2.11.In haar rapport van 26 oktober 2016 (Report # P 0173 - 3009/2016 - TL) schrijft UG onder meer:
"As a result of the investigations,
1. it was determined that the various weld anomalies observed exist since the construction
period of the pipeline (installation in 1958 - 1960). In particular, the unusual combination of
a burn through with multiple slag inclusions in almost the same plane has been present in
girth weld # 6-4167 at the 5 o'clock since the commissioning of the RRP pipeline (1960).
2. it is judged that the likelihood of occurrence of a combination of a severe burn-through and,
on top of it, multiple slag inclusions which have fragmented is very remote. This is
substantiated by the observation that, according to RRP, this is the first time a similar leak
originating from a HAZ pinhole has been found in the concerning pipeline.
3. it is believed that an occasional pipe deformation event (such as a small soil settlement,
which may occur given the peaty nature of the soil) facilitated the fragmentation of the hard
and brittle (Si02 and Ti02) slag inclusions in the fill and cap passes of the girth weld (i.e.,
'above' the 3.30 mm high / deep burn-through anomaly) and the ductile failure of the
remaining metal ligament, which caused the leak.
4. since no traces of an oil products spill / loss have ever been observed at that particular location, it can be concluded that the leak has occurred shortly before or on June 8, 2016.
SCANNING ELECTRON MICROSCOPIC (SEM) EXAMINATION AND ENERGY-DISPERSIVE X-RAY (EDX) ANALYSIS OF THE FRACTURE FACES
(…)
Apparently the slag inclusions have fragmented (note that welding slag is a relatively brittle
substance) and have been 'washed out' by the pressurized oil products being transported, so as
to create the leak to the outside.
The macro / microphotographs of the polished and etched cross section taken out through the
leak and presented below confirm that the weld deposit effectively contained slag inclusions in
the immediate vicinity of the leak. The embedded (buried) slag inclusion had a height (in the
through-thickness direction) of 1.60 mm and was not open to the outer pipe wall (see Appendix
F).
[opmerking rechtbank: hierna volgen twee afbeeldingen, die zijn weggelaten]
To summarize, it can be concluded that the leak is a result of the presence of slag inclusions in
the fill and cap passes of the girth weld (i.e., 'above' the 3.30 mm high / deep burn-through).
These anomalies exist since the construction period of the pipeline (installation in 1958 - 1960).
This combination of a burn-through with multiple slag inclusions has been present in girth weld
# 6-4167 at the 5 o'clock since the commissioning of the RRP pipeline (1960). Since slag
inclusions are relatively hard and brittle (SiO2 and TiO2), these have fragmented and 'washed
out', thus creating leakage from the 'wormhole like' anomalies. Since no traces of an oil
products loss have ever been observed at the particular location (RRP girth weld # 6-4167), it
must be concluded that the leak has occurred shortly before or on June 8, 2016.
(...)
OPTICAL METALLOGRAPHIC (MACRO -M ICRO) EXAMINATION OF GIRTH WELD # RRP 6-4167 AT THE LOCATION OF THE LEAK
(...)
In this Appendix, the findings of the optical metallographic examination of two macrographic
sections, taken out close to the leak are presented and briefly discussed. A first cross section (#
4167 - L1) was extracted through the deepest point of the burn-through defect (see Figure F.1),
whereas the second section (# 4167 - L2) was made through the HAZ pinhole leak (see Figure
F.5). Both sections were polished to 1 micron finish, micro etched in 2 % nital and subsequently
studied under the optical microscope. As an illustration, macrophotographs and a series of optical
microphotographs (magnification: x 35) have been taken, which are presented in the following
Figures:
Cross section # Macrophotograph Optical microphotographs in Figures
4167 -L 1 F.2 F.3 &F .4
4167 -L 2 F.6 F.7 to FA 1
The examination of section # 4167 - L1 has not provided additional useful information, except
that the maximum height / depth of the burn-through defect (3.30 mm) could [be] measured with
precision. As expected, the 'artificial' fracture (provoked by opening up in LNZ) had propagated
through the weld metal in a fully brittle manner (cleavage fracture).
The study of section # 4167 - L2 has yielded more relevant features. The optical microphotograph
in Figure F.7 provides evidence that no plastic deformations had occurred at the pinhole
leak. This excludes the hypothesis that the leak might have been a result of a severe accidental
overload. Of particular relevance is also the microphotograph in Figure F.8, portraying a slag
inclusion in the grain refined weld deposit towards the cap of 1.60 mm high. This confirms that
the wormhole like defects identified under the SEM (see Appendix E) were originally / initially
(i.e., at the time of construction) filled with slag inclusions, which have been washed out over
time."