Available online at www.sciencedirect.

com

ScienceDirect
Procedia Materials Science 3 (2014) 861 – 866

20th European Conference on Fracture (ECF20)

Estimation of the fracture toughness of structural steels by means of

the CTOD evaluation on notched small punch specimens
T.E. Garcíaa,*, C. Rodrígueza, F.J. Belzuncea, I. Peñuelasa, I.I. Cuestab
a
University of Oviedo, IUTA, Campus de Gijón, 7.1.17, 33203, Gijón (Spain);
b
Structural Integrity Group, Escuela Politécnica Superior, C/Villadiego s/n, 09001 Burgos, Spain

Abstract

By means of the experimental analysis of two structural steels, one with a ductile behaviour and the other one with
brittle behaviour, this paper compares the results obtained by means of small punch test (SPT) performed on notched
samples with those obtained in standard fracture toughness tests, in order to obtain a relationship between them and
to analyse the suitability of the SPT for estimating the fracture toughness. With the aim of analysing the evolution of
the defect tip opening displacement (δSPT) during the test, loading was interrupted at different levels of punch
displacement (d) and these specimens were analysed under SEM.
In the case of the ductile steel, δSPT at the onset of crack growth was larger than the value of CTOD obtained in
conventional fracture tests, but this could be expected due to the lower constraint condition of the SPT. By the other
hand, the brittle steel exhibits ductile fracture micromechanisms in the SPT, so fracture mechanisms have changed
from the standard to the small punch test, due to the different load conditions. Relationship between d and δSPT
seems to be material independent.
©
© 2014
2014Published by Elsevier
The Authors. Ltd. Open
Published access under CC BY-NC-ND license.
by Elsevier Ltd.
Selection and peer-review under responsibility of the Norwegian University of Science and Technology (NTNU), Department
Selection and peer-review under responsibility of the Norwegian University of Science and Technology (NTNU), Department of
of Structural Engineering
Structural Engineering.

Keywords: CTOD; Small punch test; Structural steel.

* Corresponding author. Tel.: +34-985-181-967; fax: +34-985-182-433.

E-mail address: garciatomas@uniovi.es

2211-8128 © 2014 Published by Elsevier Ltd. Open access under CC BY-NC-ND license.
Selection and peer-review under responsibility of the Norwegian University of Science and Technology (NTNU), Department of Structural Engineering
doi:10.1016/j.mspro.2014.06.140
862 T.E. García et al. / Procedia Materials Science 3 (2014) 861 – 866

1. Introduction

The small punch test (SPT) is a miniature non-standard test which employs very small specimens (generally 8
mm diameter and 0.5 mm thickness). It was developed in the early 80’s with the aim of determining the post
irradiation mechanical properties of materials used in nuclear industries (Manahan et al. (1981)). The specimen is
clamped between two matrixes and deformed by the action of a punch, which pass through a hole located in the
lower matrix. The load versus punch displacement record gives information that allows estimating mechanical
properties, such as the yield strength or the ultimate tensile strength. Good correlations have been found between the
SPT parameters and these properties for a wide range of materials, making the SPT a suitable test when structural in-
service elements must be analyzed (Mao et al. (1987), Rodríguez et al. (2013), García et al. (2014)), or when
standard specimens cannot be extracted from small regions of the structure (Rodríguez et al. (2009), Dymácek et al.
(2013)). Although the SPT is still a non-standard test, a European Code of Practice (CEN Workshop Agreement
(2006)) was published, providing general recommendations for its application.
One of the key required properties when analyzing structural components, especially components in energetic
applications (neutron irradiated or thermal aged materials), is the fracture toughness. In a conventional SPT test an
un-cracked and very thin specimen is used, so test conditions are very far from plane strain as it is required in the
standard fracture toughness tests, and the estimation of this parameter by means of the SPT still remains a
controversial subject.
Different strategies for estimating the fracture toughness were employed since the appearance of the SPT. The
first one was proposed by Mao et al. (1987). It consists on the measurement of a parameter called the biaxial strain at
fracture, εqf, which relates the original specimen thickness with the one at the thinnest part of the fracture region.
Quite good correlations with the fracture toughness were found by many authors (i.e Guan et al. (2011) or García et
al. (2014)), but very different empiric expressions were developed and they also have a clear dependence on the
material. Other strategies are, for example, the measurement of the energy contained under the SPT curve until a
20% load drop after the maximum load, CEN Workshop Agreement (2006), or the use of neural networks for fitting
the parameters of a damage model by means of the SPT curve combined with the numerical simulation of a standard
fracture toughness test using these parameters (Abendroth et al. (2006), Alegre et al. (2011)). In recent years, the
trend of different researchers, such as Cuesta et al. (2011) or Rodríguez et al. (2013), is the use of notched SPT
specimens for estimating the fracture toughness. This kind of specimen seems to be more suitable, since an initial
defect is introduced. One of the proposals using notched SPT specimens is based on the measurement of the tip
opening displacement (δSPT), similar to the CTOD (δ) concept used in standard fracture test (Lacalle et al. (2012)).
By means of the experimental analysis of two different structural steels, one with a ductile behavior and the other
one with brittle behavior, this paper compares the results obtained by means of the SPT with those obtained in
standard fracture toughness tests (δSPT vs δ), in order to obtain a relationship between them and to analyze the
suitability of the SPT for estimating the fracture toughness of both, ductile and brittle materials.

2. Materials and conventional characterization

A 108 mm thick plate of 2.25Cr-1Mo-0.25V steel (SA 542 Grade D-Class 4) was used as the base material. It
was normalized and then, quenched and tempered. The chemical composition of the steel is shown in Table 1.
A weld coupon with a length of 1300 mm and a width of 600 mm was produced. The weld metal also included a
de-hydrogenation treatment of 4 hours at 350 oC. Table 1 gives also the chemical composition of the weld metal.

Table 1. Chemical composition of the SA 542 Grade D-Class 4 steel (base metal) and weld metal [%wt]
Material %C %Si %Mn %Cr %Mo %V %Ni
Base metal 0.15 0.09 0.52 2.17 1.06 0.31 0.19
Weld metal 0.08 --- --- 2.28 0.93 0.24 0.03
 T.E. García et al. / Procedia Materials Science 3 (2014) 861 – 866 863

2.1. Tensile characterization

Tensile tests were performed according to the ISO 6892-1:2009 standard. Three specimens of each material were
tested, and the tensile stress–strain curves were obtained. The average results of these tests are shown in Table 2.

Table 2. Results of tensile and fracture tests

Material E (GPa) σys (MPa) σut (MPa) δini (mm) δIC (mm) C1 C2 JIC (kJ/m2) KIC (MPa m0.5)
Base metal 200 595 711 0.214 0.417 0.62 0.53 555 -
Weld metal 236 1034 1121 0.011 0.011 - - - 85

2.2. Fracture characterization

Fracture toughness tests were performed using single edge notched bend specimens, SE(B), with a crack length
to width ratio: a/W=0.5, and following the ASTM E1820 standard. Specimens were fatigue pre-cracked to the
required nominal a/W using a load ratio of 0.1, and they were subsequently side-grooved.
The single-specimen method, based on the use of the elastic unloading compliance technique, was used to
determine the δ-Δa resistance curve of the base metal (ductile material). The results thus obtained were corrected
using the physical measure of the crack determined at the end of each test by means of a suitable low magnification
microscope. The value of δ of each unload (δi) was determined after splitting up its elastic and plastic components
(expressions 1-3). The elastic component was obtained from the stress intensity factor, K (expression 2), in which ν
is the Poisson’s ratio. The value of rp is given in the ASTM E1820 standard (rp = 0.44) (νpl is the plastic component
of the crack mouth opening displacement). Experimental points δi - Δai were fitted by means of the power law of
expression 4. In the case of the weld metal (brittle material), the δIC value was assessed making use of expression 2.

‫ܭ‬௜ ଶ ሺͳ െ ߥ ଶ ሻ ‫ݎ‬௣ ሺܹ െ ܽ௜ ሻ‫ݒ‬௣௟ ௜

ߜ௜ ൌ ߜ௘௟ ௜ ൅ ߜ௣௟ ௜ (1) ߜ௘௟ ௜ ൌ (2) ߜ௣௟ ௜ ൌ (3) G ൌ ‫ͳܥ‬οܽ஼ଶ (4)
ʹߪ௬௦ ‫ܧ‬ ‫ݎ‬௣ ሺܹ െ ܽ௜ ሻ ൅ ܽ௜

Results of the fracture tests are shown in Table 2, while Fig. 1a shows the δ-Δa curve of the base metal and Fig.
1c shows the load-COD record of the weld metal. The values of K IC and JIC were also assessed in compliance with
the ASTM standard. Since crack growth measurement is not possible in the SPT, CTOD crack initiation values (δini)
were defined as the value of δ when the δ-Δa curve separates from the blunting line (ductile material) or at the point
of instability (brittle material, δini = δIC). It is important to remark the different fracture behavior exhibit for the two
materials analyzed. The base metal exhibited a totally ductile behavior, showing the typical fracture
micromechanism, consisting in the nucleation, growth and coalescence of micro-voids, as it is shown in Fig. 1b. By
the contrary, the fracture of the weld metal was mainly brittle, but with a certain amount of ductility. Fig. 1d shows a
detail of the mixed fracture, cleavage and dimples.

Fig. 1. (a) δ-Δa curve of Base metal (b) Load-COD record of Weld metal (c) Base metal’s fracture surface (d) Weld metal’s fracture surface.
864 T.E. García et al. / Procedia Materials Science 3 (2014) 861 – 866

3. Small punch test

Fig. 2a shows the typical SPT load-displacement record of a ductile steel. At the beginning (zone I), the material
undergoes an elastic deformation, accompanied by punch indentation. The initiation of plastic deformation takes
place at zone II, and this deformation becomes generalized at zone III. Near the maximum load, the curve slope
decreases due to the necking and crack initiation. Finally, in the region of the maximum load, the developed crack
attains the total specimen thickness, causing the final breakage of the sample.
Fig. 2b shows a scheme of the small punch test device, which was connected to a universal testing machine. The
specimen was placed onto the lower matrix, which had a 4 mm diameter hole with a 0.2 mm fillet radius, and it was
firmly clamped by means of a threaded fixer. Load was applied by means of a hemispherical punch with a diameter
of 2.4 mm. An extensometer placed outside the experimental device was employed to accurately measure the punch
displacement. A high stiffness material was previously tested in order to measure the machine and equipment
stiffness to correct the obtained SPT curves. All the tests were carried out at a cross head speed of 0.2 mm/min.

Fig. 2. (a) Typical SPT record (b) Scheme of the SPT device (c) Notched specimen and criteria employed for the δSPT measurement.

Notched 10x10 mm2 square specimens with a thickness (t) of 0.5 ± 0.01 mm were used (Fig. 2c). The samples
had a 0.3 notch length to thickness ratio (a/t). The notches were extended from the center of one side to the center of
the opposite side and they were machined by means of a 30 o micro-milling tool, with a 100 μm tip radius. It was
observed by Peñuelas et al. (2012) that this kind of notch is able to act as a crack initiator, and the micromachining
assures its reliability (a uniform shape and depth along all the specimen length). Fig. 2c shows a scanning electron
microscopy (SEM) image of the profile of this kind of notch.
Regarding the definition of the CTOD gave by Rice (1968) - displacement at the intersection of a 90o vertex with
the crack flanks - the measurement of the notch tip displacement, δSPT, was took as the notch mouth opening
displacement, since it is very close to the Rice’s definition (Fig. 2c).
In order to analyze the evolution of the δSPT parameter during the sample loading, interrupted tests at different
values of the punch displacement were performed, as well as complete tests until the specimen failure. Three
complete tests were carried out for each material, and 11 tests were interrupted between punch displacements of
0.25 and 1.25 mm, in order to follow the deformation of the sample until the maximum load. Measurements of δ SPT
were performed by means of SEM and an image analysis software. The values of the punch displacement and the
experimental measurements of δSPT are shown in Table 3. After testing, failed specimens were brittle fractured in
two halves inside liquid nitrogen, with the aim of looking for the fracture micromechanisms developed during the
test. Fig. 3 shows the SPT curves of both materials, with some SEM images of deformed specimens.

Table 3. Results of the interrupted SPT tests. Punch displacement (d) and experimental measurements of δSPT
d (mm) 0.28 0.38 0.47 0.57 0.67 0.77 0.86 0.98 1.06 1.15 1.25
Base metal
δSPT (mm) 0.047 0.081 0.113 0.137 0.140 0.147 0.172 0.177 0.255 0.231 0.284
d (mm) 0.26 0.35 0.47 0.54 0.62 0.71 0.80 0.89 0.98 1.06 1.17
Weld metal
δSPT (mm) 0.066 0.112 0.128 0.136 0.142 0.148 0.171 0.185 0.197 0.227 0.273
 T.E. García et al. / Procedia Materials Science 3 (2014) 861 – 866 865

Fig. 3. (a) SPT curves of Base metal with SEM images for δ SPT measuring. (b) SPT curves of Weld Metal with a detail of crack at d=0.47mm
(c) Relationship between δSPT and punch displacement for both base and weld metal

4. Discussion

Fig. 4 shows the fracture surfaces of various specimens tested until different levels of punch displacement and
finally broken in liquid nitrogen. In the case of the Base metal, no signs of ductile crack growth were observed until
a punch displacement of 1.25 mm (Fig. 4b). At this level of displacement, a value of δ SPT of 0.284 mm was obtained.
This value is larger than those obtained in the standard test (δSPT = 0.214 mm), but this could be expected due to the
lower constraint condition of the SPT.
In the case of the brittle material, it is important to remark that while in the standard test the specimen behaves
mainly brittle (Fig. 1c and 1d), in the SPT the specimen exhibits ductile fracture micromechanisms (Fig. 4c). As it
was observed by García et al. (2014), fracture mechanisms can change from the standard fracture test to the SPT due
to the different load conditions (plain strain in the standard test versus plain stress in the SPT). However, for this
material crack growth was observed from a punch displacement much lower, close to 0.5 mm (Fig. 4c). At this level,
a value of δSPT of 0.128 mm was measured, quite larger than the one measured in the standard test (δSPT = 0.011
mm). The aforementioned differences between the standard and the SPT can explain these results.
The observation by SEM of the specimens broken in liquid nitrogen reveals that the small grooves observed
during the δSPT measurements were in fact actual cracks present in the material in the notch region (comparing both
Fig. 3b and 4c, punch displacement of 0.47 mm).

Fig. 4. SEM pictures of specimens broken in liquid nitrogen:

(a) Base metal (d=1.15mm) (b) Base metal (d=1.25mm) (c) Weld metal (d=0.47mm).

Another important finding of this paper is that δSPT depends only on the geometry of both notch and device, but
not on the material, as it can be seen in Fig. 3c, where δSPT vs punch displacement is compared for both materials.
Moreover, the relationship between δSPT and the displacement can be linearly fitted. This could be interesting for the
866 T.E. García et al. / Procedia Materials Science 3 (2014) 861 – 866

development of a methodology based on this δ SPT concept, in which crack initiation must be observed during the
test, for example by using a small camera located in the lower die hole.

5. Conclusions

A new methodology for estimating the fracture toughness by means of SPT specimens with a longitudinal non-
trough notch was developed.
Although testing other different materials is already needed, the measurement of the δSPT can be a good strategy
for estimating the CTOD at the onset of crack growth in ductile materials.
Since the behavior of brittle materials in the SPT and in the conventional fracture toughness test is quite different,
it does not seem possible to obtain the fracture toughness of brittle materials with the small punch test, even using
notched specimens.
At least with the notch morphology employed in this paper, the δSPT seems to be material independent and it has a
linear relationship with the punch displacement.

Acknowledgements

The authors are grateful for the financial support for this study provided by the Spanish Ministerio de Ciencia e
Innovación, through project MICINN-12-MAT2011-28796-C03-03. T.E. García is also grateful for financial
support from the Principado de Asturias Government through the Severo Ochoa programme (contract BP12-160).

References

Abendroth, M., Kuna, M., 2006, Identification of ductile damage and fracture parameters from the small punch test using neural networks,
Engineering Fracture Mechanics, 73, 710-725.
Alegre, J.M., Cuesta, I.I., Bravo, P.M., 2011, Implementation of the GTN damage model to simulate the small punch test on pre-cracked
specimens, Engineering Procedia, 10, 1007-1016.
ISO 6892-1:2009, “Metallic materials -- Tensile testing -- Part 1: Method of test at room temperature”, ISO European Standards.
ASTM E 1820-08a, “Standard test method for measurement of fracture toughness”, American Society for Testing and Materials.
CEN Workshop Agreement, 2006, CWA 15627:2006 E, ‘‘Small Punch Test Method for Metallic Materials’’, CEN, Brussels.
Cuesta, I.I., Rodríguez, C., Belzunce, F.J., Alegre, J.M.. 2011, Analysis of different techniques for obtaining pre-cracked/notched small punch
test specimens, Engineering Failure Analysis, 18, 2282-2287.
Dymácek, P., Dobes, F., 2013, Investigation of fracture in pure aluminium after ECAP by means of small punch test, Acta Metallurgica Slovaca,
3, 57-64.
García, T.E., Rodríguez, C., Belzunce, F.J., Suárez, C., 2014, Estimation of the mechanical properties of metallic materials by means of the small
punch test, Journal of Alloys and Compounds, 582, 708-717.
Guan, K., Hua, L., Wang, Q. et al, 2011, Assessment of toughness in long term service CrMo low alloy steel by fracture toughness and small
punch test, Nuclear Engineering Dessign, 241, 1407-1413.
Lacalle, R., Álvarez, J.A., Arroyo, B., Gutiérrez-Solana, F., 2012, Methodology for fracture toughness estimation based on the use of small punch
notched specimens and the CTOD concept, In: 2nd International Conference SSTT, Ostrava, Czech Republic.
Manahan, M.P., Argon, A.S., Harling, O.K., 1981, The development of a miniaturized disk bend test for the determination of postirradiation
mechanical properties, Journal of Nuclear Materials. 104, 1545-1550.
Mao X., Takahashi, H., 1987 Development of a further-miniaturized specimen of 3 mm diameter for TJZM disk (0 3 mm) small punch tests,
Journal of Nuclear Materials, 150, 42-52.
Peñuelas, I., Rodríguez, C., Belzunce, F.J., García, T.E., Betegón, C., 2012, Analytical and experimental determination of the fracture toughness
of structural steels thorugh small punch test specimens, notched by micromachining, laser and femtolaser techniques, In: 2nd International
Conference SSTT, Ostrava, Czech Republic.
Rice, J.R., 1968, A path independent integral and the approximate analysis of strain concentration by notches and cracks, Journal of Applied
Mechanics, 35, 379-386.
Rodríguez, C., García, J., Cárdenas, E., Belzunce, F.J., Betegón, C., 2009, Mechanical properties characterization of heat-affected zone using the
small punch test, Welding Journal, 88, 188-192.
Rodríguez, C., Cárdenas, E., Belzunce, F.J., Betegón, C., 2013, Fracture Characterization of Steels by Means of the Small Punch Test,
Experimental Mechanics, 53, 385-392.