Page 86 - Handbook of Materials Failure Analysis

80 CHAPTER 4 Creep damage of high alloyed reformer tubes

respectively), while stresses were greater than design ones. In order to obtain short
rupture time, test stresses were chosen greater than 17.5 MPa, that is the hoop stress
corresponding to the design internal pressure of 32.8 bar.
Samples of tube wall were submitted to metallographic preparation by an auto-
matic grinder-polisher system using a series of grit papers and a finishing diamond
polish.
Metallurgical analysis was performed by optical and scanning electron micros-
copy (SEM). The utilized SEM apparatus was equipped with backscattered electrons
detector and microanalysis device for energy dispersive X-ray analysis (EDX).
Samples were electrolytically etched by a solution of 10% oxalic acid for 8 s at
5 V before being examined by optical microscopy.
For SEM observations, samples were etched by a solution containing 15% of
glycerol (100% of concentration), 45% of HNO 3 (65% of concentration), and
40% of HCl (37% of concentration).

2.2 RESULTS OF THE EXPERIMENTAL INVESTIGATIONS
2.2.1 Internal diameter measurements and hardness tests
Results of LOTIS measurements are given in Figure 4.9a by means of an internal
diameter growth tube map of the reforming furnace. This map allows to choose
the tubes to put out of service whether their internal diameter deformation exceeds
the limit value ε¼1.5%, established for the plant under examination. Anyway, tubes’
internal radius is not uniform along the tube axis (Figure 4.9b).
By way of example, results of LOTIS internal inspection along tubes, respec-
tively, after 90,000 and 114,000 h, are given in Figure 4.10. The solid and the dashed
line show that internal diameter is not uniform along the tube length: the upper seg-
ment of the tube results undeformed being the internal diameter equal to the nominal
value, while the maximum internal diameter growth was recorded in the tube bottom
where the highest temperatures are reached.
Vickers hardness results, performed across the thickness of different samples, cut
from the tube decommissioned after 114,000 h (samples B2-B6, see Figure 4.8),
were put in comparison with diameter deformation.
From Table 4.3, it can be noticed that at high diameter deformation corresponds
small Vickers microhardness: samples B3 and B6 (cut from segments of the lower
part of the tube where high diameter deformation was recorded) have smaller hard-
ness values than the other samples and this can be considered as an effect of creep
damage.

2.2.2 Metallurgical analysis—optical microscopy observations
Microscope observations were carried out on metallographic sections taken transver-
sally to the tube wall. As pointed out in Refs. [30,31], metallographic features of
tubes are characterized by austenitic dendritic grains oriented in radial direction
(Figure 4.11), providing greater strength and better creep resistance to the circum-
ferential stress generated by the inner pressure.