dr inż. Jerzy Stanisław Kowalski

dr inż. Marek Mazur

dr hab. inż. Wojciech Wojciechowski

Institute of Materials Engineering

Cracow University of Technology

Pad welding as a means of protection against corrosion

Abstract

The article presents the possibilities that the arc pad welding creates in protecting from corrosion various materials used in construction of equipment operating in power industry. By application of pad welding, a coating (layer) is deposited which not only protects the material from corrosion caused by the flow of more or less aggressive media (e.g. water, waste gases) but also prolongs the life of pad welded elements since padding welds are usually characterised by high resistance to abrasive wear. Pad welding is specially suitable for repair of parts damaged or suffering standard wear and tear during operation in water installations, power supply systems, etc.

INTRODUCTION

Failure of pipelines handling industrial water to power plants or water from abyssal strata is in the majority of cases caused, apart from the „human” factor, by corrosion and erosion. This is due to the fact that water transported by pipelines contains dissolved gases (including gases that speed up the corrosion process), mineral compounds (e.g. calcium carbonate, chlorides of sodium, calcium, or magnesium), and organic compounds. Additionally, water from the subterranean springs often contains the acidic sodium carbonate, carbon dioxide, sulphates and chlorides in concentrations much higher than the water in surface layers. Apart from the effect of purely chemical nature, this also influences the pH value and the electrode process that takes place at cathode spots with reactions of hydrogen or oxygen depolarisation. The processes of erosion that take place on the internal surfaces of steel pipes are accelerated by the water-carried particles of undissolved quartz and argillaceous minerals which abrade the deposits growing on the internal pipe walls. The next factor that makes the pipeline degradation process more intense is the water flow rate. It is generally assumed that increasing the rate of water flow usually speeds up the corrosion process, caused by the corrosion products washed out in a strong jet of the flowing liquid. In the latter two cases, the processes of corrosion and erosion are observed to proceed at a rate definitely higher in these sections of the pipeline where there are some elements of the technical infrastructure restricting the free water flow, like reducing pipes, orifices and, particularly, elbows (bends).

An important factor that affects the service life of pipelines is the temperature of the handled water. This refers to both pipelines supplying water to power plants and those which serve for drawing out and handling of abyssal and geothermal waters. In the latter case, of particular importance is the occurrence of temperature gradients that favour the formation of corrosion cells.

The above mentioned factors responsible for the corrosive and erosive degradation of water pipelines mainly depend on the properties of the handled water, and the tools that the technical supervising personnel have to control the processes of corrosion are in this case very limited. The possibilities to control these phenomena are definitely greater at the stage of designing and selecting materials for the pipeline construction, specially as regards the use of effective methods for intraoperational upgrading. Failure of pipeline is always a serious problem in terms of both material and economy. The repair requires large financial outlays for replacement of the damaged section and the necessity of arresting the operation of the whole pipeline during the repair. There are known various methods to reduce losses caused by the erosion of water pipelines. The choice of the best method depends on detailed analysis of the pipeline operating conditions. In most cases, the aim of all the methods applied currently is to prolong the life of the structural elements of a pipeline using for their construction clad plates, steel plates after thermal-chemical treatment, plates surface etched with acids containing corrosion inhibitors, or plates coated or painted with anticorrosive agents. The sections of the pipelines operating under the most trying conditions of the corrosion-threatening environment are made from the very expensive stainless steels, and in some, economically justified, cases from the titanium-based alloys. Nevertheless, modern technology has no efficient technical means to totally eliminate the corrosive wear and tear of water pipelines. Therefore, the problem of current repair of these elements is very important. Effective and relatively cheap means of upgrading the pipeline sections damaged by corrosion and erosion offers the technique of pad welding. The said technology is very effective in making up for the losses of material and filling the pittings in parent metal, due to the application of a layer of proper material selected together with the welding electrode. Additionally, in the case of e.g. plumbing fittings fabricated by the technique of powder metallurgy, pad welding ensures the removal of surface porosities present in these products. Another advantage of the pad welding process is the possibility to use this method in repair of castings, specially at the places where, due to a long-lasting operation, the sub-surface defects, like pores and voids, have been detected.

    In the study presented here, a method of pad welding of the damaged surfaces of the steel plates has been proposed as a relatively simple and effective means for their upgrading, ensuring further failure-free operation of pipeline.

All over the world, corrosion is a serious problem. It destroys approximately 2 – 5 % of the gross national product, is responsible for the contamination of natural environment and creates numerous hazards to the human health and life. In 1989, in Poland, the losses due to corrosion and the related contamination of natural environment reached 32% of overall material losses. The effects of corrosion are present everywhere. Corrosion affects concrete, ferroconcrete, harbour facilities, industrial devices, military infrastructure and military equipment. As much as 80% of all financial means assigned for the repair of bridges and viaducts are as a matter of fact assigned for combating the effects of corrosion which attacks the industrial facilities and distribution systems.

CORROSION

Corrosion means all these processes during which metal or alloy used as a structural material undergoes, under the effect of environmental factors, a transformation from the metallic state to a chemically bounded state.

Corrosive reactions are not always harmful to the environment. The use of numerous reactive metals (e.g. titanium, niobium, tantalum, and rare earth metals) is possible in some environments due to the formation of a thin layer of corrosion products that forms a barrier impeding further progress of reaction.

Depending on the mechanism by which the corrosion process is proceeding, two basic types of corrosion are distinguished, viz. the chemical and electrochemical corrosion.

In aqueous solutions, the majority of corrosion processes take place by electrochemical route, which means that during reactions the electric charges are crossing the phase boundary. In terms of structure, the surface of metal is not homogeneous but consists of minute electrodes electrically shortcircuited by the metal. As long as the metal remains dry, the current is not passing and the corrosion does not take place. As soon, however, as it enters in contact with water solution or with moisture contained in the humid atmosphere, the local cells start acting and transform the metal into a product of corrosion. The impurities present in metal enhance the corrosion, because acting as microcells they exert a definitely stronger effect, and for this reason, pure metals usually corrode much more slowly than metals with some admixtures.

WELDING OF METALS vs CORROSION

The thermal processes that take place during joining of metallic materials cause different phase transformations which, to a smaller or greater extent, may occur within the heat affected zone. This is specially true in the case of welding, partly also in the case of pressure welding and soldering. The transformations change the mechanical properties and corrosion resistance of thus produced joints.

WELDING

Depending on the welding technique, the type of auxiliary materials and welding parameters, the welding process can change in a different way the physico-chemical nature of metals and alloys. The changes involve chemical reactions that proceed more quickly at higher temperatures and precipitation of compounds, like carbides, oxides, nitrides, sulphides, phosphorides, and intermetallic compounds. Moreover, during the process of welding, there is a recrystallisation of metal, the growth and deformation of crystal grains, sometimes the formation of new structural phases which move and agglomerate, giving rise to lattice defects and internal stresses of uni- or biaxial character, caused mainly by the shrinking power of the welded metal.

There is always some degree of inhomogeneity between the welded joint and parent metal. Its symptoms are differences in structure, mechanical properties and the state of stress inside metal. The inhomogeneity may enhance the corrosion rate, specially as regards the electrochemical and stress-induced corrosion. To eliminate or reduce this inhomogeneity, additional thermal or mechanical treatments are applied.

THE CORROSION OF WELDED STEEL JOINTS

In the majority of cases, the welded joints are less resistant to corrosion than the parent metal. Normally, all welded structures made from carbon and low-alloy steels are protected from the atmospheric corrosion with special coatings applied by painting. Quite often, however, it happens so that on the welded joints the varnish is spalling and falling off, exposing the bare metal surface, which is now running the greater risk of corrosion. To avoid spalling of paints and enamels, the welded joints and their neighbourhood should be cleaned mechanically before painting with, e.g. hot water, and dried.

The laboratory tests of the corrosion resistance of the joints welded from carbon and low-alloy steels are carried out in accordance with PN-76/H-04601 to 04604.

In welded structures operating in the aqueous media containing, e.g., alkaline solutions, soon after the installation has been introduced to use, the cracks of orientation transverse to the weld appear and are penetrating to the parent metal. It has been observed that corrosion of this type may take place only under the combined effect of mechanical stresses and chemical reaction of the hydroxide solution. Therefore, to arrest the development of corrosion, the welded joint should be stress-released by annealing at a temperature of 650^oC, or possibly by subjecting it to the thermal and mechanical treatment. The water preheaters in modern power boilers and pipe systems through which the aggressive media are flowing are composed of the pipe sections joined together by welding. Often in these joints some perforations appear due to the selective corrosion which usually occurs near the weld end.

To avoid corrosion of this type, it is recommended to make the girth weld with several runs instead of one single run, laying the runs in such a way as to make the beginning of the new run and the end of the previous one shift in respect of each other.

The more numeous are the runs, the thinner is the electrode, and the lower is the welding current, the more reliable is the joint. The welded joints in pipes expected to operate under the conditions of pressure should be, in principle, subjected to heat treatment.

For the construction of some pipelines, specially those handling potable water and water used by food industry, chromium steels of the 0H13, 0H13J, 0H17 and 0H17T grades are commonly used nowadays, though the austenitic 0H18N9 and 1H18N9 grades are also observed to gain popularity. These steels are usually welded with electrodes giving the weld metal of an austenitic structure in grades E 13 B 22, E 17 B 22, E 19 9 R 22 or E 19 9 B 22 (acc. to EN 1600). In some environments, the welded joints of these steels are susceptible to intercrystalline and stress corrosion.

If the welded joints in the above mentioned steels are expected to offer some resistance to the intercrystalline corrosion, they should be subjected to heat treatment which consists in annealing at a temperature of 800^oC for 2 hours (when the weld has been made with austenitic electrodes), or 4 hours (when a ferritic filler has been used), followed by slow cooling together with furnace to a temperature of 600^oC and rapid cooling in the air, next. The necessity to use so complex heat treatment seriously limits the range of application of these steels in construction of chemical apparatus, specially because of the grain growth, and hence higher brittleness of the material.

The risk of occurrence of the intercrystalline corrosion in a welded joint depends on the carbon content in welded steel and in metal deposited from the electrode used for welding; it also depends on the heat volume supplied in a time unit. To produce welded joints resistant to the intercrystalline corrosion it is recommended to use the steel of low carbon content or steel stabilised with titanium. If the joint should be deposited with several runs, the layer of the weld contacting directly the aggressive medium should be made with an electrode of low carbon content (0,02...0,03%C), or it should be stabilised with, e.g., niobium.

The knife-line corrosion attack is of an intercrystalline nature and it occurs specially in welded joints made from the chromium-nickel-molybdenum steels. Corrosion of this type occurs mainly at the welded joint-parent metal interface, and it may destroy in a relatively short time the whole joint. As preventive means it is recommended to use electrodes of low carbon content, possibly electrodes that will give the weld metal of a ferritic-austenitic type, or to apply a stabilising heat treatment..

The welded joints made in chromium-nickel and chromium-nickel-molybdenum steels with the commonly used electrodes are less resistant to stress corrosion than the parent metal. In some cases, the resistance of the welded joint to the stress corrosion is only 10% of the parent metal resistance.

PAD WELDING

Pad welding is the process of application of metallic coatings where a characteristic feature is remelting of the substrate material on which the padding weld has been deposited.

When for pad welding the material of properties different than the properties of parent metal is used, there are some side effects resulting from mixing of the base (pad welded) metal with the fused padding metal. The degree to which these two materials get mixed together is the greater, the greater is the depth of penetration of the padding weld into the base metal.

The content of carbon in the individual layers of a padding weld depends on the content of carbon in parent metal and in the filler, and also on the penetration depth. Therefore, to obtain the highest possible corrosion resistance, both parent metal and filler metal should have the lowest possible content of carbon with the highest possible content of elements, like chromium and nickel.

Among the numerous methods available, the smallest penetration depth into the pad welded metal gives the process of pad welding with shielded arc and strip electrodes, the highest – the pad welding in protective gas atmosphere with consumable electrodes, and with shielded arc and one wire electrode.

Pad welding with strip electrodes is widely used by industry in the manufacture of heavy-wall tanks (e.g. for the nuclear power industry), which should have a very high corrosion resistance in their inside part.

METHODS TO TEST THE CORROSION

The practical aim of corrosion testing is to determine the life of metals or alloys during performance under given conditions. The results of these tests should give an answer to the question in what way the metals or alloys will behave on performance.

At present, the following tests are conducted:

1. Laboratory tests under the conditions created artificially by means of various devices.

2. Field tests under actual operating conditions.

3. Laboratory tests carried out on real facilities and under the real operating conditions.

Numerous methods of corrosion measurement are available, but they all can be divided in qualitative and quantitative methods.

QUALITATIVE METHODS consist in observations of the object exposed to corrosion and are divided into:

a) direct (visual observations with unarmed eye)

b) macroscopic (observations under magnifying glass)

c) microscopic (observations under microscope)

d) evaluation of changes in mechanical properties

All these methods are characterised by great simplicity. The last method is very easy and handy in determination of the internal (hidden) corrosion.

QUANTITATIVE METHODS

These methods consist in the determination of:

a) the time when on a given surface the first corrosion spots appear and the corrosion spots count in a time unit;

b) the change in specimen thickness and depth of pits, or loss of weight;

c) increase of weight when the corrosion products are insoluble and stick to the specimen. If this is the case, the corrosion is expressed as an increment in weight in a time unit, e.g. g/m²·year.

d) mechanical properties during the tensile test (R_m and A), or technological properties during the bending test (the number of flexions). In this case the corrosion is expressed as a percent drop in the value of the tested property during one year.

Besides the tests mentioned above, there are also numerous other, often very complicated, tests, like the determination of the volume of the soluble corrosion products in a solution, of the volume of evolved hydrogen or absorbed oxygen, changes in electric resistance, changes in the thermal effect, changes in the ability to reflect light, etc.

OWN INVESTIGATIONS

TESTING OF THE PADDING WELDS CORROSION RESISTANCE

Testing of corrosion resistance was carried out on the steel specimens with and (for the sake of comparison) without padding welds.

The aim of the investigations was to evaluate the behaviour of pad welded elements (the pad welded specimens) under the performance conditions. To achieve this goal, simulation tests were carried out under the laboratory conditions.

Two methods of the evaluation were applied:

1) visual – the evaluation of changes in the appearance of the specimen surface,

2) evaluation of changes in the specimen weight and dimensions.

Evaluating the surface condition, it is necessary to evaluate the type and colour of the corrosion products, the severity of corrosive attack, and the symptoms of corrosion. The qualitative evaluation should indicate the type of corrosion, its character and if there are any local damages caused by the corrosion or not.

TEST MateriaL FOR INVESTIGATION OF THE CORROSION RESISTANCE OF PADDING WELDS

Corrosive environment: water; aqueous solutions of H₂SO₄, HNO₃, HCL,

NaCl, simulating the saturation of water from the abyssal or geothermal springs with the above mentioned compounds.

Basing on the reference literature and the results of own investigations it has been assumed that the 5% concentration of corrosive media is corresponding with great probability to the real conditions of operation of a steel (metal) structure, e.g. of a pipeline for handling of the abyssal and geothermal waters, taking into account the factors, like water temperature, flow rate, gas saturation ratio, and erosive wear during an uninterrupted operation for 25 years.

Table 1. Chemical composition of steel S235JR (St3S);
in accordance with a standard EN 10025: 1993

C	Mn	Si	P	S	Cr	Ni
0÷0,22	0÷1,1	0,1÷0,35	0÷0,05	0÷0,05	0÷0,3	0÷0,3

Table 2. Chemical composition of weld metal [%]

Designation	Type	C	Si	Mn	Cr	Ni	Mo
1	E19 9 B 22	0,07	0,3	1,2	19,5	9,0
2	OK16.31	0,03	0,6	1,1	19,0	12,0	2,5
3	OK 61.30	0,03	0,8	0,6	19,0	10,0	0,75
4	OK 12.51	0,1	0,9	1,5
0	No weld

Table 3. Corroding medium.

Designation	5% of water solution
S	H2SO4
N	HNO3
H	HCl
C	NaCl

TEST METHODS

Tests and investigations were carried out on the specimens of S235RJ steel. On some of the specimens, the padding welds were deposited with wire electrodes and with coated electrodes, following guidelines adopted in the research program. The padding welds were deposited semi-automatically under gas shielding and manually with coated electrodes (MAG proccess).

PARAMETERS OF THE PAD WELDING PROCESS

Pad welding under gas shielding:

Arc voltage - U₁=18,3 [V], Welding current - I₁=128 – 132 [A]

Wire feed rate - v_e=3,45 [m/min], Protective gases - CO_2, Argon

Manual pad welding with coated electrodes:

Electrode diameter - 4,0 [mm], Welding current: 114 [A]

PROGRAM OF RESEARCH

On S235JR steel plates, altogether 16 padding welds were deposited – four with each type of the electrode and wire, and then the pad welded specimens and the four specimens without padding welds were subjected to the attack of aqueous solutions of H₂SO₄, HNO₃, HCL and NaCl.

Table 3. Project of tests in accordance with the table 2 - 3

Weld metal	Corroding medium
	H2SO4	HNO3	HCL	NaCl
E 19 9 B 22	1S	1N	1H	1C
OK. 16.31	2S	2N	2H	2C
OK 61.30	3S	3N	3H	3C
OK 12.51	4S	4N	4H	4C
No weld	5S	5N	5H	5C

(Detali of the diagram above)

Fig. 1. Mass decrement of surfaced plate specimen; water solution H₂SO₄

(Detali of the diagram above)

Fig.2. Mass decrement of surfaced plate specimen; water solution HNO₃

(Detali of the diagram above)

Fig.3. Mass decrement of surfaced plate specimen;

water solution HCl

(Detali of the diagram above)

Fig. 4. Mass decrement of surfaced plate specimen;

water solution NaCl

THE RESULTS AND CONCLUSIONS

1. The corrosion resistance of the steel and padding welds depends on the type of the corrosive medium (the contamined water).

2. The pad welding with wire electrodes, which give the weld metal of the same properties as the common steel, makes no sense. As results from the plotted graphs, the corrosion resistance of such padding welds is lower than that of the steel itself.

3. The highest corrosion resistance had the specimens with padding welds deposited with an Autrod OK. 16.31 wire electrode containing 2,5 % Mo. The results almost identical were obtained when electrodes giving the weld metal of much lower Mo content were used, which is quite important from the economic point of view.

4. Slightly inferior results were obtained in the case of padding welds deposited with E 19 9 B 22 (ES 18-8 B) electrodes. The drop of corrosion resistance is due to the effects that accompany the reactions taking place between the components of the electrode coating (specially if they are of a basic reaction) and parent metal. An important factor improving this resistance is without any doubt the exact removal of slag originating from the melted electrode coating and periodical (according to the technical possibilities) cleaning of the padding welds or welded joints.

5. The greater losses – independent of the type of the padding weld – occurred when water was acidified with sulphuric acid and with hydrochloric acid, which is probably due to depolarisation and its type related to either oxygen or hydrogen.

6. It seems advisable to use the technology of pad welding in structures and facilities that should offer high corrosion resistance, specially as regards current repairs and emergency repairs (the latter ones being of major importance), instead of replacing the whole sub-assemblies. Pad welding is fully justified from the point of view of economy. It provides, moreover, some opportunities to reduce the volume of the accumulating waste parts.