The issue, as summarized by AREVA, was related to maintaining local core cooling, a coolable geometry and long-term core cooling after the initiation of a LOCA. The melting temperature of the silver-indiumcadmium absorber material is well established as 1470 °F. If the material were to exceed this temperature and melt, it would be contained within the clad of the rodlet.
However, if the rodlet cladding were to be in contact with a guide tube, made of a zirconium-based alloy, a eutectic reaction could begin at approximately 1715 °F. AREVA cited severe accident tests reported in a 1985 conference paper(2) that indicated that the eutectic reaction could result in localized melting of the zircaloy guide tube and the stainless steel control rod cladding at approximately 2138 °F.
This localized melting of the control rod resulted in expulsion of essentially all of the molten Ag-In-Cd absorber material above the elevation of failure. After this occurred in the test, the molten absorber material reacted with the cladding on the fuel rods, flowed downward, and solidified when it reached the lower regions of the assembly where the coolant temperatures were lower.
For this severe accident test, this led to localized flow blockage within the fuel assembly.
As noted in AREVA’s Interim Report(1), Section 6.15 of NUREG-1230(3) (Control Rod Performance) cited several severe accident tests that demonstrated eutectic temperature reactions for core materials.
This section discusses experimental results as noted below:
“Recent information from experiments in the NIELS facility at Kernforschungszentrum Karlsruhe (KfK) in Germany (6.15.4), using zircaloy guide tubes and stainless steel control rods show that, at low pressure, failure can occur around 2138 °F (1443 K). This failure is probably due to a eutectic reaction between stainless steel and zircaloy.
The iron-zircaloy and nickel-zircaloy phase diagrams predict that a eutectic reaction can happen around 1736 °F (1220 K). Although the observed zircaloy guide tube failures are less violent than those that would occur in a stainless steel guide tube, they may, in the long run, cause more damage to the rest of the core because the failure has been observed to be a small hole in the zircaloy and stainless steel that allows the Ag-In-Cd alloy to spray out onto the hot fuel rods.
This alloy is liquid at about 1430 °F to 1520 °F (1050 to 1100 K).
All the liquid above the hole was expelled during the experiment and then was observed to flow to the bottom of the rod bundle and block the inlet. Ag also has a eutectic reaction with zircaloy and has been observed to erode or melt the zircaloy cladding.”
While the facts stated above about eutectic reactions are valid, care must be used in reviewing severe accident test results (papers). This point will be elaborated on in more detail later in this paper. As noted in the “Concluding Statement of Position of the Regulatory Staff”(4) on “Acceptance Criteria for Emergency Core Cooling Systems for Light-Water Cooled Nuclear Power Reactors”:
“A 2300 °F limit is also sufficient in the staff’s present opinion to limit cladding damage by eutectic formations, even though the staff Supplemental Testimony suggested 2200 °F limit to preclude a damaging amount of zirconium-nickel or zirconium-iron eutectic (Exhibit 1113, Section 19). The staff clarified that earlier suggestion by stating in response to questioning that if effects of grid spacer flux depression, cladding preoxidation, and other factors were considered, a peak cladding temperature of 2300 °F would be sufficiently low to limit damage by eutectics (Transcript 20, 538-41).”
While the final rulemaking for the Emergency Core Cooling Systems (ECCS) Acceptance Criteria settled on 2200 °F for a peak cladding temperature, the above statement shows that the AEC was aware of eutectic reactions∗. In addition, the mentioning of oxidation, while associated with the fuel cladding in this context, will be shown to have an effect on the zirconium-nickel/zirconium-iron eutectic reactions between guide thimble tubes and control rod cladding.
This paper will describe the various control rod designs that exist in Pressurized Water Reactors (PWRs) in the U. S. and their susceptibility to eutectic reactions. A review of more recent severe accident papers will also provide an insight that these eutectic reactions will not result in loss of local or long-term core cooling or a loss of coolable geometry (i.e., refer to the discussion in the Assessment section).
However, if the design basis for the ECCS Acceptance Criteria is exceeded (i.e., entering severe accident regime), than eutectic reactions could cause significant core degradation.
Control Rod Designs:
There are several control rod designs that exist today in U. S. PWRs. These designs were supplied by the original NSSS vendors. These designs have been duplicated by current fuel vendors (i.e., the same design has been maintained for replacement control rods) or newer replacement designs have been developed that are equivalent to the original control rods in terms of form, fit, and function.
The following is a general listing of these designs followed by a discussion of their susceptibility to eutectic reactions.
1. Full Length Ag-In-Cd absorber with a Stainless Steel clad rodlet
2. Full Length Ag-In-Cd absorber with a Stainless Steel clad rodlet that has Chrome plating for Ion-Nitride plating.
3. Full Length Ag-In-Cd absorber with an Inconel 625 clad rodlet
4. Hybrid Designs:
– B4C absorber with Ag-In-Cd absorber in the lower portion of the control rod with a Stainless Steel clad rodlet
– B4C absorber with Ag-In-Cd absorber in the lower portion of the control rod with an Inconel 625 clad rodlet
5. Full length B4C absorber with an Inconel 625 clad rodlet
6. Full length Hafnium absorber with a Stainless Steel clad rodlet (no longer in use)
There may be other variations on the above designs, but the key to identifying the designs are the materials used. As noted previously, the eutectic reactions of concern are zirconium-nickel and zirconium-iron.
Therefore, the Stainless Steel clad rodlets and the Inconel 625 clad rodlets are treated the same, since both have the nickel or iron necessary for a eutectic reaction. The other key aspect to identifying the various designs is the absorber material used and how it will react at the elevated temperatures experienced during a design basis Loss of Coolant Accident (LOCA).
Some of the designs also have either a plated coating or an ion-nitride surface treatment. The presence of either of these surface treatments can slow or stop the eutectic reaction rates when there is no wear through the protective layer.
The Stainless Steel or Inconel 625 rodlets all have the nickel/iron necessary for eutectic reactions. As documented in Reference 5 (experimental results), the lowest eutectic reaction observed was between Fe–Zr at approximately 1742 °F† on the Zr rich side. The eutectic temperature on the Fe rich side is at about 2373 °F. The eutectic reaction between Ni-Zr occurs between 1760 °F and 2138 °F. These reactions and their corresponding reaction rate equations will be examined in the next section. Again, it is important to remember that the eutectic reactions are between zirconium-nickel/zirconium-iron, thus Stainless Steel and Inconel are treated as the same (since both materials have the nickel and iron) and Zircaloy-4, M5™ and ZIRLO™ are treated the same (since these materials are all 98% zirconium).
The Ag-In-Cd absorber material melts at approximately 1470 °F(6), but is contained within a Stainless Steel or Inconel rodlet. The Ag-In-Cd absorber does not eutectically react with the rodlet materials; however, it does react with zirconium. The Ag-In-Cd absorber in hybrid designs is typically located in
the bottom of the control rod and should remain below the melting point of the Ag-In-Cd absorber material during a LOCA.
The B4C absorber material melts at approximately 4477 °F(6). The B4C absorber does have a eutectic reaction with Stainless Steel or Inconel that commences at 1880 °F; however, rapid liquidification does not occur until approximately 2330 °F(7). This The chrome-plated control rod design would have a eutectic reaction with zirconium at approximately 2336 °F(8)∗. The ion-nitride plated control rod design does not have a eutectic reaction with the
zirconium. Examination of plated control rods, during plant inspections, has shown some limited wear of the plating. The size of the affected area varies by plant, but is usually about 0.5 inches in length.
However, the plating does potentially provide some additional margin to eutectic reactions between the control rod and the zirconium based guide thimble tube if it has not been worn through.
[toggle_simple title=”Related articles” width=”600″]
- October 27th, 2011 – North Anna Power Station, Units 1 & 2 – Dominion Again Quotes EPRI Reports in Documents For Restart Readiness Determination Plan (enformable.com)
- Damaged Spent Fuel Pool No. 4 just had 204 “new” fuel rods inserted before quake + Scientists say another 9.0 megaquake may hit at year’s end = “Fukushima is still on the edge” (enenews.com)
- Doctor finds Uranium and Zirconium in Tokyo resident’s fingernails – “We are becoming nuclear fuel rods” (enenews.com)
- AREVA to supply safety and operational equipment to Chinese Nuclear Stations (enformable.com)
- March 26th, 2011 – Fuel Coolant Interaction Analysis – Worse Case Melt Scenario (enformable.com)
- Palisades Nuclear Station – Loss of both main feedpumps (enformable.com)
- EDF says New Flamanville 3 Nuclear power plant delayed until 2016 after 2 serious accidents (enformable.com)
- Clinton Nuclear Station Loss of RHR Cooling due to incorrect water level indiciation (enformable.com)