- Email: [email protected]
The effect of dissolved oxygen on the terminal solubility of hydrogen in alpha zirconium
JOURNAL OF NUCLEAR MATERIALS 4, No. 1 ( 1961) 1 1 0 - 1 1 2 , NORTH-HOLLAND PUBLISHING CO., AMSTERDAM
T H E E F F E C T OF DISSOLVED O X Y G E N ON T H E T E R M I N A L S O L U B I L I T Y OF H Y D R O G E N IN A L P H A ZIRCONIUM A. BROWN C. A. Parsons & Company Ltd., Nuclear Research Centre, Fossway, Newcastle upon Tyne 6, UK
and D. HARDIE Department el Metallurgy, University el Durham, King's College, Newcastle upon Tyne l, U K
Received 16 March 1961 There has been considerable doubt about the effect which the presence of oxygen has upon the solubility of hydrogen in the hexagonal zirconium. It is certain that the oxygen atom, because of its size, can only occupy the octahedral sites in the lattice and it seems that such occupation can prevent hydrogen from filling the neighbouring tetrahedral interstices. Martin and Rees 1) suggest that as a result, three sites are blocked per oxygen atom in dilute solutions of oxygen in zirconium, whereas in a 50 at °/o alloy (which cannot in fact be obtained owing to the limited solubility of oxygen in zirconium) one site would be blocked per oxygen atom. Hall, Martin and Rees 2) have investigated the effect on the hydrogen absorption isotherms of oxygen present in solid solution in the zirconium and found that the volume of hydrogen absorbed decreased as the oxygen content increased, but the equilibrium pressure of hydrogen increased. However, they carried out their measurements at comparatively high pressures in the range 0.5 to 760 mm of mercury, and in fact determined the fl/(fl+(~) and (fl+~)/~ phase boundaries. They were therefore dealing with a bodycentred cubic structure instead of hexagonal close packing and their arguments concerning saturation solubility no longer strictly apply. Their work, which is in agreement with the later results of Edwards, Levesque and Cubi-
ciotti 8) merely indicates that the solubility of hydrogen in fl zirconium is reduced b y the presence of oxygen in solid solution. Subsequent w o r k e r s - E l l s and McQuillan 4) and Gulbransen and Andrew 5 ) _ h a v e actually determined the terminal solubility of hydrogen in the a phase at temperatures respectively above and below the eutectoid, where equilibrium pressures are in the range 10 -2 to 1 mm of mercury, but they made no systematic study of the effect which the presence of oxygen might have on this solubility. In fact it has been generally assumed that the original reasoning of Martin and Rees can be directly applied to the a solid solutions. Work on the variation of solubility of hydrogen in both the a and fl phases of zirconium when various alloying elements (including oxygen) are present has now definitely revealed that oxygen increases the terminal solubility in the z phase though it does reduce it in the fl phase. Using a high vacuum apparatus in which accurately m e a s u r e d amounts of hydrogen could be added to a reaction system containing zirconium and the equilibrium pressure determined by means of a Pirani type gauge, equilibrium pressure concentration isotherms have been plotted for zirconium containing 1200, 4200, 5600 ppm oxygen in the temperature 110
THE
EFFECT
OF DISSOLVED
OXYGEN
range 530-790 ° C. (The original zirconium contained the following impurities: Fe 275, C 100, O 1200, N 170, A1 150, Cr 30, P b 40 ppm b y weight, and H f 1.9 wt %). The results are most readily assessed when plotted as the logarithm of the equilibrium pressure against reciprocal temperature for solutions of various hydrogen concentration. Since in pure zirconium the isothermal equilibrium pressure is invariant with composition in a two-phase region, the and fl fields on such a plot are generally separated b y a single line. In the presence of oxygen however there is slight pressure variation with composition in the (~+fl) field and this is then more truly represented by a band rather than a single line. The boundary shown in fig. 1 is in actual fact therefore the o~/(~+fl) boundary and represents the terminal solubility of hydrogen in the ~ phase. The terminal solubility at any particular temperature is given by the intersection of an hydrogen isostere with this boundary. Although the pressure at the ~/(~+fl) phase boundary for any particular temperature remains constant as the amount of oxygen present is increased, there is simultaneous movement of the isostere3 in such a a::~W = 0.2
.
U3 E 0 . 0
/
/
/
'/ /
b--+...x
/
/
/
Y /
~
/'
/,:oo~
o as recelved[~r-O / /
+ plus 4 4 0 0 p p m O /
x
/
ON T H E
TERMINAL
SOLUBILITY
111
w a y that the terminal solubility of hydrogen at any particular temperature in this region increases with the oxygen content of the zirconium. The heat of solution, as determined from the relationship AH=R \51/T]c shows no significant variation and has a value of 23.7 -4-0.5 Keals/g mole in all cases. These results also seem to clarify interpretation of microstruetures obtained in ~ zirconium after corrosion in high temperature
Fig. 2. Sample of zirconium -- 1 w t % copper alloy after corrosion in s t e a m at 1 atm. pressure at 600 ° C for 6 days, showing absence of h y d r i d e precipitation n e a r the m e t a l surface. ( P h o t o m i c r o g r a p h p r o v i d e d by the Metallurgy Division, A E R E , Harwell). X 400
uJ - 0 . 2 tY
n -0.4
o
E. - 0 8
_. -1.0
T
o-1.6
1'° at °/° H
(-9 -1.8 O ~ -2.0 : 9.0
10.0 RECIPROCAL
11.0 12.0 TEMPERATURE
104 T°K
Fig. 1. The effect of o x y g e n c o n t e n t ( p p m b y weight) u p o n the e q u i l i b r i u m p r e s s u r e - t e m p e r a t u r e relationships for solutions of h y d r o g e n in a-zirconium.
Fig. 3. Zirconium d i l a t o m e t e r specimen containing 112 p p m b y weight of hydrogen, showing absence of hydride p r e c i p i t a t i o n near the surface due to t h e increased solubility in this region resulting f r o m the presence of o x y g e n in solution, x 70
112
A. BROWN
AND
water or steam. Such structures frequently show a much greater quantity of precipitated hydride in the centre of the specimens compared with regions near the surface (fig. 2). This phenomenon can be explained without recourse to a hydrogen concentration gradient, since there is inevitably a higher oxygen concentration towards the surface and the decreased precipitation near the surface m a y simply be due to the increased solubility of hydrogen in the presence of dissolved oxygen. Such a result is also clearly demonstrated in the phot0micrograph of a section from a dilatometric specimen, fig. 3. Due to prolonged heating in the dilatometer the specimen has absorbed some oxygen into the surface layers despite the vacuum of < 10 -5 mm of mercury. As a result, on hydriding there is no precipitate apparent in the surface layers but there is copious precipitation in the rest of the specimen, due to the lower hydrogen solubility there. A similar effect has been observed in the contamination of titanium by water vapour 6). Some preliminary investigations by means of
D.
HARDIE
vacuum dilatometry at temperatures below the eutectoid indicate t h a t the terminal solubility is also increased in this region and this is being examined further. This work was carried out in the Department of Metallurgy, King's College, Newcastle upon Tyne, with financial and material support from the UKAEA, Atomic Energy Research Establishment, Harwell, to whom we are very grateful. References
1) S.L.H. Martin and A. L. G. Rees, Trans. Faraday Soc. 50 (1954) 343 2) M.N.A. Hall, S. L. H. Martin and A. L. G. Rees, Trans. Faraday Soc. 41 (1945) 306 3) R. K. Edwards, P. Levesque and D. Cubicio~ti, J. Amer. Chem. Soe. 77 (1955) 1307 4) C. E. Ells and A. D. MeQuillan, J. Inst. Metals 85 (1956-57) 89 a) E. A. Gulbransen, and K. F. Andrew, Trans. Amer. Inst. Min. Metall. Engrs. 203 (1955) 136 6) P. C. Hughes and I. R. Lamborn, J. Inst. Metals 80 (1960-61) 165
T H E E F F E C T OF DISSOLVED O X Y G E N ON T H E T E R M I N A L S O L U B I L I T Y OF H Y D R O G E N IN A L P H A ZIRCONIUM A. BROWN C. A. Parsons & Company Ltd., Nuclear Research Centre, Fossway, Newcastle upon Tyne 6, UK
and D. HARDIE Department el Metallurgy, University el Durham, King's College, Newcastle upon Tyne l, U K
Received 16 March 1961 There has been considerable doubt about the effect which the presence of oxygen has upon the solubility of hydrogen in the hexagonal zirconium. It is certain that the oxygen atom, because of its size, can only occupy the octahedral sites in the lattice and it seems that such occupation can prevent hydrogen from filling the neighbouring tetrahedral interstices. Martin and Rees 1) suggest that as a result, three sites are blocked per oxygen atom in dilute solutions of oxygen in zirconium, whereas in a 50 at °/o alloy (which cannot in fact be obtained owing to the limited solubility of oxygen in zirconium) one site would be blocked per oxygen atom. Hall, Martin and Rees 2) have investigated the effect on the hydrogen absorption isotherms of oxygen present in solid solution in the zirconium and found that the volume of hydrogen absorbed decreased as the oxygen content increased, but the equilibrium pressure of hydrogen increased. However, they carried out their measurements at comparatively high pressures in the range 0.5 to 760 mm of mercury, and in fact determined the fl/(fl+(~) and (fl+~)/~ phase boundaries. They were therefore dealing with a bodycentred cubic structure instead of hexagonal close packing and their arguments concerning saturation solubility no longer strictly apply. Their work, which is in agreement with the later results of Edwards, Levesque and Cubi-
ciotti 8) merely indicates that the solubility of hydrogen in fl zirconium is reduced b y the presence of oxygen in solid solution. Subsequent w o r k e r s - E l l s and McQuillan 4) and Gulbransen and Andrew 5 ) _ h a v e actually determined the terminal solubility of hydrogen in the a phase at temperatures respectively above and below the eutectoid, where equilibrium pressures are in the range 10 -2 to 1 mm of mercury, but they made no systematic study of the effect which the presence of oxygen might have on this solubility. In fact it has been generally assumed that the original reasoning of Martin and Rees can be directly applied to the a solid solutions. Work on the variation of solubility of hydrogen in both the a and fl phases of zirconium when various alloying elements (including oxygen) are present has now definitely revealed that oxygen increases the terminal solubility in the z phase though it does reduce it in the fl phase. Using a high vacuum apparatus in which accurately m e a s u r e d amounts of hydrogen could be added to a reaction system containing zirconium and the equilibrium pressure determined by means of a Pirani type gauge, equilibrium pressure concentration isotherms have been plotted for zirconium containing 1200, 4200, 5600 ppm oxygen in the temperature 110
THE
EFFECT
OF DISSOLVED
OXYGEN
range 530-790 ° C. (The original zirconium contained the following impurities: Fe 275, C 100, O 1200, N 170, A1 150, Cr 30, P b 40 ppm b y weight, and H f 1.9 wt %). The results are most readily assessed when plotted as the logarithm of the equilibrium pressure against reciprocal temperature for solutions of various hydrogen concentration. Since in pure zirconium the isothermal equilibrium pressure is invariant with composition in a two-phase region, the and fl fields on such a plot are generally separated b y a single line. In the presence of oxygen however there is slight pressure variation with composition in the (~+fl) field and this is then more truly represented by a band rather than a single line. The boundary shown in fig. 1 is in actual fact therefore the o~/(~+fl) boundary and represents the terminal solubility of hydrogen in the ~ phase. The terminal solubility at any particular temperature is given by the intersection of an hydrogen isostere with this boundary. Although the pressure at the ~/(~+fl) phase boundary for any particular temperature remains constant as the amount of oxygen present is increased, there is simultaneous movement of the isostere3 in such a a::~W = 0.2
.
U3 E 0 . 0
/
/
/
'/ /
b--+...x
/
/
/
Y /
~
/'
/,:oo~
o as recelved[~r-O / /
+ plus 4 4 0 0 p p m O /
x
/
ON T H E
TERMINAL
SOLUBILITY
111
w a y that the terminal solubility of hydrogen at any particular temperature in this region increases with the oxygen content of the zirconium. The heat of solution, as determined from the relationship AH=R \51/T]c shows no significant variation and has a value of 23.7 -4-0.5 Keals/g mole in all cases. These results also seem to clarify interpretation of microstruetures obtained in ~ zirconium after corrosion in high temperature
Fig. 2. Sample of zirconium -- 1 w t % copper alloy after corrosion in s t e a m at 1 atm. pressure at 600 ° C for 6 days, showing absence of h y d r i d e precipitation n e a r the m e t a l surface. ( P h o t o m i c r o g r a p h p r o v i d e d by the Metallurgy Division, A E R E , Harwell). X 400
uJ - 0 . 2 tY
n -0.4
o
E. - 0 8
_. -1.0
T
o-1.6
1'° at °/° H
(-9 -1.8 O ~ -2.0 : 9.0
10.0 RECIPROCAL
11.0 12.0 TEMPERATURE
104 T°K
Fig. 1. The effect of o x y g e n c o n t e n t ( p p m b y weight) u p o n the e q u i l i b r i u m p r e s s u r e - t e m p e r a t u r e relationships for solutions of h y d r o g e n in a-zirconium.
Fig. 3. Zirconium d i l a t o m e t e r specimen containing 112 p p m b y weight of hydrogen, showing absence of hydride p r e c i p i t a t i o n near the surface due to t h e increased solubility in this region resulting f r o m the presence of o x y g e n in solution, x 70
112
A. BROWN
AND
water or steam. Such structures frequently show a much greater quantity of precipitated hydride in the centre of the specimens compared with regions near the surface (fig. 2). This phenomenon can be explained without recourse to a hydrogen concentration gradient, since there is inevitably a higher oxygen concentration towards the surface and the decreased precipitation near the surface m a y simply be due to the increased solubility of hydrogen in the presence of dissolved oxygen. Such a result is also clearly demonstrated in the phot0micrograph of a section from a dilatometric specimen, fig. 3. Due to prolonged heating in the dilatometer the specimen has absorbed some oxygen into the surface layers despite the vacuum of < 10 -5 mm of mercury. As a result, on hydriding there is no precipitate apparent in the surface layers but there is copious precipitation in the rest of the specimen, due to the lower hydrogen solubility there. A similar effect has been observed in the contamination of titanium by water vapour 6). Some preliminary investigations by means of
D.
HARDIE
vacuum dilatometry at temperatures below the eutectoid indicate t h a t the terminal solubility is also increased in this region and this is being examined further. This work was carried out in the Department of Metallurgy, King's College, Newcastle upon Tyne, with financial and material support from the UKAEA, Atomic Energy Research Establishment, Harwell, to whom we are very grateful. References
1) S.L.H. Martin and A. L. G. Rees, Trans. Faraday Soc. 50 (1954) 343 2) M.N.A. Hall, S. L. H. Martin and A. L. G. Rees, Trans. Faraday Soc. 41 (1945) 306 3) R. K. Edwards, P. Levesque and D. Cubicio~ti, J. Amer. Chem. Soe. 77 (1955) 1307 4) C. E. Ells and A. D. MeQuillan, J. Inst. Metals 85 (1956-57) 89 a) E. A. Gulbransen, and K. F. Andrew, Trans. Amer. Inst. Min. Metall. Engrs. 203 (1955) 136 6) P. C. Hughes and I. R. Lamborn, J. Inst. Metals 80 (1960-61) 165