Presented at NACE Corrosion 2009, Paper 09577

Atlanta, GA, March 2009.

CORROSION ASPECTS FOR STAINLESS STEEL SURFACES

IN THE BREWERY, DAIRY AND PHARMACEUTICAL SECTORS

Troels Mathiesen, Jan Elkjaer Frantsen

FORCE Technology
Park Alle 345
DK-2605 Brndby, Denmark
[email protected]

ABSTRACT

Based on practical cases, tests and available literature the paper reviews the relation between
surface condition and corrosion properties of stainless steel used in breweries, dairies and
pharmaceutical processing plants. Aspects related to specification, corrosion testing and post-
treatment (e.g. pickling and passivation) are also discussed.

Keywords: Corrosion testing, CPT, pitting potential, standards, accept criteria, surface
roughness, post-treatment, sensitization, brewery, dairy, pharmaceutical.
 INTRODUCTION

Stainless steels for processing plants in the brewery, dairy and pharmaceutical sectors are
normally specified as either AISI 304 or AISI 316L type material. When supplied from the steel
mill, the material is documented with an inspection certificate (e.g. EN 10204-3.1), which
ensures that the most crucial parameters affecting corrosion resistance are fulfilled. This
implies spectral analysis of the chemical composition as well as intergranular corrosion testing
of the heats to ensure that the microstructure is healthy. The main corrosion form in question
for the product side is localized corrosion such as pitting and crevice corrosion, since the
equipment is exposed to chloride-containing and potentially aggressive products and cleaning
agents.

To a great extent the steel certificate ascertains that the basic corrosion properties of the
stainless steel are met. For the brewery, dairy and pharmaceutical sectors a well documented
surface condition of the supplied steels is required as well to meet the requirements for
hygiene and cleanability as discussed in another paper /1/. The surface condition is as
minimum characterised by the manufacturing route /2/ but normally implies additional
requirements to the surface roughness (Ra).

The subsequent handling of the semi-finished products during fabrication always involves risks
of compromising the corrosion resistance unless the basic guidelines for handling stainless
steel are followed strictly. For instance, the surface finish may be contaminated or damaged
during cold forming. Moreover, hot forming or subsequent heat treatments may alter the
microstructure, and likewise heat tinting or geometrical defects may be introduced during
welding. Although these issues are well-described in the relevant standards, disputes occur
frequently where the consequences and measures for re-establishing the corrosion resistance
are questioned.

The paper reviews the most common corrosion-related issues that are encountered during
commissioning of processing plants for the brewery, dairy and pharmaceutical sectors. As a
material testing institute, FORCE Technology is often involved in such cases, either as an
independent third party or when contracted by one of the companies. Consequently, the
following review is mainly based on our experience from practical cases, but tests and
available literature are included as well. In cases where fabrication defects compromise the
corrosion resistance, it is our intention to illustrate the consequences and discuss available
methods for re-establishing the surface.

STANDARDS TO SPECIFY AND RE-ESTABLISH SURFACE REQUIREMENTS

Besides from choosing the correct stainless steel grade providing sufficient corrosion
resistance, it is equally important to specify the right surface condition of materials used in
brewery, dairy and pharmaceutical sectors. This parameter obviously influences the
cleanability and hygienic properties of the equipment. Moreover, the surface condition affects
the corrosion resistance to a certain extent, which means that requirements may possibly be
met by specifying the proper finish rather than upgrading the chosen alloy.

Several standards are available for specifying the surface requirements of stainless steel
equipment. The ASME-BPE standard /3/ and ISPE baseline guides are widely applied for
pharmaceutical plants. For the dairy and brewery sector the 3A and EHEDG guidelines are
commonly used but individual company standards are also frequently used. Along with these
standards, international standards are available on specific items like tubes (e.g. DIN 11850
and DIN 11866).

Depending on the specific use of the equipment, the specified finish ranges from 2B/2D
(pickled) to mechanically polished and electropolished finish. The typical minimum
requirements for the equipment are listed below, but great care should always be taken when
specifying and interpreting such criteria:

- Surface roughness, Ra 0.6 m

- Metallic bright and clean surface
- Surface defects are only acceptable to a specified level
- Final passivation

The interpretation of the surface roughness is thoroughly discussed in another paper /1/.
Likewise, the allowable degree of local surface defects is often an issue for dispute. The
ASME-BPE standard /3/ provides a fairly clear specification of this, whereas other standards
often lack relevant details. One example is the DIN11850 standard concerning stainless steel
tubes for the food industry. This standard refers to EN 10217-7 regarding technical delivery
conditions, e.g. allowable surface imperfections. When read literally, this standard allows
manufacturing-related surface imperfections up to a depth of the specified minimum wall
thickness, which is 90 % of the nominal wall thickness. To avoid any subsequent discussions,
it is recommended to make additional specifications for the allowable surface defects.

Regarding the term passivation, reference is normally made to the ASTM A380 and A967
standards. For electropolished surfaces, the ASTM B912 has recently become available.
According to these standards, it is no longer compulsory to perform a strong nitric acid
passivation that actively builds up the protective oxide film and improves the Cr/Fe ratio of the
outer layers. Today, the passivation treatment may be performed with milder non-oxidizing
acids like citric acid as long as it leaves the surface perfectly free from inclusions and metallic
contaminants (like iron) that otherwise might hinder the spontaneous passivation process of
the stainless steel.

Although the mentioned standards provide a good basis for obtaining the desired corrosion
resistance, they seldom provide specific methods or acceptance criteria for assessing the
corrosion resistance of the finished product.

METHODS TO CHARACTERISE CORROSION RESISTANCE

The possibility of on-site non-destructive testing methods is limited for directly measuring the
obtained corrosion resistance of the finished equipment. Assessment of the pitting resistance
of stainless generally requires a semi-destructive test that generates pits in the material either
by electrochemical polarisation or a temperature ramp test. Consequently, this testing
approach has never found wide use for on-site evaluation, since it would require subsequent
repair of the tested area.
Simpler and non-destructive tests are available for obtaining an indicative measure of the
corrosion resistance. The ASTM A380 and ASTM A967standards describe several test
methods for this purpose, including the water wetting and high-humidity tests for detecting
highly susceptible areas as well as the ferroxyle and copper sulphate tests for detecting free
iron. Short-term salt spray testing (2 hours) may also be used to evaluate the condition. Such
tests will effectively reveal any iron contamination or local areas with equally low corrosion
resistance that present a possible risk of initiation sites for pitting. However, the tests do not
provide a sufficiently fine graduation to reveal defects that are related to geometrical surface
issues or sensitization of the metal.

Another non-destructive approach is to measure the corrosion potential of the surface in a well-
defined test solution. This method is available as commercial instruments under the names
Oxilyser or Passivation Tester. A micro electrochemical cell (ec-pen) for measuring pitting
potentials in very small areas (1.5 mm2) is also available as a commercial instrument. Although
both techniques can provide valuable data, a thorough basis for establishing clear acceptance
criteria with these methods is still desirable.

In extreme cases, where sensitization is suspected as a result of improper heat treatment (e.g.
during welding) there is a range of test methods available for evaluating the risk of
intergranular corrosion. The oxalic acid etch test according to ASTM A262 can be made on-site
with portable polishing and etching equipment. The polished and etched microstructure is
either evaluated on-site with microscope or, alternatively, a replica cast is taken for later
examination in the laboratory.

Another way of quantifying the degree of sensitization can be obtained using the ASTM G108
standard test method based on Electrochemical Reactivation (EPR). This technique has also
been adapted to provide on-site evaluation /4/.

The simplest approach to evaluate the presence of sensitized areas (e.g. in HAZ) is to perform
a mild pickling treatment of the welds or suspected surface. Any sensitization will be revealed
as severely etched areas that can easily be detected by a specialist.

We have experience with most of the above methods from practical cases in the brewery, dairy
and pharmaceutical sectors. To evaluate how fabrication-related surface issues affect pitting
resistance we have applied two main approaches.

Provided the fabrication procedures and surface treatments are well documented, a
reasonable estimate of the obtained corrosion resistance can be based on surface
characterisation and corrosion data (see next section). Non-destructive assessments may be
applied to characterise the surface topography better, e.g. by surface roughness
measurements or replica castings of the surface for subsequent examination in a scanning
electron microscope. When this information is combined with the known effects of surface
finish and chemical post-treatment, a fairly good estimate of the resulting corrosion resistance
can be established.

When disputes get stuck and harder evidence is needed, we have occasionally tested samples
cut from the plant to verify the corrosion resistance. This approach may seem as a last resort
but on the other hand it provides the answers quickly without too much delay in the project. As
an example, such disputes may arise from various discolouring phenomena (possible heat
tinting) that represent borderline cases according to common standards.

In order to test against well-accepted criteria, we apply standard methods whenever possible.
The ASTM G150 CPT method is perhaps the only standardized electrochemical method that
allows comparison of pitting data between testing laboratories, Figure 1. As the standard
contains extensive statistical data for grades AISI 316L and better, it is clear what to expect
from a healthy stainless steel. However, the harsh testing conditions of the G150 method (1 M
NaCl and -700 mV SCE) do not allow testing of the AISI 304 material, which is frequently used
for dairies and breweries.

The conditions of the G150 test are furthermore far from the typical service conditions in food
and pharmaceutical processing plants. In order to test under conditions closer to this level, the
pitting potential is determined by using cyclic polarisation in milder chloride solutions. Based on
ASTM G61 standard, this technique has been further improved to allow testing of plate and
tube sections without cut-faces and with "crevice-free" mounting /5,6/. This ensures a high
degree of reproducibility. The previously published work has formed a good basis for
evaluating the corrosion resistance of AISI 304 and AISI 316L materials in practical cases
where the quality was questioned.

100
AISI 316
1200
SAF 2205
80
Electropolished
904L
254SMO
1000 Pickled
Ground
800 As-extruded
60
CPT (C)

600

40 400

200
20
+700 mV SCE
0
1 M NaCl
1C/min
0 -200
20 25 30 35 40 45 0 20 40 60 80 100
PRE (=%Cr x 3.3%Mo x 16%N) Temperature (C )

FIGURE 1 Relationship between alloying and FIGURE 2 - Relationship between pitting potential
CPT determined with ASTM G150 (1 M NaCl, and temperature of different surface conditions in
+700 mV SCE). 5 % chloride solution /5,6/.

DEPENDENCE BETWEEN SURFACE CONDITION AND CORROSION RESISTANCE

The literature provides several data about mechanical and chemical surface treatments in
respect to corrosion resistance. However, the way data is obtained varies due to the lack of
standardized testing techniques as discussed above. This makes direct comparison of different
datasets difficult.
The results of a project on the influence of various surface conditions on pitting resistance
have previously been published /5,6/. Despite the very different surface topographies, tubes
with ground, extruded and pickled finish came out with nearly the same pitting resistance when
tested in passivated condition, Table 1. As expected, the electropolished finish showed
superior pitting resistance in comparison to this. The study included both pitting potential
determination at various chloride levels (Figure 2) and CPT-testing according to ASTM G150.

TABLE 1.
Comparison of critical pitting temperatures (CPT, C) of AISI 316L tubes (60-63 mm)
obtained by using either polarization technique or ASTM G150 /5,6/.
CPT Polarization ASTM G150
in C Cl-, mg/l Cl-, mg/l
Surface condition Ra, m 500 5,000 50,000 35,500
Ground, 0.19 14 9 6 72
Extruded 5.4 13 <10 <10 11 4
Pickled 0.48 17 14 9 14 5
Electropolished 0.12 65 45 45 57 13

In comparison to the surface finish, the weld quality is more decisive for the corrosion
resistance. It has earlier been shown that the oxygen content of the shielding gas strongly
influences the pitting resistance of standard stainless steel grades, Figure 3. At 20-30 ppm
oxygen in the shielding gas the pitting resistance of AISI 304 is significantly impaired, whereas
AISI 316L and UNS 08904 tolerate slightly higher oxygen levels. These results have formed
the basis for our colour reference atlas for welds /8/ that is widely used to specify weld accept
criteria in the brewery, dairy and pharmaceutical sectors.

CPT

Oxygen
FIGURE 3 - Relationship between oxygen content FIGURE 4 Breakdown potentials as function of
in shielding gas and pitting temperature of TIG temperature of 904L with two different surface
welded AISI 304 /7/. finishes in 1 M NaCl /11/.
Several papers in the literature support the above statements. Ericsson et al have published
pitting potential data for AISI 316 that emphasizes the importance of pickling subsequent to
abrasive treatment in order to improve resistance /10/.

Moayed et al have shown that the critical pitting temperature (CPT) of 904L increases with
increasing surface smoothness obtained by abrasive treatment, Figure 4.

Data based on exposure in ferric chloride (ASTM G48) show a similar tendency, Figure 5. The
beneficial effect of decreasing surface roughness has furthermore been proven by quantifying
the nucleation rate of metastable pitting and electrochemical noise measurements /13,14,15/.

Another approach has been to characterize the dependence between chemical composition of
the passive layer (i.e. Cr/Fe ratio) and abrasive and chemical surface treatment /16,17/. It is
shown that common passivation treatments provide only temporary improvement of the
passive film. In few days the passive film adjusts itself and reaches an equilibrium state that
mostly depends on the environment /17/.

Based on the literature as a whole, it is commonly accepted that electropolished finish provides
better resistance than pickled finish, whereas mechanically ground or polished finishes fall in
the least corrosion resistant group. Presumably, this ranking can be related to the resulting
amount of free inclusions and surface roughness related to the different surface treatments.

A wide range of data is also available for stainless steel surfaces exposed to atmospheric
conditions /18,19,20/. Generally the same ranking is reported as for the fully immersed
exposure conditions above. However, other parameters are more decisive here, such as
wetting properties and the ability to collect dirt.

FIGURE 5 Critical pitting temperature in ferric chloride (per ASTM G48) in base metal, HAZ and weld
areas of UNS N08926 with different surface finishes /12/
 CORROSION RELATED DEFECTS ENCOUNTERED DURING COMMISSIONING

As a material testing and consulting institute we often see the worst cases when it comes to
fabrication defects that are observed during final commissioning. The key questions are
possible consequences and remedies to re-establish the surface.

A typical defect type is mechanical scratches or tool marks like those shown in Figure 6. Since
this kind of defect often involves micro crevices it is unacceptable in an industry where
cleanliness is a key issue. Besides from compromising the cleanability, the crevice significantly
lowers the resistance against chloride-induced corrosion. Depending on the severity of the
micro cavities and the accessibility the surface, different methods may be considered for re-
establishing the surface. Mechanical grinding to remove the defect is the first choice. However,
on-site grinding should always be carefully planned to ensure complete removal of the grinding
dust, since this could otherwise introduce a secondary problem.

Grinding should preferably be followed by pickling and passivation in order to dissolve any
micro slag particles uncovered by the grinding process. Pickling without prior grinding might be
considered if it can be demonstrated (e.g. by corrosion testing) that the cavities are "opened",
and thereby made less susceptible to corrosion as a result of the chemical treatment.

Iron contamination is another frequent issue that may occur as a result of tool marks or
grinding dust in the fabrication area, Figure 7. Except from the formation of rust, the stainless
surface is usually unaffected unless the environment is simultaneously highly contaminated
with aggressive compounds like chloride. In that case micro pits may be observed in the
surface. In most cases an acidic decontamination is sufficient for removal of the iron and rust
products.

FIGURE 6 Mechanical scratches on stainless FIGURE 7 Iron contamination from airborne

steel surface. grinding dust on the exterior of tubing.

It has been shown that severe heat tints from welding have a negative influence on the pitting
resistance of stainless steel, Figure 8. This is further confirmed by numerous corrosion failures
where through-wall pitting developed in heat tinted welds or HAZs after a short exposure
period. For this reason it is common practice to specify maximum allowable heat tint levels of
welds according to colour reference atlases /8,9/. When proper shield gas protection has been
applied during welding, there is usually no need for additional post-treatment. In cases where
severe heat tinting occurs and the welds are free from geometrical defects (like incomplete root
penetration) pickling may be applied. However, a quality level that obviates pickling should
always be aimed at, since this treatment affects the surface roughness and may be quite
complicated to perform in a fully completed plant.

FIGURE 8 Severe heat tinting in circum- FIGURE 9 Rust spot on the inside of a
ferential weld as seen with endoscope. sensitized tube.

When sensitization is observed, it is usually a result of inadequate degreasing prior to welding.

However, in few cases this phenomenon has also been ascribed to improper procedures
during casting or fabrication of e.g. tubes from semi-finished plate products. Figures 9 and 10
show extreme examples of longitudinally welded tubes that were severely sensitized, possibly
due to incomplete lubricant removal prior to annealing during tube fabrication. The problem
was recognised as rust and soot contamination during the final endoscopy of the
circumferential tube welds. Metallographic cross sections and corrosion testing according to
ASTM A262 revealed sensitization to a depth of 50-100 m along the inner tube surface,
Figure 11. On this basis, there was no other possibility than replacing all the tubes in order to
avoid the possibility of intergranular corrosion. Pickling was not an option given the depth of
the affected material.

100 m

FIGURE 10 Soot contamination on the inside of FIGURE 11 Cross section along inner side of
a sensitized tube. sensitized tube. El. oxalic acid etched.
The above examples of fabrication defects undoubtedly have a negative influence on the
corrosion resistance if not dealt with before final commissioning. However, we seldom observe
such defects as the prime cause for corrosion failures in systems operating at the nominal
service conditions. In most cases failures are either related to improper conditions during
hydrotesting or unforeseen changes towards more aggressive service conditions, e.g. higher
temperature. Although some of the mentioned defect types may appear insignificant in this
respect, there is no doubt that such defects lower the tolerance against corrosion and
compromise the hygienic requirements.

CONCLUSIONS

Stainless steels for processing plants in brewery, dairy and pharmaceutical sectors are
normally specified as either AISI 304 or AISI 316L type material. When supplied from the steel
mill, the material is documented with a certificate. This certificate ensures the fulfilment of the
most crucial parameters affecting corrosion resistance, such as chemical composition,
microstructure and surface condition. During subsequent fabrication of the equipment there are
several risks of compromising the corrosion resistance unless the basic guidelines for stainless
steel are strictly followed. The paper has reviewed the most common aspects that might impair
the corrosion resistance of the final equipment. On this basis, a number of important issues to
ensure optimum corrosion resistance have been summarised below:

- Carefully specify the requirements for weld and surface quality according to common
standards such as ASME-BPE. Ensure that the standards include possible defect types or
alternatively specify additional requirements.
- Preserve the original surface as far as possible during fabrication.
- Plan the fabrication process to minimize the amount of subsequent mechanical grinding.
- Specify a maximum allowable heat tint level of welds in order to minimize or avoid the need
for post-treatments like grinding and pickling.
- Ensure complete draining after hydrotesting or, alternatively, use high purity water for this
purpose.
- Always perform a final decontamination (or passivation) to remove particulate
contaminants, but do not rely on this treatment as a measure that removes coarse defects
related to surface finishing and welding, or provides a permanent improvement of the
corrosion resistance.

REFERENCES

1. Jan Elkjaer Frantsen, Troels Mathiesen, Specifying Stainless Steel Surfaces for the Brewery,
Dairy and Pharmaceutical Sectors; Paper 095373, NACE Corrosion 2009.
2. ASTM A 480 / A 480M - 01; Standard Specification for General Requirements for Flat-Rolled
Stainless and Heat-Resisting Steel Plate, Sheet, and Strip.
3. ASME Bioprocessing Equipment, ASME BPE-2007.
4. Jargelius R. F. A. et al.; Evaluation of the EPR technique for measuring sensitization in type 304
stainless steel; Corrosion, 1991, vol. 47, no 6, pp. 429-435.
5. Troels Mathiesen, Torben S. Nielsen, Jan Elkjaer Frantsen, Influence of various surface
conditions on pitting corrosion resistance of stainless steel tubes type EN 1.4404, Paper 06095,
NACE Corrosion 2006.
6. Torben S. Nielsen, Troels Mathiesen, Jan Elkjaer Frantsen, Pitting corrosion resistance of
electropolished seamless stainless steel tubes type EN 1.4404, Paper 07194, NACE Corrosion
2007.
7. J.V. Hansen, E. Maahn, The effect of oxygen level in the shielding gas on the root side corrosion
resistance of welded stainless steels; Proc. 11th Scandinavian Corrosion Congress NKM 11,
Stavanger 1989.
8. Reference colour charts - for purity of purging gas in stainless steel tubes, FORCE Technology
report 94.34.
9. Specification for welding of austenitic stainless steel tube and pipe systems in sanitary (hygienic)
applications, American Welding Society, AWS D18.1:1999
10. Ericsson, R.; Schn, L. og Walln, B.; Effect of some oxide removal treatments on the corrosion
resistance of 316 stainless steel; 8th Scandinavian Corrosion Congress. Proceedings, Vol. 1 pp.
321-329; 1978.
11. Moayed M.H.; Laycock N.J.1; Newman R.C.; Dependence of the Critical Pitting Temperature on
surface roughness; Corrosion Science, Vol. 45, No. 6, June 2003, pp. 1203-1216(14).
12. Tuthill, A.H. og Avery, R.E.; Specifying stainless steel surface treatment; NIDI Technical Series
No. 10068; 1992.
13. Yu Zuo, , Haitao Wang and Jinping Xiong; The aspect ratio of surface grooves and metastable
pitting of stainless steel; Corrosion Science, Vol. 44, Issue 1, January 2002, pp. 25-35.
14. G.T. Burstein and P.C. Pistorius, Surface Roughness and the Metastable Pitting of Stainless
Steel in Chloride Solutions; Corrosion, May 1995, pp. 380.
15. A. Burkert, K. Schilling, A. Heyn; Einfluss der Schleifbehandlung auf das Korrosionsverhalten von
Chrom-Nickel-Sthlen; Materials and Corrosion, Volume 55, Issue 10, pp. 787 793.
16. JC Tverberg; Conditioning of stainless steel surfaces for better performance; Stainless Steel
World, April 1999, pp. 36-41.
17. Lena Wegrelius and Birgitta Sjdn, Passivation Treatment of Stainless Steel; ACOM 4-2004,
Outokumpu Stainless AB.
18. Karlsson, A. og Olsson, J.; Atmospheric corrosion of stainless steel; 7th Scandinavian Corrosion
Congress, Trondheim, pp. 71-86, 1975.
19. D. Wallinder, I. Odnevall Wallinder, and C. Leygraf; Influence of Surface Treatment of Type 304L
Stainless Steel on Atmospheric Corrosion Resistance in Urban and Marine Environments;
Corrosion, vol. 59, nr. 3, pp. 220-228, 2003.
20. Houska, C.; Stainless Steels in Architecture, Building and Construction; Reference Book Series,
No. 11024. Nickel Development Institute (NiDI), 2001.