Available online at www.sciencedirect.

com

Procedia Engineering 55 (2013) 82 – 87

6th International Conference on Cree

ep, Fatigue and Creep-Fatigue Interaction [CF-6
6]

Hot-Tensile Data and Creepp Properties Derived There-from for

f
316L(N) Stainless Steell with Various Nitrogen Contents
R. Sreenivasan∗
P.R
Metallurgy and Materials Group, Indira Gandhii Centre for Atomic Research, Kalpakkam, Tamilnadu-603102. India

Abstract

Many researchers have proposed equivalence betweeen hot-tensile data and creep rupture properties based on Larson n-Miller
or similar time-temperature parameter approaches. T This is attractive from the point of view economy of time and cost of
testing as compared to the long-time creep tests. B Basic idea behind such a comparison is that the plastic deforrmation
behavior controlling both the hot-tensile and creep pproperties are the same, provided the microstructure does not undergo
u
significant changes for the time periods under coonsideration. Particularly, such a procedure will be useful in alloy
development and preliminary screening. Such a tim me-temperature parameter approach is adopted in the follow wing for
predicting the stress-rupture behavior of 316 L(N) sstainless steel from reported hot-tensile properties and the results are
compared with actual creep test data for almost a simmilar steel at two temperatures. It is found that the predicted dataa follow
the ASME Code Case minimum stress-rupture data for 316 SS (stainless steel) and thus are very conservative. Prredicted
creep rates are also much larger than the actual onees for the one case where comparison has been made. The con nclusion
drawn is that considering the microstructural stabilitty of the low-carbon nitrogen bearing SSs (at least for lower nitrogen
n
contents), the method of equivalence of hot-tensile annd stress-rupture data seems feasible. But test results at differen
nt strain
rates, rather than the single value considered here, along with the more accurate method suggested would yield d better
results. In addition, using some old results from thhe literature, relation for variation of yield stress with temperature,
nitrogen content and grain-size for the present heats oof steel (with various nitrogen contents) has been derived.

© 2013The
© 2013 The Authors.
Authors. Published
Published by Elsevier
by Elsevier Ltd.under
Ltd. Open access Selection and/orlicense.
CC BY-NC-ND peer-review under responsibility of the Indira
Gandhi
Selection Centre for Atomic
and peer-review Research. of the Indira Gandhi Centre for Atomic Research.
under responsibility

Keywords: Hot-tensile test; creep test; 316 L(N) stainless steeel; Larson-miller parameter

1. Introduction

Creep is an important consideration for strucctures and components operating at high temperatures: say, for
example, boilers and steam generators in electricc power plants and structures and components of nuclear power
plants. Long term creep tests involve considerabble investment in time and cost. Hence, ability of prediction of
useful creep properties from short-term hot-teensile tests is an attractive proposition. This is based on the

∗
Corresponding Author:
E-mail address: sreeprs@yahoo.co.in

1877-7058 © 2013 The Authors. Published by Elsevier Ltd. Open access under CC BY-NC-ND license.
Selection and peer-review under responsibility of the Indira Gandhi Centre for Atomic Research.
doi:10.1016/j.proeng.2013.03.223
 P.R. Sreenivasan / Procedia Engineering 55 (2013) 82 – 87 83

premise that the fundamental plastic deformation characteristics are the same under both hot-tensile and creep
tests. Then, there is likely to be an equivalence between short-term hot tensile and long-time creep properties.
This will be particularly so for materials showing good microstructural stability and little degradation during
the time periods of creep tests. Following Steen [1], Levi de Oliveira Bueno (hereinafter referred as Bueno) [2]
and co-workers (see references in [2]) have demonstrated the equivalence between hot-tensile properties and
creep testing data for a 2.25Cr-1Mo steel.
Their approach involves, apart from long-time creep tests, hot tensile tests over a range of strain rates
varying over three orders of magnitude (6.7x10-6 to 1.3x10-2 s-1) and at various temperatures and then finding
equivalence between the properties of the two tests through an activation energy and strain rate parameter
approach. Theoretically such an approach is more satisfying and likely to be more accurate. Hung-Kuk Oh
(hereafter, Oh) [3] has used a rather simpler approach which does not involve such a wide range of hot-tensile
tests as in the case of the procedure adopted by Bueno et al. and demonstrated the applicability of his method
for 304 stainless steel (SS) in predicting stress-rupture data and minimum creep rates. Which method can be
used depends partly on the data available also. In the present paper, the hot-tensile data at a single strain rate
reported for a fast reactor nuclear grade 316 L(N) SS by Ganesan et al. [4] have been analysed by the Oh
method and the results discussed in comparison with some actual creep data available for a similar grade of
steel. In addition, using some old results from the literature, relation for variation of yield stress with
temperature, composition and grain-size for the present heats of steel (with various nitrogen contents [4]) will
be derived.

2. Materials and data

The material is a low-carbon 316 SS with nitrogen content varying from 0.07 to 0.22 wt%. The four heats
with nitrogen 0.07, 0.11. 0.14 and 0.22 wt%, respectively, and reported in [4], are designated as 7N, 11N, 14N
and 22N, respectively. Their composition is essentially the same except for the nitrogen content and the heat
designated as 7N is almost of the same composition as has been adopted for the construction of the Indian
PFBR. All the hot tensile results reported were conducted at a single strain rate of 3.10-3 s-1. The hot-tensile
properties are given in [4]. All the heats were manufactured through air induction melting (AIM) and electro-
slag refining (ESR) route. Reference [4] may be consulted for full composition and fabrication details.

3. Data reduction, results and discussion

3.1. Creep data

As outlined in the introduction, there are two approaches available for data reduction: (i) Bueno and (ii) Oh
methods. For details of the Bueno method, the reader is referred to references [1] and [2] and the appended
references. As only hot-tensile data at one strain rate are available in the present case, the simplified method of
Oh [3] is followed here.
The basic assumptions of the Oh method [3] are:
• The UTS is related to the Larson-Miller Parameter (LMP), P, and P is given by T(K).(20 + log(tR)),
where T is the test temperature in K and tR is the rupture time in h.
• In a uni-axial tensile test, UTS (ıU) represents the rupture stress (ıR) and the time to UE, the rupture time.
• Hence from the test strain rate, knowing UE, the tR can be calculated, that is uniform strain over tensile
strain rate.
• From the above, by plotting log(UTS = ıU) vs. P, the basic calibration curve is obtained.
• Another assumption made is that creep deformation is equal to the difference between the TE and UE (if
necessary, each taken as lower bound values).
• Creep deformation (expressed as strain, not %elongation) divided by rupture time gives the strain rate or
minimum creep rate.
Justification for the assumption (5) above may be that UE is similar to instantaneous deformation during
loading at the start of creep test and the strain after UE (in tensile test) is similar to the strain accumulated in
84 P.R. Sreenivasan / Procedia Engineering 55 (2013) 82 – 87

creep when undergoing extensive deformation damage. The calculated rupture times were obtained as per item
(3) listed above using a tensile strain rate of 3.10-3 s-1 and uniform elongation. Now using the ıU, tR and
temperature data in [4], the basic LM plot obtained for the 7N steel is given by Eq. (1). The calculated rupture
times from the uniaxial tensile tests vary from 0.01 to 0.05 h. Oh makes a general assumption that for tensile
tests the time to UE is 0.1 h. However, in this paper the actual data have been used. Following Oh [3], the creep
strain (that is, [(%TE – %UE)/100]) as a function of temperature is given by Eq. (2).
log(ıU) = 2.9801 – 3.9259*10-4*X + 5.2101*10-8*X2 – 3.4519*10-12*X3 + 1.1827*10-16*X4 –
1.8207*10-21*X5, where X = P, the LMP (1)
-6 2 -9 3
Creep Strain = -0.3538 + 0.0027*X - 5.163*10 *X + 3.0141*10 *X
where X = T in K (2)
Now, Sasikala et al. [5] have determined the creep properties for a 316 L(N) SS of almost the same
composition as the 7N steel. The steel reported by Sasikala et al. [5] is referred as ‘reference steel – RS’. The
composition of the RS is (wt%): C-0.023, Ni-12.21, Cr-17.12, Mo-2.31, Mn-1.65, Si-0.29, S-0.003, P-0.024,
Cu-0.1, B-0.0012 and N-0.086. The creep rupture properties of the RS at 923K and 873K have been reported in
[5], spanning a rupture life of less than 100 h to more than 10000 h, and compared with the ASME Code Case
lower bound for standard 316 SS.
Hence, this can be used to verify the Oh’s procedure by applying it to derive the creep-rupture data from the
7N tensile data. For example, at 923K, for 10000 h rupture life, the LMP, P is 22152 (that is 923*(20 + 4)),
which from Eq. (1), for P = 22152, gives a log(ıU) value of 1.0943 and a corresponding ıU = 12.424 ksi (=
85.664 MPa). Similarly, values are obtained for other assumed rupture times of 100, 1000 and 100000 h. The
data are plotted as log(ıR) vs. log(tR) in Fig. 1 and compared with the data for the RS at 923K along with the
ASME Code Case lower bound for standard 316 SS. It seems that the estimated data by the Oh’s method
almost follows the ASME lower bound curve while the actual test data lie at a much higher level. Similar is the
case for the data at 873K also as shown in Fig. 1. Thus the Oh’s method seems to give very conservative stress-
rupture predictions.
2.6 2.6

2.4 2.4
log(Stress, MPa)

2.2 2.2

2.0 2.0
ASME LB for 316 SS at 923K
Oh's Method Data at 923K
RS test Data at 923K
1.8 Oh's Method Data at 873K 1.8
ASME LB for 316 SS at 873K
RS test Data at 873 K

1.6 1.6
0 1 2 3 4 5 6

log(rupture life, h)

Fig.1. Comparison of rupture data for the 7N steel predicted by the Oh’s Method
with those from actual tests for the RS steel.

Now, applied stress vs. minimum creep rate data for the RS at 923K are available in [5]. To compare with
the Oh’s method, from Eq. (2), the creep deformation at 923K is given to be 0.11 which when divided by the
rupture time in h gives the corresponding strain rate, that is, equivalent to the minimum creep rate. Such
calculations for 100, 1000, 10000 and 1000000 h along with the corresponding stress from Fig. 1 are plotted
 P.R. Sreenivasan / Procedia Engineering 55 (2013) 82 – 87 85

along with the actual data of minimum creep rate vs. rupture stress at 923K for the RS in Fig. 2 (shown as Oh’s
Method-I). The Oh’s method predictions give much higher rates.
Creep Strain = 1.3463 - 0.0078*X + 1.5311*10-5*X2 – 9.6743*10-9*X3
where X = UTS, ıU (3)
Since in Eq. (1), UTS (ıU) has been related to LM type parameter, it is a better idea to relate the creep
deformation to the corresponding UTS at each tempetrature. Such a fit is given in Eq. (3). Then corresponding
to each of the UTS values for 100, 1000, 10000 and 100000 h rupture lives, the corresponding strains from Eq.
(3) are used to calculate the respective minimum creep rates. Here, unlike the Oh Method-I, different strains
are obtained for different UTS at the same temperature. Hence, this method is called Oh Method-II. The strain
rate (minimum creep rate) vs. applied stress data for the Oh Method-II shown in Fig. 2 are much higher than for
both the Oh Method-I and actual test results for RS. Hence there is much scope for improving the accuracy and
reducing conservatism. Perhaps, this can be achieved by multiple strain rate hot-tensile data coupled with the
improved analysis procedures suggested by the Bueno method (Method-(i)). As making fits similar to Eqs. (1)
to (3) for 11N, 14N and 22N steels using the data in [4] is a simple matter, the same has not been given.

3.2. Expression for the yield stress of 316 L(N) SS as a function of N content, grain size and temperature

Norstrom [6], through Hall-Petch analysis of tensile data for 316 L SS with varying N content of 0.05 to 0.18
wt%, had found an analytical expression for the 0.2% YS as a function of grain size (GS), D (mm), T (K) and N
content (wt%) in the following form:

YS = σ 0 + σ Ni + σ ND + σ Di (4)
where ı0 is the friction stress without N including the contribution of dislocations, σ N is the true-solution
i

hardening from N at infinite GS; σ N is the hardening contribution from N via the grain size and σ D is the
D i

grain size contribution to hardening in the absence of N. After analyzing the test data in [6], Norstrom had
given the final expression in the form of Eq. (5) as given below:
-2
RS data at 923K
7N Oh's Method-I Data at 923K
7N Oh's Method-II Data at 923K
-3
log(Strain Rate/h)

-4

-5

Eye-Fit Line to 7N data Oh's Method-I Data

-6

Eye-Fit Line to 7N data Oh's Method-II Data

-7
1.6 1.8 2.0 2.2 2.4 2.6

log(stress/MPa)

Fig.2. Comparison of stress-strain rate (minimum creep rate) data derived from hot-tensile data
for 7N steel with actual creep test data for RS.

−1
33000 1690 − T 1
YS = 15 + + 65( )( N 2 ) + [78 N + 7]D 2 (5)
T T
86 P.R. Sreenivasan / Procedia Engineering 55 (2013) 82 – 87

In Eq. (5), the first two terms represent ı0, the third term represents σ N , the fourth and fifth terms represent
i

σ and σ Di , respectively. Examining the details of the analysis in [6] and considering the similarities between
D
N
the steels in [6] and [4], it appears that the two cases will differ only in the friction term, ı0. Norstrom had
arrived at the friction term (that is, the first two terms in Eq. (5)) by subtracting the other terms of Eq. (5) from
the 0.2% YS at various temperatures. The friction term so obtained was plotted against (1/T) and a linear
regression made. A similar approach seems to be adequate here.
Figure 3 gives a plot of the YS-T data for the four steels listed in [4] along with the corresponding fitted
lines (black). Figure 4 gives the plot of friction term against reciprocal of temperature obtained for the steels in
[4]. Individual data points for each nitrogen content as well as all data together without distinction of N (the
filled black circle) are also shown. The line fitted is to all the data points. Based on this fit, Eq. (5) is modified
for the present 316 L(N) SS as follows:

−1
39080 1690 − T 1
YS = 18.56 + + 65( )( N ) + [78 N + 7]D 2
2
(6)
T T

The slightly larger friction term for the present 316 L(N) may arise from the different alloy content.
Prediction based on Eq. (6) for the 7N and 22N steels are also indicated in Fig. 3 in red colour. Considering the
scatter in the basic data in Fig. 3, the agreement seems to be good.

180

7N steel
160
11N steel
14N steel
140 22N steel
Friction Stress, σ0/MPa

All data
Fit to All Data
120

100

80

60

40

20
0.5 1.0 1.5 2.0 2.5 3.0 3.5

(1/T)/K-1*1000

Fig.3. Variation YS with temperature for 316 L(N) SS with Fig. 4. Variation of friction stress with reciprocal of
varying nitrogen. temperature, indicating strong thermal contribution

4. Conclusions

• Considering the microstructural stability of the 316 L(N) stainless steels, the method of equivalence of hot-
tensile and stress-rupture data seems feasible through an appropriate time-temperature parameter.
• Based on the literature, two methods, namely, Bueno Method and Oh Method have been suggested: Bueno
Method is more elaborate and is more fundamentally based and, hence, is likely to produce more accurate
results; but, the Oh Method is simpler to apply.
• Because of the availability of only a single strain rate hot-tensile data for the 316 L(N) SSs, in the present
paper, the simpler Oh Method has been applied for predicting the stress-rupture properties from hot-tensile
data.
• It is found that the predicted data follow the ASME Code Case minimum stress-rupture data for 316 SS
(stainless steel) and thus are very conservative. Predicted creep rates are also much larger than the actual
ones for the one case where comparison has been made.
• The conclusion drawn is that test results at different strain rates, rather than the single value considered here,
along with the more accurate Bueno Method suggested would yield better results.
 P.R. Sreenivasan / Procedia Engineering 55 (2013) 82 – 87 87

• In addition, using some old results from the literature, relation for variation of yield stress with temperature,
nitrogen content and grain-size for the present heats of steel (with various nitrogen contents) has been
derived.

Acknowledgement

The author thanks Director, Metallurgy and Materials Group, and Director, IGCAR, Kalpakkam-603102 for
encouragement and support.

References
[1] M.Steen, International Journal of Pressure Vessels and Piping, Vol.14, Issue 4, 1984, pp. 201-225.
[2] Levi de Oliveira Bueno, “Creep Behaviour of 2.25Cr-1Mo Steel - An Equivalence between Hot Tensile and Creep Testing Data.”,
ECCC Creep Conference, 12-14 September 2005, London – Proc.Creep & Fracture in High Temperature Components –Design &
Life Assessment Issues – Ed.by I.A.Shibli, S.R.Holdsworth, G.Merckling, DEStech Publ., USA., 2005, pp. 969-980.
[3] Hung-Kuk Oh, "Determination of rupture time and strain rate in creep by means of the uniaxial tensile test", Journal of Materials
Processing Technology, Vol. 59, 1996, pp. 294- 296.
[4] V.Ganesan, M.D.Mathew and K.Bhanu Sankara Rao," Effect of Nitrogen on Tensile Properties of 316LN Stainless steel: Report
IGC-294, 2007”, Indira Gandhi Centre for Atomic Research (Government of India, Department of Atomic Energy), Kalpakkam
603102, Tamil Nadu, India.
[5] G.Sasikala, M.D.Mathew, K.Bhanu Sankara Rao and S.L.Mannan, “Assessment of creep behavior of austenitic stainless steel
welds”, Indira Gandhi Centre fo Atomic Research (Government of India, Department of Atomic Energy), Kalpakkam-603102,
Tamil Nadu, India, pp.9.
[6] L.A.Norstrom, “The influence of nitrogen and grain size on yield strength in Type AISI 316 L austenitic stainless steel”, Metal
Science, June 1977, pp. 208-212.