SCIENCE FOR PRODUCTION

MECHANISM OF INHIBITION OF STRESS CORROSION AND CORROSION CRACKING OF HIGH STRENGTH STEELS AND ALLOYS IN ACID MEDIA UDC 620.193.7

E. S. Ivanov

The efficiency of corrosion cracking inhibitors of high-strength steels and alloys in acid media is controlled by their capacity to suppress local corrosion processes on the stressed metal and depends on the nature of the inhibitor, its surface activity, the anion composition of medium, and the sign and potential of the stressed metal. In this work, investigations were carried out into the adsorption of inhibitors of various nature on the stressed metal and also into the effect of polarization and the magnitude of stresses on corrosion of steels in acid chloride media. 30KhGSA, 30KhGSNA, and 65G steels and 36NKhTYu alloy were examined. The majority of tests were carried out on specimens of 30KhGSA steel (0.30 wt. % C, 0.98 Cr, 1.0 Mn, 1.05 Si, 0~ wt. % Ni) in the form of 56 mm diameter disks quenches from 900~ • 10~ and tempered in oil at 200~ for two hours (o B = 1660-1740 MPa, HRC 49-50). The tensile biaxial stresses were developed by the method described previously in [i], the working the surface of the specimen was 1 cm 2, the volume of the solution 4 ml. The corrosive medium was in the form of a 4M solution of HCI (chemically pure), and inhibitors were represented by individual substances of the cation (diethylamine, diethanolamine, esters of di-triethanolamines) and anion (thiourea, dodecylsulfate) type, and several industrial inhibitots (BA-6, GMU, olazol). The potential was set with a P-5827 potentiostat, a silver chloride reference electrode was used, the holding time at a given potentia I was 1 h. The experiments were carried out with natural aeration, 20~ • I~ and the stationary potential was recorded after holding the electrode for 1 ho The concentration of ions of iron, chrome, manganese transferred into the solution was determined using a Perkin-Elmer-400 atomic absorption spectrometer. The selectivity coefficients Z were calculated. Adsorption of the inhibitors of chlor-ions was determined by the method of radioactive indicators [2]. The following data represent mean arithmetic values of the results of 3-5 measurements. The relative error of the experiments at a reliability factor of 0.95 is 5-7%. The experiments show that in anodic polarization, the time to appearance of pits in pure HCI solution decreases and in cathodic polarization increases (Fig. i). In the presence of thiourea (TU) no pits form in anodic polarization and in cathodic polarization their formation rate is high, i.e., TU is a pitting stimulator. In comparison with pure solutions of 4M HCI, diethynolamine slightly inhibits the nucleation of pits both in cathodic and anodic polarization. The inhibitors BA-6, GMU, and diethynolamine ester (DEA-3) suppress pitting in both cases.

The nucleation of pits under the effect of biaxial tensile stresses is associated with the fact that 30KhGSA and 30KhGSNA steels are subjected to selective dissolution in 4M HCI with mainly chrome and manganese transferred to the solution (Table i). The rate of transfer of iron is two orders of magnitude higher than that of chrome and manganese, i.e., dissolution of steels is accompanied by preferential ionization of iron~ The addition of N-containing inhibitors of the cation type to the acid reduces the rate of transfer of the components of the 30KhGSNA and 30KhGSA steels to the solution (especially chrome and manganese). The esters of di- and triethanolamine and the BA-6, GMU, and olazol All-Union Scientific-Research Institute of Protection of Metals Against Corrosion, Moscow. Translated from Fiziko-Khimicheskaya Meckhanika Materialov, No. 6, pp. 92-95, NovemberDecember, 1989. Original article submitted June 23, 1988.

0038-5565/89/2506-0633512.50

@

1990 Plenum Publishing Corporation

633

,

I

I

i

T'-"-~

_

"-,

Fig. i. Dependence of the time to appearance of pits on the polarization of stressed (~ = 196 MPa) 30KhGSA steel in 4M HCI without (i) and with additions (10 -2 M) of thiourea (2); octylester of diethynolamine (3) ; BA-6 (4) ; diethynolamine (5); GMU (6). TABLE i. Effect of Stresses on the Rate of Transfer of Components of 30KhGSA Steel into the Solution Rate of transfer of [ Stress, I ions into the solu[ Selectivity Corrosive medium, MPa I tion m 2. h ~. coefficient addition 4 M HCI Diethylamine, 5"10-2 M DEA-3 ester, 10-3 M Triethanolamine ester BA-6, 5 g / l i t e r

GMU, 5 g/liter TU, I0-2 M

0

196 294 0 196 294 0 196 294 0 196 294 0 196 294 0 196 294 0 196 294

2,66 3,51 3,62 0,788 0,652 0,640 0,48 0,41 0,42 0,49 0,53 0,56 0,53 0,55 0,61 0,24 0,18 0,22 1,31 3,80 4,80

0,031 0,032 0,035 0,08 0,052 0,032 0,10 0,08 0,08 0,018 0,012 0,012 0,036 0,018 0,019 0,020 0,012 0,012 0,016 0,061 0,060

0,034 0,030 0,082 0,048 0,023 0,026 0,08 0,08 0,08 0,006 0,008 0,009 0,008 0,010 0,011 0,008 0,006 0,005 0,018 0,064 0,051

2,0 0,84 0,86 8,82 7,22 4,63 1,52 1,78 1,65 3,47 2,13 2,10 6,44 3,15 2,93 7,53 6,13 4,94 1,20 2,45 1,80

l,I I 0,89 0,94

5,19 3,16 3,74 2,33 2,18 2,34 1,24 1,36 1,44 1,39 1,65 1,56 3,10 2,81 2,10 1,33 1,61 1,03

inhibitors reduce the transfer of chrome and manganese from the stressed steel into the solution 2-7 times more efficiently than from the nonstressed steel. This indicates that these inhibitors are adsorbed mainly on active areas of the surface of the steel, in the areas of stress concentration. In the presence of these inhibitors, no pits and corrosion cracking were detected throughout the entire experiment. In the presence of anion-active additions (for example TU) (see Table i), the transfer of iron and manganese into the solution is accelerated. In cathodic polarization this acceleration is greater than in anodic polarization (see Fig. i). Selecting the solution of chrome and manganese takes place, pits form at a high rate, and the cracking susceptibility increases, i.e., TU is stimulator of corrosion cracking. Dodecylsulfate has a similar effect. Thus, the effect of polarization on the stressed and nonstressed steel differs in relation to the surface activity of the inhibitor. In the absence of stresses, polarization has no special effect on the transfer of the components of the steel into the solution. Tensile stresses activate electrochemical processes of this solution of the steel and result in the formation of areas with a higher concentration of crystalline heterogeneities which are centers of local dissolution and of nucleation of corrosion cracks in active operation.

634

Fig. 2. Appearance of surfaces of stressed (o = 196 MPa) 30KhGSA steel after holding for 2 h in a solution of 4M HCI without (a) and with addition of 0.025 M GMU (b) and 0.05 M of thiourea (c). TABLE 2. Values of Heterogeneity Factor (f) and the Difference Between the Maximum and Minimum Adsorption Energies (AG) for 30KhGSA Steel Adsorbate

Stressed I Noustre steel (196 steel MPa~

f C1BA-6 GMU TU

" ,

AG,

kJ!m

3,0 7,35 2,35 5 , 7 1 2,50 6,28 4,16 10,20

t 2,35 2,34 2,36 2,33

AG,

~kJlm 5,71 5~68 5,75 5,65

The higher pitting susceptibility in cathodic polarization in the presence of TU can be explained by the fact that the anodic process on the cathode-polarized surface takes place only in the zone of maximum stress concentrations, i.e., at the tip of a growing crack, whereas the remaining surface with the cathodic potential does not dissolve. The pitting formation process assumes a strictly oriented direction. The investigations showed that the cation-active inhibitors BA-6, GMU, and the esters of diethynolamine are absorbed more intensively on the stressed metal and mainly on the iron component of the alloy. An increase of the stresses increases the extent of adsorption of TU but the nature of its absorption and of adsorption of cation-active inhibitors differs. Adsorption of BA-6, GMU, and TU is governed by the Temkin isotherm @ = A + i/f in C but the values of the prelogarithmic coefficient f, which determines the heterogeneity of the surface, differ for inhibitors and stimulators of corrosion cracking (Table 2) ~. TU which is a stimulator of corrosion cracking increases the coefficient f for the stressed steel thus increasing the heterogeneity of the surface and causing localization of the corrosion process. On the other hand, the inhibitors of corrosion cracking BA-6 and GMU reduced the energy heterogeneity of the surface and suppressed the local corrosion processes on the stressed metal, This conclusion is confirmed by the nature of failure in the examined media. Dissolution of the steel in HCI is accompanied by the formation of a small number of pits with a branching crack growing from each pit (Fig. 2a). In the solutions with GMU, the crack propagates from a single pit and then branches during its growth (Fig. 2b). In the presence of TU, a large number of pits forms on the surface and many cracks then start to propagate from these pits and cover the entire surface of the steel by a network (Fig. 2c). Curves of the variation of the anodic current with time form the stressed and nonstressed 30KhGSA steel in 4M HCI solutions were determined (Fig. 3). In the inhibited solutions, the currents are almost two orders of magnitude lower than in nonhibited solutions and the steady

635

t

T'x

7 o

I O

/0

oS

['-L ~ min

O

I

2

3

20

Fig. 3

Fig. 4

Fig. 3. Variation of the anodic current of stressed (196 MPa) (2, 4) and nonstressed (i, 3) 30KhGSA steel in potentiostatic holding (9 = -0.14 V): i, 2) 4M HCI without additions; 3, 4) with addition of 0.01 M DEA-3. Fig. 4. Kinetics of adsorption of DEA-3 on stressed (i) and nonstressed (2) 30KhGSA steel, 9 = -0.14 V. current is established after 10-20 min, i.e., the rate of adsorption is relatively high. The values of the current in the stressed metal are always high, i.e., the stresses accelerate dissolution. Representing the i--r curves in the Ai--ln 9 and in (i0/i)-in 9 coordinate, it can be seen (Fig. 4) that, in the first case, for the nonstressed metal in the presence of DEA-3 ester, the dependence becomes linear, i.e., DEA-3 ester inhibits the corrosion process by the blocking mechanism and is adsorbed on the uniformly heterogeneous surface [3]. On the stressed surface, the linear from the dependence is obtained in the coordinates in (i0/i)in ~, i.e., inhibition takes place as a result of the ~1-effect. The variation of the inhibition mechanism of dissolution of the steel by the cation-active inhibitors can be explained by the fact that the degree of filling of the surface of the stressed metal by the inhibitor molecules decreases and inhibition is determined by the generation of an additional ~1-potential by volume octyl radicals oriented in the direction of the solution. For the chemisorbed inhibitors (GMU, BE-6) the inhibition mechanisms remains blocking both on the stress and nonstressed metal. Thus, the cation-active compounds are effective inhibitors of stress corrosion and corrosion cracking whereas the anion-active compounds act as stimulators of these processes. The efficiency of the cation-active inhibitors is determined by their capacity to adsorb on the negatively charged local areas of the surface of the metal, in zones with the maximum concentration of tensile stresses. The cation-active inhibitors with a blocking effect (FA6, GMU) suppress the stress corrosion and corrosion cracking processes most efficiently. LITERATURE CITED I.

E. S, Ivanov and V. V. Egrov, "Method of determining the start of formation of a corrosion crack under the effect of biaxial tensile stresses," Zavod. Lab., No. i, 78-80

2.

E. S. Ivanov, "Adsorption of inhibitors and chloride ions on the surface of a stressed 30KhGSA steel," Zashch. Met., No. 6, 984-987 (1984). S. M. Reshetnikov, Inhibitors of Acid Corrosion of Metals [in Russian], Khimiya, Leningrad (1986).

(1982).

3.

636