shock;
σ ₂ - is the average value of the tensile strength when bending samples without notching before thermal shock.

According to the claimed method for determining the heat resistance of structural ceramics, it is proposed to use the criterion K _1c in the framework of the concepts of linear fracture mechanics. K _1c (MPa • m ^1/2) ) is the critical stress intensity factor at the crack tip, at the value of which the crack starts (condition of plane deformation, crack normal crack). The larger the value of the parameter K _1c , the greater the intensity of stresses at the tip of the crack will be its pulling (propagation) into the body of the sample as a result of applying an external load and, therefore, the greater the resistance to crack initiation is exerted by the ceramic structure. To determine the parameter K _1c, prismatic samples with a specially created stress concentrator (crack or notch simulating a crack) are tested for bending according to a three or four-point loading scheme. It is preferable to use a crack as a stress concentrator, however, creating a crack in a ceramic specimen with an uniquely reproduced front configuration, a given length and orientation is a laborious and difficult task. This is due to the fact that when a crack grows, its front interacts with elements of the ceramic structure (grains, grain boundaries, micropores, other phases, stressed local regions, etc.), as a result of which branching or irreproducible bending of the front may occur. Therefore, in ceramics, there is a significant variation in the measured values of the parameter K _1c due to the uncertainty in the configuration of the crack front. In the present invention, it is proposed to use an incision as a stress concentrator under the condition that there are no induced defects in the region at its apex. Such induced defects (microcracks) can occur when cutting sintered samples with a rotating diamond wheel, their sizes can be commensurate with the sizes of grains in the ceramic structure, but their presence can lead to an increase in the stress concentration at the top of the notch during testing. This can also lead to a spread in the measured values of the parameter K _1c . The absence of induced defects can be achieved by selecting the appropriate ceramic cutting mode (speed of rotation of the diamond wheel, the force of pressing the wheel against the sample, the exclusion of beats, the use of cutting fluid), taking into account its density and structure. For the final control of the absence of induced defects in the area at the top of the incision, the staining method can be used, followed by observation under a microscope. The control should be carried out on several samples to establish the optimal cutting mode. A positive result is achieved with a two-stage cutting: first cut with a diamond wheel, and then the cut is lengthened by sawing with a thin wire covered with a layer of diamond paste, forming the top of the cut, the area at which does not contain induced defects. The authors developed their own method of forming an incision in a crude sample, according to which an incision is first formed during pressing by pressing a sharpened wedge into the sample body, and then an additional 0.1 mm thick steel blade is cut from the top of the formed wedge-shaped incision (after removing the wedge) with an angle sharpening 14 ^o . After sintering, the region at the top of the notch does not contain induced defects as a result of excluding the impact of a diamond tool on the sintered sample.

In addition, when creating an incision, the condition imposed on the radius of curvature of its vertex (ρ) should be fulfilled, under which it adequately simulates a crack: ρ ≤ 10d, where d is the average grain size. It is known from practice that an overestimation of the value (ρ) leads to incorrect certification of ceramics with respect to the parameter K _1c ; it is overestimated. For fine-grained structural ceramics with values of d 1, 5, 10 μm, the values (ρ) will be 10, 50, 100 μm, respectively. Incisions with (ρ) 50, 100 μm and above can be obtained using diamond wheels. An incision with ρ ≤ 10 μm for ceramics sintered from ultrafine powders was obtained by the authors using the proprietary technique described above. In this case, taking into account the relative linear shrinkage of 15-25%, the values (ρ) are 5-10 microns.

σ₁ = 1,5•P $\genfrac{}{}{0ex}{}{\text{*}}{\text{c}}$ •L/b•(h-l)² (3)
где P_c и P*_c - значения критических (разрушающих) нагрузок до и после термоудара соответственно;
l - глубина надреза,
h - высота образца;
Y(l/h) - коэффициент, зависящий от соотношения l/h и условий нагружения;
L - расстояние между опорами;
b - ширина образца.Thus, according to the claimed method, 2 types of prismatic samples (beams) are made with an incision and without an incision. The notch is made to a depth of 0.5 of the height of the sample, the radius of curvature of the notch tip ρ ≤ 10d (d is the average grain size) is the condition for modeling a crack, the region at the notch tip does not contain induced defects to ensure reproducible results under the selected test conditions. Some of the notched samples are subjected to thermal shock by heating and subsequent cooling in a given mode (for example, 850 ^o - water, 1200 ^o - air). After trials for three-point bending of incised samples not subjected to thermal shock and after thermal shock, the values of the critical stress intensity factor before thermal shock (1), after thermal shock (2), and also the tensile strength after bending after thermal shock (3) are calculated:

σ ₁ = 1.5 • P $\genfrac{}{}{0ex}{}{\text{*}}{\text{c}}$ • L / b • (hl) ² (3)
where P _c and P * _c are the values of critical (destructive) loads before and after thermal shock, respectively;
l is the depth of cut
h is the height of the sample;
Y (l / h) is a coefficient depending on the ratio l / h and loading conditions;
L is the distance between the supports;
b is the width of the sample.

(n - количество значений измеряемых величин a_i) ε_i = a_i-a - отклонение измеряемой величины от среднего арифметического. Простую среднюю ошибку определяли как

а среднеквадратическую ошибку

Из результатов, представленных в таблице, видно, что определение термостойкости по предложенному способу характеризуется существенно меньшей ошибкой. Для оценки термостойкости по предложенному способу достаточно использовать результаты испытаний 3-6 надрезанных образцов в виду малого разброса измеряемых характеристик. Кроме того, при использовании заявленного способа наблюдается тенденция к нивелированию влияния объема испытуемого образца на термостойкость. При испытании (термоудар 600^oC-воздух) серий образцов ZrO₂- Y₂O₃ (3 мол.%), отличающихся по объемам (3х5х30 мм³ -1 тип образцов, 5х8х40 мм³ - 2 тип образцов), установлено, что средняя потеря трещиностойкости у образцов 1 и 2 типа составила 21,2% и 22,0% соответственно, а нечувствительность к надрезу для этих типов образцов составила 0,25 и 0,23 соответственно. Тогда как средняя величина потери прочности при изгибе для образцов этих типов существенно отличалась и составляла 32% и 51% соответственно.After testing prismatic samples without a notch, the bending strength (4) is calculated (the samples are not subjected to thermal shock):
σ ₂ = 1.5 • P _c • L / b • h ² (4)
The average values of K _1c and σ _{2 are} calculated as a result of testing at least 5-6 samples. Then, to determine the heat resistance, the following relations should be used: (1 - K * _1c / K _1c ) • 100% (6) and σ ₁ / σ ₂ (7).
In its physical sense, relation (6) is the value of the loss of crack resistance (in%) of ceramics after thermal shock, since K _1c is a crack resistance parameter characterizing the crack initiation resistance. Relation (7) represents the "insensitivity" (relative units) of the ceramic structure to defects formed in the region near the top of the notch as a result of thermal shock. Theoretically, the magnitude of this ratio cannot exceed unity, the larger it is, the higher the "insensitivity" of ceramics to a defect that controls failure under the influence of thermal stresses (during thermal shock). Both relations (6) and (7) are calculated after testing samples with a notch, which is the only stress concentrator, from the top of which failure is initiated under the influence of thermal stresses. Since fracture during thermal shock is initiated from the top of the created notch, as the sharpest stress concentrator, in this case the task of the invention is really fulfilled: the dispersion of the measured characteristics is reduced, as a result of which the error in determining heat resistance is reduced, it is possible to reduce the number of test samples to reliably evaluate this property, the influence of the volume factor of the test sample on heat resistance. The table shows the data on the heat resistance of partially stabilized zirconia obtained by the claimed method and the prototype. Measurements of physical and mechanical properties were carried out on a series of samples of 10 pieces. Thermal shock was created by heating samples at 600 ^° C and subsequent cooling in air at room temperature. To do this, the samples were mounted on a ceramic plate and introduced into a 600 ^° C preheated oven, kept for 1 hour, then removed from the oven. After cooling the samples, measurements of physical and mechanical properties were performed. The average value of the tensile strength in bending before thermal shock (σ ₂ ) was 420 MPa, the average value of the critical coefficient of stress intensity of the samples before thermal shock (K _1c ) was 8.3 MPa • m ^1/2 . According to the prototype, the loss of strength in bending was measured (σ ^* is the tensile strength in bending of the sample after thermal shock, σ ₀ is the average value of the tensile strength in bending of samples before thermal shock). To compare the errors of determining the heat resistance according to the claimed method and the prototype was calculated the value of the mean square and simple average error / 4 /. For this, the arithmetic mean was calculated

(n is the number of measured values a _i ) ε _i = a _i -a is the deviation of the measured value from the arithmetic mean. Simple average error was defined as

and the standard error

From the results presented in the table, it is seen that the determination of heat resistance by the proposed method is characterized by a significantly smaller error. To assess the heat resistance of the proposed method, it is sufficient to use the test results of 3-6 notched samples in view of the small variation in the measured characteristics. In addition, when using the inventive method, there is a tendency to level the effect of the volume of the test sample on heat resistance. When testing (thermal shock 600 ^o C-air) series of ZrO ₂ - Y ₂ O ₃ samples (3 mol%), differing in volumes (3x5x30 mm ³ -1 type of samples, 5x8x40 mm ³ - 2 type of samples), it was found that the average loss of crack resistance in type 1 and type 2 samples was 21.2% and 22.0%, respectively, and notch tolerance for these types of samples was 0.25 and 0.23, respectively. Whereas the average value of the bending strength loss for samples of these types differed significantly and amounted to 32% and 51%, respectively.

 Examples of the implementation of the proposed method.

 Example 1

From ultrafine powder (particle size 0.2-0.5 μm) of the composition ZrO ₂ -Y ₂ O ₃ (3 mol%), prismatic samples (3x5x30, mm) with and without notches were made by pressing and sintering. After sintering, the average grain size was 1 μm, the notch was introduced at half the height of the sample according to the method developed by the authors, the radius of curvature of the notch tip was 8 μm, which satisfies the condition for adequate modeling of the crack, the region at the tip of the notch did not contain induced defects, which was confirmed by microscopic examination. Six notched samples were subjected to thermal shock in the regime of 600 ^o C - air, then they were tested for three-point bending (distance between supports L = 20 mm, load application speed - 8 μm / s) and parameters K * _1c and σ _{1 were} calculated by formulas (2 ) and (3). After that, six samples with a notch and six samples without a notch, not subjected to thermal shock, were tested under the same conditions and the parameters K _1c and σ _{2 were} calculated using formulas (1) and (4). Based on these tests, the arithmetic mean values of K _1c and σ ₂ were determined. The value of the coefficient Y (l / h) in formulas (1) and (2) with the ratio L / h = 4 was taken equal
Y (l / h) = 1.93-3.07 (l / h) +14.53 (l / h) ² -25.11 (l / h) ³ + 25.8 (l / h) ⁴ .

Then, the ceramic resistance to loss of crack resistance after thermal shock (1- K * _1c / K _1c ) 100% was 21%, 20%, 22%, 20.7%, 23%, 21.5% (the results were obtained as a quotient of the division of six values of K * _1c per average value of K _1c ), and the insensitivity to defects after thermal shock σ ₁ / σ ₂ was 0.21, 0.23, 0.26, 0.22, 0.21, 0.30 (the results were obtained as quotient of dividing six values of σ ₁ by the average value of σ ₂ ).
Example 2

Analogously to example 1, ZrO ₂ - Y ₂ O ₃ ceramics samples were prepared (3 mol%) with the addition of metallic chromium (3, 7, 10 vol%) and subjected to thermal shock in the 600 ^o C-air mode. After testing, the average values (out of 6 values) of crack resistance with an increase in the volume fraction of metal were 15.2%, 12.3%, 1.2%, respectively, and the insensitivity to defects after thermal shock was 0.42, 0.43, 0, 51 respectively.

 Example 3

Similarly to examples 1 and 2, alumina samples with two different structural types were obtained by pressing and subsequent sintering. Samples with a fine-crystalline structure (type 1 structure, average grain size of 5 μm) were made from GLMK industrial powder (Al ₂ O ₃ +0.5 wt.% Mgo). Samples with a specially organized structure, in which primary alumina grains (average size 5 μm) are connected by porous interlayers (0.3-0.5 μm) of the secondary alumina phase (type 2 structure), were obtained by reaction sintering during the oxidation of aluminum powder included in composition of the initial charge. Notches in the samples were created by cutting with a diamond wheel, the radius of curvature of the notch tip is 50 μm. Samples of both structural types were subjected to thermal shock in the 850 ^o C-water mode. After testing, the average values (out of 6 values) of the loss of crack resistance and insensitivity to defects after thermal shock for samples 1 and 2 of structural types amounted to 32.5%; 10.5% and 0.31; 0.56, respectively.

 Thus, according to the claimed method for determining the heat resistance of structural ceramic materials, the error in determining heat resistance is actually reduced, it is possible to reduce the number of test samples tested, there is a leveling effect of the volume factor of the test sample on heat resistance. In addition, the proposed parameters (relative loss of crack resistance and insensitivity to defects after thermal shock) give a quantitative assessment of the resistance of the ceramic structure to initiation of thermal cracks and its sensitivity to defects formed as a result of thermal shock.

Sources of information
1.K.K. Strelov, I.D. Kashcheev. Theoretical Foundations of the Technology of Refractory Materials.-M.: Metallurgy, 1996, 608 p.

 2. Workshop on the technology of ceramics and refractories / Edited by D.N. Poluboyarinov and R.Ya. Popilsky.-M .: Stroyizdat, 1972, 352 p.

 3. E.S. Lukin, N.T. Andrianov. Technical analysis and control of the production of ceramics.-M .: Stroyizdat, 1986, 272 p. (prototype).

 4. I.N. Bronstein, K.A.Semendyaev. Handbook of mathematics for engineers and students VTUZ.-M .: Nauka, 1964, 608 pp.

Claims (1)

A method for determining the heat resistance of structural ceramic materials, including the manufacture of experimental prismatic samples, the creation of thermal shock by a heating-cooling cycle, and the subsequent determination of the loss of physical and mechanical properties of the samples with respect to their values before thermal shock, characterized in that the samples are made with a side notch modeling the crack and not containing induced defects at the apex, and without an incision, and heat resistance is determined by the values of the ratios
(1 - K ^* _1s / K _1s ) • 100%
and
σ ₁ / σ ₂ ,
where k $\genfrac{}{}{0ex}{}{\text{*}}{\text{1c}}$ - the critical stress intensity factor of the sample after thermal shock;
K _1s is the average value of the critical coefficient of stress intensity of samples before thermal shock;
σ ₁ - tensile strength in bending of the notched sample after thermal shock;
σ ₂ - the average value of the tensile strength when bending samples without notching before thermal shock.

Images