10 0000@www Onepetro org@conference-paper@NACE-2019-12783
10 0000@www Onepetro org@conference-paper@NACE-2019-12783
10 0000@www Onepetro org@conference-paper@NACE-2019-12783
12783
Effect of Carbon Content on Polythionic Acid Stress Corrosion Cracking Resistance for UNS
S34751 with Extra Low Carbon Content
ABSTRACT
Polythionic Acid Stress Corrosion Cracking (PTA SCC) of conventional austenitic stainless steel is a
serious problem in refinery desulfurization units. Stabilization treatment after welding, Nitrogen
blanketing and/or neutralization treatment using an alkaline solution are required during each turn
around to prevent PTA SCC according to NACE(1) SP0170-2012.
It has been known that proprietary S34751 has superior PTA SCC resistance compared with S34700 or
S32100 even without Post Weld Heat Treatment (PWHT) and neutralization treatment during turn
around. However, the difference of of PTA SCC resistance between proprietary S34751 and generic
S34751 within extra low Carbon (C) content less than 0.02 wt% has not been reported.
Therefore, the relationship between C content and PTA SCC resistance for generic S34751 was
investigated by using long term aged samples with simulated welding Heat Affected Zone.
Key words: Polythionic acid stress corrosion cracking (PTA SCC), Sensitization, Proprietary version
S34751, Hydroprocessing, Carbon content
(1) NACE International (NACE), 15835 Park Ten Pl, Houston, TX 77084, USA
Stabilized austenitic stainless steels such as UNS S32100 or UNS S34700 are widely used in hydro-
treating or hydro-cracking processes in a complex refinery. However, S32100 or S34700 are
susceptible to Stress Corrosion Cracking (SCC) due to the formation of Polythionic acid (PTA) (H2SxO6,
x = 3 - 6) which causes cracking in the coarsened Heat Affected Zones (HAZ) because of sensitization,
due to the formation of a Chromium (Cr) depleted zone by the precipitation of Cr based M23C6 carbides
along grain boundaries which occurs during actual operation.1
To prevent PTA SCC, Stabilization Post Weld Heat Treatment (PWHT) at the temperature range of
843 °C to 900 °C for 2 to 4 hours is required for S32100 and S34700 to tie up free Carbon as TiC and
NbC in the coarsened HAZ. Furthermore, several kinds of PTA SCC shutdown mitigation methods are
listed in NACE SP0170 such as nitrogen purging to exclude oxygen, alkaline washing of equipment
surfaces, and dry air to prevent liquid water formation.2 Stabilization PWHT and/or these PTA SCC
mitigation methods affect not only the fabrication cost but also the cost and schedule during
maintenance opportunities. Hence, a stabilized austenitic stainless steel with excellent resistance for
sensitization to prevent PTA SCC without the above mitigation methods would be ideal for hydro-
treating and hydro-cracking processes in a complex refinery.
It has been noted that the eliminations of stabilization heat treatment (including PWHT) seems to be
possible for generic UNS S34751 with extra low Carbon content less than 0.02 wt% and higher Nb/C
ratio over 15. Reports on proprietary S34751 indicate superior sensitization resistance within generic
S34751,3-4 and S34751 has been shown to have excellent PTA SCC resistance without both
stabilization PWHT through proven long-term service experience in actual plants of hydro-treating or
hydro-cracking processes over 20 years. 5-7 However, the difference of PTA SCC resistance between
the generic S34751 and proprietary S34751 has never been documented under extra low Carbon
content less than 0.02 wt%. Therefore, it is difficult to judge the elimination of all PTA SCC mitigation
methods for the usage of not proprietary S34751 but generic S34751.
In this study, the slight variation of Carbon content less than 0.02 wt% was focused on S34751 and the
effect of Carbon content on PTA SCC resistance was investigated using the long-term aged samples
with simulated coarsened HAZ.
EXPERIMENTAL PROCEDURE
Test material
Chemical compositions of materials are shown in Table 1. Items 1 and 2 satisfied the chemical
composition range of S34751. Item 1 contains Carbon content similar to proprietary S34751. Item 2
contains higher Carbon content than Item 1 but remain within the chemical composition range of
generic S34751. Item 3 is a reference material of S34700. After the production of plates with 20 mm
(0.7874 in) thickness using hot forging and hot rolled processes, all items were solution heat treated
over 1040 °C and water quenched to room temperature.
Table 1
Chemical compositions of materials used. (wt%)
Item C Si Mn P S Ni Cr N Nb Nb/C
1 0.012 0.40 1.40 0.030 <0.001 10.3 17.4 0.080 0.35 29.2
2 0.018 0.38 1.47 0.034 <0.001 9.95 17.1 0.078 0.29 16.1
3 0.073 0.39 1.43 0.034 <0.001 9.94 17.2 <0.01 0.78 10.7
S34751 0.005- 1.00 2.00 0.045 0.030 9.0- 17.0- 0.06- 0.20- 15
0.020 max max max max 13.0 19.0 0.10 0.50 min
Generally, PTA SCC for stabilized austenitic stainless steels such as S34700 and S32100 occur at a
coarsened HAZ involving the full solution of Nb(C,N) or Ti(C, N) during the welding heating process. To
evaluate PTA SCC susceptibility in coarsened HAZ for Items 1-3, additional thermal treatment was
applied to simulate the microstructure of a coarsened HAZ under much higher heat input such as
Submerged Arc Welding using high frequency induced heating equipment. The simulated HAZ thermal
treatment condition was applied to Items 2, 3 (Figure 1a) and 1 (Figure 1b) test samples. Sample
dimensions were 12 mm (0.4724 in) square cross section and 100 mm (3.937 in) in length. The
difference of simulated HAZ condition between Figure 1 (a) and (b) is the holding duration at 1400 °C.
The goal of a longer holding duration at 1400 °C for Item 1 was to obtain the microstructure with larger
grain size than Item 2 for accelerated M23C6 precipitation.
(a) (b)
1400 °C for 5s hold
1400 °C for 10s hold
Aging condition
To simulate the sensitization treatment, the test samples were subjected to aging heat treatments at
550 °C up to 10,000 hours. This aging temperature was chosen because it corresponds approximately
to the nose temperature of the Time-Temperature-Sensitization (TTS) diagram for S34700.8
Cross section microstructures taken from the aged test samples were etched according to ASTM(2)
A262 Practice A, which is the 10% oxalic acid etch test.9 The etched microstructures were classified
into step structures without a ditch around the grains, dual structures with partial ditches around the
grains, and ditch structures with one or more grains with fully ditched grain boundary.
Also, EPR tests were conducted to evaluate the normalized charge Pa for these cross section
microstructures according to ASTM G108.10 Pa was calculated by Equation 1.
Pa(Coulombs/cm2)=Q/[As×{5.1x10-3exp(0.35G)}] (1)
Where Q is the charge in units of coulombs measured on current integration measuring instrument, As
is the specimen area in units of cm2 and G is the grain size at 100 x measured according to ASTM
E112.11 Generally, Pa for a sensitized material shows higher than that of non-sensitized material.
(2) ASTM International, 100 Barr Harbor Dr., West Conshohocken, PA 19428-2959 USA
©2019 by NACE International.
Requests for permission to publish this manuscript in any form, in part or in whole, must be in writing to
NACE International, Publications Division, 15835 Park Ten Place, Houston, Texas 77084.
The material presented and the views expressed in this paper are solely those of the author(s) and are not necessarily endorsed by the Association.
3
Sensitization Test
Aged test samples were immersed according to ASTM A262 Practice E and bent at the center of their
simulated HAZ thermal treated zone.12 After that, the cross section microstructures of the tested
samples were observed to evaluate Intergranular SCC (IGSCC) susceptibility using optical microscopy
in the as polished condition.
Aged test samples were U-bent with a bend radii from 0.5 to 1.0 of the sample thickness. Test samples
were then immersed in a solution of 1% K2S4O6 acidified by H2SO4 to a pH of 2.0 at room temperature
for 100 hours in accordance with Hosoya, Yamamoto and Kagawa (1985).13 This test was used instead
of ASTM G35-98 because the preparation of the Wackenroder solution in ASTM G35-98 requires
caution due to using H2S.14 A penetrant test and investigation for PTA SCC through a 10-power
binocular magnifying glass was conducted for the samples removed from the beaker.
RESULTS
Table 2 shows the comparison of microstructures of Items 1 to 3 after simulated HAZ thermal treatment.
Item 1 had a greater grain size than Items 2 and 3 due to the longer hold duration at 1400 °C.
Table 2
Comparison of microstructure for Items 1-3 after simulated HAZ thermal treatment.
Item 1 (0.012%C) Item 2 (0.018%C) Item 3 (0.073%C)
Tables 3 to 5 show the microstructure test results of Items 1, 2 and 3 according to ASTM A262 Practice
A after aging at 550 °C respectively. It was found that Item 1 only had a slightly sensitized
microstructure by M23C6 precipitation at the grain boundaries even after 10,000 hours aging because of
dual structures for all aging conditions tested. However, Item 2 showed sensitized microstructures at
3,000 hours aging at 550 °C because ditch structures formed due to the precipitation of M23C6 at grain
boundaries. Item 3 showed sensitized microstructures at 1,000 hours aging at 550 °C. From these test
results, it was clarified that there was a large difference of M23C6 precipitation amounts caused by the
slight variation of Carbon content when less than 0.02% wt% and M23C6 precipitation amounts
increased with increasing Carbon content.
Table 4
Microstructures of Item 2 (0.018 wt%C) according to ASTM A262 Practice A.
550 °C for 1,000 hours 550 °C for 3,000 hours 550 °C for 10,000 hours
Table 5
Microstructures of Item 3 (0.073 wt%C) according to ASTM A262 Practice A.
550 °C for 1,000 hours 550 °C for 3,000 hours 550 °C for 10,000 hours
Tables 6 to 8 show IGSCC test results of Items 1, 2 and 3 according to ASTM A262 Practice E after
aging at 550 °C respectively. It was found that although Item 1 showed increased IGSCC susceptibility
with increasing aging duration, the maximum length of an IGSCC crack after aging for 10,000 hours
Table 6
IGSCC test results of Item 1 (0.012 wt%C) according to ASTM A262 Practice E.
550 °C for 1,000 hours 550 °C for 3,000 hours 550 °C for 10,000 hours
Table 7
IGSCC test results of Item 2 (0.018 wt%C) according to ASTM A262 Practice E.
550 °C for 1,000 hours 550 °C for 3,000 hours 550 °C for 10,000 hours
Fractured Fractured
Some indications (PT was not applied) (PT was not applied)
Tables 9 to 11 show PTA SCC test results of Items 1, 2 and 3 after aging at 550 °C respectively. It was
found that Item 1 showed heavy evidence of PTA SCC only after 10,000 hours aging at 550 °C.
However, Item 2 showed slight PTA SCC with about 20 m indications along the grain boundary after
only 1,000 hours aging and found many PTA SCC indications over 3,000 hours aging. Also, for Item 3,
all test samples had evidence of heavy PTA SCC along the grain boundaries. From these test results, it
was clear that Item 1 had superior PTA SCC resistance compared with Items 2 and 3 in spite of greater
grain size. Item 3 showed the lowest PTA SCC resistance.
Figure 2 also shows the relationship between the measured normalized charge Pa and aging duration
at 550 °C. Although the Pa value of Item 1 increased with increasing aging duration, the absolute value
of Pa at each aging duration for Item 1 was lower than those of Items 2 and 3. Also, Pa values of Items
2 and 3 increased with increasing aging duration and the absolute values of Pa for Item 2 were lower
than for Item 3. Therefore, it was found that the difference in sensitization susceptibility among Items 1
to 3 didn’t depend on grain size.
Table 9
PTA SCC test results of Item 1 (0.012 wt%C).
550 °C for 1,000 hours 550 °C for 3,000 hours 550 °C for 10,000 hours
Table 11
PTA SCC test results of Item 3 (0.073 wt%C).
550 °C for 1,000 hours 550 °C for 3,000 hours 550 °C for 10,000 hours
40
Item 3
30
20
10
0
0 2000 4000 6000 8000 10000
Aging duration (hours)
Figure 2: Relationship between normalized charge Pa and aging duration at 550 °C.
Discussion
Several test results show that PTA SCC resistance was influenced by Carbon content within extra low
Carbon content (less than 0.02 wt%) in S34751. Table 12 shows the comparison of M23C6 precipitation
amounts after 10,000 hours aging at 550 °C by replica TEM observation. Although the slight amount of
finer M23C6 precipitation was only found in the grain boundaries of Item 1 after 10,000 hours aging,
Items 2 and 3 contained much greater amounts of coarsened M23C6 precipitation in grain boundaries.
Additionally, the precipitation amount of Item 3 was greater than Item 2. These results coincided with
the microstructure test results shown in Tables 3 to 5 by ASTM A262 Practice A. Therefore, the
difference of PTA SCC resistance between Items 1 and 2 is likely influenced by the difference of M23C6
precipitation amount.
Table 12
Comparison of M23C6 precipitation behavior after 10,000 hours aging by replica TEM.
Item 1 (0.012 wt%C) Item 2 (0.018 wt%C) Item 3 (0.073 wt%C)
In order to verify the validity of the difference of M23C6 precipitation amount by the variation of Carbon
content (when less than 0.02 wt%), the equilibrium calculation at 550 °C was conducted with
Thermodynamic calculation software based on the CALPHAD (CALculation of PHAse Diagram) method
Conversely, the amount of M23C6 precipitation at the grain boundaries is determined by the combination
of the nucleation frequency of M23C6, the grain size at nucleation sites and the growth rate of M23C6.
EPR test results shown in Figure 2 indicate the correlation to the variation of M23C6 precipitation
amounts and the difference of nucleation sites among Items 1 to 3 corresponding to grain size is not
sensitive for these test results (shown in Tables 3 to 5 and Table 12). Also, since the growth rates of
M23C6 between Items 1 and 2 is determined by the growth rate satisfied by the flux balances of Cr and
Ni with much smaller diffusion rates than C, the slight variation of growth rates between Items 1 and 2
caused by the slight deviations of Cr and Ni contents is not sensitive to these test results.16 Therefore, it
is assumed that the difference of M23C6 precipitation amounts shown in Tables 3 to 5 and Table 12 is
most sensitive to the variation of M23C6 nucleation from the supersaturated austenitic phase
corresponding to the microstructure of a coarsened HAZ.
Generally, it has been known that the nucleation frequency of precipitation correlates with the driving
force of nucleation.17 Also, the supersaturated austenitic microstructure is the initiation site before aging
at 550 °C by applying thermal treatment to simulate the microstructural behavior of a coarsened HAZ.
Thus, in this research, the driving force of M23C6 nucleation from supersaturated austenitic phase was
evaluated. Figure 3 shows the schematic illustration of the driving force from austenitic phase to M23C6
for a binary system, where represents M23C6. The driving force of M23C6 nucleation from
supersaturated austenitic phase G → can be calculated by the combination of the mathematical
induced parallel plane equations of ⑦-⑧ and ⑨-⑩ and the thermodynamic parameters of chemical
potential of element i for austenite i , chemical potential of element i for dual phase equilibrium
condition of austenite and M23C6 /M23C6i and mole fractions of element i xi calculated using
Thermodynamic calculation software. Table 14 shows the calculated driving force of M23C6 from a
supersaturated austenitic phase. It was clarified that the driving force of M23C6 from supersaturated
austenitic phase for Item 1 was the smallest and that for Item 3 was the largest. Also, the driving force
of M23C6 coincided with the observed results shown in Table 12.
Consequently, PTA-SCC resistance of S34751 by the variation of Carbon content depended on the
variation of M23C6 precipitation amount at the grain boundaries of a coarsened HAZ with a
supersaturated austenitic phase. The behavior of growth of M23C6 will be solved using a numerical
model in further research.
Table 13
Volume fraction obtained using equilibrium calculation results.
Item 1 Item 2 Item 3
(0.012 wt%C) (0.018 wt%C) (0.073 wt%C)
NbC - - 0.007292
Z phase 0.006047 0.005003 -
Cr2N 0.002749 0.003645 -
Laves - - 0.002602
M23C6 0.00227 0.0034361 -
G C
M
G→
G
C
xC
xC M23C6 C
Figure 3: Schematic illustration of driving force of M23C6 from austenite for binary system.
Table 14
Calculated driving force of M23C6 from supersaturated austenitic phase.
Item 1 Item 2 Item 3
(0.012 wt%C) (0.018 wt%C) (0.073 wt%C)
4754 J/mol 5260 J/mol 7384 J/mol
CONCLUSIONS
・ PTA SCC resistance of S34751 was susceptible to slight variations of Carbon content within the
extra low Carbon content (less than 0.02 wt%) and the strict control of Carbon content less than
0.015 wt% was mandatory to achieve higher PTA SCC resistance of S34751.
・ The variation of M23C6 precipitation amount at a coarsened HAZ of S34751 was explained by the
driving force of M23C6 nucleation from a supersaturated austenitic phase at sensitized temperatures
using thermodynamic calculations. The variation in the driving force of M23C6 by changing Carbon
content agreed with the microstructure observation results.
・ From these results, the strict decrease of Carbon content within the specified Carbon range of
S34751 was one of the important factors for superior PTA SCC resistance of proprietary S34751.
1. M. Kowaka, Metal Corrosion Damage and Protection Technology, 1st ed. (Tokyo, Japan : Agune,
1983), p. 404.
2. NACE SP0170 (2012 Edition), “Protection of Austenitic Stainless Steels and Other Austenitic Alloys
from Polythionic Acid Stress Corrosion Cracking During a Shutdown of Refinery Equipment”
(Houston, TX : NACE).
3. S. A. Bradley, M. W. Mucek, H. Anada and T. Osuki, “Alloy for Resistance to Polythionic Acid
Corrosion Cracking for Hydroprocessing Applications”, Materials and Design, 110 (2016): pp. 296-
303.
5. T. Kudo, Y.Tarutani, M. Miura, Y.Sawaragi and M.Nishi, “Development of 347AP with Excellent PTA
SCC Resistance”, SUMITOMO KINZOKU, 38-3 (1986): pp.190-200. (In Japanese)
6. Y. Matsuda, J. Higuchi and H. Anada “Properties of Sumitomo 347AP Steel Tube for Hydro-treater
in Complex Refinery”, STAINLESS STEEL WORLD 2013, (Maarstricht, NL) .
8. R. W. Howe and D. V. Beggs, “Polythionic Acid Stress Corrosion Cracking of Heat and Corrosion
Resistant Alloys” , CORROSION 93, paper no. 541 (Houston, TX : NACE, 1993).
9. ASTM A262-13 PRACTICE-A “OXALIC ACID ETCH TEST FOR CLASSIFICATION OF ETCH
STRUCTURES OF AUSTENITIC STAINLESS STEELS” (West Conshohocken, PA: ASTM).
10. ASTM G108-94 “Electrochemical Reactivation (EPR) for Detecting Sensitization of AISI Type 304
and 304L Stainless Steels” (2015)
11. ASTM E112-13 ”Standard Test Methods for Determining Average Grain Size” (West Conshohocken,
PA: ASTM).
12. ASTM A262-13 PRACTICE-E “COPPER-COPPER SULFATE-16% SULFURIC ACID TEST FOR
DETECTING SUSCEPTIBILITY TO INTERGRANULAR ATTACK IN AUSTENITIC STAINLESS
STEELS” (West Conshohocken, PA: ASTM).
13. K. Hosoya, K. Yamamoto, N. Kagawa “IG SCC Susceptibility of Austenitic Stainless Steels in
Acidified K2S4O6 Solutions New Polythionic Acid SCC Test Solution”, Boshoku Gijutsu 34 (1985):
pp.568-572.
16. Sahlaoui H,. Sidhom H., and Philibert J.” Prediction of chromium depleted-zone evolution during
aging of Ni–Cr–Fe alloys”, Acta Materialia., 50-6 (2002): pp. 1383-1392.
17. T. Nishizawa, Thermodynamics of Microstructure, 1st ed. (Tokyo, Japan : Maruzen, 2005), p. 209.