threshold
stress 3 0 intensity
Kith 2 5
[MPa',/m] 20
15
10
A Z~
-zx .._.o.oir/~_. /
__ 3.5% N ~
9
9 I
:~-KIscc- Q
AI 2024 T 351 (S-L) 'O" O
O O
_specimen type
o 9
zx A DCB environment
0 air; 9 3.5 % NaCI
0 ' , I , i I , , I , = I
10- 5 10-4 10-3 10-2 10-1
displacement rate VLL [mm/h]
FIG. 7--Influence o f the displacement rate (dVLL/ dt) on the threshold stress intensity K.h 2024 T351 ).
0 0
aluminum
lower displacement rates, the threshold values might go up again as would in principle be expected for a repassivating system.
In order to include the regime of small cracks in the investigations, where the L E F M approach does not seem to be applicable, threshold nominal stresses ~r,, were measured in constant load experiments on initially smooth specimens of the same material and in the same corrosive environment. Together with the results from conventional tensile tests, the K,~ value from CT and DCB specimens, and the results of fracture toughness tests in air, according to [21] a range of sensitivity to SCC was determined for this material/environment system. This range is plotted in Fig. 8 for a specimen with a semi-elliptical surface crack.
In this graph the results from rising displacement tests on surface cracked panels conducted at three different displacement rates are also plotted. Even though crack initiation in these panels did not occur immediately after reaching the " b o r d e r line" of the SCC sensitive area, the correspondence between the prediction derived from tests on laboratory specimens (CT, DCB, tensile specimens) and the more "realistic" defect represented by a surface cracked specimen appears remarkable.
Finally, Fig. 9 proves that for this material and orientation the requirements for applying L E F M methods were satisfied. In this graph the line drawn under 45 ~ represents the rela- tionship
J = K 2 / E ' with E ' = E l ( 1 - v 2)
where E is Young's modulus and v is Poisson's ratio. This relationship is valid only in the linear elastic regime and may thus be taken as a criterion for the applicability of L E F M
DIETZEL AND SCHWALBE ON RISING DISPLACEMENT TEST 143
1 000 1
AI 2024 T351 (S-L) / 3.5% NaCI J , . . . - ~ ~-~,/1
RpO.2 = 2gO MPa "~--.~. l I )
~ / s t re sLc~ r r~176 ncr acking / / J ~ / ~ ' 2 - ~ , ~ o v~. )
10 ~ I
0.1 1 10 100
crack depth a [mm]
FIG. 8--Combined stress - stress intensity factor versus flaw size diagram predicting the SCC susceptible zone for aluminum 2024 T351; comparison with results from three tests on surface crack specimens.
formalisms in the tests. According to Fig. 9, virtually all curves, even those obtained from testing in air, start in the vicinity of this 45 ~ line. This indicates that in all cases at least the initial cracking occurred under LEFM conditions.
Steel F e E 690 T
For the higher-strength steel FeE 690 T, it was doubtful whether LEFM methodologies could be used. Therefore, the formalisms of elastic-plastic fracture mechanics (EPFM) were applied to the data evaluation. From the tests, data crack growth resistance curves (R- curves) for the J-integral as well as for the C T O D were generated. Figure 10 shows the J- R curves obtained in tests at various displacement rates spanning more than four orders of magnitude. The graph contains R-curves measured in air as well as R-curves from tests under hydrogen-charging conditions.
According to this graph, the material's resistance against further extension of a pre-existing crack decreases drastically at lower displacement rates when being exposed to the corrosive environment. The critical J-value taken from these J-R curves at 0.2 mm of ductile tearing after crack tip blunting and designated J(,2BL [17] drops from about 300 N/ram measured at a displacement rate of 30 m m / h to less than 30 N/ram at 1 I~m/h. Taking into account that for low displacement rates no considerable blunting was detected on the fracture surface, the starting value J~ for the J-integral (Aa=0) can be used to calculate the stress intensity factor K~.,~ for the initiation of environmentally enhanced cracking via
K , ~ = V ( J 9 E').
Copyright by ASTM Int'l (all rights reserved); Sat Dec 19 20:05:50 EST 2015 Downloaded/printed by
University of Washington (University of Washington) pursuant to License Agreement. No further reproductions authorized.
1 4 4 S L O W S T R A I N RATE T E S T I N G
E E
Z
r - )
20
10
AI 202/. T351 (S-L)
envJronm.
a i r
,~ NaCt 100
,,- .. 10
9 1
o 0 . 5
0 0.1
0 10
K 2/E' [N/mini
FIG. 9 - - P l o t o f J-integral versus K2t E ' for aluminum 2024 7"351.
MLL [H m/h]
0.1 / 100
20
Since the lowest initiation value measured in the tests was in the order of J~ = 10 N / m m , the corresponding threshold value should be as low as
Kj~c = 50 MPak/-mm
Such low initiation values were found only in rising displacement tests. So far, they could not be verified in tests for the same corrosion system when constant load or constant displacement were applied.
F r a c t o g r a p h y
A fractographic examination of the fracture surface of specimens tested in the corrosive environment at the lowest and the highest displacement rate applied in the investigations reveals the different fracture modes occurring under various loading conditions (Figs. lla,b).
The scanning electron micrograph taken from the specimen fractured at 30 mm/h (Fig. 1 lb) corresponds to a ductile rupture by microvoid coalescence showing dimples of various size with no indication of secondary cracking. For the specimen tested at 1 m m / h the fracture surface is dominated by items of a brittle failure, i.e., quasi-cleavage facets and numerous secondary cracks (Fig. lla). Hence, the fracture morphology confirms the assumption that