ML20205L317

Staff Research Paper (NuScale) for the ACRS Fc Meeting 07/21/2020

ML20205L317
Person / Time
Site:	NuScale
Issue date:	07/20/2020
From:	Peter Yarsky Office of Nuclear Regulatory Research
To:
Bavol B
References
NRR-2020-015-IAR
Download: ML20205L317 (13)

Text

US Nuclear Regulatory Commission Office of Nuclear Regulatory Research TER on Boron Redistribution Issues Raised by the ACRS Technical Evaluation Report on Boron Redistribution Issues Raised by the Advisory Committee on Reactor Safeguards. Work done to support NRR-2020-015-IAR.

Yarsky, Peter 7-20-2020

1 Introduction By e-mail dated July 13, 2020 (ADAMS Accession No. ML20196L845) as supplemented by e-mail dated July 15, 2020 (ADAMS Accession No. ML20198M646), the Office of Nuclear Reactor Regulation (NRR) requested that the Office of Nuclear Regulatory Research (RES) staff provide technical evaluation assistance in responding to questions raised by the Advisory Committee on Reactor Safeguards (ACRS) regarding boron redistribution phenomena during certain postulated events for the NuScale power module (Ref. 1). RES accepted this request and tracks this request under the identifier: NRR-2020-015-IAR. The current work builds on work previously requested by NRR under NRR-2020-014-IAR, which culminated in a white paper developed by the RES staff (Ref. 2). The NRR staff presented the RES white paper findings to the ACRS on July 8, 2020. Following the meeting of the ACRS members, the ACRS staff transmitted several questions for consideration by the staff, which is the subject of the current request. These questions are available in the supplemental e-mail dated July 15, 2020 (Ref. 3).

While several questions are listed, only a subset are related to the work done by RES under the predecessor IAR. These questions are reproduced below:

(1) There are different points of opinion expressed on the subject; what seems to be the main point of contention: is it a reactivity insertion rate, an impact of this insertion on fuel, or? What type of evaluation can be done to resolve these differences?

(2) Are SBLOCA [small break loss of coolant accident] sequences (CVCS [chemical and volume control system] / CFDS [core flood and drain system] injection after a prolonged ECCS [emergency core cooling system] injection or a partial ECCS failure) also analyzed with all rods inserted; is there an analysis which confirms that there are no issues with reactivity insertion due to boron dilution in that case?

(3) Was there any discussion related to instrumentation available to the operators, and type of recovery actions for diluted downcomer?

(4) Estimates of time ranges (for different conditions/events) when downcomer coolant boron concentration becomes an issue, in the cases of:

• Prolonged ECCS injection,

• Delayed ECCS injection,

• What are the NuScale sensitivity studies mentioned in PRA [probabilistic risk assessment] Slide #35.

This technical evaluation report documents the RES staff responses to these questions.

2 Question 1: Main Point of Contention The main point of contention - in the opinion of the RES staff - relates to the importance of internal recirculation driven mixing in the core and riser region. In the white paper developed by the staff under NRR-2020-014-IAR (Ref. 2), the staff argues that flow incursion following boron redistribution can be caused by several mechanisms, which occur over differing time scales.

When considering the consequences of these mechanisms, however, it is important to evaluate 1

the potential for these mechanisms to insert reactivity quickly enough relative to other processes that will counteract the accumulation of positive reactivity.

The staff did not classify the identified flow incursion mechanisms into different subcategories, but to aid in subsequent discussions, it is worth dividing the mechanisms into two broad groups.

The first group of mechanisms are those flow incursions that produce a rapid flow pulse. For example, the level swell and flush that occurs following ECCS actuation is a transient initiator that will cause a very rapid flow response - but after just a short period of time the driving force for the flow increase dissipates and the flow returns to a relatively stagnant condition. Another example is condensation void collapse in the riser driven by CVCS injection into the riser above the core. This injection will collapse voids, lead to shrinkage, and cause a flow surge - but once the void in the riser is collapsed, there is no sustained driving force to perpetuate the increased core flow. Therefore, these mechanisms can be classified as transient flow incursion mechanisms.

The second group is to be contrasted with the first group in that these mechanisms represent phenomena or processes that can result in a sustained and prolonged increase in core flow rate. An example is manual injection of water either through the CVCS or CFDS. A prolonged injection will cause a continuous increase in the water level and lead to a persistent elevated core flow.

Having differentiated between transient and prolonged mechanisms, the role of neutronic feedback and mixing can be more clearly discussed in terms of the relative importance. For transient mechanisms, the processes are generally quite rapid and, therefore, there is not an opportunity of mixing to play a significant role in the evolution of the nuclear response of the core. It may or may not be the case for the transient mechanisms in which thermal-hydraulic feedback plays a role. This is because thermal-hydraulic coupling occurs on a time scale dictated by the fuel rod thermal inertia. This time frame is commonly characterized by a fuel thermal time constant. The fuel thermal time constant appears in the heat conduction equation in an exponential term and dictates the time it takes for the surface heat flux to steady after a step change in the rod power. In the staffs estimation this time constant is between 5 and 7 seconds. Several of the transient phenomena considered occur over just a few seconds and so it is not clear how significant a role the thermal-hydraulic kinetic feedback will be in assessing event progression and consequences. Regardless, the fuel temperature kinetic feedback is prompt and acts immediately to dampen any power increase in response to flow incursion.

In the previous analysis, the staff relied on using the Fuchs-Nordheim approximation to compute the energy deposition that might result from transient flow incursions. In these calculations to reactivity insertion from the flow incursion is assumed to be completely counteracted by Doppler reactivity worth (i.e., an adiabatic approximation) and the results showed that these flow incursions were not sufficient to challenge fuel damage limits (Ref. 2).

For prolonged mechanisms, the staffs analysis approach was different. The staff estimated the rate of reactivity insertion in units of $/min. For the incursion to present a feasible challenge to the fuel damage threshold the reactor must reach a condition of prompt criticality and experience a power excursion. However, if the prolonged mechanism leads to a very slow reactivity insertion rate, the reactivity cannot accumulate to overcome the subcriticality in the core.

2

Consider the situation where the downcomer is deborated. Under this condition the boric acid concentration in the core and riser region is very high. It may be as much as 10 times the nominal boric acid concentration at the initiation of the event. If a prolonged mechanism occurs, for example, the operators initiate the CVCS to inject into the downcomer, this will cause the level in the reactor pressure vessel (RPV) to increase. The steady increase in the RPV level will cause the downcomer inventory to flow into the core. The staff translated that rate into a reactivity insertion rate. One element of confusion may be to interpret this reactivity insertion rate as a continuous ramp of reactivity insertion. The staff contends that this is not the appropriate paradigm to consider this reactivity insertion. Rather, as deborated water is transferred from the downcomer to the core region there will be an internal recirculation flow pattern that will mix the newly transferred water with the existing inventory. Later in this section of this document the staff provides some additional discussion regarding the phenomenology and experimental evidence for such mixing.

Since the existing core inventory is very highly borated, the positive reactivity inserted by the flow of the deborated water from the downcomer to the core does not accumulate at the calculated rate over time. Rather, there is a removal rate for that reactivity that is a function of the mixing. As the deborated water mixes the average boron concentration in the core decreases, but the coolant (post mixing) remains at such a high boron concentration that it is as if the potential positive reactivity introduced by the deborated coolant is completely erased.

Therefore, that reactivity insertion rate calculated by the staff in the white paper must be interpreted as being only one piece of the overall dynamic picture. In any specific scenario it would be easy to become bogged down in the details of trying to simulate all the phenomena and processes affecting the transient progression of the boron concentration, power, void, evaporation rate, and flow response. Rather than approach the problem from this angle, the white paper tries to explain that if this reactivity insertion rate is so slow that it takes longer to insert one dollar than it takes for the core inventory to mix, then it is not possible for the insertion to accumulate enough positive reactivity over time to generate a super prompt critical power excursion.

The introduction of the deborated water may increase the reactivity such that the reactor achieves criticality and this, in turn, results in the power increasing or fluctuating - the staff did not perform a detailed evaluation of these dynamic processes because the staff does not consider such power fluctuations to pose a challenge to fuel damage limits so long as the net reactivity remains below the super prompt critical threshold (i.e., below one dollar). However, any increase in power will likely promote void generation, which in turn, should promote mixing.

With this clarification, it appears that the main point of contention is whether there is a mechanism driving the internal recirculation in the core and riser region. The staff contends that internal recirculation provides the flow pattern that will cause the core and riser inventory to mix.

TRACE calculations were referenced in the white paper to illustrate such internal recirculation.

However, it appears to the RES staff that there is a question on the part of the ACRS of what phenomena or processes drive the internal recirculation and whether any systems analysis tool can be relied on to calculate that flow pattern.

The staffs white paper presents discussion about one-dimensional methods. Clearly one-dimensional methods are incapable of predicting internal recirculation. While the staff relied on three-dimensional calculations in the analysis, it is clear that modeling assumptions and practices could substantially affect a systems analysis codes prediction of the internal recirculation and the staff would expect that design basis calculations performed with the 3

applicants one-dimensional method (NRELAP-5) would predict very different results. The staff is aware of some calculations performed by the applicant using two-dimensional modeling, but such methods have not been reviewed by the staff in detail.

Therefore, it is worth considering additional evidence that such an internal recirculation flow pattern would develop. Internal recirculation and its effect on safety analysis has been considered for many years because of the importance of this phenomenon not only to NuScale, but because it is important for other passive reactor designs. The internal recirculation in passively cooled plants prompted significant experimental and analytical work to support safety analyses. A good overview of experiments is provided by Hibiki and Ishii in Ref. 4. While the focus of Ref. 4 is on the development of drift flux models for large diameter pipes, the paper summarizes well the findings over many experimental campaigns that internal recirculation occurs near pool boiling conditions in both large diameter pipe geometries as well as bundle geometries. One of the key processes that affects the three-dimensional flow field is the channeling of voids into a column which allows for a central region of the flow with high void and high upward velocity to be contrasted with the periphery of the flow field.

This channeling as well as internal recirculation was also observed by Bosemans and Berghmans where the interfacial shear was shown to produce rapid bubble velocity in the central core of the flow with liquid recirculation (Ref. 7). Bosemans, et al., were able to explain their observations with a simple theoretical model of the internal recirculation: the channel flow is a lower density due to higher void and therefore is being driven upward by gravitational potential relative to the denser, low void periphery. This driving potential recirculates the flow such that this potential is counterbalanced by turbulent shear forces and the net liquid flow across the cross-section remains at zero.

While the low flow, pool boiling data referenced by Hibiki, et al., are probably the most applicable data to demonstrate the internal recirculation phenomenon, the staff also considered some experimental data and computational fluid dynamic (CFD) simulations performed at the Large scale Investigation of Natural circulation and miXing (LINX) facility. The LINX experiments and analysis are interesting because the tests were air/water experiments and were adiabatic. The tests and analyses demonstrate that the presence of void, even without wall heat transfer, was still sufficient to produce the channeling and internal recirculation flow pattern (Ref. 5). These LINX tests are not directly applicable to the NuScale configuration, but are consistent with the theoretical framework developed by Bosemans, et al., to explain the internal recirculation. In the NuScale power module these flow patterns would very likely be enhanced by the riser wall heat conduction, which would serve to strengthen the gravitational potential driving the internal recirculation.

Additionally, such flow patterns have been observed in larger scale three-dimensional reactor geometry experiments, namely cylindrical core test facility (CCTF) and slab core test facility (SCTF). The relevant tests were performed to study steam binding phenomena during large break loss of coolant accident, but internal recirculation flow patterns between the core and upper plenum were observed in these tests. The researchers determined that this flow pattern was enhanced by the radial power distribution in the core which has a higher void generation rate in the central region compared to the periphery (Ref. 8 and 9). While TRACE has been assessed against these tests, the internal recirculation was not a specific figure of merit for the assessment, however, TRACE has able to reasonably reproduce the sensitivity of the reflood behavior to changes in the radial power distribution (Ref. 10). Since there is a similar radial 4

power distribution in the NuScale reactor core, this should also promote the internal recirculation.

Perhaps the most compelling experimental evidence is data collected from the PKL facility in 2010. The Primrkreislauf Primary Circuit Reactor Coolant System (PKL) III G5.1 test was aimed at assessing boron mixing in the core and RPV during postulated large break LOCA scenarios. In the test the ECCS injection rate was lowered to match evaporation rates in the core to sustain a prolonged boron enrichment phase whereby the boil-off in the core would result in ever increasing boron concentration. The purpose of the test was to study boron precipitation. The level was lowered so that only steam would flow to the steam generators and the core was covered by two phase mixture. The facility is instrumented with conductivity probes to measure the boric acid concentration at various axial and radial locations in the RPV around the core volume and these measurements were supplemented by sampling grab measurements. Therefore, these tests measure the evolution of the spatial redistribution of boron within the RPV directly. The experimental setup and results are provided in Ref. 11.

Because the test data are sensitive, the results are not reproduced here, but the experimenters found that there was a rather uniform boron concentration above and below the core throughout the test, which indicates rather effective mixing of the liquid in the lower region with the highly concentrated fluid in the two-phase region. Since the test includes a phase where level was lowered to prevent liquid flow to the steam generators and ECCS flow was lowered to match boil-off, these conditions are quite analogous to the expected thermal-hydraulic conditions in the NuScale RPV during postulated LOCA events.

In summary, the staff has identified a series of separate effects and integral effects tests that demonstrate the internal recirculation phenomenon. In addition, the integral effects tests have shown: (1) voiding even under adiabatic conditions drives internal recirculation (i.e., LINX), (2) the effect of radial power distribution to enhance the internal recirculation effect (e.g., CCTF) and (3) how internal recirculation homogenizes the axial and radial boron distribution (i.e., PKL).

One might argue that after a sufficiently long time, the power level may become so low in the reactor core that there is no driving mechanism to produce void and this would significantly reduce the internal recirculation. However, the decay heat can be expected to remain above 0.2 percent of rated thermal power for up to a week after shutdown (Ref. 6). Combined with the exceptionally low pressure and essentially stagnant flow conditions, this implies that there will be void generation in the core for at least this long.

One might further argue that in the long-term the downcomer might become very subcooled, and this might preclude void formation in the core if the downcomer inventory was introduced to the core. The staff finds this unlikely because the staffs own TRACE calculations have indicated that the downcomer temperature remains relatively close to saturation (Ref. 2), so even quite low power levels would lead to boiling. Operators may inject water through the CVCS or CFDS and this may be much colder - however - as discussed in the staffs previous white paper, such injection sources are highly borated. Therefore, if such low temperature water were to reach the core, for it to be cold, it would have not mixed with the existing inventory (which would raise the temperature) and therefore, the water would still have a high boron concentration anyway.

The staff could not reason what process or phenomenon could preclude the development of internal recirculation in the NuScale power module. The staffs literature review indicated experimental evidence in large diameter pipes, bundles, scaled reactor coolant system 5

geometries and even under adiabatic conditions. These tests appear to show the consistent evolution of internal recirculation in pool boiling conditions. While this appears to be the main point of contention, the staff concluded that such internal recirculation is a genuine physical phenomenon and that it could be reasonably expected to occur in the NuScale power module during the postulated events.

3 Question 2: SBLOCA with All Rods Inserted Flow incursion under fully rodded conditions is not expected to produce fuel damage consequences. The presence of a significant population of control rods inserted into the core ensures a static negative reactivity worth even if the flow incursion displaces the highly concentrated coolant in the core region. Therefore, for the flow incursion to develop some potential for core damage, it would necessarily need to be of the prolonged type. It would further have to be postulated that the prolonged mechanism would add sufficient reactivity to first overcome the negative reactivity worth of the rods, then continue to add reactivity until a net positive reactivity accumulates above one dollar. The staff finds this scenario to be exceptionally unlikely because the slow accumulation of reactivity is counteracted by internal mixing, which acts to erase the inserted reactivity over time.

This rationale is discussed at much greater length in Section 2 of this white paper, but to summarize, as deborated water flows into the core it will mix with the highly concentrated coolant in the core and riser section. This mixing will slightly decrease the boric acid concentration in the core, but will ensure that the core remains highly subcritical. It would only be possible for a prolonged flow incursion mechanism to overcome the negative reactivity worth of the rods if the downcomer water displaced the core inventory and did not mix. This absence of mixing would need to persist for a substantial amount of time, maybe longer than 10 minutes, and it is not clear to the staff what process or phenomenon would preclude mixing for that duration.

4 Question 3: Instrumentation Availability during Recovery Actions Emergency operating procedures and potential recovery actions have not been submitted for NRC review during the NuScale design certification review as such procedures are generally site-specific and are submitted by combined license applicants. Such procedures depend on site-specific considerations, for example, the operability of different systems could depend heavily on the availability of alternating current (A/C) power, which depends on the specific site.

The RES staff did not perform any technical review with respect to recovery procedures or operations. However, the staff sought to provide some context for the ACRS to facilitate their deliberations on the feasibility of using existing instrumentation to monitor reactor conditions during postulated recovery operations. The CVCS boric acid concentration instrumentation line during long-term LOCA will not be operable and cannot be used to provide real-time measurements of the reactor coolant system (RCS) boric acid concentration. However, this does not imply that there are no instruments that could be used to monitor the reactor conditions during a postulated recovery.

During start-up procedures for many plants, including the current operating fleet, nuclear instrumentation (in- and ex- core monitors) are useful in determining the subcritical margin. This is a key element in most plant startup procedures. Because of the accumulation of fission 6

products the core is producing a low level of neutron flux even after extended periods of shutdown. That background neutron flux is amplified by subcritical multiplication. As the core multiplication factor increases (during startup this may be a result of rod withdrawal or boron dilution) the operators monitor the core subcriticality using a 1/M plot technique. The 1/M plot technique uses nuclear instrumentation measurements to infer the amount of subcriticality and allows the operators to carefully approach core criticality.

It is conceivable that the same philosophy could be applied to monitor core subcriticality during postulated recovery operations without the need for any additional instrumentation dedicated to this purpose.

5 Question 4: Time Range Estimates The staff performed TRACE calculations for normal and delayed ECCS actuation to address the ACRS question. The prolonged ECCS actuation case is the base case with normal ECCS actuation (which occurs based on the RPV pressure reaching the setpoint, 900 psia in this case). The ECCS actuation in the calculations was delayed in sensitivity calculations by lowering the setpoint.

According to the staffs calculation for a 7% SBLOCA event, the time between the start of dilution (~1000 s) and ECCS actuation (~1700 s) is about 700 seconds. Right before ECCS actuation, the boron concentrations in the riser and the downcomer differed by 50 ppm (i.e.,

1036 ppm in the core and 985 ppm in downcomer). Following ECCS actuation the downcomer boron concentration in the short term increases due to the flashing of liquid inventory in the downcomer. Figure 1 shows the level response and boron concentration in the core and downcomer during the nominal case for the first hour.

7

Figure 1: Prolonged ECCS Actuation By lowering the RPV low pressure setpoint the ECCS actuation can be delayed. The staff first lowered the set point from 900 psia to 600 psia. The ECCS actuation was delayed to 2775 seconds (from 1717 s) and the difference in boron concentration in the core and downcomer increased from 50 to 280 ppm (i.e., 1202 ppm in the core and 920 ppm in downcomer). To further delay the ECCS actuation, the staff reduced the set point further to 500 psia. In order to achieve this case, the staff had to tentatively lower the High RCS temperature interlock set point from 475 F to 380 F. This was to avoid ECCS non-actuation due to the interlock. With these changes, the ECCS actuation delayed further to 3263 seconds. Further boron dilution in downcomer was observed accordingly. The boron concentration difference increased to 388 ppm (i.e., 1263 ppm in the core and 875 ppm in downcomer). After the ECCS actuation, the concentrations in the core and the downcomer continued to rise due to flashing. Figure 2 shows the delayed ECCS actuation sensitivity study results for the first 6000 seconds.

8

Figure 2: Delayed ECCS Actuation To investigate the post-ECCS boron concentration trend, the staff extended the run for another 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br />. At this time, the RCS depressurized to low pressure (around 40 psia). Figure 3 shows the TRACE results, but these calculations predict mixing between the core and downcomer inventories so there is a relatively constant concentration difference between the core and downcomer. This core/downcomer mixing arises from predicted flow oscillations between the core and downcomer and, unlike the mixing caused by the internal recirculation, there is a much higher degree of uncertainty in this mixing mechanism. Such mixing predictions would be much more highly sensitive to numerical aspects of the calculation such as node size and time step size and would be subject to numerical diffusion effects in standard LOCA calculation setups. For this reason, the staff recommends artificially blocking this mechanism in systems analysis because the contribution of flow oscillation mixing cannot be well quantified.

While not useful to the current analysis, the results are still presented to illustrate one of the challenges to using systems analysis tools to calculate the evolution of the boron concentrations during design basis accidents using production methods. An area of potential future work might be to craft a new TRACE deck to simulate this event that adjusts the loss coefficients to limit or preclude the contribution of oscillations to cross core/downcomer mixing, but such work could not be completed on the time frame needed to fulfil the scope of the current request.

9

Figure 3: Long-term Boron Concentration in Delayed ECCS Actuation Because TRACE predicts mixing between the downcomer and core regions the results are not useful in extrapolating the time when the downcomer concentration reaches the critical boron concentration or completely deborates. To meet the scope of the incoming request, the staff developed a simplified hand calculation approach to estimate the time frames of interest.

Using just the TRACE predicted concentrations immediately following ECCS actuation, hand calculations were performed to compute the time it takes to dilute the downcomer assuming that water is transferred to the core only to compensate for boil-off. Therefore, calculations were performed at various decay heat power levels and various initial concentrations. The hand calculation results are summarized in Table 1. The critical boron concentration is approximately 1000 ppm and will likely increase in the downcomer region following ECCS actuation due to flashing, but this is not credited in the hand calculation. To gauge the time scale of the boron dilution, the staff considered the time it would take to reach 90 percent dilution (i.e., 100 ppm) and to reach 99 percent dilution (i.e., 10 ppm). The methods are very approximate but appear to indicate that the downcomer concentration reaches 100 ppm within the time frame of one day to one week and then will further dilute to a concentration of 10 ppm between about one week and one to two months.

10

Table 1: Prolonged and Delayed ECCS Scenarios Downcomer Dilution Timing Approximate Time of Time to Time to DC Reactor Steaming ECCS reach 100 reach 10 Concentration Power Rate Actuation ppm ppm post-ECCS seconds ppm %RTP kg/sec days days 1700 1000 1.0 0.74 0.94 10.3 2800 900 1.0 0.74 0.83 9.3 3300 800 1.0 0.74 0.73 8.2 1700 1000 0.5 0.37 1.87 20.6 2800 900 0.5 0.37 1.66 18.5 3300 800 0.5 0.37 1.46 16.4 1700 1000 0.2 0.15 4.68 51.5 2800 900 0.2 0.15 4.16 46.3 3300 800 0.2 0.15 3.64 41.1 This timing information is provided for the information of the ACRS, but the staff notes that the preceding analysis documented in Ref. 2 presumes a fully diluted downcomer when considering manual level recovery scenarios. The staffs previous calculations showed there was no realistic fuel damage consequence following manual injection to recover level even if the downcomer becomes fully deborated.

6 References

1. Work Request Form NRR-2020-015-IAR (ADAMS Accession No. ML20196L840)

2. Yarsky, P., NuScale Core Reactivity Consequences of Boron Dilution and Incursion during Postulated SBLOCA with Coincident Rod Insertion Failure Events July 2020 (ADAMS Accession No. ML20191A069).

3. E-mail from Rosenberg, S., to Pohida, M., et al., FW: Questions for Consideration in Support of ACRS Meeting on Tuesday, July 21st, dated July 15, 2020 (ADAMS Accession No. ML20198M646).

4. Hibiki, T. and Ishii, M., One-dimensional Drift Flux model for Two-phase Flow in a Large Diameter Pipe, Intl. Journal of Heat and Mass Transfer 46 (2003) pp.1773-1790.

5. Staedke, H., Franchello, G., Worth, B., Graf, U., Romsedt, P., Kumbaro, A., et al.,

Advanced Three-dimensional Two-phase Flow Simulation Tools for Application to Reactor Safety (ASTAR), Nucl. Engr. And Design, 235 (2005) pp. 379-400.

6. E-mail from Thompson,J., to Yarsky, P., RE: Literature Review re: NRR-2020-015, dated July 16, 2020.

7. Bosemans, B and Berghmans, J., Level swell in pool boiling with liquid circulation, Intl.

Journal of Heat and Mass Transfer 38 (1995) pp.989-998.

8. NUREG/IA-0127, Reactor Safety Issues Resolved by the 2D/3D Program, July 1993 (ADAMS Accession No. ML062560279)

9. "Analysis Report on SCTF Core-I and II Reflood Test," prepared by Japan Atomic Energy Research Institute, JAERI-Memo-01-348.

10. TRACE V5.0 Assessment Manual Appendix C: Integral Effects Tests 11

11. PKL III G5.1 Test Report PTCTP-G/2011/en/0004 Rev. B, Investigation on Boron Precipitation following a Large Break LOCA, March 2011 (ADAMS Accession No. ML14099A208).

Contributors: Dr. Peter Lien and Jason Thompson made substantial material contributions to this report. Dr. Lien performed several TRACE calculations and Mr. Thompson performed a literature review.

12