ELECTRODE PROBE FOR USE IN AQUEOUS ENVIRONMENTS OF HIGH TEMPERATURE AMD HIGH RADIATION
rigid gf the invention
This invention relates to reducing the corrosion potential of components exposed to high-temperature water, i.e., water having a temperature of about 150°C or greater. In particular, the invention relates to the monitoring of electrochemical potential in a light water nuclear reactor, for example, during the injection of hydrogen into the reactor water.
Background of the Invention
Stress corrosion cracking (SCC) is a known phenom¬ enon occurring in reactor components, such as structural members, piping, fasteners, and welds, exposed to high- temperature water. The reactor components are subject to a variety of stresses associated with, e.g., differ¬ ences in thermal expansion, the operating pressure need¬ ed for the containment of the reactor cooling water, and other sources such as residual stress from welding, cold working and other asymmetric metal treatments. In addi¬ tion, water chemistry, welding, heat treatment, and radiation can increase the susceptibility of metal in a component to SCC.
It is well known that SCC occurs at higher rates when oxygen is present in the reactor water in concen¬ trations of about 5 ppb or greater. SCC is further increased in a high radiation flux where oxidizing species, such as oxygen, hydrogen peroxide, and short¬ lived radicals, are produced from radiolytic decomposi- tion of the reactor water. Such oxidizing species in-
crease the electrochemical corrosion potential (ECP) of metals. Electrochemical corrosion is caused by a flow of electrons from anodic to cathodic areas on metallic surfaces. The ECP is a measure of the kinetic tendency for corrosion phenomena to occur, and is a fundamental parameter in determining rates of, e.g., SCC, corrosion fatigue, corrosion film thickening, and general corrosion.
One method employed to mitigate intergranular stress corrosion cracking (IGSCC) of susceptible mate¬ rial is the application of hydrogen water chemistry (HWC) , whereby the oxidizing potential of the BWR environment is modified to a more reducing condition. This effect is achieved by adding hydrogen gas to the reactor feedwater. When the hydrogen reaches the reactor vessel, it reacts with the radiolytically formed oxidiz¬ ing species to reform water, thereby lowering the con¬ centration of dissolved oxidizing species in the water in the vicinity of metal surfaces. It has been shown that IGSCC of Type 304 stainless steel used in BWRs can be mitigated by reducing the ECP of the stainless steel to values below -0.230 V(SHE) . An effective method of achieving this objective is to inject hydrogen into the feedwater. The injected hydrogen reduces the level of oxidizing species in the water, such as dissolved oxygen, and as a result lowers the ECP of metals in the water.
The primary method used to quantify the levels of hydrogen injection needed to achieve IGSCC protection is the measurement of the ECP of BWR water in the spe¬ cific region of interest. Electrochemical potential monitoring is conventionally carried out employing paired electrochemical half-cell probes or electrodes which are mounted within the recirculation piping or in an external vessel which has its water source from the reactor water in the recirculation piping. The elec-
trodes are accessed to the external environment through gland-type mountings or the like. Where the electrode system of interest involves the potential from a metal corrosion electrode, then the reference electrode can conveniently be a metal-insoluble salt electrode, if the metal salt couple is chemically stable and if appropri¬ ate thermodynamic data are available. Accordingly, one of the thus-mounted probes which is configured as a reference electrode may be based, for example, on a silver/silver chloride half-cell reaction. Once the reference electrode half-cell is defined, the cell is completed with the sensing cell portion based upon a metal such as platinum or stainless steel. Verification of the reference electrode and/or the electrode pair is carried out by thermodynamic evaluation and appropriate Nernst based electrochemical calculations in combination with laboratory testing within a known environment. When the measured ECP and value calculated from thermo¬ dynamics are in agreement, verification of the reference electrode is achieved.
Half-cell electrodes developed for use in reactor circulation piping traditionally have been configured withmetal housings, high-temperature ceramics and poly¬ meric seals such as Teflon brand polytetrafluoroethy- lene. These structures have performed adequately in the more benign and essentially radiation-free environments of recirculation piping.
Over the recent past, investigators have sought to expand the ECP monitoring procedures to the severe environment of the fluid in the vicinity of the reactor core itself for the purpose of studying or quantifying the effect of hydrogen water chemistry adjustment in mitigating irradiation-assisted stress corrosion crack¬ ing (IASSC) as well as IGSCC. Within the reactor core, the monitoring electrode can be mounted, for example, with otherwise unemployed or in tandem with the travel-
ing instrumentation probe of available local power range monitors and the like. The monitors are located in a severe, high-temperature (550βF) , high-radiation (typi¬ cally 109 R (rads) per hour gamma, lθ13 R per hour neu- tron) water environments. Probe structures of earlier designs are completely inadequate for this reactor core environment, both from a material standpoint and with respect to the critical need to prevent leakage of radioactive materials to the environment outside of the reactor vessel.
U.S. Patent No.5,217,596 discloses a silver/silver chloride electrode probe for use in the high-tempera¬ ture, high-radiation zone of a nuclear reactor. This device, as well as others for related applications such as those disclosed in U.S. Patent Nos. 5,192,414 and 4,948,492, are electrochemical monitoring sensors that measure the propensity for environmentally assisted cracking in a nuclear reactor. The key part of each of these devices is the voltage sensing end where metal components are brazed to sapphire insulating bodies. In the standard process the surface of the sapphire is made compatible for brazing by a complex process, including painting, referred to as "metallizing", with a tungsten paint followed by nickel plating of the sapphire surface previously treated with the tungsten paint. Following metallizing and each separate coating step, the sapphire is fired to assure adherence of each layer. The final step is brazing of sapphire to the metal components with pure silver. The entire process must be followed with careful attention to each coating thickness, specific coverage, sintering temperature and process environment if a successful product is to be produced.
With a proven design, the longevity of the product during service depends on the adequacy of the braze and how precisely each separate layer is protected from the
high-te perature aqueous environment. The device will not provide the correct potential if chemical reactions occur between the metallizing or any of the protective coatings and the aqueous environment of the nuclear reactor.
An additional, but critical, life-limiting factor is the corrosion rate or dissolution rate of sapphire in high-temperature water. The general dissolution rate in high-temperature, high-purity water is 0.010 inch/yr. However, upset conditions such as pH shifts or erosion/ corrosion greatly increase the corrosion rate of sap¬ phire. In one plant installation the local chemistry in the installation region caused premature sensor failure by increased sapphire dissolution rate. It is clear that the application of an insulator having a corrosion resistance much greater than the corrosion resistance of sapphire would be a major bene¬ fit. However, corrosion resistance by itself is of little value unless a metal-to-ceramic joining technol- ogy is available or could be developed for the partic¬ ular ceramic. The entire process must meet the require¬ ments of chemical compatibility with the nuclear envir¬ onment. In addition, physical compatibility between the ceramic and the metal members must also be achieved.
Summary of the Invention
The present invention is a reference electrode probe for evaluating electrochemical potentials in which magnesia or yttria partially stabilized zirconia has been substituted for the sapphire of the prior art probes. The stabilized zirconia has a corrosion rate which is at least an order of magnitude lower than the corrosion rate of sapphire. Furthermore, zirconia has a corrosion rate which does not increase with pH shifts and is resistant to the erosion/corrosion that might exist in specific regions of the nuclear reactor.
In joining the stabilized zirconia to the appro¬ priate metal component, the present invention uses a technique known as active metal brazing where a small concentration of the active metal, in this case titanium in a silver paste, becomes chemically bonded to the ceramic. The usual formulations of available active metal brazes contain alloying agents such as copper to improve brazing characteristics. However, in the case of the silver/silver chloride sensors of the designs shown in U.S. Patent No. 5,217,596, the presence of alloying agents creates major voltage offsets from the desired theoretical voltage.
In accordance with the present invention, a one- or two-step brazing process is performed whereby a sil- ver-titanium paste is used to join the metal components to the magnesia-stabilized zirconia. Simultaneously or as a separate step, excess pure silver - which is a requirement for proper electrode performance - is brazed to cover the internal metal components of the sensor. Because the layering process described in U.S. Patent No. 5,217,596 has been eliminated in accordance with the present invention, voltage offsets are mini¬ mized and the manufacturing process is greatly simpli¬ fied, resulting in significant cost reduction.
Brief Description of the Drawings
FIG. 1 is a sectional view of an electrode probe according to the present invention.
FIG. 2 is a partial sectional view of the sealing retainer structure shown in FIG. 1. FIG. 3 is an end view of the electrode probe shown in FIG. 1.
Detailed Description of the Preferred Embodiments
While having utility in a broad variety of indus¬ trial monitoring functions, the electrode structure of the instant invention finds particular utility operating under the rigorous environment of the reactor core of a nuclear power facility. No elastomeric seals or polymeric components are present in its structure, which incorporates a sealing architecture of the highest integrity. In the latter regard, a brazed and welded assembly consisting only of ceramic and metal parts forms the structure of the device. The electrode finds preferable employment as a reference component of an electrode system involving a metal-metal ion couple and thus the instant electrode can conveniently be a metal, slightly soluble salt electrode. In accordance with the preferred embodiment, the device is a silver-silver chloride reference electrode which functions reversibly, i.e., provides the voltage predicted by thermodynamics. In general, such electrodes consist of a silver metal immersed in a solution containing chloride anions. The electrode reaction is:
AgCl(s) + e" - Ag(s) + Cl" At 25°C the electrochemical potential of such an electrode can be computed as:
V(SHE) - 0.2222-0.05915log10acl-
where V(SHE) is the voltage of the electrode of interest versus the standard hydrogen electrode. For a more de¬ tailed discussion in connection with the above, refer¬ ence is made to "Physical Chemistry" by G. W. Castellan, Chapter 17, "Equilibria in Electrochemical Cells", pp. 344-382, Addison-Wesley Publishing Co., Reading, Mass. (1964) .
96/22519 PCIYUS96/00452
-8-
Referring to FIG. 1, the structure of the reference electrode according of the invention is represented in general at 10 in sectional fashion. Probe 10 has a generally cylindrical structure comprised of five principal components including a cylindrically shaped cell retainer or base 12; a cylindrical end cap 14 formed to fit over base 12; and a positioning and transfer arrangement which includes: base sleeve 16; elongate cylindrical transition component or piece 18; and cable assembly or connector 20.
Retainer or base 12 is structured not only to with¬ stand the duress otherwise imposed by radiation, high temperatures, and high pressure, but also to achieve a highly reliable seal that allows for electrolytic con- tact with the electrode and ultimately to the outside reactor environment, but eliminates major incursion of water either from the inside of the reference electrode to the outside environment or vice versa. The base, in its preferred embodiment, is formed of either magnesia or yttria (Y205) partially stabilized zirconia. The zirconia material provides the requisite electrical insulation and is chemically inert. The zirconia has a corrosion rate at least an order of magnitude lower than sapphire. Further, the corrosion rate does not increase with pH shifts and zirconia is resistant to the erosion/corrosion that might exist in specific regions of the nuclear reactor. Accordingly, the zirconia material forming base 12 is ideal for the in-core environment. Base 12 is formed having cylindrical base region 22 which is terminated at its upper end by cylindrical land 24 which is adjacent to cylindrical sidewall 26. Adjacent land 24 and sidewall 26 form the end cap sur¬ face attachment region. Pedestal 30 is adjacent to land 29 which is disposed at the end of sidewall 26 opposite land 24. Pedestal 30 is integrally formed from base 12.
Referring additionally to FIG. 2, pedestal 30 is seen to extend within cavity 28 from base region 22 to flat coupling surface 32. Cylindrical bore or contin¬ uous access channel 34 extends from coupling surface 32 and through base region 22. Channel 34 serves to pro¬ vide access for electrically conductive transmission line or conductor wire 36 which may be formed of kovar or nickel and flattened at its end to form a disk-shaped head 38. Wire 36 is seen to be inserted through channel 34 and inward side 40 of disk 38 is shown in abutment with coupling surface 32 of pedestal 30. Kovar materi¬ als are a group of alloys having a characteristic of thermal expansion making it compatible with that of the zirconia material of base 12. Generally, kovar material comprises 17-18% cobalt, 28-29% nickel, with the balance being mostly iron. One representative kovar material comprises Fe - 53.8%, Ni - 29%, Co - 17% and Mn - 0.2%. This group of alloys, because they are ductile and do not become embrittled under conditions of ordinary use, including heating and annealing, are useful, for exam¬ ple, in sealing glasses.
The first of the internal seals for electrode 10 is developed with respect to the necessary electrical communication provided by wire 36 through the employment of pedestal 30 in conjunction with sealing retainer cap 42 (see FIG. 2), which also is fashioned of platinum or kovar. Retainer cap 42 is formed as a cylinder having a closed end and an internal diameter serving to provide an outwardly disposed seal at its union with a circular cylindrical sidewall 44 of pedestal 30. To achieve a sealed union of high integrity between the concave internal surface of retainer cap 42 and the sidewall 44 of pedestal 30, certain metallurgical procedures are carried out.
In joining the magnesia-stabilized zirconia to the appropriate metal component the present invention uses a technique known as active metal brazing where a small concentration of the active metal, in this case titanium (approximately 5-6 wt.% or less) in a silver paste, becomes chemically bonded to the ceramic.
In the two-step process the silver-titanium paste is placed on cylindrical surface 44 of pedestal 30 and on surface 32 adjacent to channel 34. Wire 36 is inserted through channel 34 and retainer cap 42 is placed over cover disc 38 of wire 36 and cylindrical surface 44. With the retainer cap 42 and wire 36 in place on pedestal 30, the assembly is then heated to brazing temperature in a vacuum furnace and the metal retainer cap 42 is sealed to ceramic surface 44. Simultaneously disc 38 of wire 36 is joined to surface 32. Once the brazing is completed, the entire outside surface of metal retainer cap 42 is covered with excess pure silver, either in the form of foils or with a pre- machined cap of pure silver that is fitted over and completely covers retainer cap 42. The pure silver is then brazed to retainer cap 42.
In the single-step process the joining of retainer cap 42 to surface 44 and the brazing of retainer cap 42 with excess silver are accomplished simultaneously. For proper electrochemical performance the pure silver must completely cover the outside surface of retainer cap 42.
If the retainer cap 42 is fabricated from kovar, a somewhat elaborate procedure of surface treatment is required in view of its presence within a silver chlor¬ ide environment, which is a strong oxidizing agent. It also will be observed that the ultimate coating is sil¬ ver which forms part of the electrode system. In prep¬ aration of retainer cap 42, it is cleaned and inspected, following which it undergoes a nickel strike. The cap is sintered to improve the plating bond, whereupon the
sintered part is again inspected. Alternatively, it also is possible to rhodium plate and sinter directly on the cleaned kovar cap, or platinum plate on the nickel strike or nickel plating. After each plating or sintering operation, inspections are required to assure continuity of the separate platings. Then retainer cap 42 is silver-plated and the silver plating is sintered and then inspected. Cap 42 is silver-plated again as a last step in its treatment. Returning to FIG. 1, the lower, outer cylindrical surface portion of base region 22 of base 12 is a surface attachment region, the extent of which is represented by 46. This region is also painted with titanium-silver paste and then brazed in the same manner and at the same time as the surface of pedestal 30 to provide a seal.
Shown positioned within cavity 28 of end cap 14 and base 12 is a pellet 48 of silver chloride which is shown schematically in granular form as an aqueous suspension. In a preferred arrangement, the silver chloride may be melted and formed into rods, portions or plugs of which may then be located within cavity 28.
End cap 14 is also formed of magnesia-stabilized zirconia. Cap 14 is cup-like in shape, being formed of generally circular cylindrical sidewall 50 and base 52. The cap is dimensioned so as to provide a "tight" fit around base 12 at land 24 and sidewall 26. The noted fit is one which permits electrolytic communication of the reactor coolant water with a very minimum movement or mass transfer of water or material. In effect, a diffusion junction is formed between cap 14 and land 2 /sidewall 26. Exemplary of the type of fit involved, the access opening diameter may, for example, be machined to provide a gap of only 0.0005 in. Further retention of end cap 14 is provided by transverse slots 54 and 56 (see FIG. 3) within which stainless steel wire
(shown in section at 58 in FIG. 1) , is positioned there¬ about in harness fashion and attached at lower connector 20 region of device 10. Alternatively, straps of proper dimension may be fitted into slots 54 and welded to kovar sleeve 16.
Base or retainer 12 of device 10 is initially supported by circular cylindrical base sleeve 16 which, to achieve compatibility with the magnesia-stabilized zirconia base 12 from the standpoint of the thermal coefficient of expansion thereof, is formed of kovar alloy. Note that the internal diameter of sleeve 16 is offset, for example, by counterboring at 58 to provide an acceptance portion suited for receiving and being attached to surface attachment region 46 of base region 22 of base 12 for forming an intimate seal thereat. The initially produced cylinder of kovar for sleeve 16 is prepared by cleaning and inspecting, following which a post-machining annealing procedure is carried out. Following this annealing procedure, the component is nickel-plated, sintered and inspected. Generally, the thus-prepared component is stored in sealed plastic packaging until it is utilized. An intimate seal of surface attachment region 46 of base 12 with the accep¬ tance portion 58 of sleeve 16 is provided by painting the ceramic with titanium-silver paste and then silver brazing at the same time as the brazing of cap 42 to surface 44. This arrangement then completes a highly secure second seal for electrode 10, as is required in view of the intended use thereof within the core region of a reactor.
Hollow interior 60 of sleeve 16 provides an inter¬ nal channel through which wire or conduit 36 may pass. To assure that wire 36 is insulated from the internal surfaces of sleeve 16, a circular cylindrical tube 62 made of ceramic, e.g, alumina, is inserted within chan¬ nel 60. Ceramic tube 62 provides electrical insulation
while remaining immune to the high temperatures encoun-. tered inside the reactor.
Kovar sleeve 16 is supported, in turn, by attach¬ ment to a circular cylindrical transition tube, which may be made of, e.g., Type 304 stainless steel. Tran¬ sition tube 18 has a diameter equal to that of sleeve 16 and is attached at its transition end 64 to a cor¬ responding attachment portion 66 of sleeve 16 by tung¬ sten inert gas welding, e.g., using a tube welder. Hollow interior 68 of transition tube 18 provides an internal channel representing a continuation of channel 60 of sleeve 16. Alumina tube 62 extends into channel 68. The lower end of transition tube 18 is formed in necked-down fashion to provide sealing end 70. End 70 is welded by tungsten inert gas welding to cylindrical stainless steel collar 72 of a cable connector assembly 74 which has a ceramic support component 76 through which mineral insulated cable 78 extends. Cable 78 may be provided with a stainless steel outer shell having alumina mineral insulation disposed inside with conduct¬ ing cable 80 centrally arranged therein. Mineral insu¬ lated cable 78 extends outwardly to the ambient environ¬ ment from the reactor environment region in the applica¬ tion of interest. In the ambient environment the volt- age signals from the reference electrode are obtained. To provide an electric circuit completing connection with lead 80, nickel or kovar conductor 36 is spot welded thereto at 82. To facilitate this attachment and provide a modicum of tension within the nickel or kovar conductor 36, a spring winding is formed in connector 36 as represented in general at 84. Cable assembly 74 is marketed, for example, by Reuter-Stokes, a division of Genera] Electric Company, Twinsburg, Ohio.
The preferred embodiments of the electrode probe made in accordance with the concept of the present invention have been disclosed for the purpose of illus-
tration. Variations and modifications of the disclosed structure which do not depart from the concept of this invention will be readily apparent to engineers skilled in the art of designing electrode probes. All such variations and modifications are intended to be encom¬ passed by the claims set forth hereinafter.