4.1 Regulatory Requirements-The USA Code of Federal Regulations (10CFR Part 50, Appendix H) requires the implementation of a reactor vessel materials surveillance program for all operating LWRs. Other countries have similar regulations. The purpose of the program is to (1) monitor changes in the fracture toughness properties of ferritic materials in the reactor vessel beltline region resulting from exposure to neutron irradiation and the thermal environment, and (2) make use of the data obtained from surveillance programs to determine the conditions under which the vessel can be operated with adequate margins of safety throughout its service life. Practice E185, derived mechanical property data, and (r, θ, z) physics-dosimetry data (derived from the calculations and reactor cavity and surveillance capsule measurements (1) using physics-dosimetry standards) can be used together with information in Guide E900 and Refs. 4, 10-17 to provide a relation between property degradation and neutron exposure, commonly called a "trend curve." To obtain this trend curve at all points in the pressure vessel wall requires that the selected trend curve be used together with the appropriate (r, θ, z) neutron field information derived by use of this guide to accomplish the necessary interpolations and extrapolations in space and time.

4.2 Neutron Field Characterization-The tasks required to satisfy the second part of the objective of 4.1 are complex and are summarized in Practice E853. In doing this, it is necessary to describe the neutron field at selected (r, θ, z) points within the pressure vessel wall. The description can be either time dependent or time averaged over the reactor service period of interest. This description can best be obtained by combining neutron transport calculations with plant measurements such as reactor cavity (ex-vessel) and surveillance capsule or RPV cladding (in-vessel) measurements, benchmark irradiations of dosimeter sensor materials, and knowledge of the spatial core power distribution, including the time dependence. Because core power distributions change with time, reactor cavity or surveillance capsule measurements obtained early in plant life may not be representative of long-term reactor operation. Therefore, a simple normalization of neutron transport calculations to dosimetry data from a given capsule is unlikely to give a satisfactory solution to the problem over the full reactor lifetime. Guide E482 and Guide E944 provide detailed information related to the characterization of the neutron field for BWR and PWR power plants.

4.3 Fracture Mechanics Analysis-Currently, operating limitations for normal heat up and cool down transients imposed on the reactor pressure vessel are based on the fracture mechanics techniques outlined in the ASME Boiler and Pressure Vessel Code. This code requires the assumption of the presence of a surface flaw of depth equal to one fourth of the pressure vessel thickness. In addition, the fracture mechanics analysis of accident-induced transients (Pressurized Thermal Shock, (PTS)) may involve evaluating the effect of flaws of varying depth within the vessel wall (4). Thus, information is required regarding the distribution of neutron exposure and the corresponding radiation damage within the pressure vessel, both in space and time (4). In this regard, Practice E185 provides guidelines for designing a minimum surveillance program, selecting materials, and evaluating metallurgical specimen test results for BWR and PWR power plants. Practice E2215 covers the evaluation of test specimens and dosimetry from LWR surveillance capsules.

4.4 Neutron Spectral Effects and DPA-Analysis of the neutron fields of operating power reactors has shown that the neutron spectral shape changes with radial depth into the pressure vessel wall (2, 3). The ratio of dpa/ϕt (where ϕ is the fast (E > 1.0 MeV) neutron fluence rate and t is the time that the material was exposed to an average fluence rate) changes by factors of the order of 2.0/1.0 in traversing from the inner to the outer radius. Although dpa, since it includes a more detailed modeling of the displacement phenomenon, should theoretically provide a better correlation with property degradation than fluence (E > 1.0 MeV) (1, 18), this topic is still controversial and the available experimental data does not provide clear guidance (18, 19). Thus it is recommended to calculate and report both quantities; see Practice E853 and Practice E693.

4.5 In-Vessel Surveillance Programs:

4.5.1 The neutron dosimetry monitors used in reactor vessel surveillance capsules provide measurements of the neutron fluence and fluence rate at single points on the core midplane within the reactor, and near the vessel wall; that is, at the surveillance capsule locations (1). In actual practice, the surveillance capsules may be located within the reactor at an azimuthal position that differs from that associated with the maximum neutron exposure (or that differs from the azimuthal and axial location of the assumed flaw); and at a radial position a few centimeters or more from the flaw and the pressure vessel wall (4, 5). Although the surveillance capsule dosimetry does provide points for normalization of the neutron physics transport calculations, it is still necessary to use analytical methods that provide an accurate representation of the spatial variation (axial, radial and azimuthal) of the neutron fluence (refer to Guide E482). It is also necessary to use other measurements to confirm the spatial distribution of RPV neutron exposure.

4.5.2 Given that surveillance capsules are located radially closer to the core than the surface of the RPV, they may be shifted azimuthally away from the peak exposure location in order to limit the magnitude of the surveillance capsule lead factor. The lead factor is defined as the ratio of the fast neutron fluence at the center of the surveillance capsule to the peak fast neutron fluence at the clad - base metal interface of the RPV. One adverse effect of this azimuthal shift away from the peak is that the surveillance capsule dosimetry does not "see" the part of the core that produces the peak exposure of the reactor vessel. As a result, the surveillance capsule is unable to monitor the effect of changes in the core power distribution that are made to reduce the peak RPV neutron exposure. Another adverse effect is that with larger lead factors, the capsules are rapidly exposed to a high neutron fluence. For example, with a lead factor of five, a surveillance capsule will receive an exposure in as little as 12 years that is equivalent to what the reactor pressure vessel peak may see in 60 years of operation. Practices E185 and E2215 suggest not exceeding twice the maximum design fluence (MDF) or twice the end-of-license fluence (EOLF). In this example, this would require withdrawing any remaining surveillance capsules after 24 years of operation. Thus, without taking other steps, the reactor would be operated for the remaining 36 years (of a 60-year life) with no dosimetry present.

4.5.3 New or replacement surveillance capsules should recognize and correct operating deficiencies by using improved capsule dosimetry. For example, for one class of PWR, the copper wire is cadmium shielded to minimize interference from trace amounts of cobalt. In about one third of the measurements the copper has become incorporated into the cadmium preventing separation and further processing. A simple solution to this problem is to use stainless steel hypodermic tubing to contain and separate the radiometric monitor wire inside the cadmium tubing. Example dimensions include: Typical radiometric monitor wire outside diameter = 0.020 in. (0.5 mm). Typical 19 gauge stainless steel tubing is 0.042 in. outside diameter by 0.027 in. inside diameter, 0.008 in. wall thickness. Typical cadmium tubing is 0.090 in. outside diameter by 0.050 in. inside diameter, 0.020 in. wall thickness.

4.5.4 For one class of BWR reactor, the surveillance capsule dosimetry is minimal; consisting of an iron wire and a copper wire (sometimes also a nickel wire). This dosimetry is not suitable for longer irradiations as the "memory" of the activation products is too short to measure the accumulated fluence. Practice E844 states that radionuclides with half-lives less than one third of the irradiation duration should be avoided. For example, for the iron (n,p) activation product, ⁵⁴Mn, the half-life is 312 d. For the copper (n,α) activation product, ⁶⁰Co, the half-life is 5.27 y. After three half-lives the remaining activity is on the same order as the counting statistics. The result is that the iron wire has "forgotten" everything that has happened more than two cycles ago and the copper wire has forgotten everything that has happened more than eight cycles ago. This assumes 24-month-long fuel cycles. Note that the copper (n,α) reaction is induced by high energy neutrons and that at a BWR surveillance capsule position only 1 % to 3 % of the fast (E > 1.0 MeV) neutrons are of high enough energy. This limits the value of the copper wire as a neutron fluence monitor. In order to monitor the neutron exposure of the RPV other dosimetry is needed. Installation of ex-vessel neutron dosimetry is the most reasonable and cost-effective option.

4.5.5 The neutron fluence calculation on the RPV inner surface can be further verified by means of analyzing small samples of the irradiated stainless steel RPV cladding. Analyzing RPV cladding samples has been a well-established practice for over 30 years (20-35). During the reactor shut down periods, small samples (50-100 mg) can be machined from the RPV cladding. For retrospective dosimetry purposes the measured ⁵⁴Mn, ⁵⁸Co, and ^93mNb activities are used. Because of its long half-life, ^93mNb is especially useful for integrating fluence over time periods where accurate neutron transport calculations are not available. With sample locations properly selected, the fast neutron fluence distribution and its maximum on the RPV inner surface can be determined. By comparison of these data to the dosimetry data of the surveillance capsules, the lead factor at the time of measurement can also be obtained. This technique works best if the cladding material is one of the niobium-stabilized stainless steels. Type 347 with 0.7 % niobium is one example. Retrospective dosimetry has been successfully demonstrated for ordinary Type 304 stainless steel cladding with only a trace (~ 50 ppm) of niobium (34). It is important that the cladding surface is first polished to remove radioactive corrosion products before the sample is machined otherwise competing activity may compromise the sample. The tooling used to take these samples needs to be accurately located relative to reactor landmarks in order to know the actual axial and azimuthal locations of the samples. A reasonable accuracy target is ±25 mm axially and azimuthally. The effect of the sampling position error can be estimated by examining the spatial fast neutron fluence rate gradient in the vicinity of the sample point. In general, in the areas where the fast neutron fluence is the greatest, the gradient tends to be very small; approaching flat in the case of the axial distribution opposite the middle of the core. At extreme axial positions, well beyond the ends of the core, the gradient is steep. There the positioning error could lead to an estimated fluence error of ±20 %. A similar discussion applies to the azimuthal fluence rate gradients. The tooling also needs to be designed to completely retain all machined cladding chips and to prevent cross-contamination from one sample to another. Access to the full extent of azimuthal and axial clad samples is generally limited to PWRs due to the extensive structure (jet pumps, etc.) blocking general access to the RPV cladding of many BWRs. It may be possible to take a more limited set of samples from the cladding of a BWR RPV.

4.5.6 The design and manufacture of new reactor pressure vessels should consider using one of the stainless steels or Inconel alloys that contains niobium for the purpose of cladding the inner surface of the vessel. This would result in a designed-in retrospective dosimetry system that would capture neutron exposure data from reactor startup.

4.6 Ex-Vessel Surveillance Program:

4.6.1 Ex-vessel neutron dosimetry (EVND) has also been in wide scale application in nuclear reactors for over 30 years (27, 28, 30, 32, 34, 36-96). The main advantages of EVND are the relative simplicity and the relatively low cost of the dosimetry system. Removal and replacement of irradiated dosimetry takes little time. Typical installations have dosimetry that spans the active core height and continues to cover the extended beltline region of the RPV. The extended beltline is defined as those portions of the RPV where the accumulated neutron fluence (E > 1.0 MeV) at the end of reactor operation will exceed 10¹⁷ cm^-2. Installation of dosimetry at multiple angles allows full octant coverage (for octant symmetric cores). Some EVND installations include multiple measurements at symmetric azimuthal angles to confirm symmetry in the azimuthal fluence rate distributions. Asymmetries may result from such things as non-symmetric core power distributions, differences in water temperatures from one loop to another, or ovality in the as-built dimensions for the reactor internals or RPV. Dosimetry capsules typically contain a full complement of radiometric monitors (refer to Practice E844) to ensure good spectral coverage and fluence integration. Typically, capsules are connected and supported by stainless steel wires or chains, which are, in turn, segmented and counted to provide axial gradient information.

4.6.2 In order to minimize measurement field perturbation, the dosimeter capsules should be made of a neutron-transparent material such as aluminum. This also serves to reduce the radiation dose rates encountered when removing and replacing dosimetry. The gradient chains or wires should be a low mass per linear foot material, again to reduce the dose rates encountered during handling of irradiated dosimetry.

4.6.3 An ex-vessel neutron dosimetry system needs to be accurately located with respect to well known and easily verified reactor features. A reasonable accuracy target is ±25 mm axially and azimuthally. The effect of the dosimetry position error can be estimated by examining the spatial fast neutron fluence rate gradient in the vicinity of the measurement point. In general, in the areas where the fast neutron fluence is the greatest, the gradient tends to be very small; approaching flat in the case of the axial distribution opposite the middle of the core. At extreme axial positions, well beyond the ends of the core, the gradient is steep. There the positioning error could lead to an estimated fluence error of ±20 %. A similar discussion applies to the azimuthal fluence rate gradients.

4.6.4 Ideally, the ex-vessel neutron dosimetry is installed before reactor startup so that it can provide data over the operating lifetime of the reactor. It is recommended that the ex-vessel neutron dosimetry be analyzed before and after significant plant modifications that would alter the neutron exposure of the reactor vessel. Some examples include switching from low-leakage core loading patterns back to out-in loading patterns (or vice versa), performing a significant (>10 %) uprating of the plant power, adding (or removing) core flux suppression absorbers or dummy fuel rods, or modifying the reactor internals geometry. The typical dosimetry replacement interval is between one and five 18-month-long fuel cycles (or equivalent intervals for other fuel cycle lengths).

4.6.5 Periodic measurements (either RPV cladding samples or EVND) serve to confirm neutron fluence projections and help to avoid problems that result from errors in reactor-specific calculational models (97).

4.6.6 Calculations of neutron fields in commercial reactors show that the neutron exposure (dpa) at the inner diameter of the pressure vessel can vary by a factor of three or more as a function of azimuthal position (2, 3). Dosimetry monitors in the reactor cavity outside the reactor pressure vessel are a useful tool, therefore, in determining the accuracy of the neutron field calculations at points inside the pressure vessel wall. Practice E853 recommends the use of ex-vessel reactor cavity neutron dosimetry measurements for verification of the physics transport calculations. The status of benchmark field and power reactor applications as well as studies of this approach are discussed in Refs. 1, 17, 18, 36-39, 98-111.