PSI - Issue 42

A. Laureys et al. / Procedia Structural Integrity 42 (2022) 1458–1466 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

1460

3

Table 1: Overview of the chemical compositions (in wt%) of the studied materials. Element 316 L 316 Ti Sanicro 28 Sanicro 35

Hastelloy G30 Hastelloy G35

Ni Cr Fe Co Mo Mn Cu Nb W

12 17

10-14 16-18

31 27

35 27 30

43

58

28-31.5

33.2

65.5

64.4

36.5

13-17 max 5

max 2 max 1

2.5

2.0-3.0

3.5

6.5

4-6 2.5

8.1

max 0.6 max 0.5 max 0.3

2

2

max 2.0

0.63

max 1.5

0.41

1

2

0.8

Ti Si

0.70 0.75

0.75

max 0.6

max 0.2

max 0.8 max 0.04 max 0.03 max 0.02

max 0.6 max 0.3 max 0.05 max 0.015

P C

max 0.045

max 0.045

max 0.025 max 0.020 max 0.010

max 0.025 max 0.020

0.03

0.08

S

0.03

Al

max 0.4

N

0.10

0.10

0.28

The overall goal of the investigation was to compare the corrosion sensitivity of different materials in concentrated phosphoric acid at elevated temperatures ranging from 80 °C to 160 °C. The materials were first cut, ground and polished up to 3 µm followed by rinsing with demi-water and ethanol prior to corrosion testing. Due to the variable geometry and availability of the different materials, sample sizes varied between materials. For corrosion testing samples were inserted in a glass flask with round bottom containing concentrated phosphoric acid. A condenser was mounted on top of the glass flask to limit evaporation of the hot acid. The acid was stirred at a speed of 300 rpm. The glass flask hanged in an oil bath, which was temperature controlled by a hotplate. Food grade phosphoric acid (FGPA) (Prayphos 85 wt% FGPA of Prayon) was used. After testing the samples were again rinsed with demi-water and ethanol. The test duration, i.e. 48 h, was selected as such that a balance exists between the running time of an experiment and the time necessary to break through the passive layer of a material. The effect of the acid solution on the material was measured by determining the loss of weight of the specimen. The corrosion rate was calculated as follows: Corrosion rate ( m y m ) = A K x x t W x d (3) With K = 8.76 x10 4 , t = time of exposure (h), A = total area (cm²), W = weight loss (g), and d = density (g/cm³). Weight loss measurements allow to evaluate the uniform corrosion sensitivity of a material, however, they do not define the amount of local damage. Therefore, samples were, subsequently to submerging them in phosphoric acid, analyzed by SEM to evaluate the corroded sample surfaces. The latter was complemented with an EDX analysis to study the chemical composition of the surface layer and identify possible depositions. As such, the actual corrosion mechanism and sensitivity could be determined. 3. Results and discussion 3.1. Weight loss measurements The corrosion rates obtained from weight loss measurements after testing the different materials at various temperatures for 48 h in FGPA are visualized in Fig 1a, allowing straightforward comparison between materials. 0.15 mm/y was set as a maximum corrosion rate allowable for industry, therefore, Fig. 1b shows a more detailed view of the corrosion rates in the vicinity of this number. The various materials each show a specific corrosion behavior which evolves differently with temperature depending on the stability of their passive film. The stability of the passive film strongly depends on the chemical composition of the materials (Neville and Hodgkiess (1996)). At a certain