1. Introduction.

With enhancement of technological processes intensity, requirements to improvement of instrumentation and equipment service durability and reliability, determined mostly by structural materials application, become stricter. High-alloyed steels and alloys with high corrosion resistance are of the most interest. Susceptibility to local corrosion in oxidizing media: intergranular corrosion (IGC) – in areas exposed to heating to 450^?? at welding and thermal treatment, knife-line corrosion – in the near-weld zones of welded joints, corrosion cracking – in the strained areas of structures, is the disadvantage of these steels and alloys.

Large volumes of XH28MDT-type alloys are used as a structural material for a number of rather corrosive media encountered at production of complex mineral fertilizers, sulphuric, phosphoric acids, etc. Among the range of alloys developed for sulphuric acid hot solutions (Table 1), the alloy 03XH28MDT is the most corrosion-resistant and recommended for the sulphuric acid of any concentration at the temperature not exceeding 80^??.

Table 1

Structural materials for oxidizing media

However, even at a very low carbon content (0,02%) such type alloys at certain conditions disclose susceptibility to IGC. Metal exposure in the critical temperature interval region over 3-10 minutes, depending on the quenching temperature, may alone cause development of this type of corrosion in operating conditions. Rising of the quenching temperature to 1300^?? results in reduction of resistance to IGC. The XH28MDT-type alloys reduced resistance to IGC in a number of media is explained not only by the high nickel contents, improving carbon’s thermodynamic activity rating in the solid solution, but also by unstable titanium carbides passing into the solid solution at welding [2].

Methods and strictly regulated accelerated corrosion testing technique are the integral part efficient and highly informative means of the most corrosion-resistant materials rapid search and selection. It is necessary to carry out assessment of base metal, weld metal and deposit corrosion resistance both in the laboratory conditions and in the harshest real industrial media. It should be kept in mind at the same time that the materials corrosion resistance assessment by the laboratory investigations data has only an approximate nature, as it is impossible to reproduce at such testing a number of factors influencing the course of corrosion processes (stirring, solution changing frequency, variations of corrosion reagents concentrations by technological reasons, pressure, temperature, etc.). Welding of stably austenite steels and alloys is a rather difficult task. In this respect 06XH28MDT and 03XH28MDT alloys are the most complex. Problems of such type alloys weldability have been the subject of our company’s specialists long-term investigations, since the major scope of welding works on these alloys was carried out by the manual arc welding. These researches resulted in development of several grades of electrodes for welding of chromium-nickel-molybdenum alloys and the accelerated method for base metal, weld metal and deposit IGC determination.

2. Determination of the accelerated IGC determination method.

There is a national standard for determination of steels and alloys resistance to intergranular corrosion GOST 6032-89 in Russia, containing 2 methods for determination of Susceptibility to IGC of XH28MDT-type alloys: method «? A (accelerated)» and method «?». Method «? » physics: samples are exposed in a boiling aqueous solution of the sulfurous copper with addition of the blue powder. Method «?» is designed for a prolonged cycle at the boiling-point and, besides, gives the results, not comparable with the real resistance of welded structures in real industrial media. According to testing results of the alloy 06XH28MDT, carried out by method «?», IGC, as a rule, was not observed. At the same time, intergranular fracture is observed in real industrial media. Method «? A» stipulates harsher testing conditions – steel alloy samples are exposed to aqueous solution of sulfurous iron and sulphuric acid. It allowed reducing testing duration to 48 hours. However, the testing cycle still remained rather long. Lack of efficient and reliable method inhibited investigations on development of new electrode grades.

As a result of this, a method based on application of a 50-% sulphuric acid solution with addition of 40g/l of Fe2(SO4)3 and of 9g/l fluorine in the form of hydrofluoric acid at a temperature of 75-80^?? as a pickling solution, was suggested. A plant, the appearance of which is shown in Fig.1, was designed for this aim.

Fig.1. Appearance of the plant for corrosion testing Testing was carried out in fluoroplastic beakers with covers by stirring the solution with mixer (70 rot/min) during 12 hours with heating to 75-80 ^??. About 5 cm³ of the solution corresponded to 1 cm² of the sample surface. IGC susceptibility was

determined by a standard method of 90-degree sample bending. Cracking at sample bending served as the grounds for the alloy characterization as IGC susceptible. Method development was carried out on the alloy 06XH8MDT with the following composition: C=0,06%; Mn=0,58%; Si=0,50%; Cr=24,14%; Ni=27,80%; Cu=2,84%; Mo= 2,67; Ti=0,69%. The alloy was tested upon quenching at 1050^?? during 2 hours and quenching with provoking tempering at the temperature of 700^?? during 30 minutes.

Both corrosion and electrochemical investigations were carried out. Tests in standard solutions and in production conditions in the extraction phosphoric acid (hemihydrate process) were carried out in parallel. Solutions composition is given in Table 2.

Table 2

Test results of 06??28? ?? alloy IGC susceptibility.

Solu tion No.	Solution composition	Alloy state	Testing duration	Corrosion rate, mm per year	Corrosion potential, ?
1	10%H2SO4+110g/l ? uSO4 +5g/l	Quenching		0,0048
	of blue powder at boiling (102? ?)-method ? according to	Tempering	144	0,006	+0,2
	GOST 6032-89
2	Extraction phosphoric acid 43%	Quenching		0,086
	? 2? 5, 0,64%SO3, 0,97% F, 0.45%		100		+0,35
	Fe2O3 at 95? ? (laboratory testing)	Tempering		0,28*
2?	Extraction phosphoric acid (pulp)	Quenching		0,080
	35-39% ?2? 5, 0,9-1,2% SO3, 1,6-2,1% F, at 90-95? ? (production	Tempering	1000
	testing)			0,32*
3	50%H2SO4 + 40g/l Fe2(SO4)3 at	Quenching		0,15
	boiling -method ? A according		48		+0,7
	to GOST 6032-89	Tempering		0,58*
4	50%H2SO4 + 40g/l Fe2(SO4)3 +	Quenching		0,92
	9g/l of fluorine, 80°?, stirring –		12		+0,5
	suggested accelerated method.	Tempering		3,64*
5	50%H2SO4 + 9g/l of fluorine, 80? ? (potentiostatic pickling at +0,9V).	Quenching Tempering	1	11,2 16,3*	+0,9

*Note: IGC is revealed at 90-degree sample bending.

Experimental results demonstrated that at IGC susceptibility of the investigated materials in the extraction phosphoric acid IGC is also revealed in 50% sulphuric acid-based solutions No.3, 4 (Table 2).

At testing by method ? GOST 6032-89 in solution No.1 there was no IGC. The reasons for corrosion absence can be explained by the fact that at testing in standard sulphuric-based solutions in presence of Cu²⁺ ions at Fe³⁺ high-alloyed steels are located in the region of potentials, corresponding to grains passive state [4].

Corrosion rate of a quenched alloy in studied solutions (Table 2) increases at transition from the 10% H2SO4 solution to more corrosive media – extraction phosphoric acid, 50% H2SO4 and especially 50% H2SO4 with fluoride. The difference in corrosion rate between the quenched and the tempered alloy, characterizing IGC susceptibility, also increases at transition to more corrosive media, and time required for IGC revealing reduces.

Pickling of quenched and tempered alloy samples showed that increasing of weight losses on the tempered alloy occurs due to irritating of intergranular regions followed by grains precipitation.

In weak corrosive solution of 10% H2SO4 (method ? ) grain boundaries are exposed to slight pickling, only in certain regions (Fig.2?) grains precipitation does not occur even at long-run (144 hours) tests. In these conditions grain boundaries passivate or dissolve at a minor rate comparable with the grain dissolution rate.

In 50% H2SO4 solution with fluoride and ferric sulfate at 80^??, IGC is clearly apparent already within 1 hour (Fig. 2b).

?) b)

c)

Fig. 2. Microstructure of the quenched alloy 06XH28MDT after pickling in solutions:

?) No.1 – method ?, pickling during 144 hours;( ?500)

?) No.4 – accelerated method, pickling during 1 hour;(?300)

?) No.5 – the same, potentiostatic pickling, 1 hour at +0,9V,

(?300)

Metallographic investigations showed that the quenched alloy dissolution in the region of studied potentials from–0,1 to +0,9 V is uniform, and preferential dissolution along grain boundaries was observed on the tempered alloy, growing at transition to potentials corresponding to high current densities, as well as at the medium mixing.

Potentiostatic pickling within 1 hour of the tempered alloy in the 50% sulphuric acid solution with fluoride at the potential of +0,9 V resulted in its intergranular fracture (Fig. 2c) with colouring of individual grains and cracking at 90-degree sample bending.

Since the results of the alloy testing by method «?» and those for technological media of the extraction phosphoric acid production do not coincide, method «?» can not be applied as a standard method at testing for IGC susceptibility of high-alloyed steels and alloys in oxidizing media (the allowed rate of structural materials uniform corrosion is 0,1 mm per year).

In harsher operating conditions of equipment made of 06XH28MDT-type alloys, as well as of 03XH21M4GB steel at temperatures of 90-100^?? the risk of IGC grows, which accordingly requires stricter methods for IGC susceptibility testing.

Therefore, for accelerated IGC determination in high-alloyed steels and alloys the developed by us method, based on 50% H2SO4 with addition of 40g/l Fe2(SO4)3 and 9 g/l fluorine in the form of hydrofluoric acid at the temperature of 80^?? with the medium stirring may be recommended, as it allows clear revealing of this type of corrosion and reducing the testing time from 48 hours (method «?A» GOST 6032-89) to 12 hours maximum. Experiments on industrial alloys and welded joints carried out by the manual arc welding with covered electrodes showed that at tests duration of 12 hours only 25-30% of samples pass the tests.

Experiments with variation of the cycle duration (12, 8, 6, 4, 3,) hours demonstrated that complete correlation of the data with test results in industrial media for welded joints is achieved at three-hour tests. At the same time it is possible to differentiate metal with different corrosion resistance (Fig.3).

Fig.3. Intergranular fracture of welded joints after testing by the accelerated method, 3 hours (?400)

3. Comparative assessment of accelerated methods for IGC testing.

In order to obtain comparative data, testing of more than 60 industrial melts of 06XH28MDT alloy by various methods was carried out: by method «?», method «?A» and the stated above method.

The chemical composition of melts was within the following limits: ?=0,03-0,06%; Si=0,41-0,68%; Mn=0,36-0,75%; Cr=22,6-24,9%; Ni=26,3-28,8%; Mo=2,61-3,01%; Cu=2,58-3,16%; Ti= 0,51-0,95%.

Flat samples were exposed to heat treatment – quenching at 1100-1150^?? in water followed by subsequent provoking tempering – 720^?? , 20 minutes. In every case 2 samples were used, tests were repeated in cases of doubt. Test results demonstrated that all 60 melts passed tests for IGC susceptibility by method «?». At testing in harsher conditions – by method «? A» (48 hours) and by the suggested by us method (12 hours) 15 melts of 60 did not pass testing, and 45 of them were recognized as corrosion susceptible. At the same time we failed to reveal any regularity of the chemical composition influence on IGC susceptibility. Some results of comparative tests are given in Table 3.

4. Bibiliography

1. Kakhovsky N.I.. Welding of high-alloyed steels, Kiev, «Technica»,1975.
2. Urchenko U.F, Agapov G.I. . Corrosion of welded joints in oxidizing media. Moscow, «Mashinostroyeniye», 1976.
3. Babakov A.A., Pridantsev M.V. Corrosion-resistant steels and alloys. Moscow, «Metallurgy»,1971.
4. Chigal V. Intergranular corrosion of stainless steels. Moscow, «Chemistry»,1969.
5. Tufanov D.G. Corrosion resistance of stainless steels, alloys and pure metals. Moscow, «Metallurgy», 1982.