Engineering FailureAnalysis, Vol 2, No. 3 pp.
165-174, 1995
Pergamon Copyright © 1995 Elsevier Science Ltd
Printed in Great Britain. All rights reserved
1350-6307/95 $9,50 + 0.00
1350-6307(95)00017-8
FAILURE ANALYSIS OF CARBONATE REBOILER HEAT
EXCHANGERS
R A K E S H K A U L , N. G. M U R A L I D H A R A N , T. J A Y A K U M A R ,
K. V. K A S I V I S W A N A T H A N and B A L D E V R A J
Division for PIE & NDT Development, Indira Gandhi Centre for Atomic Research,
Kalpakkam 603 102, India
R. K. D A Y A L , V. R. R A J U and J. B. G N A N A M O O R T H Y
Metallurgy Division, Indira Gandhi Centre for Atomic Research, Kalpakkam 603 102, India
and
S. P A T T U
Nagarjuna Fertilizers & Chemicals Ltd, Kakinada 533 001, India
(Received 26 May 1995)
Abstract--Unacceptably high leakage rates from a large number of type 304 stainless steel
tubes forced the scrapping of two carbonate reboiler heat exchangers from the carbon dioxide
removal system of a fertilizer plant. Extensive failure investigation carried out on the failed
heat exchangers established that the main factor responsible for premature failure of the heat
exchangers was partial expansion of the stainless steel tubes into the carbon steel tube sheet as
against the design specification for the expansion of almost the full length of the tubes inside
the tube sheet. Partial expansion of the tubes into the tube sheet left undesirable tube-to-tube
sheet gaps, which subsequently acted as crevices and led to localized corrosion of the tube
sheet as well as the tubes.
1. I N T R O D U C T I O N
eaks from a few tubes of two carbonate reboiler heat exchangers were noticed
mortly after start up of the heat exchangers. The leakage was detected by an
~normal increase in the hydrogen content of the shell-side medium, the so-called
:ocess solution. The leaking tubes were plugged and the heat exchangers were
.,started. However, the number of leaking tubes still increased. Finally, the two heat
:changers had to be scrapped as it was no longer economical to operate them with a
rge number of plugged tubes. Figure 1 shows a photograph of a heat exchanger
rnilar to the failed ones. Detailed failure analysis was carried out on the failed heat
:changers to ascertain the cause of the failure and to suggest remedial measures to
,oid such costly failures in future.
2. B A C K G R O U N D I N F O R M A T I O N
The heat exchangers concerned had been made with stainless steel (A 213-798
~-304) U-tubes of 19 m m outside diameter and 2 m m wall thickness, and a tube
eet of A 105-80 grade carbon steel. According to the information provided, a thick
rer of stainless steel had been weld-overlaid on the carbon steel tube sheet. The
rid-overlaying involved deposition of a barrier layer of A W S E 309 L on the carbon
~el followed by the deposition of A W S 308 L. The tube-to-tube sheet joint had been
166 RAKESH KAUL et al.
Fig. 1. Photograph of a heat exchanger similar to the failed ones.
made between the stainless steel overlaid portion of the tube plate and the stainless
steel tubes. The tubes had been subsequently expanded to fill the tube-to-tube sheet
gap. According to the drawings for the heat exchangers provided by the supplier,
almost the full portion of the tube inside the tube sheet was to be expanded (Fig. 2).
Tube-to-tube sheet welds were pneumatically tested at 0.2 MPa before expansion of
the tubes into the tube sheet.
It was reported that, before putting the heat exchangers into operation, passivation
was carried out for 72 h by circulating the process solution on the shell side. The
process solution was a mixture of K2CO 3 (26.5-29%), KVO 3 (0.5-0.8%) and a
solution containing diethylamine and glycine (13.5-14.5 g/l) in a ratio of 44 and 56%,
respectively. During operation of the heat exchangers, the stainless steel tubes carried
hot process gas, known as "syngas" (a mixture of nitrogen and hydrogen) while the
medium on the shell side was the highly alkaline process solution. The vanadium
concentration of the process solution was 5.8 g/l. The inlet and outlet temperatures of
the syngas were 436 and 399 K, respectively. The pressures on the tube and shell
sides were 20 and 16 MPa, respectively.
STAINLESS STEEL
TUBE SHEET OVERLAY
TUBE
Fig. 2. Design specificationfor expansion of tube into tube sheet.
Failure analysis of carbonate reboiler heat exchangers 167
3. E X A M I N A T I O N T E C H N I Q U E S A N D R E S U L T S
3.1. Visual examination
Visual examination of the failed heat exchanger revealed that the tube sheet was
heavily eaten away (Fig. 3). The damage to the tube sheet was confined to the region
:lose to the shell face of the tube sheet. In some regions, the damage was severe
enough to produce through-and-through holes. Features present on the damaged
~urfaces of the tube sheet holes are indicative of gouging of the material. Portions of
:he stainless steel tubes inside the tube sheet had also undergone substantial thinning
!rom the outside surface (Fig. 4). It was noticed that the thinned region of the
;tainless steel tube coincided with the damaged region of the mating tube sheet
;urface.
t.2. Dye penetrant testing
Dye penetrant testing (DPT) was carried out on: (i) the shell-side surface, (ii) the
velded-side face, and (iii) longitudinal sections of the cut blocks of the tube sheet
with tubes). The D P T revealed non-uniform tube-to-tube sheet gaps (Fig. 5).
Fig. 3. Extensive corrosion damage on tube sheet.
-Welded
Shell side end
side e~
Tube sheet
.........
Tube
Fig. 4. Photograph showing corrosion damage on tube sheet as well as on tube. Defect regions
are indicated by arrows.
168 RAKESH KAUL et al.
Fig. 5. DPT indications on tube sheet cut portion.
3.3. D i m e n s i o n a l m e a s u r e m e n t s
In o r d e r to d e t e r m i n e t h e e x t e n t o f e x p a n s i o n of t h e t u b e s into t h e t u b e s h e e t ,
i n t e r n a l d i a m e t e r m e a s u r e m e n t s (with an a c c u r a c y of 0.002 m m ) w e r e c a r r i e d o u t on
the p o r t i o n s o f the stainless steel t u b e s t h a t w e r e inside t h e t u b e sheet. T h e
d i m e n s i o n a l m e a s u r e m e n t s w e r e p e r f o r m e d in the r e g i o n s e x t e n d i n g a b o u t 65 m m
f r o m b o t h e n d s o f t h e t u b e sheet. T h e results o f t h e d i m e n s i o n a l m e a s u r e m e n t s ,
c a r r i e d o u t on two t u b e s h e e t b l o c k s - - b l o c k s " A " a n d " B " - - a r e p r e s e n t e d in T a b l e s
1 a n d 2, r e s p e c t i v e l y , T h e s e results clearly show t h a t t h e t u b e s h a d b e e n e x p a n d e d
o n l y u p to a d i s t a n c e of a b o u t 3 0 - 4 0 m m f r o m the w e l d e d e n d o f t h e t u b e s h e e t while
the r e m a i n i n g p o r t i o n o f the t u b e s inside t h e t u b e s h e e t h a d b e e n left u n e x p a n d e d .
3.4. MetaUographic examination
M e t a l l o g r a p h i c e x a m i n a t i o n o f the t u b e s h e e t s p e c i m e n t a k e n f r o m t h e r e g i o n a w a y
f r o m the w e l d s h o w e d a n o r m a l f e r r i t e - p e a r l i t e m i c r o s t r u c t u r e (Fig. 6). M e t a l l o -
g r a p h i c e x a m i n a t i o n o f the w e l d side p o r t i o n of the t u b e s h e e t r e v e a l e d weld-
o v e r l a y i n g of stainless steel (Fig. 7) o v e r the c a r b o n steel t u b e sheet. T h e o v e r l a i d
p o r t i o n o f the t u b e s h e e t was not f o u n d to be a s s o c i a t e d with any a b n o r m a l i t y . T h e
Table 1. Inside diameter measurement of tubes in block "A" (mm)
Distance from welded face (ram) Distance from shell-side face (mm)
Tube Tube
no. 0 20 30 45 65 60 50 25 0 no.
I 15.520 15.630 15.680 1 5 . 4 4 0 15.420 15.370 15.400 15~420 15.455 l
15.625 15.675 1 5 , 4 1 0 15.410 15.360 15.390 15.400
2 Plugged 15.380 15.400 15.420 15.380 2
15.370 15.395 15.380 15.370
3 Damaged 15.480 15,505 -- 15.470 3
15.470 15.495 15.460
4 Plugged -- 15,360 15,280 15.330 4
15.350 15,260 15.325
5 15.510 15.660 15.650 15.480 15.440 -- 15.380 15,325 15.330 5
15.500 15.630 15.465 15.435 15,360 15,320 15.325
6 15.540 15.670 15,700 15.380 15.390 15.300 15.330 15,270 -- 6
15.370 15,340 15.280 15.320 15,260
Failure analysis of carbonate reboiler heat exchangers 169
Table 2. Inside diameter measurement of tubes in block 'B' (mm)
Distance from welded face (mm) Distance from shell-side face (mm)
Tube Tube
no. 0 20 30 45 65 65 45 30 20 0 no.
1 15.800 15.800 15.810 15.370 15.380 15.350 15.330 15.350 15.300 14.870 1
15.790 15.790 15.805 15.365 15.375 15.300 15.325 15.345 16.875
2 Plugged 15.340 15.300 15.040 15.030 14.870 2
15.295 14.865
3 15.755 15.230 15.235 15.380 15.365 15.380 15.370 15.360 15.385 15.390 3
15.750 15.225 15.230 15.375 15.360 15.380
4 15.790 15.740 15.740 15.360 15.360 15.385 15.390 15.385 15.385 15.390 4
15.785 15.730 15.730 15.355 15.355 15.380 15.380
5 15.750 15.720 15.735 15.365 15.365 15.360 15.340 15.340 15.340 15.300 5
15.745 15.710 15.730 15.360 15.360
6 15.790 15.820 15.795 15.345 15.345 15.320 15.320 15.300 15.330 15.330 6
15.830 15.790 15.340 15.340
7 15.310 15.790 15.770 15.390 15.390 15.380 15.370 15.340 15.380 15.320 7
15.305 15.780 15.760 15.360 15.360
8 15.775 15.755 15.720 15.290 15.340 15.340 15.340 15.340 15.260 14.750 8
15.770 15.750 15.715 15.280 15.270 14.740
9 15.740 15.740 15.750 15.400 15.400 15.455 15.455 15.440 15.450 14.460 9
15.735 15.735 15.745 15.390 15.390 15.450 15.450
10 15.810 15.770 15.760 15.360 15.360 15.350 15.360 15.380 15.500 14.720 10
15.800 15.765 15.750 15.355 15.355 14.710
ll 15.770 15.725 15.730 15.380 15.380 Damaged 11
15.765 15.720 15.725
t2 15.800 15.780 15.780 15.440 15.440 15.400 15.390 15.390 15.400 15.400 12
15.770 15.770 15.430 15.430 15.395
13 15.820 15.785 15.720 15.400 15.400 15.390 15.380 15.390 15.410 15.420 13
15.810 15.780 15.710 15.390 15.390
Fig. 6. Microstructure of tube sheet material.
a b e - t o - t u b e s h e e t w e l d r e g i o n was also f o u n d to be q u i t e n o r m a l with r e s p e c t to its
l i c r o s t r u c t u r e (Fig. 8).
Metallographic examination of the defective region of the tube sheet material
e v e a l e d e x t e n s i v e c o r r o s i o n a t t a c k (Fig. 9). E x a m i n a t i o n o f t h e a f f e c t e d r e g i o n s o f
ae stainless steel t u b e i n d i c a t e d u n i f o r m t h i n n i n g (Fig. 10).
.5. Corrosion testing
3.5.1. Electrochemical potential measurements. O p e n circuit p o t e n t i a l s w e r e m e a s -
r e d f o r b o t h the t u b e s h e e t a n d t h e t u b e m a t e r i a l s in t h e p r o c e s s s o l u t i o n ( c o n t a i n i n g
170 RAKESH KAUL et al.
Fig. 7. Dendritic microstructurc of stainless steel overlay.
Stainless
steel
overlaid----
p o r t i o n of
t u b e steel
Weld
Stainless
steel
tube
Fig. 8. Tube-to-tube sheet weld.
Fig. 9. Damaged region on inside surface of tube sheet hole.
Failure analysis of carbonate reboiler heat exchangers 171
!
2oo
Fig. 10. Thinning from outside surface of stainless steel tube.
:he vanadium compound) at the boiling temperature. The respective potential values
!or the tube sheet and tube materials with respect to a saturated calomel electrode
zeere found to be - 5 2 0 and - 5 3 5 mV. Very little difference in the open circuit
~otentials for both materials indicates that the surface films formed on both the tube
;heet and tube materials in the process solution environment were similar.
3.5.2. Galvanic current measurements. Galvanic current measurements were made
~etween the tube sheet and tube samples (with different area ratios) in the process
;olution (containing the vanadium compound) at the boiling temperature using a zero
•esistance ammeter. The tube sheet to tube area ratios selected for testing were 1:1,
!:1 and 3:1. In all three cases, an almost zero galvanic current was recorded. This
)bservation again confirms that the surface films formed on both the tube sheet and
ube are of the same nature, and that the galvanic corrosion of the dissimilar metals
vas unlikely under normal circumstances.
3.5.3. Anodic polarization tests. Anodic polarization curves, obtained for both the
abe sheet and the tube materials in the process solution (with and without the
anadium compound) at 353 K, are presented in Figs 11 and 12, respectively. These
esults clearly indicate a lower breakdown potential for both materials in the process
~lution without vanadium compound. In other words, it can be inferred that, in the
bsence of vanadium compound, surface films formed on both materials are com-
aratively less stable and can result in localised corrosion.
3.5.4. Crevice corrosion tests using multiple crevice assembly. Multiple crevice
ssembly tests were performed on the tube sheet material in conjunction with tube
laterial as well as with Teflon in the process solution (containing the vanadium
3mpound) at 353 K. For this purpose, two multiple crevice assemblies were
tbricated as per A S T M standard G-78 [1] to obtain crevice assemblies of tube
leet/tube material and tube sheet/Teflon. No crevice attack was noticed on any of
lese two crevice assemblies after 10 days of exposure. Samples were repolished and
te crevice assemblies were re-exposed for 30 days. Metallographic examination of
te samples after this 30-day-long test indicated initiation of crevice corrosion attack
Lthe form of small pits under the crevice sites. Figure 13 shows initiation of crevice
)rrosion attack on tube sheet material in the tube/tube sheet assembly.
172 RAKESH KAUL et al,
950
WITH V20 s
LLI
U')
WITHOUT V20 s --~
.J_..%
E
l,-
450
Z
LU
I
I
0
Q_
LLI
O
0
-50~ /
rY /
I.--
UJ
._.i
W
[
I
-550-]F- -- T ~ T- -T I - - r - -
J
0.01 0.1 1 10 100 1000 10000
CURRENT DENSITY. pA/cm z
Fig. 11. Anodic polarization curves of tube sheet material in process solution at 353 K.
950
WITH yzOs
L¢)
J
E
450-
WITHOUT V20~
Z
LLI
I,--
0
0_
IJJ
a [ /
0
ILl
--J
tJ..l
-550 j | I 1 1 t I
0.01 0.1 1 10 100 1000 10000
CURRENT DENSITY. pA/cm 2
Fig. 12. Anodic polarization curves of tube material in process solution at 353 K.
4. D I S C U S S I O N
The failure of the two heat exchangers involved: (i) damage to the tube sheet in the
region close to the shell face (initiated from the inside surface of tube sheet holes),
and (ii) localized thinning of stainless steel tubes which coincided with the damaged
region on the tube sheet. Metallographic examination of both the tube sheet and tube
materials, including the weld regions, showed normal microstructures.
Dye penetrant examination indicated non-uniform tube-to-tube sheet gaps. Precise
Failure analysis of carbonate reboiler heat exchangers 173
100 gtm
Fig. 13. Initiation of crevice corrosion attack on tube sheet material in tube/tube sheet
assembly.
Leasurement of the inside diameters of the tubes established that expansion of the
tll length of the tube inside the tube sheet, a requirement of the specification, had
9t been carried out. The length of the expanded portion of the tubes (from the
elded face of the tube sheet) was found to be in the range of 30-45 ram, while the
',maining portion of the tubes had been left unexpanded. In some tubes, the length
! the expanded portion was found to be less than 20 mm.
Partial expansion of the tubes into the tube sheet left undesirable tube-to-tube
teet gaps on the shell side, the extent of which varied with local net dimensional
triations in the internal diameter of the tube sheet hole and outside diameter of the
bes. These gaps acted as the ideal crevice sites for crevice corrosion attack.
The electrochemical and crevice corrosion tests, carried out as part of the failure
vestigation, indicated that the material is susceptible to crevice corrosion attack in
e process solution. Although the crevice corrosion tests were not accelerated tests
~e to the experimental limitation of the test temperature and duration of exposure,
ey do indicate that, in the presence of a sufficient crevice gap, an active crevice site
developed after a certain incubation period, where the bulk solution cannot
-establish the passivation action of the vanadium compound resulting in increased
rrosion attack on the tube sheet material. This leads to enhanced concentration of
rrous ions in the crevice volume, which raises the electrode potentials of both the
174 RAKESH KAUL et al.
tube sheet and the tube materials in the anodic direction, thereby resulting in
dissolution/thinning of the stainless steel tubes and the tube sheet material.
The crevice corrosion, once initiated, would have progressed rapidly by cathodically
protecting its surrounding surfaces. U n d e r these conditions, the thickness of the tubes
is continuously reduced to the extent where it could no longer sustain the internal
pressure, thereby resulting in the rupture of the tubes. The jet of pressurised process
gas coming out of the ruptured tubes could have contributed to the enhanced damage
of the tube sheet by erosion. Corrosion failure of a carbon steel c o m p o n e n t due to
the decreasing vanadium concentration in the potassium carbonate solution has also
been reported elsewhere [2].
5. C O N C L U S I O N
In the light of the evidence gathered during the course of the investigation, the
cause of the failure of the two heat exchangers can be attributed to the partial
expansion of the tubes into the tube sheet which left undesirable tube-to-tube sheet
gaps. These gaps subsequently acted as crevices and led to localized corrosion of the
tube sheet as well as the tubes.
6. R E C O M M E N D A T I O N S
In order to avoid similar failures in future, the following actions are recommended:
(1) The tubes inside the tube sheet should be expanded to the designed total length
of the tubes.
(2) During the final inspection of the fabricated heat exchangers, full expansion of
the tubes inside the tube sheet should be ensured by m e a s u r e m e n t of the internal
diameter of the tubes.
Acknowledgements--The authors are thankful to Dr Placid Rodriguez, Director IGCAR, for his constant
support and guidance during the course of the investigation. The authors are also thankful to the Quality
Control & Engineering Section of IGCAR for help in carrying out dye penetrant testing and dimensional
measurements. The authors wish to thank the plant authorities concerned for providing the relevant
material and information for this failure investigation.
REFERENCES
1. ASTM standard G 78-89, 1992 Annual Book of A S T M Standards, Vol. 03.02, pp. 322-327.
2. J. J. Johnson, in Proceedings o f Corrosion 87, National Association of Corrosion Engineers (1987).