HAL Id: jpa-00228043https://hal.archives-ouvertes.fr/jpa-00228043

Submitted on 1 Jan 1988

HAL is a multi-disciplinary open accessarchive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come fromteaching and research institutions in France orabroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, estdestinée au dépôt et à la diffusion de documentsscientifiques de niveau recherche, publiés ou non,émanant des établissements d’enseignement et derecherche français ou étrangers, des laboratoirespublics ou privés.

HYDROGEN SEGREGATION AND DIFFUSION ATGRAIN BOUNDARIES

H. Birnbaum, B. Ladna, A. Kimura

To cite this version:H. Birnbaum, B. Ladna, A. Kimura. HYDROGEN SEGREGATION AND DIFFUSION ATGRAIN BOUNDARIES. Journal de Physique Colloques, 1988, 49 (C5), pp.C5-397-C5-401.�10.1051/jphyscol:1988546�. �jpa-00228043�

JOURNAL DE PHYSIQUE Colloque C5, supplhent au nOIO, Tome 49, octobre 1988

HYDROGEN SEGREGATION AND DIFFUSION AT GRAIN BOUNDARIES

H.K. BIRNBAUM, B. LADNA and A. KIMURA

Department of Materials Science, University of Illinois at Urbana-Champaign, Urbana, IL 61801, U.S.A.

ABSTRACT - The segregation of hydrogen at grain boundaries in Ni has been shown to be dependent on the nature of the boundary using SIMS techniques. This contrasts to hydrogen segregation at free surfaces which appears to be less sensitive to the crystallography of the surface. The increased H concentration due to segregation occurs over appreciable distances from the surface or grain boundaries. The diffusivity of H along grain boundaries has been determined using SIMS techniques and correlated with studies of fracture kinetics.

INTRODUCTION

While segregation and diffusion of hydrogen at surfaces and grain boundaries has been of great interest, the difficulty of hydrogen microanalysis has limited the number of studies carried out. Secondary Ion Mass Spectrometry (SIMS) techniques have shown that segregation of deuterium (2~) occurs at free surfaces of fcc Ni (1) and bcc Nb-V alloys (2). Greatly enhanced 2~ concentrations occur at distances behind the free surface of the order of 100 nm. This distribution is sensitive to the presence of monolayers of other segregated species such as S (1). In these studies, no great sensitivity of the 2~ segregation on the crystallography of the surface was noted. Segregation of 28 at uncharacterized grain boundaries in Ni was observed (1) but extensive subsequent studies have shown that this segregation is very sensitive to the crystallography of the grain boundaries. Diffusion of hydrogen along grain boundaries has usually been assumed to be faster than in the lattice but no measurements of this parameter have been available. A previous SIMS experiwent showed qualitatively that permeation occurred preferentially along grain boundaries (3) but attempts to quantitatively measure this effect were unsuccessful (4).

EXPERIMENTAL PROCEDURES

Two principal techniques were used in the present studies; SIMS (1,2,5) and fracture kinetics. SIMS measurements were carried out at 140 K using a 17 keV CS+ ion beam on Ni bicrystal specimens grown from 99.999% pure Ni which were electro- lytically charged with 2 ~ . Measurements were carried out on symmetric tilt grain boundaries having a <110> rotation axis. These crystals were prepared by Dr. M.B. Hintz ( 6 ) to whom we wish to express our appreciation. Measurements will be reported for 39' (Z=9) and for 12g0 ( E = l l ) grain boundaries. The 3g0 boundary corresponds to a relatively high energy and the 129' boundary to a low energy boundary (7.8). Auger electron spectroscopic analysis of the 39' boundary which was fractured after hydrogen charging showed a fractional grain boundary surface coverage of 0.02. The low energy 129' boundary could not be fractured even after hydrogen charging.

Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:1988546

C5-398 JOURNAL DE PHYSIQUE

Fracture k i n e t i c s were used t o study the d i f fus ion of hydrogen along g r a i n boundaries i n po lycrys ta l l ine n i c k e l specimens (9,lO). Cylindrical specimens of high p u r i t y N i (99.999%), Ni-C, o r Ni-S a l l o y s were e l e c t r o l y t i c a l l y charged with IH a t var ious temperatures followed by t e n s i l e f r a c t u r e a t 77 K. Intergranular f r a c t u r e i n the ou te r annulus del ineated t h e extent t o which H had diffused along the gra in boundaries.

EXPERIMENTAL RESULTS AND DISCUSSION

SIMS MEASUREMENTS

The e l e c t r o l y t i c charging condit ions were chosen t o form a deuter ide at the sur face of the n icke l b i c r y s t a l s t o provide a reference composition f o r the g r a i n boundary diffusion. Charging with 2~ was car r ied out a t 295 K and the d i f fus ion p r o f i l e s were probed along t h e tilt ax is d i rec t ion , t h e (110). Typical concentration p r o f i l e s f o r cathodic current d e n s i t i e s of 20 ma cm-2 a r e shown i n Fig. 1 where the deuter ide layer thickness varied between 1 and 3.5 micrometers. This v a r i a t i o n i n t h e thickness of the deuter ide layer re f lec ted d i f fe rences i n the 2~ f u g a c i t i e s a t d i f f e r e n t p a r t s of the surface. The 2 ~ / ~ i r a t i o is constant a t about 0.7 i n t h e deuter ide and t h e r a t i o decreased t o a value of 10-3 t o 7 x 10-3 i n t h e s o l i d so lu t ion adjacent t o the deuter ide a s expected from t h e phase diagram.

Analyses of 28 a t the g r a i n boundaries were performed by s t e p scanning across the boundaries (with a l a t e r a l reso lu t ion of 2 micrometers) with the t y p i c a l r e s u l t s shown i n Figs. 2 and 3 ( f o r d e t a i l s see reference 11). The i n i t i a l repeated l i n e scans show a uniform 2~ concentration i n the deuter ide layer i n t h e two adjacent g ra ins and the gra in boundary. Below the deuter ide layer the 2~ concentration decreased by about two orders of magnitude which character ized the concentration i n t h e s o l i d solution. (Scans performed through t h e deuteride-solid s o l u t i o n i n t e r f a c e show l a r g e var ia t ions along t h e scan, eg. scan 7 i n Fig. 3). For the 39O g r a i n boundary peaks i n the 2~ concentration were observed i n t h e s o l i d so lu t ion a t the pos i t ions of the g ra in boundary (Fig. 2). No enhanced 2~ concentrations were seen f o r t h e 129' g ra in boundary (Fig. 3).

The dependence of the deuter ide thickness on the char ing time a t 295 K gives a value of t h e 2~ d i f f u s i v i t y i n t h e N i deuter ide of 3 x 10-f2 cm2 1-1. This value is about two orders of magnitude lower than the d i f f u s i v i t y i n t h e s o l i d so lu t ion (4).

The grea te r depth of penetrat ion of the 2~ along t h e 3g0 gra in boundary is the r e s u l t of enhanced d i f f u s i v i t y along the gra in boundary compared t o t h e l a t t i c e . The lack of enhanced penetrat ion of t h e 2~ along the 12g0 gra in boundary ind ica tes t h a t i n t h i s case the d i f f u s i v i t y is l e s s than o r equal t o the l a t t i c e d i f f u s i v i t y . An est imate of the ZH d i f f u s i v i t y along the 39O boundary may be obtained from t h e measured concentration p r o f i l e s using the gra in boundary d i f f u s i v i t y ana lys i s presented by IeClaire (12). A constant source concentration is provided by the deuter ide layer with t h e o r i g i n of the d i f f u s i v i t y being a t t h e s o l i d solution-deuteride in te r face . The data f i t t h e x6I5 dependence on d i s tance very w e l l (11) and y ie ld values f o r s6Dgb of 3 - 6 x 1 0 - ~ ~ m3 s-l ( s is the segregat ion f a c t o r , 6 , is the gra in boundary "width", and Dgb is the gra in boundary d i f fus iv i ty ) . Assuming a value f o r 6 of 1 nm and f o r s of 1 leads t o a Dgb of 3-6x10-13 m2 s-1 f o r t h e 39O g r a i n boundary. This value may be compared t o a value f o r 2~ d i f f u s i v i t y i n the l a t t i c e of 3.6~10-I4 (4). The d i f f u s i v i t y of 2~ i n N i a t 295 K is enhanced by a f a c t o r of between 8 t o 17 f o r the high energy 3g0 gra in boundary and no enhancement is seen f o r the low energy 12g0 gra in boundary.

FRACTURE KINETICS MEASUREMENTS

Hydrogen induced in te rgranula r f r a c t u r e of N i a t 77 K has been showr t o r e s u l t from the attainment of a c r i t i c a l hydrogen concentration a t the g ra in boundaries (13,14) t h e value of which depends on the presence of other impurity species (15,16). While these c r i t i c a l concentrations a r e not known exact ly (13), the d u c t i l e t o b r i t t l e t r a n s i t i o n a s the g ra in boundary hydrogen concentration increases can be used t o measure the d i f f u s i v i t y of hydrogen along the gra in boundaries i n a po lycrys ta l l ine specimen.

Typical r e s u l t s f o r high p u r i t y N i a r e shown i n Fig. 4 where t h e "Grain Boundary Frac tu re Katio", ie. t h e r a t i o of the in te rgranu la r f r a c t u r e a rea t o the t o t a l f r a c t u r e a r e a is shown a s a funct ion of cathodic charging time and temperature. The t r a n s i t i o n between the ou te r annulus of in te rgranu la r f r a c t u r e and the inner d u c t i l e f r a c t u r e a rea was very sharp with t h e t r a n s i t i o n o f ten occurring within a s i n g l e g r a i n boundary. The broken curves on Fig. 4 show the ca lcu la ted behaivor assuming l a t t i c e d i f fus ion a s the process which t ranspor t s H from the surface. A s shown, the depth of embrittlement is much g r e a t e r than can be accounted f o r by l a t t i c e diffusion. A d i f fus ion l imi ted model can be f i t t e d t o the da ta assuming g r a i n boundary d i f fus ion a s t h e t ranspor t mechanism (10) with the r e s u l t s shown i n Table I.

Table I GRAIN BOUNDARY DIFFUSIVITY

TEMPERATURE LATTICE DIFF. GRAIN. BD. DIFF. Dgb/D1 (K) (m2 s-1) (rn2 s'l)

The value f o r Dgb a t 295 K is i n good agreement with the value obtained from the SIMS analysis . The a c t i v a t i o n enthalpy (Hgb) obtained f o r the g r a i n boundary d i f fus ion of H i s 36.1 kJ mole and the r a t i o t o the l a t t i c e d i f f u s i o n enthalpy is H /H = 0.92.

gb kimilar r e s u l t s and analyses were obtained f o r Ni-C and Ni-S a l l o y s (10) under conditions where t h e C and S were segregated t o t h e g r a i n boundaries. These r e s u l t s are summarized i n Fig. 5 and i n Fig. 6 which-shows the temperature dependence of the time t o achieve GBFR=0.5 f o r pure N i and segregated Ni-C and Ni-S al loys. It is seen t h a t C has the e f f e c t of decreasing t h e r a t e of embrittlemebt while S inc reases the embrittlement ra tes . I n t e r p r e t a t i o n of these experiments i n terms of g r a i n boundary d i f fus ion parameters (10) g ives the r e s u l t s shown i n Table 11.

Table I1 SUMMARY OF THE H GRAIN BOUNDARY DIFFUSIVITIES

ALLOY

pure N i ( l a t t i c e ) 39.2 pure N i (gr. bd. ) 36.1 0.92 N i - C 40.5 1.03 1.1 N i - S 43.9 1.1 1.2

The pre-exponential terms f o r t h e g ra in boundary d i f f u s i v i t i e s of t h e Ni-C and Ni-S a l l o y s cannot be obtained without knowledge of the e f f e c t s of segregated C and S on t h e c r i t i c a l H concentrat ion f o r in te rgranu la r f rac tu re .

ACKNOWLEDGEMENTS

This work was supported by the National Science Foundation grant DMR 86-05955. We would l i k e t o acknowledge use of the f a c i l i t i e s of the Center f o r the Microanalysis of Mater ia ls a t the Mater ia ls Research Laboratory which is p a r t i a l l y supported by the Department of Energy.

REFERENCES

(1) H. Fukushima and H.K. Birnbaum, Acta Metall., 32, 85 (1984). (2) B. Iadna and H.K. Birnbaum, Acta Metall. 34, 899 (1986). (3) T. TSUN and R.M. Latanision, S c r i p t a Metall. 16, 575 (1982). (4) J. Volkl and G. Alefeld, "Hydrogen i n Metals I" eds. G. Alefeld a n d J . Volkl,

Springer-Verlag Press , Ber l in (1978) p. 321. (5) H.K. Birnbaum, H. Fukushima, and J. Baker, "Advanced Techniques f o r

Character iz ing Hydrogen i n Metals" eds. N.F. Fiore and B.Y. Berkowitz, AIME,

C5-400 JOURNAL DE PHYSIQUE

Warrendale, PA (1982) p. 149. W. Gust, M.B. Hintz, A. Lodding, H. Odelius, B. Predel, Acta Metall. 30, 75 (1982). G. Hasson and C. Goux, Scripta Metall. 5, 889 (1971). G. Herrmann, H. Sautter, G. Baro, and H. Gleiter, "Grain Boundaries In Engineering Materials" eds. J.L. Walter, H.J. Westbrook, D.A..Woodford, Claitor's Pub., Baton Rouge (1974) p. 43. H. Kimura and H.K. Birnbaum, Scripta Metall. 21, 219 (1987). H. Kimura and H.K. Birnbaum, Acta Metall. In press. B. Ladna and H.K. Birnbaum, Acta Metall. In press. A.D. LeClaire, British Jnl. Appl. Phys., 14, 351 (1963). D. Lassila and H.K. Birnbaum, Acta Metall. 34, 1237 (1986). S.M. Breummer, R.H. Jones, M.T. Thomas, and D.R. Baer, Scripta Metall. 14, 1233 (1980). R.H. Jones, S.M. Bruemmer, M.T. Thomas, and D.R. Baer, Metall. Trans. 14A, 1729 (1980). D. Lassila and H.K. Birnbaum, Acta Metall. 34, 1237 (1986).

Norrnalizatibn Factor ( 2 ~ - / 5 8 ~ i - ) x lo5 8r Norrnalizotion Factor ( 2 ~ ' / 5 8 ~ i 7 x lo5

O.6prn /scan Grain 2

grain

L , t I n I 0 I n l . ( 1 1 t 1 1 1 1 1 . 1

2 3 4 5 20 40 60 80 100 Depth of Analysis (micrometers) Distance (micrometers)

Fig. 1 Normalized profiles of 2~ Ni crystals cathodically charged at 295K for 3 h at 2 ma The curves are indicative of repeated measurements at different points on the surface.

Fig. 2 Normalized line scans of 2~ across the 39' <110> tilt boundary in cathodically charged Ni. The numbers indicate the sequence of line scans.,

Normalization Factor (2~- /58~i - ) x lo5

I grain boundary

%.in I Grain 2 i i

0 5'0 ' Id0 ' 1;o ' 2 h ' Distance (micrometers)

Fig. 3 Normalized l i n e scans of 2~ across the 129' <110> tilt boundary i n ca thod ica l ly charged b i c r y s t a l s of N i . The numbers i n d i c a t e the sequence of l i n e scans.

"

Charging Time (minutes)

Charging Time(min)

Fig. 5 Grain boundary f r a c t u r e r a t i o a s a funci ton of cathodic charging time and temperature f o r Ni-C ( s o l i d po in t s ) and Ni-S (open points) .

273 K; 293 K; 317 K; 273 K; 296 K; 317 K.

Fig. 4 Grain boundary f r a c t u r e r a t i o a s a funct ion of cathodic charging time and temperature f o r pure N i . The broken curves a r e ca lcu la ted values based. on l a t t i c e d i f fus ion .

Fig. 6 Cathodic charging times t o a t t a i n a GBFR = 0.5 vs temperature f o r pure N i ; Ni-C ; and Ni-S

specimens.