PSI - Issue 42

Jean-Baptiste Delattre et al. / Procedia Structural Integrity 42 (2022) 886–894 Jean-Baptiste Delattre / Structural Integrity Procedia 00 (2019) 000–000

887

2

depth understanding of the link between heat treatment, resulting microstructures and mechanical properties. Many studies reported about relationships between the microstructure and fracture properties. Yet, there is no systematic investigation in the case of low alloy steels, starting from a single material. The present research aimed at filling this gap, focusing on impact toughness properties at the beginning of the ductile-to-brittle transition and at analyzing the relationships between heat treatment, microstructure, tensile properties and fracture properties Previous work on this topic (Carassou; Ma¨ntyla¨ et al.; Renevey; Rodriguez-Ibabe) revealed the critical role of in clusions and precipitates in the initiation of brittle fracture. Starting from a single industrial material, the present work focused on the impact of the microstructure of this material, as a ff ected by various thermal treatments. Practically, the selected material is a nuclear grade steel with low amounts of impurities such as sulfur or phosphorus with a limited density of inclusions of small size and is representative of actual production. This study will focus on the e ff ect of tempering conditions with a limited number of cooling rates while another study deals with the e ff ect of cooling rate for a given tempering condition (Delattre et al.).

Nomenclature

E U Upper Shelf Energy E L Lower Shelf Energy T t Transition temperature T w Half width of the transition temperature range

2. Materials and method

The material selected for this study is a block of RCC-M 20MND5 steel coming from an 133-mm-thick cylindri cal shell. Table 1 gives the chemistry of the material. The homogeneity of the chemical composition across the shell thickness was checked prior to the study. The as-received material was delivered in a “Quality Heat Treatment” con dition (i.e. Quenched and tempered), and had a tempered bainite microstructure, with a hardness of about 225 HV 0 . 2 . Eight 14-mm-thick-blaks were cut from the block, normal to the radial direction of the shell. Each of these blanks

Table 1. Chemical content (in weight %) of the starting material (bal Fe.) (product analysis) C Mn Ni Mo Si Cr P

S

Cu

Al

0.20

1.44

0.89

0.47

0.18

0.15

0.005

0.0012

0.072

0.012

was quenched in a small industrial furnace equipped with gas quenching facility with two di ff erent controlled cooling rates, respectively representative of the minimum and maximum rates locally encountered in actual components dur ing water quenching. For each cooling rate, the facility was calibrated using tests on representative dummy blanks, instrumented with thermocouple. The extreme case of mid-thickness, when the total thickness of the component is around 700 mm, and of regions close to the component free surface, were encompassed in the study. Thus, the selected cooling rates were 150°C / h and 10000°C / h. All blanks were heated at 360°C / h, hold at 880°C for 2 hours, then cooled at one of the selected each cooling rate (150°C / h or 10000°C / h). The e ff ects of temperature and holding time were ex plored using the following conditions: 660°C for 6h, 640°C for 6h, 660°C for 6h and 640°C for 20h.The austenitized blanks were reheated at 360°C / h, hold at one of the above mentionned conditions, and then cooled at 50°C / h down to room temperature. A total of eight heat treatment conditions, leading to eight di ff erent microstructures, was thus investigated. In order to compare the tempering conditions, we used the tempering parameter P , given in equation 1, where T is the tempering temperature in degrees Celsius and t is the tempering duration in hours.

T + 273 . 15 1000

(1)

(20 + ln( t ))

P =

∗