Angela Wensley Report Nueva Aldea 2010 PDF

15397 COLUMBIA AVENUE Tel.
+1-604-536-2053
ANGELA WENSLEY WHITE ROCK, BC, CANADA Fax. +1-604-536-2078
E N G I N E E R I N G V4B 1K1 Email: awensley@uniserve.com
CORROSION INSPECTIONS
DURING THE GENERAL
PLANT SHUTDOWN
AT NUEVA ALDEA
MARCH 2010
Prepared for:
Celulosa Arauco y Constitución S.A.

Planta Nueva Aldea
Autopista del Itata – km 21
Nueva Aldea
Chile
Attention: Mario Ávila G.

Jefe de Análisis Sintomático
Date: 22 April, 2010 Dr. Angela Wensley, P.Eng.

Project No. AWE10005 Pulp & Paper Corrosion Specialist
C:\Winword\Nueva Aldea\Nueva Aldea 2010\Report Text Nueva Aldea 2010.doc

ANGELA WENSLEY Corrosion Inspections During the General Plant Shutdown
E N G I N E E R I N G at Nueva Aldea – March 2010
TABLE OF CONTENTS
SUMMARY . . . . . . . . . . 4
RECOMMENDATIONS . . . . . . . . 14
1.0 INTRODUCTION . . . . . . . . 22
2.0 EVAPORATORS . . . . . . . . 23
2.1 Superconcentrator S-1 TAG 551-51-0101 . . . . 23
2.2 Concentrator Effect E-1 TAG 551-51-0102 . . . 26
2.2.1 Old Tube Bundles . . . . . . 26
2.2.2 New Tube Bundles . . . . . . 27
2.2.3 Concentrator Shell . . . . . . 28
2.2.4 Concentrator Top Head . . . . . 30
2.3 Evaporator Effect E-2 TAG 551-51-0103 . . . . 31
2.9 Surface Condenser TAG 551-51-0109 . . . . 37
3.0 DIGESTERS AND IMPBINS . . . . . . 38

3.1 Repairs to the External "Half-Pipe" Headers . . . 38
3.2 Line 1 (Pine) Digester TAG 541-51-1101. . . . . 39
3.3 Line 2 (Eucalyptus) Digester TAG 541-51-2101 . . . 41
3.4 Line 1 (Pine) Impbin TAG 541-51-1010 . . . . 42
3.5 Line 2 (Eucalyptus) Impbin TAG 541-51-2010 . . . 43
4.0 BLEACH PRESSES . . . . . . . . 44

4.1 Mechanical Damage of Rolls. . . . . . 45
4.2 Corrosion of Welds . . . . . . . 46
4.3 Modification of the Flaps . . . . . . 47
4.4 Corrosion of Flaps and Frames . . . . . 47
4.5 Corrosion of Distribution Screw Covers . . . . 48
4.6 Corrosion of Shredder/Conveyor Housings . . . 48
4.7 Corrosion of Duplex Stainless Steel Piping . . . 49
5.0 WHITE WATER TANKS FOR THE PAPER MACHINES . . 50
6.0 EXPANSION TANKS . . . . . . . 51
2
7.0 DISCUSSION . . . . . . . . 52
7.1 The Need for Quality Assurance . . . . . 53
7.2 The Need for Corrosion Testing . . . . . 54
7.3 The Need for Failure Analysis . . . . . 56
8.0 REFERENCES . . . . . . . . 57
FIGURES . . . . . . . . . . F-1
3
SUMMARY
Evaporator Effect S-1 (Superconcentrator)
There were numerous leaks in the type 2304 duplex stainless steel tubes near the
bottom of both the A and B tube bundles. The leaks occurred:
1. At welds between short vertical tubes and the large diameter condensate
header. These leaks originated at defects (pores) in the welds.
2. In the heat affected zones for welds between thin-wall Tubel® tubes and the
small diameter headers. These leaks had no evident origin (unlike those in
case (1) above) and appeared to result from liquor-side corrosion.
Feritscope inspection of the welds for phase balance revealed anomalies that
should have been resolved by quality assurance (QA) in either the design stage
or during construction. The welds in case (1) had ferrite contents ranging from
19% to 38%, while the welds in case (2) had ferrite contents ranging from 52% to
69%. The ferrite range expected for duplex stainless steel welds is between 30%
and 60%.
An elbow in a condensate drain was cracked and was removed. The Feritscope
revealed that it had a ferrite content of only 13%, well below the range of 35% to
55% expected for type 2304 duplex stainless steel tubes. The high austenite
content (87%) would have made the elbow susceptible to stress corrosion
cracking (SCC) in superconcentrator service. The low ferrite content was likely
due to an improper annealing heat treatment after the elbow was hot formed.
This is another example of inadequate QA during construction of the
superconcentrator.
4
Evaporator Effect E-1 (Concentrator)
The cracked (SCC) type 304L stainless steel tube bundles were removed from
bodies A, B, and C. The uncracked tube bundle was moved from body D to body
C. New bundles of type 2304 duplex stainless steel tubes were installed in bodies
A, B, and D. While the tubes were out, a very good attempt was made to remove
the cracks from the wall of the concentrator using flap discs. SCC was found in
the shell of bodies A, B, and C and in the top head in the following locations:
- Adjacent to the circumferential and vertical weld seams.

- In the circumferential welds in body A.
- Surrounding remnant welds for temporary attachments made during erection
of the concentrator.
Extensive SCC was also found in the dividers (non-pressure parts) between the
four bodies.
Corrosion thinning of the type 304L stainless steel shell and tubes in body A,
possibly due to intergranular attack, had also progressed at a slow rate (<0.25
mm year).
Removing the cracks from much of the high-risk locations on the type 304L
stainless steel shell and cap has bought some time; however, the SCC is liable to
re-initiate. Also, the type 304L stainless steel tubes moved to body C will most
probably develop SCC. The best long-term strategy is to replace the shell, caps,
and all remaining type 304L stainless steel tubes in the concentrator using type
2304 duplex stainless steel.
5
Evaporator Effect E-2
The shell and internals on the liquor side of effect E-2 are now entirely of type
304L stainless steel construction (the original carbon steel top head was replaced
in 2009) and are visually without any corrosion or SCC. PT (penetrant testing)
should be done in 2011 to confirm the absence of SCC.
There was no access to the upper head, although inspection in 2009 revealed
some galvanic corrosion of the carbon steel top head adjacent to the top
cylindrical ring of type 304L stainless steel. This location should be inspected in
2011.
The bottom shell, head, and internals on the liquor side are all type 304L stainless
steel and visually without corrosion or SCC.
There was no galvanic corrosion of the carbon steel top head adjacent to the
upper cylindrical ring of type 304L stainless steel.
Visual inspection through the lower manway of E-4 did not reveal any
significant corrosion of the carbon steel shell and bottom head. The thickness of
the carbon steel divider around the liquor inlet in E-4 was 10 mm, a value of
interest as the original nominal thickness (the dividers in the effects E-5, E-6, and
E-7 were noticeably thinner due to corrosion). Also of interest was the use of
stainless steel for the mist eliminator supports in E-4. The carbon steel supports
in effects E-5, E-6, and E-7 have experienced corrosion.
6
The China hat and the roof above the liquor inlet in the bottom of E-4 need to be
inspected for damage that may be similar to that which caused the failure of the
same components in effect E-5.
One of the ladder rungs in the bottom liquor zone had failed due to corrosion,
making entry unsafe. The stainless steel China hat on the liquor inlet had
separated (due to fatigue cracking) from the carbon steel liquor inlet. Without the
China hat, inlet liquor impacted the ceiling of the divider above the inlet, causing
it to fail. Although I had no opportunity to closely inspect the failures, I suspect
that preferential corrosion of high-silicon content welds in the carbon steel
divider contributed to the failure. Caustic SCC of the carbon steel divider may
also have been involved. This hypothesis needs to be confirmed by failure
analysis.
The bottom of evaporator effect E-5 should be scaffolded in 2011 so a full

corrosion assessment can be done. It is likely that the carbon steel internals in
the liquor zone will have to be replaced, and that protection of the lower
cylindrical wall and bottom head will have to be provided no later than 2012.
7
In the liquor zone, the shell and bottom head of E-6 experienced severe corrosion
thinning and also preferential attack of heat affected zones adjacent to the welds.
The carbon steel internals exposed to liquor on two sides had double the
corrosion rate of the carbon steel shell. The divider around the liquor inlet had
thinned to as low as 3 mm. The mist eliminator wall, roof, and supports for the
mist eliminator were also severely corroded. The roof of the mist eliminator had
a corrosion perforation. The liquor inlet, China hat, and possibly the floor (if one
exists) of the mist eliminator also likely experienced severe corrosion thinning.
There was preferential corrosion of carbon steel welds for the mist eliminator
likely due to the high silicon content of the welds. There was also galvanic
corrosion of carbon steel adjacent to all stainless steel components, especially of
drain pipes from the mist eliminator.
All carbon steel internals in effect E-6 should be replaced using type 304L
stainless steel. The cylindrical shell and bottom head in the liquor zone below
the lower tubesheet should be protected; for example, using anodic protection
below the liquor level and a stainless steel lining above.
There was no access to the upper head. Inspection in 2009 revealed some
galvanic corrosion of the carbon steel top head adjacent to the upper cylindrical
ring of type 304L stainless steel. This location should be inspected in 2011.
The bottom of evaporator effect E-7 should be scaffolded in 2011 so a full

corrosion assessment can be done. It is likely that the carbon steel internals in
the liquor zone will have to be replaced, and that protection of the lower
cylindrical wall and bottom head will have to be provided no later than 2012.
8
Surface Condenser
The internal carbon steel surfaces were covered with "carbuncles" that had
pitting corrosion underneath, characteristic of oxygen corrosion. If this situation
is unacceptable for operational purposes, consideration should be given to
replacement of the surface condenser using a stainless steel that is resistant to
oxygen corrosion.
Continuous Digester External "Half-Pipe" Headers
The external "half-pipe" headers on both digesters had leaks that occurred at
fatigue cracks in duplex stainless steel welds with incomplete root penetration.
Repairs were made by cutting windows in the half-pipes, beveling of the roots of
the upper and lower welds, followed by welding using type 309L austenitic
stainless steel. The welds were originally made only from the outside. The
windows were then replaced with adequate purging of the root side with argon.
Continuous Digesters
The bottom scrapers in both the digesters were visually in good condition. There
was no visual evidence of propagation of the small cracks reportedly found in
the Line 2 (eucalyptus) digester scraper in 2009, although these still pose a threat
to the serviceability of the bottom scraper in that digester.
The China hat in the Line 1 (pine) digester had erosion of the circumferential
welds that exposed the underlying carbon steel structural support. Weld repairs
were made using type 309L stainless steel, an unacceptably soft material. In
2011, the cone should be re-plated using type 2205 duplex stainless steel. The
China hat for the Line 2 (eucalyptus) digester was modified.
9
Cracking of stitch welds at screens can be minimized by using type 312 stainless
steel instead of the weaker type 309L stainless steel currently used.
The replacement carbon steel cover plates for the beam nozzles on both digesters
were in good condition. There is no imminent need to replace these covers with
duplex stainless steel as was recommended in 2009. The remaining solid type
304L stainless steel covers should be replaced using carbon steel.
There was no evident corrosion of the vapour injectors, although they were not
removed for inspection and there was no access for inspection. The injectors
should be temporarily removed for inspection in 2011.
Areas of white coloured surface on the wall near the top of the digester may be
active corrosion due to the action of vapour from the injectors. In 2011,
thickness measurements should be made in selected white areas.
Impbins
The exterior surface of the inner central pipe of the Line 1 (pine) impbin was
corroded over a length of approximately 1 m. The corrosion was observed when
a modification was made that involved shortening both the inner and outer
pipes. Reportedly, the outer pipe is not used so it is possible that corrosive wood
acids could have condensed inside, destroying the passivation of the type 304L
stainless steel. A vent hole in the outer pipe may prevent wood acid vapours
from accumulating.
10
Pine Bleach Presses
Most of the rolls experienced mechanical damage that may have occurred during
the earthquake. Most of the mechanical damage involved small areas of
deformation of the roll surface, but in one case (side 1 of pine D2 press 7) the roll
was torn. Such locations may be sites for initiation of fatigue cracking in service
and should be monitored.
No significant corrosion was observed on the type 316L stainless steel pre-bleach
presses 1, 2, and 3. The EOP-stage press 5 had corrosion in the housing for the
shredder/conveyor discharge screw on the top. ClO2 vapours from D-stage
filtrate introduced to the standpipe caused superficial corrosion that turned the
type 316L stainless steel housing blue in colour.
The Line 1 D0, D1, and D2-stage presses 4, 6, and 7, respectively, were more
corroded than their Line 2 counterparts, likely the result of higher ClO2 residuals
on the pine side. The pine D-stage presses had transpassive corrosion of nickel-
base welds made using filler metal ERNiCrMo-3 (Alloy 625). At least one plug
weld in the lining in the side 2 vat had already corroded through to the
underlying carbon steel. A welding filler metal that is resistant to both
transpassive and pitting corrosion should be used; unfortunately, this material
has not yet been invented. Iron-base materials such as Alloy G3, Alloy G30, and
Avesta P54 resist transpassive corrosion but may be susceptible to pitting
corrosion. The best way to determine the best filler metal is to do corrosion
testing in actual D-stage filtrates.
The pine D-stage presses 4, 6, and 7 all had pitting corrosion of the
superaustenitic stainless steel (6% Mo stainless steel) in stagnant areas and
crevices. These locations included the edges of the flaps, the frames for the flaps,
11
the distribution screw covers, and the housings for the shredder/conveyor
screws. Currently, the pitting corrosion is just a cosmetic problem, but after
several (5 to 10) years could affect the operation of the presses, necessitating
replacement of major components of the presses.
The flaps for the three D-stage presses 4, 6, and 7 were modified to narrow the
gap between the flap and the roll. First, strips of 6% Mo stainless steel were
installed, followed by a welded lining of 6% Mo stainless steel that was both seal
welded and plug welded. Unfortunately, the material used for welding was
ERNiCrMo-3 (Alloy 625) nickel-base alloy. The welds will corrode in service.
Eucalyptus Bleach Presses
Pre-bleach press 1 was not opened for inspection during the time I was at the
mill, due to damage to the foundation from the earthquake. Press 4 (D0) was
rebuilt. Access to presses 5 and 6 was limited due to nearby structural damage.
Press 7 (D0) had the most corrosion. In general, the Line 2 D-stage presses were
less-corroded than their Line 1 counterparts. The housing shredder/conveyor
discharge screw on the EOP press had developed a blue colouration due to
corrosion by ClO2 vapours released from D-stage filtrate added to the standpipe.
Duplex Stainless Steel Piping in the Bleach Plant
Duplex stainless steel piping containing Line 1 (pine) D-stage filtrates had
numerous leaks due to internal corrosion. Attempts to repair or patch the leaks
have resulted only in more leaks. Type 2205 duplex stainless steel does not have
sufficient resistance to pitting corrosion when the Cl2 and ClO2 residuals are
high. No leaks have (yet) been observed in duplex stainless steel piping on the
eucalyptus side. Suitable alternatives for replacement of the corroded lines are
12
titanium (Grade 2), fibre reinforced plastic (FRP), or 6% Mo stainless steel

(welded with a filler metal with better corrosion resistance than Alloy 625).
Stainless Steel White Water Tanks for the Machines
Microbiologically-assisted corrosion (MIC) continues to be a problem in stainless

steel white water tanks on the pine side. In tank 572-22-1155 some type 316L
stainless steel wall plates with through-wall leaks were replaced. In tank 572-22-
1156, the "carbuncles" were removed and the metal surface was polished. Weld
repairs were made in 2009 using an over-matching filler metal (type 309MoL
stainless steel) that appeared to survive better than previous weld repairs made
using E316L. If the bacteria cannot be killed, then the tanks will have to be
replaced using an alloy and filler metal that are more resistant to MIC.
Expansion Tanks for the Turbogenerator
Severe pitting was observed in the bottoms of two expansion tanks (565-23-901
and 565-23-902) of carbon steel construction. The pitting appeared to be
conventional "oxygen pitting" due to air ingress in hot water and was not
characteristic of MIC.
13
RECOMMENDATIONS
I recommend that:
For the Superconcentrator:
S1. The sample of failed Tubel® tubes should be examined by a competent

metallurgist (Angela Wensley) to provide the following information:
- Elemental composition of the tubes and welds.

- Microstructure of the tubes and welds, especially of the heat affected zones
(look for precipitation of secondary phases).
- Microhardness traverses across the tubes, welds, and heat affected zones.
- Phase balance (austenite/ferrite) measurements in the tubes, welds, and
heat affected zones.
- Evidence of the failure mechanism (corrosion, cracking).
S2. The failed elbow from the condensate drain should be examined by a
competent metallurgist (Angela Wensley) to provide the following
information:
- Elemental composition of the elbow.

- Microstructure of the elbow (look for precipitation of secondary phases).
- Microhardness traverses across the wall of the elbow.
- Phase balance (austenite/ferrite) measurements.
- Evidence of the cracking mechanism (SCC?).
14
S3. Corrosion testing should be done to determine the effect of heat treatment
and welding on the corrosion of type 2304 duplex stainless steel in
superconcentrator black liquor. Angela Wensley can do this work.
S4. In 2011, the tubes and welds in the lower part of the superconcentrator
should be inspected for defects and corrosion.
S5. In 2011, the exterior surfaces of the steam inlets (type 2304 duplex stainless
steel) in the top of the superconcentrator should be inspected for SCC using
PT. The vapour inlets should also be visually inspected for corrosion
thinning.
For the Concentrator:
C1. The shell of the concentrator should be replaced using type 2304 duplex
stainless steel.
C2. The type 304L stainless steel tubes currently in body C should be replaced
using type 2304 duplex stainless steel.
C3. QA guidelines developed by Angela Wensley Engineering for construction

of duplex stainless steel pressure vessels should be followed if the shell and
tubes of the concentrator are replaced.
C4. Samples of liquor from bodies A, B, C, and D should be analyzed for

corrosive compounds (including NaOH, Na2S, Na2S2O3, Na2SO4, Na2CO3,
NaCl, and % solids). The best company to do this is Econotech in Canada.
15
C5. Corrosion testing of austenitic and duplex stainless steels should be done in
samples of liquor from bodies A, B, C, and D to determine their
susceptibility to SCC and intergranular attack. Angela Wensley can do this
work.
C6. A sample of a cracked tube from the concentrator body A should be

metallurgically examined to determine whether it was sensitized (due to
improper heat treatment). Angela Wensley can do this work.
For the Evaporators:
E1. Selected weld tees in evaporator effect E-2 should be closely inspected for
SCC using PT. Flap discs should be used for preparation of the surface for
±25 mm on either side of the welds.
E2. The junctions between top heads and upper stainless steel rings in
evaporator effects E-3 and E-7 should be inspected for galvanic corrosion; if
necessary, the galvanic corrosion should be repaired using type 309L
stainless steel welding consumables.
E3. The China hat and the roof above the liquor inlet in the bottom of evaporator
effect E-4 should be inspected for preferential weld corrosion, fatigue
cracking, or SCC that may have contributed to the failure of the ceiling in
evaporator effect E-5.
E4. A sample for the failed divider from E-5 should be examined metallurgically
for caustic SCC in the fracture area. Angela Wensley can do this.
16
E5. The bottom of evaporator effect E-5 should be scaffolded so that a full
corrosion inspection can be made. The carbon steel wall, head, and internal
structures in the liquor zone may be expected to be in similar condition to
those in evaporator effect E-6 in 2010 (galvanic corrosion, preferential
corrosion of welds and heat affected zones).
E6. The carbon steel structural internals in the liquor zone of evaporator effect E-
6 should be replaced using type 304L stainless steel. These include the liquor
inlet, China hat, divider and ceiling around the liquor inlet, wall and ceiling
of the mist eliminator, supports for the mist eliminator, and condensate
drains from the mist eliminator.
E7. The bottom of evaporator effect E-7 should be scaffolded so that a full
corrosion inspection can be made. The carbon steel wall, head, and internal
structures in the liquor zone may be expected to be in worse condition than
those in evaporator effect E-6 in 2010 (galvanic corrosion, preferential
corrosion of welds and heat affected zones).
E8. The lower shell and bottom head of effects E-5, E-6, and E-7 must be
protected against corrosion thinning. Priority should be given to effect E-6.
Alternatives are:
1. Complete replacement of the bottom using type 304L stainless steel.

2. Stainless steel lining using type 304L stainless steel.
3 Thermal spray coating (metallizing) using nickel.
4. Anodic protection.
5. Weld overlay.
17
Since anodic protection is effective only in immersion conditions (up to 200

mm above the circumferential wed joining the bottom head to the cylindrical
shell) another protective measure such as stainless steel lining must be used
for the splash-vapour zone in the cylindrical shell above.
E9. Since application of weld overlay on substrates below 17 mm thick may

cause unacceptable deformation, prospective weld overlay contractors
should be required to demonstrate that they can apply overlay on a 15 mm
thick carbon steel substrate without significant distortion.
For the Digesters and Impbins:
D1. The external "half-pipe" headers around the bottom of both digesters should
be monitored for leaks.
D2. The central pipe in the Line 1 (pine) impbin should be inspected to see if the
corrosion had continued. If corrosive wood acids are trapped between the
central pipes, perhaps a vent hole can be made.
D3. The arms and paddles of the bottom rakes in the digesters and the impbins
should be 100% inspected using PT.
D4. The external plates on the China hat in the Line 1 (pine) digester should be
replaced using type 2205 duplex stainless steel.
D5. The remaining type 304L stainless steel cover plates for beam nozzles should
be replaced using SA516-Grade 70 carbon steel.
18
D6. The inserts in the steam injectors should be removed for corrosion
inspection.
D7. Selected white areas on the wall in the upper parts of both digesters should
be selected for thickness monitoring.
D8. Type 312 stainless steel should be used for repairs to the screen welds and to
the China hats.
D9. The thickness of the bottom heads of both impbins should be monitored for
erosion thinning.
For the Bleach Plant:
B1. Locations of pitting corrosion and crevice corrosion of the 6% Mo stainless

steel parts of the D-stage presses should be cleaned, polished, and pickled.
B2. Corroded plug welds in the D-stage presses should be repaired using an
alternative weld filler metal with better corrosion resistance than
ERNiCrMo-3 (Alloy 625). This alternative weld filler metal should be used
for all welding 6% Mo stainless steels in the D-stage presses and piping.
This requires assessment by corrosion testing in D-stage filtrates (Angela
Wensley can do this). Possible acceptable weld filler materials are Alloy G3,
Alloy G30, and Avesta P54. Nickel base fillers such as Alloy 625, Alloy
C276, Alloy C22, and Alloy 59 must be avoided in D-stage service.
B3. All type 2205 duplex stainless steel piping associated with the presses and
standpipes in the pine side should be replaced. Acceptable replacement
materials are titanium (Grade 2), FRP with an epoxy vinyl ester resin such as
19
Derakane 470-300 or equivalent, and 6% Mo stainless steel welded using a

filler metal resistant to transpassive corrosion (see recommendation B-2
above).
B4. A section of 6% Mo stainless steel pipe with a circumferential weld should be

cut out so that corrosion of the root side of the weld can be examined for
corrosion.
B5. Locations of mechanical damage to the rolls in the presses should be

monitored for the development of fatigue cracking.
For the White Water Tanks:
W1. All "carbuncles" should be removed and the locations cleaned, polished, and
passivated. If necessary, weld repairs should be made using an over-
matched filler metal such as 309MoL, 317L, or 904L.
W2. Temporary patches to locations of leaks may be made using overmatched

plate and welding filler metals such as type 317L stainless steel. This will
buy some time before new leaks develop.
W3. The use of biocides and/or boil-outs to prevent MIC in type 316L stainless
steel white water tanks should be investigated.
For the Turbogenerator Expansion Tanks:
T1. Consideration should be given to replacing the carbon steel expansion tanks
with new tanks of type 2205 duplex stainless steel. For duplex stainless steel
20
in hot water service, it is important that the chloride content of the water not
exceed approximately 10 ppm.
General:
G1. Celulosa Arauco y Constitución should support QA and corrosion testing to

ensure that the proper materials are used for critical equipment. The
equipment suppliers do not provide these services. Angela Wensley can
provide these services. There is a potential to save millions of dollars each
year that is currently being lost due to repairs or replacement of corroded
equipment that was not subjected to rigourous QA or proper material
selection.
G2. Celulosa Arauco y Constitución should purchase a portable "alloy analyzer"

for materials identification during general shutdowns and for QA during
manufacture of critical equipment.
21
1.0 INTRODUCTION
I was at the Nueva Aldea mill from 24 March through 1 April 2010 during
the annual general plant shutdown. The 2010 shutdown was
extraordinary as it was in the aftermath of a major earthquake. I
previously attended the general plant shutdowns in November 2007 [Ref.
1] and January 2009 [Ref. 2]. The mill started up in 2006.
I was based in the contractor's room on the second floor of the main office
building. I made corrosion inspections of equipment in the evaporators,
digesters, impbins, and the bleach plant presses. I also made cursory
examinations of white water tanks for the Line 1 (pine) pulp machine and
the expansion tanks for the turbogenerator. My exit meeting was on 1
April 2010.
Paul Praszkier and Sergio Stahl of Safe Control inspected the recovery
boiler fireside. Emory Hull of Sheppard T. Powell Associates made
waterside inspections. Paulo Ormeneze of Metso supervised work in the
digesters and impbins. Marcelo Silveira and Fabrezio Contin of Metso
supervised work in the superconcentrator and concentrator. Alejandro
Roa of BASF inspected FRP equipment. Several representatives of GL&V
including Micael Bertilsson and Alessandro Costa Barbosa were on site for
modifications to the bleach presses.
22
2.0 EVAPORATORS
Nueva Aldea has a Kvaerner (Metso) seven-effect falling film evaporator

train with a superconcentrator. The liquor and vapour flows are:
WEAK STRONG
COOL
BLACK BLACK
VAPOUR
LIQUOR LIQUOR
54C
19% 77%
28% 24% 21% 39% 45% 73%
7 6 5 4 3 2 1 S1
65C 74C 83C 91C 102C 114C 142C
34%
HOT
175C
STEAM
2.1 Superconcentrator S-1 TAG 551-51-0101
The cylindrical shell and internals of the

superconcentrator were constructed using type
2304 duplex stainless steel (green colour in the
drawing at left); the top and bottom heads (pink
colour) were type 2205 duplex stainless steel. The
superconcentrator has two bodies (A and B)
within one vessel.
E-S1 A/B
Figure 1 shows a vapour inlet below a liquor

recirculation inlet at the top of the
superconcentrator. Figure 2 shows superficial
corrosion of the duplex stainless steel vapour inlet
was observed.
23
Hydro-testing revealed several leaks in both the A and B bodies (Figure 3).
Leaks were found in two locations:
1. In the thin-wall (reportedly 0.9 mm) Tubel® tubes above welds to the
small diameter headers. These locations could not be repaired, so
entire platens were taken out of service by cutting and plugging
(Figure 4). Figures 5 and 6 are views of a sample removed with two
perforated Tubel® tubes and a section of the small diameter header.
There was no welding defect associated with the leaks that evidently
resulted from corrosion of the heat affected zone above the welds.
2. In the lower welds for the short vertical pipes between the small
diameter headers and the large diameter condensate headers. Leaks
occurred at original weld defects such as pores (Figure 7). In at least
one case, the leak washed an adjacent drain tube almost to the point of
perforation (Figure 8) necessitating plugging. Most of the weld defects
were successfully repaired using type 2209 duplex stainless steel filler
metal (Figure 9).
The filler metal originally used for the tube-to-header welds is not known.
I was told it could have been either "2209" or "2304." The ISO designations
for these filler metals are 23 7 NL and 22 9 NL, respectively. There may be
a large difference in corrosion resistance between these two materials.
Since there was no portable alloy analyzer on site, it was not possible to
ascertain which filler metal was used.
I used a Fischer Feritscope Model MP-30 to measure the ferrite contents of

the tubes and welds in the superconcentrator. The ferrite content range
24
expected for duplex stainless steel tubes is 35% to 55%, while the ferrite
range expected for duplex stainless steel welds is 30% to 60%. The tubes
and headers had ferrite contents consistent with 2304 duplex stainless
steel; however, the ranges measured for the welds were not within the
expected 30% to 60% range. Figure 10 shows some of the weld ferrite
measurements written on the adjacent tubes and headers. The welds
between the short vertical tubes and the large-diameter header had 19% to
38% ferrite. The welds between the Tubel® tubes and the small-diameter
headers had 52% to 69% ferrite. This wide variation in ferrite contents
suggests that there was a QA problem with materials selection (What
welding consumables were used? Were the tubes properly heat treated?)
and welding qualification during original design and fabrication of the
superconcentrator. Welds having low and high ferrite contents will have
greater susceptibility to corrosion and SCC in service.
One of the elbows in the condensate drain below the large-diameter

headers was cracked and had to be removed (Figures 11 and 12).
Inspection of the elbow with the Feritscope revealed it to have a very low
ferrite content of 13%. This value is well below the 35% to 55% expected
for a tube of type 2304 duplex stainless steel and is more characteristic of
an austenitic stainless steel (like 304L) that is more susceptible to SCC.
Inspection of other parts of the condensate drain system, including all of
the other elbows (including the replacement elbow) revealed normal
ferrite contents (Figures 13 and 14).
My hypothesis is that the elbow was not hot formed at the correct
temperature (900º to 1150ºC) or was not subsequently annealed at the
correct temperature (950º to 1050ºC).
25
2.2 Evaporator Effect E-1 TAG 551-51-0102
The concentrator has four bodies (A, B, C, and D)

within a single effect. The shell of the concentrator
was constructed entirely of type 304L austenitic
stainless steel, including the top and bottom heads
and all of the cylindrical rings (blue colour).
Originally, all the tubes were 304L stainless steel,
with duplex stainless steel being used for only the
E-1 vapour inlets at the top and large-diameter
A, B, C, D
headers and drain tubes in the bottom.
In 2009, SCC was found in the 304L tubes in

bodies A, B, and C. SCC was also found
associated with the circumferential and vertical
welds in the 304L shell.
2.2.1 Old Tube Bundles
During the 2010 shutdown, the top head was removed (Figure 15) so that
the tube bundles could be changed. The 304L tube bundles from bodies
A, B, and C were removed and scrapped. The 304L tube bundle for body
D was moved to body C where it will now be at risk of SCC. New 2304
tube bundles were installed in bodies A, B, and D.
Figure 16 shows the 304L tube bundle removed from body A. Visual
inspection confirmed that the tubes had extensive SCC. Many tubes had
SCC in a spiral pattern (Figures 17 and 18) produced by the tube
straightening operation after the annealing process. The straightening
26
operation introduced significant tensile stresses into the tubes that in turn
caused SCC in service. Cracking along the weld seam (Figure 19) is
evidence that the tubes were not properly annealed during their
manufacture. Also apparent in the 304L tube bundle removed from body
A was SCC of structural parts (Figure 20). In this case the tensile stresses
to support SCC were residual stresses from welding.
Figure 21 is a view inside the old large-diameter header showing lack of

fusion and slag pockets in the welds for the vertical drain tubes. This
same situation (that should have been found by QA) may exist with the
new tube bundles.
Visual inspection revealed the 304L tubes from body B also had SCC. I did
not have an opportunity to inspect the tubes from bodies C and D.
2.2.2 New Tube Bundles
Figure 22 shows the new type 2304 duplex stainless steel tube bundle for
body A. The Tubel® tubes (Figure 23) had the stenciled identification
"IRSA EN 10217-7 TC1 ISO D4/T3 0 32 x 1.2 1.4362 HEAT 490708 OF
511815 (48)". I have never heard of IRSA as a supplier of duplex stainless
steel tubes and they have never been approved by me. A very important
part of any QA program is qualification of the product. In this case I
would have insisted on metallurgical confirmation of the composition,
microstructure, phase balance, and heat treatment.
I used the Feritscope to measure the ferrite contents of some of the tubes
and welds (Figure 24). The tubes had ferrite contents in the range of 44%
to 51%, appropriate for type 2304 duplex stainless steel. The welds
27
between the tubes and the small-diameter headers, however, had 62% to
80% ferrite, well above the expected range of 30% to 60%. This suggests
that independent QA was not carried out during the design and
fabrication of the new tube bundles.
The short tubes between the small- and large-diameter headers were
made by Sandvik, a well-known producer of high-quality duplex stainless
steel. These tubes had very uniform ferrite content of 51% to 54% (Figure
25).
2.2.3 Concentrator Shell
Figure 26 shows body A after the old 304L tube bundle was removed. The
curved part in Figure 26 is the shell of the concentrator, while the straight
sides are the partitions, or dividers (non-pressure parts) between the A
and C bodies and between the A and D bodies. The cylindrical shell
contained 11 rings, numbered from bottom to top. The circumferential
welds were also numbered 1 through 12 from bottom to top. In ring 1,
there was an extra circumferential weld that I named 1A. For
convenience, the circumferential welds were further identified by their
body. Circumferential weld A1 was that portion (90º) of weld 1 that was
in body A.
Corrosion of the 304L shell at the bottom of body A gave it a bright,

almost mirror-like appearance (Figure 27). Original flat welds were now
slightly prominent by approximately 0.5 mm, indicating that the welds
were more resistant to corrosion (Figure 28). At higher elevations in body
A there was still evidence of corrosion of the shell (Figure 29). The most
28
plausible corrosion mechanism is intergranular attack. The shell in the

other bodies did not exhibit comparably significant corrosion.
In the A, B, and C bodies, visual inspection of the dividers revealed SCC

associated with welds (Figures 30 through 32). The most spectacular SCC
was associated with grinding removal of temporary welds from
attachments used during construction and then removed. No visual SCC
was observed on the sides of the dividers exposed in body D.
Originally, it was hoped that all SCC could be removed from the shell of
the concentrator. The large number of remnant welds with SCC, however,
made complete elimination impractical. It was decided instead to focus
on removal of SCC from the welds and adjoining shell for at least ±25 mm
on either side. The reasoning for this is that there is no driving force for
SCC adjacent to a remnant weld to propagate across the thickness of the
wall, whereas a driving force for through-wall leaking existed at the
circumferential and vertical welds.
The original procedure for finding and removing the SCC was:
1. Visual inspection (as in Figures 33 through 37).

2. Power wire brushing of the welds ± 25 mm.
3. PT (Figure 38).
4. Removal of SCC using 60- or 80-grit flap discs.
5. Polishing using 120-grit flap discs (Figures 39 and 40).
When it became apparent that the cracking in the shell beside the
circumferential and vertical welds was almost continuous (see Figure 38)
29
and that most of the cracking could be removed within a depth of 1 mm,
the procedure was modified to flap disc polishing to a depth of 1 mm on
either side of the major welds. There was some confusion as to whether
the welds were also to be polished, despite the clear evidence that the
welds, at least in body A, had SCC. I left the site before the removal of
SCC from the shell was completed. Also, the bottom head had not yet
been inspected.
Removal of the cracks should prolong the life of the concentrator shell.
Although it is hoped that the SCC will not return, it is probable that new
cracks will re-initiate. I recommend that the shell of the concentrator be
replaced using type 2304 duplex stainless steel. The new shell should
have meaningful QA, following the guidelines for duplex stainless steel
pressure vessel construction previously provided to Nueva Aldea by
Angela Wensley Engineering.
2.2.4 Concentrator Top Head
The top head was cut off 150 mm below circumferential weld 12. It was
moved to a location where it could be scaffolded internally for inspection
(Figure 41). The top head had three sections joined with circumferential
welds 12, 13, and 14. The sections in turn had segments separated by
radial welds. PT revealed extensive SCC associated with circumferential
weld 12 (Figure 42), remnant welds on the head (Figures 43 and 44)
circumferential welds 12 and 13, and the radial welds (Figures 45 and 46).
I left the site before the removal of SCC from the top head was completed.
After re-welding the top head to the shell there will be a new
circumferential weld called 11A.
30
The body of the second effect evaporator E-2 was

constructed using a mixture of SA516-Grade 70
carbon steel (yellow) in the vapour zone and type
304L stainless steel (blue) in the upper liquor zone,
the lower liquor zone, and the top of the steam
chest in the vapour zone.
E-2 In August 2007 there was a leak at the entrance of

the recirculation line at the top of E-2 due to
galvanic corrosion. The inlet of the recirculation
line was replaced using type 304L stainless steel.
In November 2007, rapid corrosion thinning of the
carbon steel top cone in the vapour zone above the
top stainless steel ring was observed.
In 2009, the carbon steel top head was replaced with a new head of type
304L stainless steel. In 2010, hydroblasting limited access to E-2 to
viewing from the upper and lower manways. A view of the new top head
through the upper manway revealed no visible problems (Figure 47).
Figure 48 is a view of the bottom head from the bottom manway where
there were no visual problems. Since all "tube side" locations (exposed to
liquor) are type 304L stainless steel, no corrosion or SCC are expected. To
confirm the absence of SCC, PT at selected weld tees should be done in
2011.
31
The body of the third effect evaporator E-3 was

constructed using a mixture of SA516-Grade 70
carbon steel (yellow) for the top head and in the
vapour zone, and type 304L stainless steel (blue)
in the bottom of the upper liquor zone, the lower
E-3
liquor zone, and the upper part of the steam chest
in the vapour zone.
Inspection of the top head in 2009 revealed slight

galvanic corrosion of the carbon steel top head
above the upper stainless steel ring in the upper
liquor zone. In 2010, hydroblasting limited access
to E-3 to viewing from the bottom manway.
Figure 49 is a view of the upper part of the mist eliminator in the bottom
liquor zone. All parts were type 304L stainless steel. There were no visual
problems.
Figure 50 shows the China hat over the liquor inlet in the bottom of E-3.
China hats of type 304L stainless steel construction often have problems
with fatigue cracking. Access for closer inspection is required to
determine whether any fatigue cracks have developed.
32
The body of the fourth effect evaporator E-4 was

constructed predominantly using SA516-Grade 70
carbon steel (yellow) with type 304L stainless steel
(blue) rings in the bottom of the upper liquor zone
and at the top and bottom of the steam chest in the
E-4 vapour zone. Evaporator effect E-4 sees the
weakest black liquor of any of the evaporators, so
is expected to have few corrosion problems with
the carbon steel components.
There was no galvanic corrosion of the carbon

steel top head above the upper cylindrical
stainless steel ring in the vapour zone (Figures 51
and 52). Figure 53 is a view of the carbon steel
bottom head.
Figure 54 is a view of the bottom of the mist eliminator in E-4. The

supports for the mist eliminator are stainless steel. The use of stainless
steel for the mist eliminator supports in the effect with the weakest liquor
is strange, especially considering that the supports in E-5, E-6, and E-7 are
carbon steel and experiencing corrosion.
Figures 55 and 56 are views of the carbon steel divider around the liquor
inlet. The thickness appeared to be the original nominal thickness of 10
mm. Since the roof of the comparable divider in E-5 has failed, it is
important to verify the condition of the divider and the China hat in E-4 in
2011.
33
The body of the fifth effect evaporator E-5 was

and at the top and bottom of the steam chest in the
vapour zone.
E-5
steel top head in the vapour zone (Figures 57 and
58). The roof of the divider above the liquor inlet
had failed (Figures 59 and 60). The China hat had
failed and may have impacted the roof of the
divider (Figure 61).
The failure of the China hat was evidently due to fatigue. Selective
corrosion of high-silicon content welds may have contributed to the
failure of the divider (Figure 62). The fractures in the divider were brittle
in appearance with little deformation as would be expected for such a
ductile material. A sample of the divider should be examined
metallurgically for caustic SCC.
Figure 63 shows the carbon steel bottom head of E-5. I did not go inside
because a rung for the ladder had failed due to corrosion, making entry
unsafe (Figure 64). In 2011, the bottom of E-5 should be scaffolded so a
complete corrosion inspection can be done.
34
The body of the sixth effect evaporator E-6 was

and in the bottom and top of the steam chest in the
E-6 vapour zone.

steel above the upper stainless steel ring in the
upper liquor zone (Figures 65 and 66).
The bottom liquor zone of E-6 was scaffolded in

2010 so that a thorough corrosion inspection could
be done.
Figures 67 and 68 are views of the bottom liquor zone before the
scaffolding was installed. There was severe corrosion of the carbon steel
in the bottom liquor zone below the lower tubesheet. Figure 69 shows the
China hat and the top of the divider around the liquor inlet. The divider
in E-6, which did not have a roof as did the dividers in E-4 and E-5, was
severely thinned, to as low as 3 mm (Figure 70). If the original thickness
of the divider was 10 mm (as was measured in E-4) this means there has
been a loss of 7 mm in less than 4 years due to corrosion from both sides,
unlike the shell that sees liquor on only one side. The stiffeners on the
outside of the divider were similarly thinned and many of the welds had
failed. It is unlikely that the divider would survive for 1 year more.
35
Above the liquor level in the bottom liquor zone of E-6 there was
preferential corrosion of high-silicon welds resulting in the complete loss
of some welds (Figures 71 and 72). The carbon steel roof of the mist
eliminator was perforated due to corrosion (Figures 73 and 74). The
carbon steel supports for the mist eliminator were corroded (Figure 75).
(Recall that the supports in E-4 were stainless steel.)
There was also significant galvanic corrosion of carbon steel adjacent to

stainless steel (Figures 76 through 78), most dramatically observed with
the carbon steel bottom sections of the drains from the mist eliminator.
(What was Kvaerner thinking to change the metallurgy to carbon steel for
immersion service?). At present the galvanic corrosion of the shell is no
more than 2 mm deep, and does not require repair.
After I had left the mill I received a report that corrosion of the weld heat
affected zones had been found in the bottom head.
I recommended repairs to the corroded carbon steel internals of E-6. In

2011, the carbon steel internals should be replaced using type 304L
stainless steel. The shell should be protected from corrosion. Protective
measures include: anodic protection, stainless steel lining, thermal spray
coating with nickel, or weld overlay. The alternatives were discussed in
my 2009 report [Ref. 2]. Overlay on a substrate less than 17 mm thick may
cause unacceptable distortion of the shell.
My recommendation is to install an anodic protection system for the

immersion zone (up to 200 mm above the bottom head to shell weld) with
a stainless steel lining above, all the way up the stainless steel conical ring
below the lower tubesheet.
36
The body of the seventh effect evaporator E-7 was

and in the top and bottom of the steam chest in the
vapour zone.
E-7
Due to excessive heat, I was only able to inspect
the top through the upper manway in 2010
(Figures 79 and 80). In 2009, slight galvanic
corrosion of the carbon steel was observed above
the upper stainless steel ring in the upper liquor
zone.
In 2009, the lower liquor zone of E-7 had the most severe corrosion of
carbon steel of any evaporator at Nueva mill. Scaffolding was installed in
2010 but I had left the mill before it was completed. Figures 81 through 83
are views through the lower manway. The carbon steel shell and internals
of E-7 were likely in a condition at least comparable to those in E-6.
Figure 84 shows galvanic corrosion of a drain pipe for the mist eliminator
where it transitioned from stainless steel to carbon steel.
2.9 Surface Condenser TAG 551-51-0109
Figure 85 is a view through the upper manway. The carbon steel wall was
covered in "carbuncles" that appeared to be from oxygen corrosion. The
carbon steel underneath was pitted (Figure 86).
37
3.0 DIGESTERS AND IMPBINS
The shells of the two continuous digesters were constructed using type
2205 duplex stainless steel. The major work done in the digesters in 2010
was the repair of the external "half-pipe" headers providing dilution to the
bottom heads of both digesters. Also, the China hat in the Line 2
(eucalyptus) digester was modified.
The impbins were both constructed using solid type 304L stainless steel.
They operate at atmospheric pressure (plus static head) and at a lower
temperature than the continuous digesters. As with all solid type 304L
stainless steel vessels, the impbins are at risk for external SCC beneath wet
thermal insulation. The bottom scrapers in both impbins are solid type
2205 duplex stainless steel for resistance to the higher wear expected in
the bottoms of the impbins. In 2011, the central pipe in the Line 2
(eucalyptus) impbin was modified (shortened).
3.1 Repairs to the External "Half-Pipe" Headers
Figure 87 shows the location of the type 304L stainless steel "half-pipe"
header around the bottom of the type 2205 duplex stainless steel Line 1
(pine) digester. After the 2009 shutdown, leaks developed in the upper
welds of the half pipes in both digesters. The temporary repair in both
cases was to construct a box around the leaks (Figure 88). During the 2010
shutdown the boxes were removed. Figures 89 and 90 show the leak in
the half-pipe on the Line 1 (pine) digester; Figures 91 through 93 are views
of the leak in the half-pipe on the Line 2 (eucalyptus) digester. It was clear
that the leaks occurred by the process of fatigue cracking.
38
Inspection through opened nozzles on the outside of the half-pipes

revealed that neither the top weld nor the bottom weld (both made only
from the outside) had full penetration. The incomplete penetration or lack
of fusion caused fatigue cracks to initiate from the roots of the welds. The
crack grew under the influence of pressure variations in the header.
It was decided to cut windows in the half-pipes (Figure 94) to give access
to the root sides of the upper and lower welds (Figures 95 and 96). The
insides of the half-pipe were slightly beveled and new weld metal was
added on the roots (Figures 97 and 98). Metso used type 309L stainless
steel for the new welds. My Feritscope revealed that the original welds
between the half-pipes and the digester had approximately 44% ferrite -
characteristic of duplex stainless steel, probably type 2209 filler metal.
Type 309L is acceptable for joining the type 304L stainless steel half-pipes
to the type 2205 duplex stainless steel digester wall. Type 2209 duplex
stainless steel welding consumables are preferred since 2209 is harder and
more resistant to cracking than 309L.
After repairs in a window were completed, the window was replaced

using argon gas purging inside the half-pipe (Figure 99). I left the mill
before the half-pipe repairs were completed. I presume that the welding
filler metal used for the windows was type 308L stainless steel,
appropriate for welding type 304L stainless steel.
3.2 Line 1 (Pine) Digester TAG 541-51-1101
In 2008, there were two failures of the bottom scraper in the Line 1 (pine)
digester. The origins of the failures were small cracks in martensitic
carbon steel that formed when the stainless steel filtrate pipes were
39
welded onto the trailing edges of the scraper arms without preheat. The
small cracks grew across the arms by a fatigue mechanism. A new bottom
scraper, reportedly built using proper preheat, was installed in the Line 1
(pine) digester during the shutdown in 2009. No problems were found
visually in 2010 (Figures 100 through 102).
Welds on the type 304L stainless steel China hat experienced erosion that
exposed the carbons steel supports at the circumferential supports (Figure
103). Carbon steel was also exposed around the bottom (Figure 104).
Ultrasonic thickness testing (UT) revealed loss of thickness of the type
304L stainless steel plates. The eroded welds were weld repaired using
type 309MoL stainless steel that is too soft to resist erosion. I had
recommended using type 312 stainless steel but none was available. The
China hat should be re-plated in 2011 using type 2205 duplex stainless
steel, welded using 2209 duplex stainless steel consumables.
There were numerous cracked stitch welds in the screens (Figures 105 and
106). These welds were probably originally made using type 308L
stainless steel which has low-strength and is prone to hot cracking. I
recommended using type 312 stainless steel welding consumables for the
stitch welds, but none was available. Type 312 stainless steel is much
stronger and more resistant to cracking than either 308L or 309L.
There are two vapour injectors in the top of the Line 1 (pine) digester. The
nozzles for the injectors are type 2205 duplex stainless steel. The external
welds were inspected using PT (Figure 107). There are replaceable inserts
in the nozzles (Figure 108) that I believe are also type 2205 duplex
stainless steel. In other digesters the inserts experience corrosion in the
40
annular space between the nozzles and the inserts. In 2011, the inserts
should be removed from the injectors to see if any corrosion has occurred.
In the top of the digester (at the elevation of the top separator) there was a
white colour on the wall that was clearly related to flow of vapour from
the injectors (Figure 109). There may be some thinning of the wall in the
white areas. Monitoring of the thickness on a reproducible ultrasonic
testing (U)T grid is recommended. There was no visual corrosion
problem with the top separator screw (Figure 110).
3.3 Line 2 (Eucalyptus) Digester TAG 541-51-2101
No visual problems were found with the bottom scraper (Figures 111 and
112). Since the bottom scraper is similar to the one that failed in the Line 1
(pine) digester, it likely contains small cracks in martensitic areas created
by the lack of preheat during welding of the dilution pipes. The existence
of microcracks was confirmed by a metallurgical sample taken in 2009. If
a severe stress cycle were to occur such as the dropping of a stuck plug,
any existing microcracks could grow by a fatigue mechanism. The Line 2
(eucalyptus) digester scraper should be inspected carefully for cracking
every year using PT. The China hat did not exhibit the high amount of
wear as the China hat in the Line 2 (eucalyptus) digester. A modification
was made in 2010 involving welding a pipe onto the top of the China hat
(Figure 113).
The new row of screens added in 2009 was intact (Figure 114). Any
cracked stitch welds should be repaired using type 312 stainless steel filler
metal.
41
In 2009, many of the type 304L stainless steel cover plates for the beam
nozzles on both digesters were found to have SCC on their external
surfaces. Most of the covers were replaced using carbon steel (Figure 115)
with a lining of duplex stainless steel on the inside (Figure 116). Although
the carbon steel covers were supposed to be a temporary measure until
duplex stainless steel covers could be made, the good condition of the
carbon steel covers in 2010 suggests that they will give good service for
many years. The priority should be replacement of the remaining type
304L stainless steel covers instead of the carbon steel covers.
In the top of the Line 2 (eucalyptus) digester, some white areas were
observed (Figures 117 and 118). A UT monitoring grid should be
established so the wall thickness can be monitored. The steam injectors
appeared intact (Figures 119 and 120). The inserts should be removed in
2011 so they can be inspected for corrosion.
3.4 Line 1 (Pine) Impbin TAG 541-51-1010
Figure 121 is a view of the bottom scraper in the Line 1 (pine) impbin.
Although no visual problems were observed, the scrapers should be PT
inspected annually for cracking. The PT should include 100% of the
accessible surfaces of the arms and paddles.
In 2010, the central pipe was modified (shortened). This involved cutting
back both the inner and outer pipes (Figure 122). During this work, severe
corrosion of the outer surface of the inner pipe was observed (Figures 123
and 124). The space between the inner and outer pipes is not used,
allowing corrosive wood acids to collect in this area. One possible way to
stop the corrosion would be to cut a vent hole in the outer pipe
42
3.5 Line 2 (Eucalyptus) Impbin TAG 541-51-2010
Figures 125 and 126 are views of the duplex stainless steel bottom scraper
in the Line 2 (eucalyptus) impbin.
The bottom head had many locations of severe mechanical deformation

from attempts to make the bottom head fit to the shell; these attempts also
required installation of a narrow ring in the bottom head (Figure 127).
Figure 128 shows the central pipe that was not modified.
43
4.0 BLEACH PRESSES
There are seven Aker Kvaerner (now GL&V) Compact Presses in the Line
1 (pine) bleach plant and seven presses in the Line 2 (eucalyptus) bleach
plant. The presses are of twin-roll design:
PULP
OUT
DISTRIBUTION SCREW
PULP IN
SHREDDER
FLAP
ROLL ROLL DEWATERING
ZONE
WASH
LIQUOR
IN FILTRATE OUT
Both the pine and eucalyptus lines have three pre-bleach presses, followed
by a D0-stage (press 4), an EOP-stage (press 5), a D1-stage (press 6), and a
D2-stage (press 7). The principal material of construction for presses 1, 2,
3, and 5 is type 316L stainless steel. The principal material of construction
for the D-stage presses 4, 6, and 7 is 6% Mo stainless steel. Most of the
piping is either FRP or 6% Mo stainless steel, but some piping is type 2205
duplex stainless steel.
In 2009, corrosion problems were found in the D-stage presses and piping
where significant residual ClO2 existed. The Line 1 (pine) D-stage presses
44
were more corroded than there Line 2 (eucalyptus) counterparts. The

most severe corrosion was found in type 2205 duplex stainless steel piping
carrying Line 1 (pine) D-stage filtrate.
In 2010, there was much earthquake damage to the foundations of the

Line 2 (eucalyptus) presses, in particular press 1. There was also
considerable damage to reinforced concrete structural beams and FRP
piping. GL&V personnel were on site to supervise modifications to the
flaps on the Line 2 (eucalyptus) D-stage presses. The Line 2 (eucalyptus)
D0-stage press 4 was disassembled.
4.1 Mechanical Damage of Rolls
Most, if not all, of the rolls had mechanical damage. Since no mechanical
damage was noted in 2009, it may have occurred during the earthquake.
Most of the mechanical damage produced deformation of the roll surfaces
(Figures 129 through 133), but at least one example of rip was observed
(Figure 134). In service, a rip may grow as a fatigue crack. The dents need
to be monitored to see whether they develop into cracks.
In the absence of a drawing, I cannot speculate as to the success of

repairing by welding. In suction press roll shells on paper machines (that
operate at higher velocities and stress loading than the rolls in twin-roll
presses) welding is forbidden as the residual welding stresses promote
cracking.
45
4.2 Corrosion of Welds
Corrosion of plug welds was apparent in the 6% stainless steel lining in

the vats in all the D-stage presses on the pine side (Figures 135 through
137). Loss of these welds means that filtrate can gain access to the carbon
steel support underneath. Carbon steel has no resistance to corrosion in
D-stage filtrates. Corrosion of weld seams was also observed on the 6%
Mo stainless steel rolls in D-stage service (Figure 138).
The welds are corroding because the nickel-base welding filler metal
ERNiCrMo-3 (Alloy 625) has poor resistance to a phenomenon called
transpassive corrosion in D-stage filtrates. Unlike pitting, transpassive
corrosion causes the entire surface of the material to corrode uniformly.
Beginning in 1989, I published a number of papers on the poor corrosion
resistance of nickel-base alloys in chlorine dioxide bleaching [Refs. 3 to 5].
This information is well known throughout the pulp and paper industry
and most companies forbid the use of nickel-base alloys in their bleach
plants.
Until very recently, there was no matching filler metal for 6% Mo stainless
steel. Avesta (now Outokumpu) developed filler metal P54 for their alloy
254 SMO (a 6% stainless steel) but it is not recommended for welding
thicknesses greater than 6 mm due to its tendency for hot cracking. A
compromise has been to use either Alloy G30 or Alloy G3 for welding 6%
stainless steel. Alloys G30 and G3 contain only 5% Mo and are slightly
more susceptible to pitting corrosion than 6% Mo stainless steels – but are
not susceptible to transpassive corrosion. In the absence of corrosion
testing, I would recommend Alloy G30.
46
4.3 Modification of the Flaps
The flaps from the Line 1 (pine) D-stage presses were removed so they
could be modified. Some of the flaps showed a discolouration (Figures
139 and 140) whose cause was unknown. To me, the colours resembled
those caused by frictional overheating.
The modification involved welding strips of 6% Mo stainless steel onto the

flaps (Figure 141) followed by installation of a complete lining of 6% Mo
stainless steel (Figure 142). Unfortunately, the welding filler metal used
was ERNiCrMo-3 (Alloy 625) that will experience transpassive corrosion
in service. For maximum corrosion resistance of welds and heat affected
zones, it is also very important to remove heat tint from welds that will be
in corrosive service. Removal of heat tint can be done by mechanical
treatment, pickling, or (preferably) a combination of both [Ref. 6]. I did
not have an opportunity to verify if this was done.
4.4 Corrosion of Flaps and Frames
While the curved surfaces of the D-stage flaps appeared to be polished

and free from pitting corrosion, the tops and sides of the flaps (that are in
stagnant areas) had deposits of corrosion product as well as pitting
corrosion (Figures 143 and 144). The frames for the flaps (also stagnant
areas) also had corrosion deposits and pitting corrosion (Figures 145 and
146).
There is little that can be done to prevent pitting corrosion of 6% Mo

stainless steel in stagnant areas. Regular cleaning and passivation will at
least inhibit pitting corrosion, if only for a short time. Currently, the
47
pitting corrosion is only a cosmetic problem as it does not affect the

operation of the presses. In the future (5 to 10 years), however, the
corrosion may become so bad as to require replacement of the major parts
of the presses.
All efforts to keep the ClO2 residuals as low as possible will increase the
service life of the D-stage presses.
4.5 Corrosion of Distribution Screw Covers
The bolted covers for the 6% Mo stainless steel distribution screw

housings create an ideal environment for crevice corrosion (a form of
pitting corrosion). Rust bleeding from the covers is a good indication that
crevice corrosion is occurring (Figures 147 and 148).
Some of the 6% Mo stainless steel covers were removed, confirming the

presence of crevice (pitting) corrosion (Figures 149 and 150). The same
recommendations as for the corroded flaps and frames, namely cleaning
and passivation, apply for the distribution screw covers.
4.6 Corrosion of Shredder/Conveyor Housings
The 6% Mo stainless steel housings for the shredder/conveyors are

located on top of the presses where the dewatered pulp is discharged into
a vertical standpipe. In the D-stage presses, the flange surface on top of
the housings have developed severe crevice (pitting) corrosion where they
are covered with flat plates of 6% Mo stainless steel (Figures 151 through
153). The same recommendations as for the corroded flaps, frames, and
48
distribution screw covers, namely cleaning and passivation, apply for the
shredder/conveyor housings.
The type 316L stainless steel housings for the shredder/conveyor screws
on the EOP presses had a blue colour (Figures 154 and 155) that was
evidently caused by ClO2 vapours escaping from the standpipes where D-
stage filtrate was used for dilution. Scratches in the Line 2 (eucalyptus)
EOP press had initiated corrosion (Figure 156).
4.7 Corrosion of Duplex Stainless Steel Piping
Duplex stainless steel piping for D-stage filtrate service in the Line 1 (pine)
side is experiencing corrosion to the point of through-wall leaking. The
corrosion is most apparent at heat exchanger 547-58-1400 (Figures 157 and
158). A horizontal line under the press floor has myriad leaks (Figures
159 and 160); attempts at repairs by patching have been unsuccessful
(patches create conditions for crevice corrosion that is easier to initiate
than pitting corrosion). Replacement of the duplex stainless steel piping
using titanium, FRP (such as Derakane 470-300 or equivalent), or 6% Mo
stainless steel (welded using Alloy G30 or Avesta P54) is recommended.
Corroded duplex stainless steel ring headers were also observed at

standpipes for Line 1 (pine) presses 5 and 6 (Figure 161 and 162). These
headers were repaired in 2009 and will soon have to be repaired again.
Replacement of the duplex stainless steel headers using titanium FRP, or
6% Mo stainless steel (welded using Alloy G30 or Avesta P54) is
recommended.
49
5.0 WHITE WATER TANKS FOR THE PAPER MACHINES
In 2009, severe MIC was found in three white water tanks for the Line 1
(pine) pulp machine. The tanks were presumably of type 316L stainless
steel construction. Two of the tanks were opened during my time at the
mill.
Pine side tank 572-22-1155 (Figure 163) had through-wall leaks in 2009. In
2010, several plates were replaced. Figure 164 shows one of the old plates
with external rust bleeds from through-wall leaks. If the plate was
replaced using the same material (type 316L stainless steel) and the same
welding filler metal (type 316L stainless steel) as the original tank, it will
develop new leaks very quickly. At the very least, a higher-alloy welding
consumable should have been used, such as type 309MoL stainless steel,
type 317L or 904L stainless steel. Use of higher alloy plate such as type
317L stainless steel or type 2205 duplex stainless steel would also give
improved resistance to MIC.
The "carbuncles" inside pine side tank 572-22-1156 had been removed
before I had an opportunity to inspect the tank. There were numerous
locations of pitting corrosion where nearby weld repairs (made in 2009
using (type 309MoL stainless steel?) were not corroded (Figure 165) and
other locations where the welds (previous repairs using type 316L
stainless steel?) were attacked (Figure 166).
Removing the "carbuncles" and polishing the surface will slow the
corrosion but MIC is liable to re-initiate. If possible, regular boil-outs and
biocide applications could kill the bacteria responsible for the MIC.
50
6.0 EXPANSION TANKS
In 2009, two expansion tanks for the turbogenerator had internal corrosion
characterized by "carbuncles". The corrosion was attributed to oxygen, a
normal corrosion mechanism for carbon steel tanks in hot water service.
Tank 565-23-901 (Figure 167) had more "carbuncles" in the bottom (Figure
168) than in 2009, suggesting that the corrosion was progressing. A spot
was selected for examination before (Figure 169) and after (Figure 170)
removal of the scale. The black deposit found beneath the scale appeared
to be magnetite (Fe3O4). In 2009, the underlying deposit was browner in
colour. Magnetite deposits indicate better control of the oxygen in the
water.
Tank 565-23-902 (Figure 171) had "carbuncles" in the bottom (Figure 172)
comparable with those observed in 2009, suggesting that the corrosion
was not progressing rapidly. A spot was selected for examination before
(Figure 173) and after (Figure 174) removal of the scale. The black deposit
found beneath the scale appeared to be magnetite (Fe3O4). In 2009, the
underlying deposit was also black in colour. Magnetite deposits indicate
better control of the oxygen in the water.
51
7.0 DISCUSSION
In my 2009 report [Ref. 2] I commented extensively on the mechanisms of

corrosion in the concentrator, evaporators, bleach plant presses and
piping, and the white water tanks. The reader should refer to that report
for more information on these topics.
This year, my sermon will be to promote the better use of materials and
corrosion resources. When I began working in the pulp and paper
industry in 1977, I was part of the Materials and Corrosion Engineering
Section located in the MacMillan Bloedel Technical Centre in Vancouver,
Canada. We served several pulp and paper mills of MacMillan Bloedel,
located across Canada, the USA, and Europe. During my time with
MacMillan Bloedel (1977 to 1992) we documented savings to the company
in excess of twenty million dollars per year. The same (or greater) amount
could be saved by Celulosa Arauco y Constitución if the materials and
corrosion function was expanded.
Let me be clear: the suppliers do not necessarily act in the best interest of
Celulosa Arauco y Constitución. They do not have the capability of doing
the necessary research to validate the materials selected for their
equipment. This is why we now have corrosion in the superconcentrator,
concentrator, evaporators, and bleach presses, to name just some of the
equipment. Suppliers rely on their past experience which often does not
help when process parameters such as black liquor solids or chlorine
dioxide residuals are being extended. The customer must have a greater
role in specifying the proper materials and welding consumables.
52
7.1 The Need for Quality Assurance
While suppliers have their own QA personnel, these people act in the
service of the supplier and not necessarily in the service of the customer
(Celulosa Arauco y Constitución). Their QA people are mainly acting to
keep costs down, often to the detriment of the customer. An independent
(of the supplier) third-party QA person can act in the interests of the
customer.
Good QA begins at the design stage, and may involve such issues as
materials selection, welding consumables, heat treatment, corrosion
allowances, post-weld cleaning, and many more subjects. An error in
materials selection caught in the design stage (selection of type 304L
stainless steel for the concentrator, selection of carbon steel for the
evaporator effects handling intermediate-solids black liquors, and
selection of ERNiCrMo-3 (Alloy 625) for welding of the D-stage bleach
presses) can prevent much more expensive repairs after the equipment
has been built.
Next, good QA requires enforcement by regular presence in the shop

where the equipment is being built (or in the field, as the case may be).
Two essential tools that I use are an alloy analyzer and a Feritscope. Few
suppliers have such instruments. Metso and GL&V are unlikely to have
them. When I used to work for a MacMillan Bloedel we had two alloy
analyzers in full-time use. They were particularly effective at catching the
incorrect use of materials when the equipment was being built in the shop
of a supplier. Prevention of a single failure can more than justify the cost
(approximately US$30,000) of purchasing an alloy analyzer.
53
There were many times when an alloy analyzer would have been valuable
during the 2010 shutdown at Nueva Aldea. Identification of the material
used in the tube-to-header welds in the superconcentrator and
concentrator are two examples. Others include identification of the filler
metal used for the "half-pipe" headers on the digesters, the welding
consumables used for the bleach presses, and the piping in the bleach
plant.
My Feritscope proved to be an invaluable resource during the 2010

shutdown at Nueva Aldea. It allowed me to quickly identify that there
had been a heat treatment error with the cracked elbow removed from the
condensate drain in the superconcentrator. A Feritscope is particularly
useful in confirming that stainless steel welds have the correct ferrite
content for the filler metal used. Many of the 2304 duplex stainless steel
tube-to-header welds tested during the shutdown had ferrite contents
outside the expected range. This suggests future problems. These
inconsistencies are best caught in the shop where they can be dealt with
rather in the field after the equipment has already been delivered.
7.2 The Need for Corrosion Testing
Another of my functions during my time with MacMillan Bloedel was

corrosion testing or research. Over 15 years I published many technical
papers on corrosion in alkaline liquors, corrosion of recovery boilers,
corrosion in the bleach plant, corrosion of digesters, and corrosion in
white waters. Most of this work was based on corrosion testing such as
corrosion coupons installed in tanks or other equipment.
54
While I still have access to a world-class corrosion laboratory, there is little

demand for corrosion testing from the pulp and paper industry. This is
foolish. The equipment suppliers carry out no corrosion research,
preferring their equipment to be the "test coupons" to the chagrin of their
customers.
There is a long list of subjects from Nueva Aldea alone that would merit
corrosion testing, including:
- Corrosion testing of duplex stainless steels in superconcentrator and

concentrator black liquors.
- SCC testing of austenitic and duplex stainless steels in
superconcentrator and concentrator black liquors.
- SCC testing of bottom scraper materials in digester extraction liquors
- Corrosion testing of welding consumables for 6% Mo stainless steel in
D-stage filtrates.
- Corrosion testing of materials and welding fillers for resistance to MIC
in white water.
Good information can be gained from chemical analysis of process

liquors, filtrates, and waters.
Laboratory corrosion testing can be used to evaluate the effects of

proposed process changes on existing materials of construction, or to
evaluate new materials for service in existing or proposed new
environments. A little money spent up front on corrosion testing can save
a lot of money later on for the repair of equipment that corroded or failed
due to "unanticipated" problems. With research, many problems can be
anticipated.
55
7.3 The Need for Failure Analysis
Failure analysis is an important part of any materials and corrosion

engineering function. When I worked with MacMillan Bloedel, we had a
fully-equipped failure analysis laboratory including a scanning electron
microscope. The problem with sending out samples of failed equipment
to outside laboratories is that they are not familiar with the equipment or
the process or the context of the failure with respect to the history of the
equipment. They often generate a pretty report that does not solve the
problem.
A good example is the leaking "half-pipe" headers for the digesters. A

report was generated but was not helpful. All it took was one look to find
the cause of the leaks was fatigue cracking initiating at weld roots that
were not fully penetrated.
Giving the samples to the equipment supplier is better than providing

them to a laboratory unfamiliar with the equipment, but the supplier will
still act in self-interest to minimize their liability.
Independent laboratories have done some very good work, although it

will be several years before they develop the knowledge necessary to
understand the samples that they receive. The best option is to have
failure analysis done by a third-party independent of the suppliers yet
familiar with the industry. I have access to an excellent laboratory for
failure analysis. It may be more expensive but you will get what you pay
for: an answer that will help solve the problem.
56
8.0 REFERENCES
1. Wensley, A., Inspections of the Digesters and the Evaporators at

Nueva Aldea, Angela Wensley Engineering Report AWE7032, 17
December, 2007.
2. Wensley, A., Inspections during the Shutdown at Nueva Aldea,
Angela Wensley Engineering Report AWE8025, 15 February, 2009.
3. Wensley, D.A., Reid, D.C., and Dykstra, H., Corrosion of Nickel-Base
Alloys in Chlorine Dioxide Washer Service, Paper No. 537 presented
at the Corrosion 90 Conference (1990).
4. Wensley, A., Reid, C., Brown, D., and Christie, D.W., Corrosion of
Stainless Alloys in a Chlorine Dioxide Bleached Pulp Washer, Proc.
TAPPI Engineering Conf., pp. 499-509 (1991).
5. Wensley, A. and Reid, D.C., Transpassive Dissolution of Nickel-
Base Alloys, Paper No. 433 presented at NACE Corrosion 99 (1993).
6. Christie, D.W., Effect of Post-weld Cleaning on Corrosion
Resistance of Austenitic and Duplex Stainless Steel Weldments in
Bleach Plant Service, Proc. 7th Intl. Symposium on Corrosion in the
Pulp and Paper Ind., pp. 87 - 95 (1992).
57
Figure 1. Vapour inlet in the top of the superconcentrator.

Arrows show direction of liquor splashing from above.
Figure 2. Erosion of the vapour inlet in the superconcentrator.
F-1
Figure 3. Leak in the bottom of the superconcentrator detected

during a hydro test.
Figure 4. Condensate drain cut below a small diameter header

where leaks had developed above the tube welds in the
superconcentrator.
F-2
Figure 5. Sample of tubes and small diameter header removed

from the bottom of the superconcentrator. Arrows
show leaks above the welds.
Figure 6. The leaks in above the small-diameter headers in the

superconcentrator had no evident origin other than
liquor-side corrosion of the heat affected zone.
F-3
Figure 7. Vapour leak at a pore in the weld between a condensate

drain tube and the large-diameter header in the
superconcentrator.
Figure 8. Erosion of adjacent tube from a vapour leak at a pore in

the weld between a condensate drain tube and the
large-diameter header in the superconcentrator.
F-4
Figure 9. Weld repairs made to the welds between the

condensate drains and the large-diameter header in the
superconcentrator.
Figure 10. Ferrite measurements (%) of welds in the bottom of the

superconcentrator. Acceptable values are in the range
30% to 60%.
F-5
Figure 11. Elbow from the large condensate drain in the bottom of
the superconcentrator removed due to cracking. The
elbow had only 13% ferrite, not characteristic of 2304.
Figure 12. Closer view of cracks in the elbow removed from the
superconcentrator. The cracks may be SCC.
F-6
Figure 13. The ferrite content of the replacement elbow (38%) and
the to welds (48% and 48%) in the superconcentrator
were acceptable for duplex stainless steel.
Figure 14. Another elbow (original) in the superconcentrator had

52% ferrite, acceptable for duplex stainless steel.
F-7
Figure 15. View inside the concentrator after the cap had been
removed.
Figure 16. Old 304L tubes removed from body A of the

concentrator. This view shows the bottom of the
bundle with the two main condensate drains.
F-8
Figure 17. Many of the 304L tubes from concentrator body A had
a spiral pattern of SCC from stresses introduced by the
tube straightening process during tube manufacture.
Figure 18. Closer view of a type 304L stainless steel tube from
concentrator body A with SCC aligned in a spiral
manner.
F-9
Figure 19. SCC aligned with a weld in a type 304L stainless steel
tube from concentrator body A.
Figure 20. SCC of a support for a header on concentrator body A.
F-10
Figure 21. View inside a large diameter header at the bottom of

concentrator body B. The arrows show incomplete
fusion of the welds between the tubes and the header.
Figure 22. New bundle of type 2304 duplex stainless steel tubes
for concentrator body B.
F-11
Figure 23. New type 2304 duplex stainless steel tubes with
stenciled identification "IRSA EN 10217-7 TC1 ISO
D4/T3 0 32 x 1.2 1.4362 HEAT 490708 OF 511815 (48)".
Figure 24. Ferrite contents of the 2304 duplex stainless steel tubes
(upper row) and the welds between the tubes and the
small diameter header (lower row) for body B.
F-12
Figure 25. Ferrite contents of the Sandvik 2304 duplex stainless

steel drain tubes between the small and large diameter
headers in the new bundle for concentrator body B.
Figure 26. View inside concentrator body A after removal of the

tubes.
F-13
Figure 27. The lower wall of concentrator body A was bright in

appearance, characteristic of uniform active corrosion.
Figure 28. The originally-flat welds in concentrator body A were

now standing proud by almost 1 mm.
F-14
Figure 29. Appearance of the wall in ring 7 of concentrator body

A.
Figure 30. Divider between bodies A and D of the concentrator.
F-15
Figure 31. Locations of SCC on side A of the divider between

bodies A and D of the concentrator.
Figure 32. Closer view of SCC on side A of the divider between

bodies A and D of the concentrator, associated with a
remnant weld made during construction.
F-16
Figure 33. Location of SCC in ring 3 of the wall of concentrator

body A, just above circumferential weld A3.
Figure 34. Closer view of SCC in ring 3 of the wall of concentrator

body A, just above circumferential weld A3.
F-17
Figure 35. Closer view of SCC in ring 3 of the wall of concentrator

body A. The multiple branched appearance is
characteristic of SCC.
Figure 36. SCC of circumferential weld A8 in concentrator body

A.
F-18
Figure 37. Closer view of SCC of circumferential weld A8 in

concentrator body A.
Figure 38. Locations of SCC adjacent to circumferential weld A2,

revealed by PT.
F-19
Figure 39. Removal of the SCC adjacent to circumferential weld

A4 in concentrator body A.
Figure 40. The thickness of the wall after removal of the SCC
below circumferential weld A6 in concentrator body A
was 19.62 mm.
F-20
Figure 41. Cap of the concentrator.
Figure 42. PT revealed extensive adjacent to circumferential weld

12 in the cap.
F-21
Figure 43. Area of intense SCC in the cap of the concentrator.
Figure 44. Closer view of intense SCC in the cap.
F-22
Figure 45. PT of the longitudinal welds in the cap.
Figure 46. SCC adjacent to a longitudinal weld in the cap of the

concentrator.
F-23
Figure 47. View inside the top of evaporator effect E-2. The shell
and top head are type 304L stainless steel.
Figure 48. View inside the bottom of evaporator effect E-2. The
shell and bottom head are type 304L stainless steel.
F-24
Figure 49. Bottom tubesheet and mist eliminator in evaporator

effect E-3. The shell and bottom head are type 304L
stainless steel.
Figure 50. Stainless steel "China hat" above the liquor inlet in
F-25
Figure 51. Above the upper tubesheet in evaporator effect E-4.
Figure 52. There was no galvanic corrosion of the carbon steel top
head of evaporator effect E-4.
F-26
wall and bottom head are carbon steel.
Figure 54. Mist eliminator in evaporator effect E-4. Note the

supports and wall of the mist eliminator are stainless
steel.
F-27
Figure 55. Intact roof above the liquor entrance in the bottom of
Figure 56. The thickness of the carbon steel divider around the
liquor inlet in evaporator effect E-4 was 10 mm..
F-28
F-29
Figure 59. The roof above the liquor inlet in the bottom of
evaporator effect E-5 had failed.
Figure 60. The roof above the liquor inlet in the bottom of
evaporator effect E-5 had failed.
F-30
Figure 61. The remainder of the roof and the failed China hat
above the liquor inlet in evaporator effect E-5 fell to the
bottom inside the divider around the liquor inlet.
Figure 62. Preferential corrosion of a weld in the roof above the

liquor inlet in the bottom of evaporator effect E-4.
F-31
wall and bottom head were carbon steel.
Figure 64. Failed rung from the bottom of evaporator effect E-5.
F-32
F-33
Figure 67. Lower tubesheet and mist eliminator in evaporator

effect E-6 before the scaffolding was completed.
Figure 68. View inside the bottom of evaporator effect E-5 before
the scaffolding was installed. The wall and bottom
head were carbon steel.
F-34
Figure 69. Intact China hat above the liquor inlet in the bottom of
Figure 70. The divider around the liquor inlet in evaporator effect
E-6 had thinned to 3 mm (from an original 10 mm?).
F-35
Figure 71. Some welds in the frame of the mist eliminator in

evaporator effect E-6 had corroded completely away.
Figure 72. Some welds in the frame of the mist eliminator in

evaporator effect E-6 had corroded completely away.
F-36
Figure 73. Corrosion perforation in the ceiling of the mist

eliminator in evaporator effect E-6.
Figure 74. Closer view of a corrosion perforation in the ceiling of

the mist eliminator in evaporator effect E-6.
F-37
Figure 75. The carbon steel supports for the mist eliminator in
evaporator effect E-6 had evidence of thinning.
Figure 76. Galvanic corrosion at a stainless steel to carbon steel

transition in a drain pipe from the mist eliminator in
F-38
Figure 77. Galvanic corrosion at a stainless steel to carbon steel

transition in the structure of the mist eliminator in
Figure 78. There was galvanic corrosion at the stainless steel to

carbon steel transition below the lower tubesheet in
F-39
Figure 79. Upper dome of evaporator effect E-7.
F-40
wall and bottom head were carbon steel.
Figure 82. Intact China hat above the liquor inlet in the bottom of
F-41
Figure 83. Lower tubesheet and mist eliminator in evaporator

effect E-7.
Figure 84. Carbon steel support and carbon steel drain pipe for
the mist eliminator in evaporator effect E-7.
F-42
Figure 85. "Carbuncles" covered the carbon steel walls above the
lower tubesheet of the surface condenser.
Figure 86. Pitting corrosion associated with the "carbuncles" in the

surface condenser.
F-43
Figure 87. The arrows show the external "half-pipe" dilution

header around the bottom of the Line 1 (pine) digester.
Figure 88. Temporary leak management box for the external "half-
pipe" dilution header around the bottom of the Line 1
(pine) digester.
F-44
Figure 89. View of the external "half-pipe" dilution header around

the bottom of the Line 1 (pine) digester after removal of
the leak prevention box.
Figure 90. The leak in the external "half-pipe" dilution header

around the bottom of the Line 1 (pine) digester was a
crack between the upper weld and the digester shell.
F-45
Figure 91. View of the external "half-pipe" dilution header around

the bottom of the Line 2 (eucalyptus) digester after
removal of the leak prevention box.
Figure 92. Location of the leak in the external "half-pipe" dilution

header around the bottom of the Line 2 (eucalyptus)
digester.
F-46
Figure 93. Closer view of a fatigue crack that caused the leak from
the external "half-pipe" dilution header around the
bottom of the Line 2 (eucalyptus) digester.
Figure 94. Window removed from the external "half-pipe" dilution

header around the bottom of the Line 2 (eucalyptus)
digester.
F-47
Figure 95. The upper weld on the "half-pipe" header did not fuse
completely through to the root side.
Figure 96. The lower weld on the "half-pipe" header did not fuse
completely through to the root side.
F-48
Figure 97. New upper fillet weld made on the inside of the "half-
pipe" header.
Figure 98. New lower fillet weld made on the inside of the "half-
pipe" header.
F-49
Figure 99. Root-side dams installed prior to replacing the window

from the "half pipe" header.
Figure 100. View of scraper arm "A" and the China hat in the
bottom of the Line 1 (pine) digester.
F-50
Figure 101. View of scraper arm "B" and the China hat in the
bottom of the Line 1 (pine) digester.
Figure 102. Scraper arm "A" in the bottom of the Line 1 (pine)
digester.
F-51
Figure 103. Eroded weld in the middle of the China hat in the
Line 1 (pine) digester, exposing the carbon steel
support underneath.
Figure 104. Carbon steel at the bottom of the China hat in the Line
1 (pine) digester.
F-52
Figure 105. Cracked stitch welds in the lower screens of the Line 1
(pine) digester.
Figure 106. Closer view of a cracked stitch weld in the lower

screens of the Line 1 (pine) digester.
F-53
Figure 107. PT of the external weld for a vapour injector at the top
of the Line 1 (pine) digester.
Figure 108. Internal view of a vapour injector at the top of the

Line 1 (pine) digester.
F-54
Figure 109. Wall of the Line 1 (pine) digester below the top
separator.
Figure 110. Upper part of the top separator screw in the Line 1
(pine) digester.
F-55
Figure 111. Scraper arm "A" in the bottom of the Line 2

(eucalyptus) digester.
Figure 112. Scraper arm "B" in the bottom of the Line 2

F-56
Figure 113. Modified China hat in the bottom of the Line 2

Figure 114. Three rows of screens in the bottom of the Line 2

F-57
Figure 115. Carbon steel cover plate for a beam nozzle for the
Line 2 (eucalyptus) digester.
Figure 116. Interior view of the duplex stainless steel lining on the
inside of a carbon steel cover plate for the Line 2
F-58
Figure 117. View below the top separator in the Line 2

Figure 118. Light-coloured area below the top separator in the

F-59
Figure 119. PT of a vapour injector at the top of the Line 2

Figure 120. Interior view of a vapour injector at the top of the

F-60
Figure 121. China hat and scraper arm in the bottom of the Line 1
(pine) impbin.
Figure 122. Bottom of the central pipe in the Line 1 (pine) impbin
after it was shortened by 1 m.
F-61
Figure 123. Section of the interior central pipe removed from the
Line 1 (pine) impbin.
Figure 124. Corrosion of the interior central pipe removed from

the Line 1 (pine) impbin.
F-62
Figure 125. China hat and scraper arm in the bottom of the Line 2
(eucalyptus) impbin.
Figure 126. Other scraper arm in the bottom of the Line 2

(eucalyptus) impbin.
F-63
Figure 127. Mechanical damage (upper arrows) in the bottom of

the Line 2 (eucalyptus) impbin. The lower arrow
shows a narrow ring in the bottom head.
Figure 128. Bottom of the central pipe in the Line 2 (eucalyptus)

impbin.
F-64
Figure 129. Mechanical damage to the roll on side 2 of the Line 1

(pine) press 1.

(pine) press 2.
F-65

(pine) press 4.

(pine) press 4.
F-66

(pine) press 6.

(pine) press 7.
F-67
Figure 135. Dilution nozzles in a plug-welded lining in the side 2

vat in the bottom of the Line 1 (pine) press 7.
Figure 136. Corroded plug weld in the lining of side 2 vat in the
bottom of the Line 1 (pine) press 7.
F-68
Figure 137. Another corroded plug weld in the lining of side 2 vat
in the bottom of the Line 1 (pine) press 7.
Figure 138. Corroded longitudinal weld in a roll from the Line 1

(pine) press 4.
F-69
Figure 139. Discolouration of the flap from side 2 of the Line 1

(pine) press 6.
Figure 140. Discolouration of the flap from side 2 of the Line 1

(pine) press 6.
F-70
Figure 141. Spacers installed (prior to lining) on the flap from side
1 of the Line 1 (pine) press 6.
Figure 142. Lining installed on the side 2 flap from the Line 1
(pine) press 4.
F-71
Figure 143. Pitting corrosion of the edge of the side 2 flap from
the Line 1 (pine) press 7.
Figure 144. Closer view of pitting corrosion of the edge of the side
2 flap from the Line 1 (pine) press 7.
F-72
Figure 145. Pitting corrosion of the frame for the flap on side 2 of
the Line 1 (pine) press 7.
Figure 146. Closer view of pitting corrosion of the frame for the
flap on side 2 of the Line 1 (pine) press 7.
F-73
Figure 147. Corrosion product leaking from bolt holes for a cover
plate on side 1 of the Line 1 (pine) press 7.
Figure 148. Closer view of corrosion product leaking from bolt

holes for a cover plate on side 1 of the Line 1 (pine)
press 7.
F-74
Figure 149. Inside of a cover plate removed from side 1 of the

Line 1 (pine) press 7.
Figure 150. Closer view of pitting corrosion on the inside of a

cover plate removed from side 1 of the Line 1 (pine)
press 7.
F-75
Figure 151. Top view of the shredder/conveyor screw and

housing on the top of the Line 1 (pine) press 6.
Figure 152. Closer view of pitting corrosion of the

shredder/conveyor housing on the top of the Line 1
(pine) press 6.
F-76
Figure 153. Another view of pitting corrosion of the

shredder/conveyor housing on the top of the Line 1
(pine) press 6.
Figure 154. Shredder/conveyor housing on the Line 1 (pine)

press 5. The blue colour is from D-stage ClO2
escaping from filtrate injected into the standpipe.
F-77
Figure 155. Shredder/conveyor housing on the Line 2

(eucalyptus) press 5. The blue colour is from D-stage
ClO2 escaping from filtrate injected in the standpipe.
Figure 156. Corrosion at scratches in the housing on the Line 2

(eucalyptus) press 5.
F-78
Figure 157. Corrosion of a duplex stainless steel pipe at a flange

at heat exchanger 547-58-1400 on the Line 1 (pine)
side.
Figure 158. Bottom of the heat exchanger showing a 2009 repair

to a leaking pipe (A) and a leak though a duplex
stainless steel spool piece (B).
F-79
Figure 159. Leaks in the duplex stainless steel pipe between the
heat exchanger and the Line 1 (pine) press 5
standpipe.
Figure 160. Leaks in the duplex stainless steel pipe between the
heat exchanger and the Line 1 (pine) press 5
standpipe.
F-80
Figure 161. Corrosion of a duplex stainless steel header for Line 1

(pine) standpipe 547-22-1401.
Figure 162. Closer view of corrosion of a duplex stainless steel

header for Line 1 (pine) standpipe 547-22-1401.
F-81
Figure 163. Line 1 (pine) white water tank 572-22-1155.
Figure 164. Leaking plate removed from the Line 1 (pine) white
water tank 572-22-1155.
F-82
Figure 165. MIC of Line 1 (pine) white water tank 572-22-1156,

after cleaning. A 2009 weld repair was not corroded.
Figure 166. MIC of Line 1 (pine) white water tank 572-22-1156,

after cleaning. A weld repair was corroded.
F-83
Figure 167. Expansion tank 565-23-901.
Figure 168. The internal surfaces in expansion tank 565-23-901

were covered with "carbuncles."
F-84
Figure 169. Close-up view of the surface deposit in expansion

tank 565-23-901.
Figure 170. Underneath the deposit the surface was black,

characteristic of magnetite.
F-85
Figure 171. Expansion tank 565-23-902.
Figure 172. The internal surfaces in expansion tank 565-23-902

were covered with "carbuncles."
F-86
Figure 173. Close-up view of the surface deposit in expansion

tank 565-23-902.
Figure 174. Underneath the deposit the surface was black,

characteristic of magnetite.
F-87

Angela Wensley Report Nueva Aldea 2010 PDF

Uploaded by

Copyright:

Available Formats

Angela Wensley Report Nueva Aldea 2010 PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Angela Wensley Report Nueva Aldea 2010 PDF

Uploaded by

Copyright:

Available Formats

15397 COLUMBIA AVENUE Tel.

Celulosa Arauco y Constitución S.A.

Attention: Mario Ávila G.

Date: 22 April, 2010 Dr. Angela Wensley, P.Eng.

C:\Winword\Nueva Aldea\Nueva Aldea 2010\Report Text Nueva Aldea 2010.doc

3.0 DIGESTERS AND IMPBINS . . . . . . 38

4.0 BLEACH PRESSES . . . . . . . . 44

5.0 WHITE WATER TANKS FOR THE PAPER MACHINES . . 50

6.0 EXPANSION TANKS . . . . . . . 51

Evaporator Effect S-1 (Superconcentrator)

Evaporator Effect E-1 (Concentrator)

- Adjacent to the circumferential and vertical weld seams.

Evaporator Effect E-2

Evaporator Effect E-3

Evaporator Effect E-4

Evaporator Effect E-5

The bottom of evaporator effect E-5 should be scaffolded in 2011 so a full

Evaporator Effect E-6

Evaporator Effect E-7

The bottom of evaporator effect E-7 should be scaffolded in 2011 so a full

Continuous Digester External "Half-Pipe" Headers

Pine Bleach Presses

Eucalyptus Bleach Presses

Duplex Stainless Steel Piping in the Bleach Plant

titanium (Grade 2), fibre reinforced plastic (FRP), or 6% Mo stainless steel

Stainless Steel White Water Tanks for the Machines

Microbiologically-assisted corrosion (MIC) continues to be a problem in stainless

Expansion Tanks for the Turbogenerator

For the Superconcentrator:

S1. The sample of failed Tubel® tubes should be examined by a competent

- Elemental composition of the tubes and welds.

- Elemental composition of the elbow.

For the Concentrator:

C3. QA guidelines developed by Angela Wensley Engineering for construction

C4. Samples of liquor from bodies A, B, C, and D should be analyzed for

C6. A sample of a cracked tube from the concentrator body A should be

For the Evaporators:

1. Complete replacement of the bottom using type 304L stainless steel.

Since anodic protection is effective only in immersion conditions (up to 200

E9. Since application of weld overlay on substrates below 17 mm thick may

For the Digesters and Impbins:

For the Bleach Plant:

B1. Locations of pitting corrosion and crevice corrosion of the 6% Mo stainless

Derakane 470-300 or equivalent, and 6% Mo stainless steel welded using a

B4. A section of 6% Mo stainless steel pipe with a circumferential weld should be

B5. Locations of mechanical damage to the rolls in the presses should be

For the White Water Tanks:

W2. Temporary patches to locations of leaks may be made using overmatched

For the Turbogenerator Expansion Tanks:

G1. Celulosa Arauco y Constitución should support QA and corrosion testing to

G2. Celulosa Arauco y Constitución should purchase a portable "alloy analyzer"

Nueva Aldea has a Kvaerner (Metso) seven-effect falling film evaporator

28% 24% 21% 39% 45% 73%

65C 74C 83C 91C 102C 114C 142C

The cylindrical shell and internals of the

Figure 1 shows a vapour inlet below a liquor

I used a Fischer Feritscope Model MP-30 to measure the ferrite contents of

One of the elbows in the condensate drain below the large-diameter

2.2 Evaporator Effect E-1 TAG 551-51-0102