ML20129J482: Difference between revisions

From kanterella
Jump to navigation Jump to search
(StriderTol Bot insert)
 
(StriderTol Bot change)
Line 1: Line 1:
{{Adams
#REDIRECT [[IA-84-928, Issue 2.1.1 - Plant Characterization]]
| number = ML20129J482
| issue date = 05/01/1984
| title = Issue 2.1.1 - Plant Characterization
| author name = Eng T
| author affiliation = NRC
| addressee name =
| addressee affiliation =
| docket =
| license number =
| contact person =
| case reference number = FOIA-84-928
| document report number = NUDOCS 8506100103
| package number = ML20129J459
| document type = NRC TECHNICAL REPORT, TEXT-SAFETY REPORT
| page count = 6
}}
 
=Text=
{{#Wiki_filter:. t"  .
Issue Number    2.1.1 Title      Plant Categorization
                                                                                                          ~
Prepared by: T. Eng                    Signature of Author: INJ              S v    v Draft i    2 Date    4/30/84
      ^
Contractor / Consultants  F. Harper /SNL, D. Kunsman/SNL Review Process (List here contractor who will supply a statement on issue relative toIDCORwork)
Initial                      Data Branch Chief            M              Y /3* / t'r Asst. Div. Dir.. D                    v/a,/rv Division Dir.      #                  f// /N Tech. Series Div. Director _
f///M i                    Date Sent to ACRS
'                    Date Sent to IDCOR 8506100103 850104                                            fg / $ . gLf
* j 2,$
PDR sNLL$s4-928 A[%
 
e ISSUE 2.1.1 PLANT CATEGORIZATION
: 1. Description of the Issue The proposed NRC Severe Accident Policy Statement states that the existing plants pose no undue risk to public health, safety and property given what we know about them and provided that certain activities go on. In support of this statement, ongoing nuclear safety programs are being conducted to provide additional assurance that this is so. One method for providing additional assurance is to use the insights from existing PRAs and infer them to all LWRs.
This extrapolation of PRA insights to all LWRs necessitates that all LWRs be categorized into generic classes in order to enhance the manageability of the suppordng research. The NRC Accident Sequence Evaluation Program (ASEP) is extending the reference plant concept that is presently used by the NRC by incorporating :'ere plant characteristics and using the probabilistic approach to plant categorization.
: 2.      Implications of the Issue to Regulatory Questions The ability to develop and justify the grouping of plants has a great effect on the NRC regulatory decision process. Since it is desired to understand nuclear safety and to confim the policy statement on severe accidents on as generic a basis as possible for all operating and near tem operating plants, the issue of developing plant classes is highly important. There is a direct relation-ship between the issue and the regulatory questions. The position of the issue can affect the answers to:
                  -      How safe are the existing plants with respect to severe accidents?
                  -      How can the level of protection for severe accidents be increased?
                  -      What additional research or infomation is needed?
                  -      Is additional protection for severe accidents needed or desirable?
: 3. Subissues
                  -      Is it possible to group plants having similar dominant accident sequence risk characteristics?
                  -      Do the existing PRAs (12) provide enough infomation to make inferences on all existing plants?
                  -      Can insights drawn from plant categorization be used in severe accident rulemaking? If so, how should uncertainties be considered?
                  -    Given the associated uncertainties, can plant categorization be used to identify weaknesses in plant design?
                  -    What level of information is appropriate for plant categorization--function, systems, or component level?
                  -    What characteristics of plants should be used in the categorization?
                  -      Can event trees for plant classes be fomulated? If so, how useful are they?
Fot A 14 ~%i A/2.
 
    ,. o 2
i 4
          -      Plants are categorized based on dominant accident sequences and plant systems needed to mitigate the accidents, consequences resulted from these incidents are not considered. Does this have any impact on regulatory 4
decisions?
          -      What are the other potential uses of plant categories?
: 4. Status of Understanding The NRC and IDCOR currently are using a similar deterministic approach to plant categorization. A reference plant is chosen, usually a plant with a PRA, to represent a group of plants that have similar containment design. The reference plants selected by the NRC are: Peach Bottom (BWR Mark I); Limerick (BWR Mark II); Grand Gulf (BWR Mark III); Zion (PWR Large Dry); Sequoyah (PWR Ice Condenser); and Surry (PWR Subatmospheric). These reference plants cover i          the containment types of all the operating and near tem operating plants. The
  >        IDCOR reference plants are a partial set of the NRC reference plants and they are Zion, Sequoyah, Peach Bottom and Grand Gulf.
Since the reference plant concept depends solely on the engineering assumption that plants can be grouped by containment type, the ASEP work extends the reference plant concept, uses a more systematic approach to plant grouping, and accounts for more plant characteristics besides containment type. This produces
-          a hierarchy of plant groups that are fomulated based on similarity of plant function / system response to the accident initiators, risk characteristics of l
the dominant accident sequences, and systems design differences. At each level,
!          the plant groups can be further categorized by their containment types. The l          attached figure represents a possible plant class hierarchy.
ASEP fomulates the plant groups from the " top down" and " bottom up" perspectives. The " top down" approach groups plants using the event tree methodology and the " bottom up" approach groups plants based on similar system I
design and similar system reliability characteristics. The first step in plant categorization is to divide the LWRs into PWR and BWR groups. From the " top down" perspective, the PWR and BWR groups are subdivided according to findings generated in the development of functional event trees. Functional event trees are developed for the PWR and BWR groups in response to the general initiating events of LOCA and transient. Plant groups are formed at this level based on the difference of the preventive or mitigative functional responses or the sequence end-states (success or core damage). ASEP does not expect this level of plant categorization to result in many plant groups since over the past 15 years or so the philosophy of both PWR and BWR accident prevention and mitigation has not fundamentally changed. That is, the needed functions are generally recognized as being the same. To develop these functional event trees and descriptions of the functions and sequences, ASEP analyzes and synthesizes the functional event trees presented in existing PRAs. At least four functional event trees are developed, two for each of the PWR and BWR groups. One of each group is for the general initiating event of LOCA, and the other, for the general initiating event of transient. Plant grouping at this level is not definite until the grouping at the lower level (e.g., systemic event tree approach) is completed since the total accident sequence delineation process is iterative.
foiA ~14-933 A/2.
 
3 ASEP PLANT CLASS HIERARCHY
* m 1    LWR                                                                                                  Light Water Reactors (LWRs)
I                                                              I BWR                                                          PWR                                                                            Two types of LWRs I
I BWR                                                      BWR                                                                                              Plant Classes based on similar Lt.1                                                         Ll .2                                                                                        functional response to initiating events 1
* BilR                                            BWR -                                                                                    Plant Classes based on similar L2.5                                            L2.6                                                                                      system resoonse to specific i
initiating events
                                                                                                              .I BWR                                                            BWR-                                                                                               Plant Classes based on similar L3.3                                                          L3.4                                                                                                risk characteristics of dominant accident sequences i
BUR                                          .
BWR                                                                                        Plant Classes based on diversity l                                                L4.4                                            L4.5                                                                                        of designs of key defense-in-g depth systems necessary to mitigate
                                                      ,                                                                                                                                          dominant accident sequences l                                                                                                                                                        ~ ~ ~
            ~"                                                                              ~                                                                                "                    ~                                                                    ~
BilR 3              BWR h                                B'W'R )!        BWR l.
BWR'1                                                'Soecific plants
:          I                                        '
Fot A- 34 -928 l
A/1.
l                            *At each level, plant groups can be further categorized by containment types.
i                      .      . .                      .              ... _.                . .        .            .....                                                                                                                                              .
 
      .                                                                                                                                    l 4
,                  The next level in the " top down" approach is the grouping of plants based on their system response to the initiating events. As mentioned earlier, the i
philosophy of accident prevention and mitigation is generally recognized as being the same for both the PWRs and BWRs, but the strategy to achieve this philosophy is different among plants. By strategy, we mean the systems to perfonn the functions. Classes of plants are developed based on the system combinations required to either prevent or mitigate a core melt ensuing a given specific initiating event (or set of similar initiating events).
l                  Further refinement in plant grouping is considered since system dependencies vary among plants. For example, containment spray recirculation at PWRs can be performed by the spray injection pumps at some plants, low pressure injection pumps at others, and is a completely separate system at some others. Another source of refinement is the possible different end states for a given accident i
sequence; that is, the success at one plant may be core damage at another.
Plant categorization using the " top down" approach is not completed until the merging of plant classes from the " bottom up" approach since the subsystem
,                  characteristics of the lower plant class levels may affect the higher level
!                  plant groupings (e.g., the structure of the systemic trees can differ depending l                  on whether or not main feedwater pumps are motor or turbine driven; in the
:                  latter, closure of the MSIVs or loss of the power conversion system causes 4
failure of main feedwater).
The " bottom up" approach uses twelve existing PRAs as the starting point to
,                  develop plant classes. Dominant accident sequences are identified as well as the plant systems and support systems needed to mitigate these dominant acci-dent sequences. A plant survey is conducted to obtain accident mitigation fluid and electrical system for all the systems and support systems for as many plants as possible. After the system drawings were collected, they are simpli-fied based on past PRA insights in tenns of major flow / energy paths, major
;                  active components, important passive components, etc. After the system had
,                              lifi l
been systemssimp (e.g.ed,  generic
                                    , compare      system all AFW  systems)    configurations consideringare  fonnulated significant          by comparing differences    in            all the redundancy, diversity, or support system dependencies for each system of interest. For example, all AFW systems are grouped into 20 generic AFW confi-I                  gurations.
l                  After the generic system configurations were formulated, plant groups are for-l                  mulated in accordance with the diversity in designs of the key defense-in-depth
!                  systems necessary to mitigate the dominant accident sequences. The PWRs having i
certain design features in connon were organized into 29 plant groups and BWR i
into 15 plant groups.
l                  Some insights can be drawn from this level of plant categorization. First, for j                  a particular system, the number of configurations for the surveyed plants can i                  be large. For example, 20 different AFWS configurations are identified for the
!                  surveyed PWRs and 7 RHR systems for BWRs. Service water systems are found to I
be essentially all plant-specific. Without accounting for service water system Fora-94 48 A/2.
 
5 variations, the differences in systems still resulted in 29 PWR pant groups (from 72 PWRs) and 15 BWR plant groups (from 31 BWRs). This is a dramatic evidence of the lack of standardized designs in the U.S.
At the next level, the plant groups based on system design differences are            l coalesced based on their risk characteristics computed through accident sequence quantification, sensitivity, uncertainty, and recovery analyses.
These characteristics include the best estimate, their upper and lower bounds, and the dominant factors that drive the sequence likelihood.
The plant groups derived from the " top down" approach, which are developed deductively by safety philosophy and strategy, are then merged with plant groups derived from the " bottom up" approach, which are developed inductively by safety tactics (subsystem traits) and dominant risk characteristics (sequence
,                likelihood range and dominant factors). Ideally, there should be a smooth meshing of the top down and bottom up approaches. Practically, this may not initially occur since either approach may overestimate or underestimate the importance of the multitude of possible similarities and differences. It may be beneficial to have a plant in one plant group for one type of accident and
,                in another for another type of accident. A coherent plant categorization hierarchy is to be fomulated after interactive refinement among all levels of plant groups. This plant categorization hierarchy can be expanded by SARRP to incorporate containment types for each plant group at various levels.        .
: 5.      NRC Position Since the research on plant categorization is still in progress, the NRC cannot provide a position on this issue at this time. Therefore, the NRC is relying on the reference plant concept (deteministic approach) in its policy making until the research results are completed in late 1984. The reference plant concept is consistent with the IDCOR approach to plant categorization.
: 6.      Continuing Confirmatory Work Research on plant categorization will continue to support the NRC position on
,                plant categorization, either through further categorization work in ASEP or through more plant-specific PRAs.
l l
l l                                                                                            FolA- 14-%
A/L
 
3 NUREO/CR-3025 SAND 82-2477 R7 Printed August 1984
                                                                          -r      ko\@      @/i -9 Q.2 k8 High-Pressure Melt Streaming (HIPS) Program Plan W. Tarbell, J. Brockmann, M. Pilch o8  a  ermore, Cahforrua 94550 under Contract DE-AC04-760P00789 99 h h253        -!
Prepared for U. S. NUCLEAR REGULATORY COMMISSION                                                Fo/A-S 6 778 AlI
 
NOTICE This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government not any agency thereof, or any of their em.
          , makes any warranty, ex ressed or implied, or assumes any egalliability or responsibili for any third party's use,or the to of such use, of any in rmation, apparatus product or process disclosed in this report, or represents that its use by such third party would not infringe privately owned rights.
Available from GPO Sales program Division of Technical Information and Document Control U.S. Nuclear Regulatory Commission Washington, D.C. 20555 and National Technical Information Service Springfield, Virginia 22161
 
SAND 82-2477 NUREG/CR-3925 R7
. HIGH-PRESSURE MELT STREAMING (HIPS) PROGRAM PLAN W. Tarbell J. Brockmann M. Pilch Sandia National Laboratories Albuquerque, NM    87185 1984 fp / A - TY-ilI AlI
 
b 7
i ABSTRACT l                The Zion Probabilistic Saf ety Study (EPSS) envisions acci-
:          dent sequences that could lead to f ailure of the reactor vessel while the primary system is pressurized.        The resulting ejection of molten core material into the reactor cavity followed by the blowdown of steam and hydrogen is shown to cause the debris to enter into the containment region.
The High Pressure Melt Streaming (HIPS) program has been developed to provide an experimental and analytical investigation of the scenario described above.      One-tenth linear scale models
    . of the Zion cavity region will be used to investigate the debris
..        dispersal phenomena. Smaller-scale experiments (SPIT-tests) are
    ,    also used to study high-velocity jets, jet-water interactions, and 1/20th scale cavity geometries.      Both matrices are developed using a f actorial design approach.
The document describes certain aspects of the EPSS ex-vessel phenomena, the experimental matrices, test equipment, and instru-mentation, and the program's analytical ef forts.              Preliminary data from SPIT testing are included.
i I
l l
l l
l                                                                            Fo/A 128 l
A/I
 
tn TABLE OF CONTENTS PAGE I. Introduction....................................                                            1 II. Review of ZPSS Ex-Vessel Analyses...............                                            8 III. Alternative Ex-Vessel Accident Phenomena. . . . . . . .                                    22 IV. Scaling Analyses................................                                            33 V. E x pe r ime nt al Pr o g r am s . . . . . . . . . . . . . . . . . . . . . . . . . . .      60 VI. System Pressure Ej ection Tests (SPIT) . . . . . . . . . . .                                66 VII. H I PS E x pe r ime n t al Pr o g r a m . . . . . . . . . . . . . . . . . . . . . . . 138 VIII. Summary.........................................                                          154 APPENDIX A - Phase I SPIT Test Program. . . . . . . . . . . . . . . . .                            156 APPENDIX B - Gas Blowdown of SPIT Appa ratus. . . . . . . . . . . .                                169 APPEN DI X C - Je t S t r eam Hea t Lo s s . . . . . . . . . . . . . . . . . . . . . .            176 APPENDIX D - Frequency Response of Embedded Thermocouples 185 APPENDIX E - Error In Melt Velocity Measurement........                                            187 REFERENCES.............................................                                            190 4
A/I
 
9 LIST OF FIGURES figuIn                                                          EAS.g
: 1. Schematic View of Zion Reactor Containment Building                                                      9
: 2. Plan View and Typical Dimensions of Zion Reactor Cavity                                                      le
: 3. Details of the Zion Reactor Cavity Region                    11
: 4. Instrumentation Nozzel - Vessel Weld                        14
: 5. Material Removal Mechanisms                                  18
: 6. Mass Discharged for Multiple Tube Failures                  29
: 7. Penetration Aerosol Relationship to the source Term                                                        55
: 8. Particle Loss By Impaction                                  59
: 9. SPIT Melt Generator                                          67
: 19. Schematic  Showing Relationship of                    Melt Generator  and Interaction Chamber                          70
: 11. Pressure System Schematic                                    71
: 12. Diagram of a Submerged Jet                                  85
: 13. Slug-Type Heat Flux Calorimeter                              91
(      14. Thermal Conductivity of ATJ Graphite                        94
: 15. Bubble and Cavity Pressure Histories for a l
Water-Filled Cavity                                        111          ,
: 16. SPIT 1/20th Scale Cavity                                    119
: 17. SPIT Apparatus Placed in the Interaction Chamber                                                    123
(      18. Embedded Thermocouple Details                              126
: 19. Melt Velocity Sensor                                        129
: 26. Electrical Circuit for Melt Velocity Sensor                131 i
i v
 
Figure                  LIST OF FIGURES CONT.
EASR
: 21. Multiple-Shield, Assembly High Velocity Thermocouple 134
: 22. Schematic of HIPS Melt Generator                      139
: 23. HIPS Melt Generator Assembly 141
: 24. HIPS Test Article 142
: 25. Assembled HIPS Test Apparatus 145
: 26. HIPS Test Apparatus Placed In the Experiment Interaction  Chamber 146
: 27. Logic Decision Network for the HIPS Test Strategy      ISO
: 28. Melt Ejection Seg'1ence 159
: 29. Flash X-ray Photographs of Jet Stream 160
: 30. Size Distribution of Aarosol Produced During Pressurized Melt Ejec+ ion 164
: 31. Electron Micrographs of SPIT Aerosol Samples          166
: 32. Vapor Pressure of Iron and Aluminum Oxide as Functions of Temperature                              167
: 33. Elemential Analysis of 65-Micrometer Aerosol Particles 168
: 34. Calculated SPIT Pressure Vessel Blowdown History for Variation in Orifice Size and Discharge Coefficient  172
: 35. Effect of System Volume on Blowdown History for Variation in the Interconnecting Piping              174
: 36. Calculated Jet Temperature as a Function of Initial Temperature 181
: 37. Calculated Melt Jet Temperature as a Function of Emissivity and Cone Angle 182
: 38. Calculated    Melt Jet    Temperature as a of  Jet  Velocity                          Function 183 vi J
 
LIST OF TABLES Table                                                    EAgm
: 1. Dissolved Hydrogen in Molten Iron                    24
: 2. Solubility of H 2 and H 2 O in an Oxidic Melt        25
: 3. Summary of Scaling Analyses                          33
: 4. Parameter Values from the ZPSS and Experiment        35
: 5. Catoff Criteria for Debris Removal Mechanisms        37
: 6. Thermal Processes in the Decomposition of Concrete                                              52
: 7. SPIT Phase II Jet and Aerosol Characterization Test Matrix                                          61
: 8. SPIT Jet / Water Interaction and 1/20th Scale Cavity  Test Matrix        ,                        62
: 9. Critical Debris Dispersal Characteristics            63
: 10. HIPS Test Matrix                                      64
: 11. Informational Sought from the SPIT Jet and Aerosol Characterization Test Matrix              74
: 12. Instrumentation for SPIT Characterization Tests      76
: 13. Aerosol Measurements and Instrumentation              97
: 14. Importance Ranking of Fission Products (Total Dose)  99
: 15. Fission Product Simulants                            99
: 16. Aerosol Analysis Techniques                          101
: 17. Independent and Dependent Test Variables            102
: 18. Probability Points of the t - Distribution          105
: 19. Material Properties for Impedance Calculations      113
: 20. Chemical Compositions of Concrete                    120 I    21. SPIT Cavity Test Informational Requirements          121
: 22. Instrumentation for the SPIT Cavity Tests            124 vii
 
Table                  LIST OF TABLES CONT.
rage
: 23. Correlation of Accident and Test Phenomena and the Effect on Test Outcome 4                                                              147 a
: 24. Details of SPIT Phase I Tests t!    25.                                                      157 ei jf Parametric Values for Jet Temperature Calculations  179 1:
C e
J P
L n
b
)
3 5
k t
2 viii r
 
I. Introduction Until the late 1978's the WASH-1499 Reactor Safety Study          ,
(RSS) (Ref. 1) represented the most thorough analysis of reactor        !
safety ever conducted. The report demonstrated the use of event tree analysis to show the relationship between the sequence of events and the consequence of each. A number of critiques have judged the basic methodology to be sound for application to other plants (Ref s. 2 and 3) .
Since the publication of the RSS, the Nuclear Regulatory Commission has explored ways of applying probabilistic risk as-sessments (PRA) to specific nuclear power plants. The key prod-uct of these analyses is a quantification of the risk to the public in operating the plant. The intent is to provide a clear indication of the events and equipment contributing to the risk and thus provide a means for assessing actions designed to reduce the risk.
A PRA involves the identification of the events and sequence i    of events conceptionally possible during an accident. Each of
:    these events is quantified in terms of f requency of occurrence during plant operation.      Contained within the quantification process is the determination of the uncertainty associated with 1    the assigned values.      Sophisticated mathematical techniques are then employed to develop the risk associated with each accident.
  ,        This report is organized into eight sections: Section I provides a general review of the Zion Probabilistic Safety Study (Z PSS) (Ref 4) and the objectives of the HIPS Program, Section II reviews the major ex-vessel events and analyses as given in the EPSS, Section III contains a discussion of other phenomena that may be important to the ex-vessel events, and Section IV is the scaling analysis used to design the HIPS experiments. Section V presents the test matrices for the HIPS experiments and the smaller scale SPIT tests.      Sections VI and VII give detailed descriptions and analysis of the SPIT and HIPS equipment and test techniq ues. Section VIII concludes the report.
Very little is presently known about the phenomena involved in high-pressure melt ejection accident scenarios. It is antici-pated that the combined experimental and analytical work under-taken in the HIPS program will likely cause the direction and
    ; scope of the effort to be altered. This document attempts to l consider all aspects of the problem to avoid large scale changes.
1
 
I.1    Eion Probalistic Safety Study The ZPSS was performed primarily in response to the NRC's perception that sites near high-population centers create unusual and excessive risk. This perception resulted principally f rom the transfer of the base PWR case of the RSS analysis to a high-population location.      The stated obj ectives of the ZPSS are:
quantify the risk of the Zion plant as constituted and operated, identify the key contributors to risk, and evaluate the potential risk reduction if alternate features are considered.
The study considers three primary areas:      the plant, its containment, and the site.      The ZPSS develops a set of scenarios f or each of these areas and matches the resulting sequence to achieve outputs.      The outputs are then used to develop a quanti-tative assessment of risk. The goal is to assemble the analyses into a set of risk curves representing each of the identified health effects (early f atalities, thyroid cancers, man-rems, e t c.) . Each curve then gives the f requency of exceeding a pre-scribed level of damage.
The ZPSS is the result of an extensive analysis development and evaluation effort.      The authors of the document believe that it advances the state of the art in the probabilistic risk assessment in several areas beyond the basis provided by the RSS.
In particular, the development and quantification of the contain-ment event tree recognizes dif f erences in the response of the system to dif f erent core melt scenarios. The study also shows that the uncertainty in the fission product source term is the major contributor to the uncertainty in the risk analysis.
The RSS identified a number of initiating events that would cause loss of coolant and result in heating and possible melting of the core. The assumption was made that the degradation of the core structure proceeds uniformly to give a molten pool in the lower plenum of the reactor pressure vessel.      The highest proba-bility event was assumed to be loss of system pressure (large break LOCA) and eventual thermal weakening of the lower head, ultimately causing a " hinge-type" f ailure. With the failure of the vessel, the molten core material pours into the cavity under the acceleration of gravity.      The study did identify other fail-ures that could occur while the system was at elevated pressure, but the results of these sequences did not significantly deviate f rom the large-break LOCA behavior. A key feature of the RSS is that containment f ailure can be caused by steam explosion or missile generation. If these mechanisms do not cause f ailure, then the long-term core / concrete interaction or the hydrogen gas burning will eventually overpressurize      the containment or pene-tration of the basemat structure will occur.        The analysis sug-gests a very low probability that the accident will terminate short of containment f ailure.
2
 
The ZPSS recognized several differences f rom the RSS in the key phenomena associated with the events leading to vessel f ail-ure. Detailed analysis of the core melt progression showed that coherent downward movement of large quantities of molten core material is unlikely. This conclusion results in a probability of 0.9 that less than 25% of the melted core will be in the lower plenum early in the accident. The analyses also recognize that core degradation (incoherent melting) can occur for transient and small-break LOCA sequences where the primary system will be at elevated pressure.
Failure of the vessel by steam explosion or by overpressuri-zation is assumed to be physically unrealizable with a very low frequency of occurrence. The ZPSS conservatively estimates a probability of 0.1, that the accident will be terminated by in-vessel cooling of the core debris. Without permanent cooling, the debris will attack and f ail the weld of one or more instru-mentation tube penetrations in the lower head of the vessel. The loss of weld integrity causes the tube to slip out of the lower head and the molten debris to be ejected under pressure into the reactor cavity.
If the_ accident proceeds to the point where molten core material is ' ejected under pressure into the cavity, the ZPSS predicts the material will be " dispersed" into the containment building. The analysis considers that the wide distribution of the material, combined with the availability of water, assures that the configuration is coolable. By minimizing concrete at-tack and consequently gas release, the probability for the loss of containment integrity is considered low.
The ZPSS containment event tree is comprised of 19 nodes or branch points chosen in an iterative process during which the principal phenomena affecting containment integrity are address-e d. Each node represents a decision characterized by a "yes-no" input. Probabilities assigned to each of the branches are in-cluded in the determination of the ultimate risk associated with the paths emanating f rom the branch. The probability of each node  is conditional plant state.
on preceding events, includingtheinpg The resulting event tree is comprised of 2 (524,288) output nodes. During the analysis, many of the branch-es were removed (pa red) as being " technically illogical" f or the path f ollowed. After the " paring" process, the containment event tree is reduced to a total of 1950 output nodes.
In order to disperse the core debris, the ZPSS proposes a number of hydrodynamic mechanisms induced by the blowdown of the primary system. The principal driving force is the energy pro-l  vided by the high-pressure jet of steam and hydrogen f rom the l  primary system f ollowing the ejection of the core material.
!  Establishing the existence of the dispersal mechanism is essen-i  tial in verifying the calculated risk factors.
l i
;                                    3
 
I.2      HIPS Program Objectives The HIPS program is an experimental and analytical investi-gation of the phenomena associa.ted with the behavior of molten core debris streaming into reactor cavity configurations. The experiments will involve high-temperature melts created at real-istic      pressure levels and injected into scaled reactor cavity geometries.        We objectives of the HIPS program are as follows:
: 1. Confirm the existence of the debris          dispersal mecha-nism.
: 2. Determine and  assess      other melt jet phenomena such as jet geometry, gas solubility, and aerosol generation.
: 3. Assess events discounted in the ZPSS,                such as melt / water interaction, energetic concrete decomposi-
        ,                    tion, and the combined effect of one or more events.
L                    4.      Reduce uncertainty in probability estimates of the fis-sion product source term and dispersal mechanism.
The ZPSS analysis is hampered by the lack of experimental information concerning the ejection and behavior of high-tempera-ture materials in confined geometries. We first objective of the HIPS        program is to experimentally confirm the debris dispersal mechanisms given in the ZPSS.              Despite the lack of confirmatory inf  ormation,    the authors  of
            ;of 0.9 with a range of 0.8 to 9.99 the ZPSS    have assigned a probability that material dispersal will occur (positive branch of the event tree at Node I).                          All sequences that proceed from the negative branch at Node I are
      ' . assigned the "placekeeper" probability of 0.9991, causing their
          - calculated f requencies to be small and thus consideration of subsequent events is neglected.
                /  If the probability of dispersal is shown experimentally to be lower than (9.9), then the values assigned to the negative branches must be correspondingly increased. Higher f requencies for the negative, non-dispersed branch will then require consid-eration of the probability of a non-coolable debris bed configu-ratio'n in the cavity and ultimately, overpressurization of con-tainment f rom concrete decomposition. The HIPS tests are specif-ically directed to confirm the existence of material dispersal mechanisms,        and to establish      probability estimates for complete dispersal.
    %              If the experimental results show that the material is com-1pletely dispersed, then the uncertainty associated with this node (can be verified or reduced. Conversely, if no material exits the cavity region, the results can then be used directly to reassess the probabilities associated with the node branches. Partial dispersal cannot be treated in the same manner.
4
 
2 7
t          Because the ZPSS event tree is constructed of binary (yes-no
, i    only) decision points (nodes) a par tial or incomplete dispersal j    of material is not directly considered in the node probability.
If the experiments show that only partial dispersal occurs, the
,      results of the HIPS tests can be used to suggest a probable outcome by applying the ZPSS " gravity-drop" analyses to the portion of material not removed.
i
    !          The second objective of the HIPS program will be to identify
  ;    damage modes other than the 19 possibilities in the Z PSS con-I    tainment matrix. Failure to consider a damage mode may mean that paths at subsequent nodes in the analysis are not representative
,  f    of the actual events.      Partial dispersal of the debris is an
  }
example of this possibility. The exact yes-no decision logic used in the ZPSS does not account for situations where portions of the debris are not removed f rom the cavity.
Melt crust formation is not considered within the context of I
the 19 damage modes presented in the ZPSS. The analyses assumes
  !    that coolability is guaranteed during a Large-Break LOCA se-l    quence. Formation of a stable crust over the dispersed debris that may prevent cooling, despite the presence of water.        Li-i    mited experimental evidence exists (Ref. 5), to suggest that the I    formation of a stable crust of decomposed concrete could prevent coolant from reaching the lower portions of the debris bed.
; i            The ef f ect of melt aerosol is a third e,xample of a possible event not considered in the ZPSS. Experiments reported here have
    ,    shown that high-pressure melt ejection sequences are accompanied by large aerosol generation. In an accident, these aerosols will contain radioactive fission products that will enter into the containment atmosphere.        The aerosols will also pose a threat to containment saf ety features because of their potential-ly high heat content and may diminish the effectiveness of heat j        transfer processes within containment.      Sufficiently large con-centrations of aerosols could ultimately eliminate containment heat removal capability.
The third objective of the HIPS program is to determine if ex-vessel phenomena neglected in the ZPSS are of consequence to
;        the calculated risk. Specifically, the ZPSS ascertains that a l
water-filled cavity has no effect on the dispersal of debris i  during transient and small-break LOCA scenarios. The analysis
      ;  assume that the water is " pushed" out of the cavity by a steam bubble formed by quenching the melt jet.        A large number of assumptions are made in this analysis including: heat transfer
:    )  between the melt and water is instantaneous, inteference of
      !  expansion waves does not occur,. and steam generation is isother-i  mal. Experimental evidence of these assumptions is not provided l
in the ZPSS.
j              The ZPSS analysis does not consider the ef f ect of concrete decomposition on the      debris removal mechanism.      Numerous s
i
 
i.
l b
e i      experimental investigations (Refs.        6,  7                                  i material    contacting                            have    shown that  molte
:r l      ration and    concrete concrete spallation.is accompan)ied by vigorous gas genen        i As a consequence, the dynamic le subsequent dispersal may not occur. the ZPSS may not form and debris  configuration    envisioned    in                                        ,
A o
.r              Within the context of identified, but is the combined effect of two or more phenomena. neglected, phenomena the core material under pressure f rom the vessel will likelThe ejection of involveexplosion, steam    a simultaneous vessel blowdown (of steam / water / hydrogen)y concrete decomposition, core debris quenching,          ,
and oxidation of Zirconium and iron.
these phenomena separately except the possibility of oxidation inThe ZPS the cavity environment (100% Zr oxidation in-vessel is assumed)              .
The ef fects of chemical oxidation are threefold:
oxidation of Zr and steel is an additional hydrogen sourc(1)e, the          (2) the heats of reaction cause additional pressurization, and (3) the particles may, act as a distributed hydrogen ignition source These effects will combine to cause more hydrogen to be available                .
than the ZPSS assumes and the likelihood of burning to be great-er.
tainties in the ZPSS analyses.The fourth objective of the HIPS program i in the ZPSS analysis is identified as the radionuclide sourceThe larges term for the site matrix calculations.
assume that the major portion of the fissionBoth              the RSS and Z PSS nreduct    source term is  in the explosions. form  of fine particles generated      during    ex-vescel steam tribute study very little to the radioactive source term. Concrete        Neither decompo identifies the possibility of aerosols formed during melt deposi tion.
fission product    The  HIPS program will address this area by including mocks in the melt composition, particulate material in the cavity region. aerosol f ormation mechan I.3      Scope of HIPS Program The scope of the HIPS program considers the ex-vessel events occurring scenarios. during the small-break LOCA and transient accident The large-break LOCA or " gravity-drop" sequence does not    involve in considered        gas-driven    debris dispersal and is therefore not the program.      The in-vessel processes of fuel melting outlined in  andthevessel ZPSS.attack    are assumed to occur in the manner used as the starting          The  single-tube  f ailure assumption will be sequence.                      point in initiatiing the experimental 6
 
Phenomena outside the cavity region are highly dependent on the details of the containment structure. The behavior of the debris, gas, and aerosols escaping the cavity will be monitored to assess their contribution to the containment loading. Decay heating and long-term coolability of the debris will not be included in the initial HIPS test matrix.
I f
i i
L-4 i
r i
t r
l h
k
        \
i
        \
        \
7
 
II.
Review of EPSS Ex-Vessel Analyses A PRA such as the ZPSS involves an extensive analysis of the many reactorprocesses accident.and events that may occur during a postulated involved and lengthy.        The resulting document is theref ore highly This section presents a review of the parts of the ZPSS document pertinent to the sequence of events from vessel provide            failure to final debris disposition. The intent is to a convenient        basis for subsequent discussion. Following a brief description of the plant, the ordering is the same as thatuen Seq    used ces. " in the ZPSS beginning with "Section 3.2, Accident II.1 Description of the Zion Reactor Plant                                          .
miles north of Chicago, Illinois, near Lake Michigan.The            Commercial two-service 1974      dates (Ref. 10) .for The the units are in December 1973 and September nuclear-steam-supply system for each of the Zion  units consists of a four-loop Westinghouse pressurized water reactor.
produces 1985 megawatts of electrical output.Each reactor is rated (FigureEach 1)  reactor is housed in an individual containment building 1/4-inch-thick  consisting (6-mm) of      a post-tensioned concrete shell over steel liner.
The internal volume of the a design pressure of 0.5 MPa (62 psia). containment structures is ap strength (149 psia). is estimated in Appendix 4.4.1 of The        lower bound the ZPSS          ultimate to be 1.0  MPa The concrete consists of Portland-t porating limestone coarse aggregate and commonype ga te (Ref. 6) .                                              sandcement    incor-fine aggre-The reactor pressure vessel (RPV) of each unit is placed in a cavity located tation-tube        placement.in      the floor of the containment building.      The      ca the inclined " keyway" leading to the floor of the containmentThe ac structure.
the tapered cross-section of the access tunnel.Much                  Figure    2 gives of the  ing two view showin containment improve clarity        equipment    has been of the major          omitted f rom Figures 1 and 2 to features.
l 1
are          The ex-vessel postulated              interactions to occur              described in the cavity, tunnel,later in thisand keyway,      section con-tainment            building.        Figure 3 is a more detailed view of these areas showing the instrumentation tubes and sump pit. The 56 instrumentation tubes emanating f rom the RPV are arranged in a rectangular array that terminates in the cable seal room in the containment building.
stainless-steel biological shield (6-mm thick)Not obvious in the figures is a conti ty        and  tunnel    regions    at  a depth of  0.5 t o  linerminbelow 0.75      the cavi-the concrete surface.
8
 
                                                            =                42.7 M                        :-
am\
l        n N
gy i
01                                                      %
1 h
POLAR                  j                                                      j CRANE                  t                                                      ;
1                                              \s  ;                                                      e.
S37M (N                                                ,
j STEAM                                                            A                                  [ 49.1 M
                                                        ),
x GENERATORS Q-'                                      _c
[
j    (        ~
w 3
i
                                                                                                                ?
REACTOR                      :$    . Q.-
{) ctb (".[m)).              .
_ ll VESSEL                        by                _                  _    ,
L
                                                                                                                                          )
:$    '.EPN /^\-                      -
IN-CORE i-}        .
x    h          ((]
INSTRUMENT                                s 4'~                  l[          }    . 4 gj.l u    v TUNNEL                                    [D;9Mt                .$ p gyget.6*:(;ndvidid]
SUMP PIT-                            L              ;gj REACTOR CAVITY 2.7 M THICK BASEMAT e
d i                          Figure 1. Schematic View ZION Reactor Containment Building                                          ,
i i                                                                                                                              t
        !                                                                                                                              i I
l j                                                                          9
 
i
          ->      2.64
* TUNNEL
                                  > ~2 5*
CAVITY  \s          i
                                                                                  ,/
r
                                                                  'N.          /
Jk 7.18 JL i
U 3.13                                                                  4.50 v      ir                                                      \
y
                                <                    9.39 I
KEYWAY                TUNNEL                        CAVITY I
Y                                -
2.29 -                  --
                                            /.                [
                                                                      -+                -
R x
i        2.60 R Figure 2.
Plan View and Typical Dimensions of Zion Reactor Cavity (Dimensions in Meters) 10
 
:                  4 6 '-6 "                  ;
4                                                    9 SEAL                THlMBLE GUIDE TABLE g--f REACTOR t
              -    :r-
[    A-                                a.
          )      l l
j LADDER 1,
i j                          \
i*
                          "        \
ns      I. e              o I    '<
:g..          ,
h aam;o;
(
:.                                                      >                                                    /
y i
n                .
9 CAVITY 7
      ;                                      \                                                              /              <
x l      i                                                          INSTRUMENTATION
      ?                                                                TUBES i
f l      k
(
Figure 3. Details of ZION Reactor Cavity Region l
I l                                                      11
 
a II.2      Accident Sequences that    The ZPSS considers a total of 68 accident-initiating events, the event  are assigned    logic diagram.
to one  of 13 initiating event catagories in The sequence of events leading to degraded core scenarios are varied and depend on the initiating event and the operation or f ailure of plant saf ety systems. A common cooling      system,characteristic                  is the loss ofofcoolant causing overheating            the core.f rom the primary i'
gories,  In general, accidents are classified in one of three cate-time of the  depending      accident.          upon the primary system pressure during the sequences are:                                      With one exception (interf ace LOCA), the Large-Break LOCA (pressure < 1.4 MPa), small-break LOCA (pressure < 7 MPa), or transient events (up to the
'        set-point of the pressurizer relief valves - 17 MPa) The ZPSS analysis recognizes differences in the behavior for the various classes of accidents, such as the t! ming of events, the amount of material involved, and the resulting interactions.
For the large-break LOCA sequence, a failure in the primary system is assumed to cause rapid depressurization allowing the primary system to discharge to an equilibrium condition with the containment atmosphere. Hydrogen generated by the oxidation of
!      the this For      coresequence, cladding is also released into the containment building.
the ZPSS considers the primary system and con-i tainment psi) .
building                      pressure  following blowdown to be 0.3 MPa (44.1 i                  The small-break LOCA sequence analysis assumes that the core is uncovered by loss of coolant through a smaller break. For this case, core degradation (loss of coherent structure) can occur while pressure exists within the primary system. The gas released through the break raises the containment atmospheric pressure slightly.
The range of pressures in the primary system proposed in the ZPSS for this accident is 1.4 to 7.0 MPa (200 to 1929 psia) with the value of 7.0 MPa used for the calculations.
The transient event occurs when safety systems fail to respond to an accident-initiating event. In this situation, the system pressure and temperature continue to rise until the relief valves open.
the core may become                    If sufficient    coolant is lost through the valves, uncovered. The transient sequence results in i
exposure conditions. of the core at pressures close to system operating Pressures of 7.0. to 17.0 MPa (1929 to 2500 psia) are considered for this sequence. The upper limit corresponds to the i
set point of the pressurizer saf ety valves, and is used f or the ZPSS analysis.
The three accident sequences result in two event chains in the transient analysis section of the ZPSS. A " Gravity Drop Model" is used for accidents where the system pressure is less than 1.4 MPa (large-break LOCA).                                          In the analysis, the release 12 I_      _. __      _ . _ _ _ _ . . _ _ _ . _ - .
 
l rate of core material into containment is related to the static head of molten material in the vessel. This treatment parallels that associated with the lower head f ailure given in the WASH-1400 analysis. The second event chain, the " Time-Phased Disper-                                ;
r      sive Model" is used where the primary system pressure is above 1.4 M Pa. The sequence of events and their relative time scales
  ,      characterize the small-break LOCA and transient events.                                          In j      general, the model considers the high-velocity gas / melt stream                                    ,
exiting the RPV breach, combined with the steam / hydrogen blow-down, has a significant affect on the final debris configuration and distribution.
II.3  Vessel Failure The onset of vessel failure represents the culmination of the in-vessel events and the entry state leading to the ex-vessel
(
interactions. The ZPSS finds that portions of the core will melt
: ;      causing material to f all or flow into the reactor vessel's lower i :      plenum. If the supply of coolant to this region is inadequate,
  -j      the large thermal energy of the core mass will cause attack of 1    the vessel's steel wall.        Internal structures such as the lower j    core support and the in-core instrumentation supports may also be g    attacked.
f            The ZPSS theorizes that molten core material entering into j      the lower plenum region immediately begins to attack the vessel
[    wall. Any existing water will be vaporized or displaced by the 4
y    core debris and additional water will vaporize of f the upper
[    debris surface.      In order to consider the range of possible j  system characteristics, the ZPSS analysis considers a range of variables including:      system pressure levels of 0.1, 7.0 and 17.0 h  MPa,    heag (100,000 fluxes onand the200,000 inside B/
surf    acef of) 630  kW/m                                  h r-  t the
                                                                ,  and vessel three      of separate 315 and
[
      ;  debris accumulation scenarios.        Two      types of failure modes are
,    [  identified; the first considers the partial penetration welds
      !  retaining the instrumentation tubes (Figure 4) that fail when the
  .I    temperature of the weld material exceeds 1100 0C (2000 0F). The l- f orce exerted by the static debris head and the system pressure I  (if present) causes the tube to " slip out" of the vessel head.
The second f ailure sequence occurs when a portion of the vessel head wall is heated to a temperature above 1100 C (2000 0F) causing a plastic flow zone to emanate horizontally f rom the weakest location along a circumferential line around the bottom head. The wall begins to tear along this line f rom the weakest Point and causes the head to " hinge" downward as the tear pro-
.        gresses.
The ZPSS analysis finds that the molten core material will
:        be  uniformly distributed in the vessel head. The authors of the ZPSS point out that a unif orm distribution is unlikely due to the influence of non-coherent melting and the core support structures.
13
 
l 2                              s i          1 g          1 5                                                    I            !
6                                      ,            ,          . r 3.9 7 cm O. D. -->                l            l    4--    GUIDE TUBE i'
i            i        TUBE WELDMENT I            I i
                                                    '            l i                    INNER VESSEL L
V                    ,,M    _ _ . _
f      _I      __  h  ,,,,        SURFACE 2.54 cm                  #{
                        /            %  _ _-j-          _i_ _ _
f A                                                  '
l                                  VESSEL WALL i          I I            I 13.00 cm                                  '            '
III                              j l          i I          l Y
                                                                    ~'
I
                                        ~_~                      -
1          1
                                              !          I e          !
I          I l
liI
                                                          \
i          I I
l I
e
                                  .        l          ,'
8 Figure 4. Instrumentation Nozzel - Vessel Weld 14 l
?'  .      - .. .            -
 
They further state that their expectation is that the core mate-rial will accumulate primarily in the central portion of the plenum region. The combination of core debris accumulation in the central region, the thickness of the partial penetration welds (relative to the head-thickness), and the shear stress loading on the welds will result in the f ailure of one or more instrument tubes long bef ore other vessel failures can occur.
Thus, the instrumentation tubes provide the principal manner and location for vessel failure.
In the eighteen cases analyzed, all but one showed " tube penetration" f ailures to occur.      In twelve cases, only the in-strumentation tubes f ail, with the debris dispersed through the holes created by the tube ejection. In four other cases, tube ej ection is followed by bottom head f ailure.      For these latter cases, debris is lost through a combination of the holes in the head and the tear in one side of the vessel.      In the one remain-ing case, bottom failure precedes tube failure.
A f ailed in-core instrumentation tube will cause an initial i      vessel breach 4 cm in diameter. The ZPSS analysis indicates that l      the heat flux of the flowing core debris will be sufficient to a
cause additional melting of the steel structure surrounding the breach. As a result of the ablation process, the available flow area of the breach will increase substantially during the period of discharge. For the sequences considered, the final breach size is assumed to be approximately 40 cm.
.f      II.4  Jet-Stream Configuration i
i            The ZPSS does not give a detailed description of the con-figuration of the jet stream emanating f rom the breach in the reactor pressure vessel. The analysis predicts that the jet will be composed of completely liquid core material with a diameter equivalent to the breach dimension. Additionally, the stream
{
does not expand from the point of discharge till contact with the cavity floor. Bernoulli's equation is used to calculate the stream velocity with values of 70 m/sec f or the transient se-quence (17.0 MPa), 42 m/sec f or the small-break LOCA (7.0 MPa) and 3 m/sec for the depressurized, large-break LOCA case.
f  II.5  Ex-Vessel Melt / Water Interactions i
h The design of the Zion plant will cause, in general, an j accumulation of water in the reactor cavity during accident sequences. The amount of water accumulated is dependent on a number of circumstances during the accident such as the size of the break in the primary system (if any) and the availability or use of engineered safety systems. Conceivably, the extent of the water pool can range f rom a dry to a f ully-filled cavity region.
The ZPSS surmises that knowing the exact amount of water is not 15
                                                                              < u
 
necessary. It is only required to know if a " substantial amount" of water exists bel ow the reactor vessel. Calculations are 3
presented    in the ZPSS for a fully-filled cavity and also a water depth of 0.5 m (6 40 0 gal) f or the partially-filled cavity case, b        The former is discussed in Section II.10.
f 3            The Z PSS cites several references to establish that a steam explosion the will occur when the melt stream exits    the RPV, pene-g trates        pool,    and contacts the cavity floor. The amount of melt in the water at the time of floor contact is calculated to be approximately 7 kg for all partially-filled cases (0.5-m water depth).
The interaction of this quantity of material with the water results in mechanical work (water vapor expansion) of
}        negligible consequence to the containment integrity. The inter-action may displace the water and a portion of the debris y
the cavity region. The ZPSS concludes that the major influence f rom
)
4        of the limited work will be on the detailed disposition of water 5
and core debris in the lower regions of the containment building.
i              The dispersal of core debris and water out of the cavity L
[
w ill result in rapid quenching and steam production. It is
(
assumed in the analysis that all core debris material outside the i
vessel ing        at the time of the interaction is involved in the quench-process.
imately 70 kg. For      the sequence analyzed, this amounts to approx-Additional steaming will occur as more core mate-rial and additional water enter the cavity.
II.6    Jet-Concrete Interaction In the pool occurs,    dry cavity case or when displacement of the water concrete basemat. the  melt jet stream will impinge directly on the The extent of concrete ablation is evaluated fusing or thethe  imposed transient      heat event    flux of (22,20  0 the kW/jet.2) The heatthe m because      fluxconvective is highest heat    transfer coefficient is propor tional to the square root of the j et velocity. The coefficient is also inversely related to the diameter of the stream, but an average diameter is used in all calculations. The imposed heat flux is assumed to cause l      rapid heating and melting of the concrete surface.
The depth of melt jet erosion determined in the ZPSS analy-sequences,f rom sis  varies          4 to 18 cm f or the transient and large-break LOCA respectively. The longer duration of the melt dis-charged at the lower driving pressure (80 seconds versus 4 sec-onds), causes a larger depth in the latter case.
state that the analysis is conservative and overpredicts      The authors the extent of erosion because the heats of decomposition and heat of fusion of the concrete are neglected. Als o, the concrete melting temperature is assumed to be 1100 C, representing a lower bound of the expected spectrum of melt temperatures.
16
 
The amount of concrete decomposed by the jet is assumed to i          be the volume of material represented by the product of twice the i          jet diameter (to allow reversal of the j et) multiplied by the calculated depth of erosion.
  ;                                          The pressure rise due to the amount of non-condensable gas released is negligible compared to the j        total containment capacity.
II.7    Dynamic Configuration of Molten Core Debris I
j                The ZPSS ascertains that material discharged from the ves-sel impinges as a high-velocity jet on the concrete cavity floor and spreads outward. Initially, the flow is radial, but then the influence of the cavity walls causes the material to be directed
    ,      into the instrument tunnel. It is estimated that, for the tran-j        sient sequence, a 0.1-meter-deep flow on the cavity floor will achieve a maximum velocity of 17 m/sec upon entering the tunnel.
The ZPSS finds that the high velocity of the flow will cause a
            " hydraulic j ump" to occur somewhere in the instrument tunnel f        region. The j ump is depicted as a transition f rom a high-velo-
  !        city, low-head state to a low-velocity, high-static-head be-l      havior. The calculations show the j ump height to be relatively j      insensitive to the initial depth of the flow. The small-break g      LOCA and transient sequences demonstrate similar behavior. The
* w ave amplitude in the large-break LOCA case is considerably diminished relative to the other sequences. The configuration of the debris f ollowing the "j ump" is used in all subsequent anal-
    ;      yses, i
[      II.8  Core Debris Relocation
    }            A unique aspect of the ZPSS analyses relative to previous
[    PRA's is the assumption that the high-pressure discharge of steam and hydrogen f rom the primary system will strongly aff ect the l
final disposition of the core debris. Four hydrodynamic phenom-ena are proposed that can result in. debris material deposition
>>        outside the reactor cavity. The mechanisms have been character-
'[        ized as: particle levitation, film sweepout, film entrainment and
      <    splasho ut. These mechanisms are shown schematically in Figure 5
      ;    and are discussed in more detail in subsequent paragraphs.      As a
,{        result of these mechanisms, the ZPSS containment model assumes t    that the degraded core material is removed from the cavity very
[    shortly (on the order of one second) af ter the initiation of the
      ;    gaseous discharge.
I l        The large-break LOCA, how ev e r, does not produce a high-
[  velocity gas discharge to disperse the material.        Consequently, except for the material potentially displaced during a stean            p explosion, the core material will accumulate on the floor of the cavity and instrument tunnel during this sequence.
17 b                                                                                s u
 
4 i                                                                                                                                                                                                                    Nr A. Particle Levitation                                                                                                                                              L
                                                                                                                                                                    - ;        Qg,w,Jndr                                        5
                                                                                                                        *
* a ., *    -
o , e ,,
                                                                                                                                            .o*.
: 8. Ftn Sweepout                                                                                        o~ *, s                                    ,
                                                                                                                                                                                                .*e                                *
                                                                                                                                                                                                                                .,a
                                                                                                                                                            , , . , . . . .                  a                        '
                                                                                                                                                                                                                              *~*
                                                                                                                                                                ,o.o.          . . . ~. . , , .. , l ,o; o,.....*o.. a., .                          l( .
                                                                                                                                                                      ..:. . ^
                                                                                                                                                                .;yn,y,.~ aw, z r .;w cm'"  e o,i. . i. ,
i.~,
                                                                                                                                                  'g s y s RIQ''
                                                                                                                                                      **. -: ~
C. Film Entrainment                                                                                          5 m,1'9 i,'st . Y/ ;pt
                                                                                                                                                        .' ' ~ . ' . , , . . . , i,.
                                                                                                                                                                      ...~,,,,,:s,~,,,$p?.I,
                                                                                                                                                                        ~''' ' . ~a,'* *' P,
                                                                                                                                                                                                        -.                  J-, * '?
1 b                    m 'i s , ' . , . , . ' ,- { \s, s
c ,s*1 pn.
3, y,[.
* P~._',y,',.eo::- i e k})df y:cy.
b
::,e .- . - ,
Sm w L*E*3m '"3
_--g \ r,i'y-s-
Figure 5.      Material Removal Mechanisms 18
    <r a    _ _ _ _ _ _ _              _ _ _ _ _ _ _ _ . _ _ _ _ . _ _ _ _          _ . _ . _ _ _ _ . . . . _ , _ _ _ _ _ - - _ , . _ _ _ _ _ _ _ ..                                      .        _ _ _ _ . , _ _ . . . _ . , , _ , _ _ _ _ _ . _ . . . . _ - , .
 
i af 1
                                                      *\0%
                                                            . . . . _ . .i) i                              PI l D. Splashout                                        I          .
1 j;
                                                  ^SbNMMBik N        )
I                                                                          En I
l
:                              se 4                                              c d                                              ~
                                          %v.
(                                                .
                                                'i t
I r
b i{                      Figure 5. (Continued)
I 19 i
 
II.8.1  Particle Levitation b
Particle levitation occurs when the high-velocity gas mix-ture impinges    on the surface of the melt pool causing the flow to stagnate and divert at right angles (Figure Sa). The downward pressure displaces the liquid to form a high-amplitude " ring" of melt material. The greatly distorted liquid surface combined with the high-velocity gas flow        causes the ring to breakup and form particles.      If the hydrodynamic drag force exceeds the gravity force, then particles can be entrained within the gas stream. The ZPSS analysis indicates that large particles (25-cm diameter in one example) can be entrained.
II.8.2  Film Sweepout Film sweepout will occur when the hydrodynamic drag force exerted by gas blowing across the melt surf ace is greater than the surf ace tension of the degraded core material (Figure Sc).
Small fragments of melt entrained in the gas stream are carried into the containment where they fall out of suspension as the gas sl ows.
II.8.3    Film Entrainment Film entrainment arises when the gas stream causes waves on the melt pool surf ace.        If amplitude of the wave crest is too large, fragments erosion  or breakup    of the wave tip will occur and the will be entrained in the stream (Figure Sc).
II.8.4    Splashout Splashout occurs when the pressure exerted by the j et in-duces a radial movement at the base of the pool (Figure 5b). The movement may cause a major f raction of the degraded core material to ingbe  transported down the instrument tunnel to the upward slop-keyway.      Velocities greater than 12 m/sec are needed to cause material to leave the tunnel region and enter the contain-ment area. Calculations in the ZPSS give velocities of tens of meters per second under the conditions of small-break LOCA or transient accidents.
II.9    Debris Cooling on Containment Floor The ZPSS assumes water (up to 15 cm deep) exists on the floor of the containment building when the degraded core material is dispersed f rom the reactor cavity. Large quantities of steam are produced as the debris material is quenched. A combination of localized steam explosions, normal water boiling processes, and radiation to the condensate on the walls and structures and 20 i
 
                                                                              ]
to the entrained water droplets will cause energy to be lost.
The major influence of the steam explosions will be to disperse the water and core debris f rom the locality of the occurrence yielding a major relocation of the constituents around the con-
  ,      tainment floor. During the explosion process, a substantial amount of thermal energy is transferred f rom the melt to the water. The estimated energy extraction is about 3400 MW (time interval of 1 millisecond).
  !            The f raction of the melt not participating      in the steam explosions is assumed to be submerged in the water pool or have a line-of-sight path to airborne water droplets or to the contain-ment structures covered by condensate.          Based on the pla er surf ace area of the containment inside the crane wall (700 m ),
the energy transfer rates by these mechanisms are estimated to range f rom 6 80 to 1900 MW/sec. These rates result in quenching times of 33 to 93 seconds for the small-break LOCA and transient cases, respectively.
In all cases (except gravity-drop), the ZPSS finds that quenching will occur over an interval comparable I      to the blowdown of steam and hydrogen f rom the ' breach in the reactor vessel ( i. e. , less than a few tens-of-seconds).                i h
II.18  Debris Interaction for a Water-Filled Cavity
    ;            In sequences where extended water injection occurs (with f ailure or recirculation), the reactor cavity and instrument f    tunnel could be filled with water. The discharge of core debris g
directly into the water results in substantial hydrodynamic f rag-      f i    mentation of the melt.        The ZPSS model assumes that the core j    material is rapidly quenched to form a high-pressure steam bubble        '
in the immediate vicinity of the vessel f ailure. The growth of s  the bubble causes compression waves in the water to initiate j  water movement away f rom the f ailure location.
    ;                                                        The release of degraded core materials f rom the pressure vessel is predicted to      I j  generate sufficient steam flow to extensively void the instrument
'    g  tunnel before the high pressure steam / hydrogen gas flow begins.      '
j  The ZPSS concludes that the behavior of the water-filled cavity e
will  essentially be the same as the fully-voided (dry) cases described in previous sections.
4
      ;      For the large-break LOCA sequence, the debris would pene-        (
trate the water and remain in the reactor cavity and quench at a i      i rate governed by a critical-heat-flux limitation at the surface of the debris bed.
I e
21
{
 
{                          III.
:                                      Alternative Ex-Vessel Accident Phenomena Experimental evidence concerning some of the aspects of ex-4 vessel at the time  interactions of the document'swas not available  inception.                to the authors of the ZPSS the discharge of dense molten material at high                                                  Littletemperature is known about and pressure.
(Ref s. 6 and        Most  10)published reactor-safety analyses and experiments melts with water or concrete. consider only interactions of gravity-driven Therefore, experiments specifical-ly designed to study a pressure-driven interaction scenario may identify other phenomena not recognized by the ZPSS analysis.
;                          The material phenom      ena,                    in this section identifies potentially pertinent primarily as an aid to developing a comprehensive test plan for the experimental programs.
III.1        Melt Temperature
;                                      The condition of the melt in the RPV directly affects the potential vessel failure modes and consequently the input condi-tions for the subsequent ex-vessel interactions. The ZPSS analy-sis for tube does                        f ailureindicate  and thermal                  attack            of the vessel structure
'                        be inf not erred        explicitly by reviewing                    the melt                      temperature.
the ZPSS plot of the non-dimensional The value can
'                        parameters involved in the vessel attack (Fig. 3.1.7-2).
300    C and 1500 C for the initial and melting temperature of theUsing pressure vessel, temperature is on the order of 2100 C.respectively, indicates that the core debris that used in other sections of the report.This value corresponds to Catton of UCLA (Ref. ll) has proposed that the temperature of the degraded core material may, in fact, be just above the steel melting temperature.
This opinion is based on the assump-tion that the energy required to melt the reactor internals, combined the    debris temperaturewith the dilution                    of the fuel / clad mixture will cause to be depressed.
A lower influence                    debris temperature subsequent                    accident events. than assumed in the ZPSS will First, the reduced heat flux ZPSS, the    will cause                the vessel lengthening                  theattack time to    tofailure. be slower than visualized in The increase in time may then cause the vessel temperature profile (both laterally and axially) to be more uniform increasing the potential for more than one tube penetration to f ail.
simultaneous tube penetration f ailures cannot be discounted.The possibility                                                If of m a number of instrument tubes f ail, the initial and possibly the final breach diameters and hence flow rates will be greater than the values assumed in the Z PSS. The longer time to f ailure will also allow further depressurization of the primary system.
A second consequence of a lower melt temperature is the j
ef fect on the debris relocation mechanism. Although the exact
;                  composition of the debris is not known, it is expected that the 1                                                                                      22
 
mixture properties will behave similarly to other ternary compo-
  ,                            sitions.                            Thus, the density, viscosity, and surface tension will be related to the melt temperature.                                                                        Each of these terms is considered in the equations governing the relocation mechanism.
;                              A lower melt temperature will also reduce the heat flux of the                                                                                                    1 j                          jet stream af ter exit f rom the vessel.                                                                    Thus, the potential and                              !
energetics of steam explosions, concrete decomposition, and steam
'l  8 generation will decrease with lower temperatures.                                                                                                                l
'  f                                  A lower. debris temperature will also increase the potential a                          for f orming a stable crust above the molten material.                                                                                    A stable crust layer will disrupt the dispersal mechanisms that rely on removing material from the surface of the molten pool. The crust 3                          will also form an effective insulating layer that will reduce the heat transfer from the upper surface of the pool. This will a
cause the downward and sideward heat flux to increase and the l                          concrete interaction to be more vigorous and longer. A longer
;i                            concrete interaction period will increase the pressure load on containment by the release of water vapor and non-condensable
: f.                        (possibly flammable) gas species.
III.2                Gas Solubility in Melt I
Several of the Zion accident sequences depict the devel-opment of a molten pool of core debris while pressure exists in the reactor vessel. The pressurizing gas is, assumed to be prin-cipally water vapor and hydrogen from the oxidation of fuel l                            cladding and structural materials.
f The ZPSS analysis does not consider the potential for gas in solution with the melt and the l      [                      effect of the dissolved gases on the ex-vessel phenomena.
l Gas solubility f or metals and metal oxides is generally a function of both temperature and pressure. Reactor accidents must consider the potential for both hydrogen and oxygen disso-                                                                                                  '
      ,                      ciated f rom water to be soluble in the molten debris species. As y                    an example, consider the solubility of hydrogen in iron that is known to obey Sievert's law:
1/2 H2 (gas)                we                  [H] dissolved in Fe i                            where:
i
[H] = concentration of dissolved atomic hydrogen.
!                        ' The equilibrium constant is:
1 4
K=  I l
(PH}l/22j j
i 4
23
.p.
                                                                                                                                                                                  ^
            ,y,--,y,,----.----        . - . . . , , . _ , _ , , . . ,        .--,    , , -  ,,,---,--,..,.-.-..-..---,--.,,-n---        -
                                                                                                                                                    . - - - - -    -e._---    -    _. ~? 3 O
 
or
                                                                    -1637                  (%H) x 104 log K =              + 2.3126 = log f T
h2) where:
T = the melt temperature (K)
P H2    = the partial pressure of hydrogen (atms) .
Then:
                                                                          -1637                  1 log (10 4%H) =                  + 2.3126 +  p log Pg The      ra tio    P      /Ptotal may range from 0.01 to 1.0 for an accident                    H2 situation.        Representative values for conditions of interest to the accident are given in Table 1.
TABLE 1 Dissolved Hydrogen in Molten Iron Temp          PH                      [H ]
(K)                          cm  3(STPf/100gFe                        [H ]
MPa)                                        cm 3(STPf/cm 3 p, 1810          15.0                    3 83 7.5                                                    26.8 27 1                            19.0 4.1                    200 0.1                                                  14.0 31                              2.2 2800          15.0                      800 7.5                                                  55.9 565                          39.5 4.1                    418 0.1                                                  29.2 65                              4.5 It is apparent f rom the results in Table 1 that the solu-bility of hydrogen is a strong function of the melt temperature.
The    concentration is less af f ected by the variation in system pressure.
The amount of dissolved gas represents a large volu-metric f raction of the melt composition at ambient conditions.
Upon release to the atmosphere during ejection, the gas will nucleate of  the jet. as  bubbles and attempt to migrate towards the boundary 24
 
The data given in Table 1 are prototypic considering the quantity of steel expected to be present in the molten core debris composition.        Reactor accident studies must also consider gas solubility in the oxidic phase of the melt.            Data are scarce on the subject of gas solubility in molten oxides, but Blander's correlation (Ref.12) f or gas solubility in molten salts can be used as a first approximation:
5 In  RT p
                                                  =    -9.184 x 1916 1
where:
l                                R = gas constant T = absolute melt temperature
:~
[C]  = concentration of gas dissolved in the oxide P  = partial pressure r  = radius of the gas molecule o  = melt surface tension.
Assuming the radius of an H O      2 and H2 molecules to be 2 A and 1.2 A,    respectively and the surf ace tension of a prototypic oxidic melt to be 200 dynes /cm, estimated solubilities can be
[                        obtained.      The results of the analysis (assuming no chemical interaction between the gas and oxide) are given in Table 2.
TABLE 2
      ;                                      Solubility of H2 and H2 O in an Oxidic Melt Concentration Temperature          P              P                          HO H2              HO 2            H2        2 (MPa)            (MPa)        (1 STP/1 of oxide) 2000            7.5              7.5            3.27      0.59 1.5            13.5            8.65      1.06 9.2          14.8            0.97      1.17 l
1880              7.5              7.5            3.11      0.22 1.5            13.5            0.62      0.39 9.2            14.8            9.06      9.43 The smaller size of the hydrogen molecule causes it to be more                i soluble than water vapor at equivalent conditions. The solubili-ties estimated by this correlation are lower than that for iron.
The amounts appear; how ever, to be adequate to cause disruption              j of the melt if the gases come out of solution.
I 25
 
i I
i The implication of the above analysis is that molten core
'                                      materials, under pressure, have a significant potential for the cover gas to go into solution with the melt. The dissolved gas i
i                                      will cause disruption of the melt stream as the gas bubbles i
nucleate and attempt to diffuse out of the melt and into the environment. The melt stream must then be modeled as a two phase mixture of gas and liquid. The gas diffusion may also cause portions                        of the melt to be removed from the stream as the gas bubbles expand, burst and fragment.                                                      The fragments represent a source term for aerosol particles and fission products, discussed in more detail in Section III.4.
III.3                Vessel Failure In the ZPSS analysis, the bottom head was assumed to be covered uniformly by the core debris.                                                        The 56 instrument-tube
!                                      penetrations are concentrated in the central portion of the lower plenum,                where the drainage and accumulation of debris is expected to occur.                                    The ZPSS predicts that all tubes are attacked in a i
similar            manner, head at the individual locations.
with only a slight variation in the debris static Because the tube penetrations are identical, it is logical to assume that the thermal attack of the partial welds is nearly equivalent at all locations. This behavior seems to be supported by the ZPSS results given                                                                                              in Appendix 3.4.6 that indicate for at least 12 (possibly 16) of the cases studied, the mode of f ailure is when instrument "t u be s "
slip out of the vessel.
The consequence of two or more tubes f ailing simultaneously will f rom    bethe  to vessel. alter the ZPSS assumption of a single jet emanating For multiple f ailures, one or more jets will develop, depending on the relative locations of the f ailures.
The attack of the vessel wall, or breach growth, will then pro-ceed at will debris                morebe                      than        one location initially    larger and    andthe willflow increase        area available              in sizetoatthe a
higher rate than assumed in the analysis. The flow area for vessel blowdown will be changed in a similar manner. The subsec-
* tionsdebris the                belowdispersal.                attempt to determine the effect of these changes on III.3.1                      Mass Flow Rate can be estimated by:
The mass discharged out of the vessel, m, for a jet stream 4
lit      =OAU pc where:
j                                                    pp = density of melt Ac *                      "I
                                                                                        " "Ifo + Bt)2 = the area of the breach 26 w-,-, - -,                            .._,.n,.
                                                        , , , . , _ . , - - . , . .    --.,,,m-.
                                                                                                          -    -, -              , . - ~ . ~ . - - - , - - , _ _ , , , . , ,          ,  -
 
B = dr/dt = constant = the breach growth rate r o = initial breach radius 2(P o -P)  a U=                            = jet disdiarge velocity P
F Po ,Pa = Pressure of vessel and containment, respectively.
The expression for A n  is developed from the form given in theincompressable for  ZPSS analyses steady  while Uflow.
is determined f rom Bernoulli's equation The total mass by integrating      in from  discharged time e to time at any t i point in time is then found mi =              pp u(ro + Bt) 2 U(t) dt 0
where U(t) may be a f unction of time if flow through the breach is not choked.
For multiple f ailurer,, the area term in the above equation must be modified according to:
n
(=              A1 = n w(ro + Bt)2 l
where n = number of tubes failed.
To determine the integral  equation.the        total mass discharged at any point in time, becomes:
o m[ =                pp n n(ro + Bt)2 U(t) dt The effect ger-multiple            of additional increase        in the tube areafailures availableis to forcause  an inte-material  dis-charge as compared to the single-failure case. Assuming that the                          e density and velocity of the stream are constant, the mass dis-charge is then given by:
(
B2 t'3) mt " P nn F    t o  t'  +  ryt'2    +
3 )U l
where t' is the time from start of discharge (t=0) .
27
 
LOCAThis      behavior accident,        is illustrated comparing mass in Figure 6 for a small-break 2, 4 and 19.                            discharge versus time for n = 1, The plot indicates that increasing the number of f ailed tubes greatly shortens the discharge time. The most significant change occurs when n goes f rom 1 to 2 with the ef fect ic value asless becoming          pronounced represented  by as  n increases, approaching an asymptot-n=10.
III.3.2      Interaction of Multiple Breach Locations The behavior modeled in Figure 6 assumes that the various tube f ailures do not interact.
fail,                                      If two or more adjacent tubes the  breach  patterns  may spacing is approximately 40 cm, so that grow  to intersect. The average tube failures in adjacent tube locations will overlap in less than 0.5 second for the transient case (dr/dt = 41 cm/sec) and sli break LOCA (dr/dt = 26 cm/sec). ghtly over 1 second for a small-Assuming this behavior occurs, adjacent tube f ailures will evolve into an asymmetric single breach independent with failures.
a somewhat smaller eff ective flow area than two If the diameter of the j ets emanating f rom multiple loca-stions um es)is
                    , equivalent to the size of the breach (as the ZPSS as-the jets will not interact until the breach locations grow to intersect.
Upon impact with the cavity floor, the be-havior of the jets becomes more complex. Motions in the melt pool induced by the stagnation of multiple jets may destructively or constructively interact, the debris may be affected. so that the dynamic configuration of III.3.3 Vessel Blowdown and Material Removal The average pressure exerted by the jet on the concrete floor provides the driving force that induces the radial movement of the debris and subsequent splashout. This pressure is in-1 versely related to the area of the j et.                Multiple f ailures will involve more surface area and hence a reduced mean pressure and less radial force on the debris pool.                    The decreased radial movement causes the wave height growth to be lessened and the traveling velocity of the wave to decrease. These changes would also serve to mitigate the material removal by causing the trav-eling velocity of the wave to decrease relative to the material velocity.
Multiple tube f ailures will also alter the blowdown of the vessel      following melt ejection because the flow area for the discharge is increased.
will cause the gas velocity          The in increased the tunnelflow areaoutto of  the vessel increase di-rectly affecting the film sweepout, particle levitation, and film entrainment mechanisms.            Each    mechanism      is proportional to the dynamic pressure of                the  gas,      i . e. , pU g g      where U g 28
!L__.____.---_--------------~~-                                    ~~  ~~~~
 
i              1 6
A 1                            C i
                                          =
O                          '
L n
K    s Pa Pa
                                                                        /
i A
E  mMM                  I 5
R  5 8 1 B  4 6 0                                s
                                                                      -  =      =
e L        =                              r L  V    o a                          lu 2                                                        '
ia A        P P                          F
                                      =
M n                            S                                        e b
u i                                                                                I 4      T l
e i
p lt i                                                                                  '            u 4                                                                      M
                                    =                                                                )      r s    o n                                                                (    f I
3      d E
I e
g M      r I
T      a 0                                                                              h 1                                                                                c i
s
                            =                                                                              D n                                                                                s s
a I
2    M 6        .
e r
i                                                                                    '
u i
F g
I                                                                                    '
1 i                                                                                                      _
0            _
0            0          0          0          0        0        0        0              O 0            0          0          0          0        0        0        0 0            0          0          0          0        0        0          0                              _
0            0          0          0          0        0        0        0                              .
8            7          6          5          4        3        2          1 m
ec5 mo<2t re o
i I      3j                                                                                                c
 
and p
  ;                                are the velocity and density of the gas in the tunnel, respe8tively. Thus, an increased flow of gas out of the vessel will cause enhancement of three of the material removal mecha-
  ,                      nisms, but. the duration of the blowdown will be correspondingly reduced.
L 4
The ZPSS considers
(                        material to be the principal removal mechanism. breakup                                and entrainment of t initial dynamic processes, the velocity required forFollowing                                  the entrainment is given by:
3.7      g u (pp-o)                    g V4 Ue"
                                                  ~
V2 (pg) where the      u is the surf acceleration            due      acetotension      gravity.of the molten material and g is If the gas velocity in the tunnel, Ug , exceeds U, then material will be entrained in the gas stream.
In order for the particle to escape intact, its horizontal l                      velocity      must be converted sloping instrument                      tunnel. to a nearly vertical velocity in the For this to occur, the gas stream velocity must not change during the transition from the tunnel to the oxit into containment. The criteria for escape is given by
!                                        D<
k 4L1 where L1 and L2 are horizontal dimensions of the tunnel and L 3 is the vertical distance to the containment floor.
The material entrainment mechanism is only dependent on the l                    gas velocity in the tunnel exceeding the entrainment velocity.
Multiple tunnel than    tubeaf single ailurestube                        will cause f ailurehigher at gas velocities in the corresponding pressure conditions, thus enhancing the material entrainment removal mech-anism.
I In summary, the net effect of multiple tube failures on the the criteria for material removal. debris relocation appears to increase the This effect is balanced by a corresponding decrease in the duration of the blowdown of the RPV.
Both ef fects must be considered when attempting to deter-mine the extent of material removed. The ZPSS analysis showed
!                  only that the criteria for each mechanism were met, but did not carry out the calculations necessary to show that the combined influence to                is suf be removed.      ficient to cause all the material in the cavity t
.                                                                                      30
 
g                            III.4    Aerosol Production During Melt Discharge The ZPSS considers aerosol production during in-vessel pro-
                              , cesses, steam explosions and core / concrete interactions in the cavity region. Aeroso1 ' formed during the melt discharge is not mentioned during the analyses.      The high pressure melt ejection experiments (SPIT tests) . described in Appendix A have shown significant aerosol generation during the melt ejection sequence.
The formation mechanisms active in the melt jet have tentatively been identified as condensation of vaporized melt species, bubble nucleation and breakup, and atomization. These processes may also be active in accident sequences.
Melt jet aerosols are important because of their potential for releasing fission products into the containment atmosphere.
Secondary ef f ects include the potential heat load f rom rapid oxidation of metallic particles, the burden placed on engineered saf ety devices by aerosol deposition, and the added potential to act as ignition sources for flammable gas species within the containment atmosphere.
III.5    Energetic Concrete Decomposition High-temperature melts in contact with concrete produce very energetic reactions      (Ref s. 6 and 7). Gases thermally released f rom the concrete rapidly expand and are forceably driven through the melt to produce turbulent mixing.          The magnitude of the forces has been observed to cause levitation of large melts.
Violent decomposition will affect the heat transfer in the cavity and the dispersal of the debris.          A simple stagnation heat transf er correlation, as employed in the ZPSS, does not account for the influence of gas generation nor does it include the mechanical energy (velocity) of the impinging stream. The combination of large heat transfer rates and the high velocity of l                              the jet may prove to be very efficient in eroding the basemat material.      Enhanced erosion will increase gas generation and j                              potentially disrupt the hydrodynamics of      melt dispersal.
Melt / concrete interactione in other areas of the cavity can also influence the accident phenomena.        As the melt flows over the cavity floor, the emerging gas f rom the floors and walls may l
cause disruption of the established flow patterns increasing the pressure loading on the containment.      The upper sidewalls above the melt and the roof of the tunnel and keyway will also decom-pose due to the radiant heat flux f rom the melt,      contributing to the overall pressure loading. If the gases react with the
;                              m el t, the inventory of flammable gas species will be significant-l ly altered.
31
 
    ),
I III.6    Dynamic Debris Configuration 2.
>~                    small-break specific            LOCA configuration        analyses for subse              are very dependent up discharge of steam and hydrogen.quent removal of material by the The energetics of the melt / con-a quiescent pool. crete interactions may be capable of disrupting the formation ing gases)            The two-phase nature of the pool (with emerg-may reloca f orces without    make ittion capable of theof dissipating the j et pressure configuration shown in the ZPSS document  debris. The dynamic pool form  because    of Helmholtz  instabilities.  (Fig. 3.2.8-1) may never 6                the stagnation pressure will be reduc If multiple jets exist, ference of the pool dynamics may occur.ed and destructive inter-III.7    Geometric Features i                  The purpose of the reactor cavity, tunnel, provide access for the instrumentation tubes to the bottom of theand keyway is
                -RPV.
The array formed by the tubes occupies a portion of the cross-sectional area of the tunnel and keyway. The presence of and gas stream are ignored in the ZPSS.the tube array and its in
    ?>
ly reducing the mean velocity of the material. fluid                      over the tu eventual-of the the fluence      tunnel    near the junction with the keyway does not dispersal (0.6m by 0.6 m) that location. is small compared to the width of the tunnel atmecha other pool.
debris      three  mechanisms are predominant on the surf ace of th provide a short-circuit flowThe annular gap between the RPV and concrete c path for the blowdown gases.
A        biological shield and associated insulation are placed around theA vessel, effectively minimizing the available flow area. If these sion of the flow could result. materials were damaged or displaced    diver-        during th nthe gap would reduce the energy available f or debris removalFlow into
                      \
                  *.                                  32 9
s                                                                            . . . .
 
i k
;      IV. Scaling Analyses The main objective of the HIPS experiments is to verify the debris relocation mechanism, and therefore the scalin is based on satisfying the debris removal criterion. gThe    analysis high-pressure ejection of molten material into a concrete cavity involves a number of physical processes occuring within the same time f rame. These processes include, but are not limited to; the movement and dispersal of the core debris, aerosol generation from the jet and from the melt / concrete and melt / water interac-tions, gas generation due to the decomposition of concrete and steam evolution, and the water quenching process.
IV.1    Program Strategy The SPIT and HIPS test programs involve 1:20 and 1:10 linear scaled experiments, respectively. The results f rom the tests provide experimental information concerning the debris dispersal arguments presented in the ZPSS and will allow valida-tion of the scaling analyses. The data will also furnish insight into other phenomena of interest and how they scale to the reac-        '
tor situation.
                                                                              )
IV.2    Summary of Scaling Analyses of Sections IV.3 through IV.5 contain detailed scaling analyses    j the phenomena expected to be important during a high-pressure melt ejection into a reactor cavity. The analyses are presented f or the 1:10 linear scaled model, with additional inf ormation concerning the 1:20 scale where required. Table 3 summarizes the phenomena and results of the analyses. All        of the inf ormation,  '
except ratios.
debris removal mechanisms, are given in terms of scaling The removal teria established    by themechanisms ZPSS. Theare    evaluated in terms the cri-evaluations are based on the parameter values for the experiments and accidents (given in Tabl e 4) .                                                              f TABLE 3 Summary of Scaling Analyses Phenomena                          Scaling (Test / Accident)              '
SPIT                    HIPS Length                          1/20                    1/10 i
Area                            1/400                    1/100 Volume                          1/8000                  1/1000 Mass                            1/8000                  1/1000 33
 
TABLE 3 (Cont.)
Phenomena            Summary of Scaling Analyses Scaling (Test / Accident)
                                      .SEIT                      HIPS Time:
Melt Ejection Blowdown 1/1975              1/150 to 1/1400 1/20                      1/10 Erosion Rate 1/(0.2 to 0.7)          1/(0.3 to 0.7)
Total Erosion:
By Jet By Pool                  1/(6 to 14)              1/(3 to 7) 1/400                      1/100 Thermal Mass                      1/3200                    1/398 Gas Generation:
Jet Contact                  1/1788 Pool Contact                  1/8000 1/316 1/1000 Aerosol Particle                  1/4.5 Size Escaping                                            1/3.2 Cavity Debris Removal Mechanism Film Sweepout:
Film Entrainment:        Exceeds Criteria          Exceeds Criteria Exceeds Criteria        Exceeds Criteria if Po 11.8 MPa Particle Levitation:      Exceeds Criteria        Exceeds Criteria Splashout:
Exceeds Criteria        Exceeds Criteria if wave thick-          if wave thick-ness less than          ness less than 27 cm                    54 cm The scaled.
correctly  results show that the dimensions, mass and time are with the test underpredictingThe thermodynamic processes are less accurate mena.                                all but one of the accident pheno-
: ments, but when combined with the time scaling,                  Erosion rate is sl the scaled total erosion is less in the experiment.
The aerosol particle size scaling is significa'nt because the material    escaping the experiment    thanthe in cavity  will have a smaller size range in the accident.
sion that the size distribution measured in the experiment willThis leads to th most probably be a lower bnund on that expected for the accident.
Scaling of the debris dispersal mechanism insures that the ZPSS cutoff criteria are exceeded in the experiments. The ZPSS 34
 
TABLE 4 Parameter Values From the ZPSS und Experiment Parameter (Symbol)
ZPSS Value              Experiment (Note a)
Density of Melt (pp) 7000 kg/m3              5900 kg/m3 Pressure in Tunnel (P) 0.3 MPa                  0.1 MPa Melt Surface Tension (a)                                  0.5 N-m Ratio of Specific Heats (k)                                                        0.6N-m 1.28-steam and 11 2      1.30 - CO 2 Acceleration of Gravity (g)                                                        1.40 - N2 9.8 m/sec2              9.8 m/sec2 Height from Cavity to Containment Floor (h)                8, (maximum value)
Initial Area of Breach (Ar)                                                        0.4 m (0. 2m) g  Impact Area of Jet (Aj )                                  1.9 x 10-3 to 1.25m
* 2 5.1 x 10-4 m2**
w                                                            2 Times jet diameter      1.7 times jet diameter Cross-sectional area of the Tunnel (A)                    7.2 - 10.0m2 (varies    0.06 - 0.09m2 with length)
Thickness of Wave (A/2)                                                            (0.015 - 0.023) 0.3m                    ?
Width of Tunnel (w) 3.2 - 2.3m (tapered)
Height of Tunnel                                                                  0.3 - 0.2m (0.15 - 0.1) 3.1m Density of gas (pg)                                                              0.3m (0.15m) 600 kg/m3 t1790 kg/m3 (CO2 )
Temperature of Cas (To)                                                              1250 kg/m3 (N2 )
(T)                                  625 K (In-vessel)        ?
573 K (In tunnel)      ?
Pressure in Vessel (Po )
1.4 - 17.0 MPa          1.4 - 17.0 MPa Particle Fragmentation Constant (C)                        12 12 Aerodynamic Drag Coefficient (Cd) 0.5                      ?
Dependent on breach size, assuming breach grows with time Can be varied to fit scaling criteria t      STP.
Notes:
: a. - Values are given for 1/10th scale, 1/20th scale values when different are given in ( )        -
 
considers that exceeding the criteria is sufficient to insure debris removal from the cavity.
IV.3 Scaling of Debris Dispersal Mechanisms The four debris relocation mechanisms identified in the ZPSS are described briefly in Section II.8. The analyses given in the ZPSS establishes cutoff criteria for each mechanism to determine if the conditions for existance are met in the accident situa-tions.
The calculations and postulated              are made based on specifics of the plant accident conditions.
The cutoff criterion for each mechanism is expressed in terms of dimensionless parameters        in Table 5. Film sweepout, film entrainment, and particle levitation are all based on one dimensionless grouping while the remaining mechanism, sweepout, is based on a second parameter.
The definition less terms show that they are comprised              of the dimension-of parameters    related to the geometry of the cavity, the conditions in the RPV at the time breach  occurs, debris and  gas. and  the thermophysical properties of the molten      i Subsequent sections discuss the scaling of each of the removal mechanisms.
IV.3.1  Time Scaling Evaluating the cutoff criteria for each , mechanism does not by itself establish the amount of material removed. It is as-sumed that the extent of material removed f rom the cavity is directly exceeded.
related to length of the time the cutoff criteria are This suggests that the experiment not only must exceed the criterion but that the length of time the value is exceeded must also be scaled to that of the accident. The experiment to accident time scale can be determined by considering the dis-charge of the melt and the blowdown of gas. For the melt ej ec-tion period, the total mass discharged is given by:
mf =        Pfgr v A dt          (IV-1) o where:
of = density of the melt Vf = velocity of the melt through the breach Ar = area of the breach l              te = time required for melt ejection Assumingperiod, the discharge    that the pressure in the vessel is unchanged during
;  constant.
the density and velocity are considered Using  the  relationship rate, then equation (IV-1) developed for the breach growth becomes:
l l
36 l
l
 
TABLE 5 Cutoff Critorio for Debris Removal Machanisms g    Removal Mechanism              Dimensionless Group                  Cutoff Criterion i
;      Film Sweepout                                R1
  !                                                                        Ry 1 9
'}    Film Entrainment                            R1                    Ry 1 13.7 Particle Levitation                          Ry                              4 C    1/2 Ry1      3 C D.
Splashout                                    R2                    R 11
        ---------------------------            --------------------2 ---------------
0Vgg                        Po        Ar -2
                  =
R1    -                      '
R3 1/2            gyp    1/2    .A.
VL                  Po        -  -
                  *
* 2 2      2gh R3
                                          ~gho' n
                                                        .g_      AW k+1 k-1 Po R3  "
T          2
    ;                            k P      To      .
g  .
                                                                                                  +
Definition of Terms P - density V - velocity i
g - acceleration of gravity
    ,  A - area P - pressure T - temperature                                                Subscripts:
g    gas stream w - width of tunnel                                          L - liquid K - ratio of specific heats r - breach h - height f rom tunnel to containment                      o - in-vessel o - liquid    surface      tension A - wave thickness                                          j - jet impact area Cd - aerodynamic drag coefficient C - constant                                                                            1 R - gas constant                                                                          ,
37
 
mg = O f (Vol) g = gf  P V [ " n (ro + Bt)2dt                        (IV-2) where:
Volg = volume occupied by the mass
  ;                ro B
                          = initial breach radius
                          = breach growth rate Performing the integration yields:
I Vol g = Vg n \ r 2t ,
B2 t,3 )
o      + r oBt, +              l 3      /
ratio The  scaling (IV-3) of equation    criterion can then be established by taking the for the experiment and accident.
I                2    B t,3 9
          )E    Vol)E f        Vf n( or t, + rapt, +        3      // E        I M 4) d)A          f A      f*    f o t, + roBte 2
                                                        +Bt e          I
                                    \
3      /) A Equation (IV-4) suggests that the time scaling for the experiments radii and growth  is dependent rate.            on the length scale, initial breach various accident and experiment        The equation conditions. can be solved for the analyses show that the scaling is principallyThe                        de results of the length scale and the size of the breach opening. pendent                          on the These results are summarized as follows:
1:20 scale tE /t g = 1/1975 for to = 1.25 cm 1:10 scale tE /t A = 1/245 for to = 1.25 cm tE /t A    "
1/ 981    f0f fo = 2.5 cm The resulting scaling ratios are strongly influenced by both the length dimension ratios and the initial size of the breach.
}  The two values of the breach growth rate (4.1 and 2.6 cm/sec) have an insignificant influence on the calculated ratios.
The time scaling analysis indicates that the desired time scale the      ratio at a given experimental            size initial        scale radius.
breach      may be achieved by adj usting The HIPS experimental apparatus c m.          allows a range in the radius of approximately 1 to 3 This range results in time scale ratios of 1/150 to 1/1400, respectively.
scaling values of phenomena in the experiment.This variationthe                        may be empl 1
38
 
A second time scale can be found by considering the blowdown of the pressure vessel to be given by the expression                                for the mass flowrate through the breach:
      .                          k r *P g Vr Ar                                (IV-5)
:        where:
f                  sr        =    mass flowrate through the breach og        =    density of gas Vr        =    velocity of gas Ar        =
breach cross-sectional area The time required for blowdown can be found f rom the total mass in the system:
                    " total        "            b*
r t,
where:
tt        =
time when dispersal criteria are no longer
  ,                                        exceeded Assuming isentropic process, the density and velocity are constant for nearly the entire blowdown interval.
t e
mtotal " 5r t                = ogrrV A (tt - t,) =grr    o V A (tgg)        (IV-6) e Scaling is based on assuming that the linear dimensions of the  experiment (E) are a known fraction of the accident (A) :
              " total                  DVAt g r r b1 m d w nh A
                                  =
f                                f                  (IV-7)
              " total    l            gvat r r Mwh            E
                        >E                                >                                          '
The gas density in the experiment and accident are nearly                              i equivalent. Assuming that the pressure in the system is the same, then the velocity (from choked flow) through the breach is the same for both situations. Cancelling the common terms and rear-ranging to solve for t blowdown,                          equation (IV-7) becomes:
                      \                                8
                                                                    ,      8          8 t
b1wdown,l E              ^f lA    ," total)E          L jA, L          L jEI
                              .                                            3 E, t
u,g,    i A,),
meoe,11, tz, y,>
g3                3 39
 
The gas blowdown time is therefore scaled in the same manner as'the length ratio, either 1:20 or 1:10 to that of the reactor accident.
IV.3.2    Film Sweepout Film sweepout is identified as a liquid film that is dragged by the out of the cavity by the high-velocity gas stream caused up  and reactor vessel blowdown.              This mechanism occurs when the drag  f  orce  on film against gravity.
the      wavy    film  surface  is sufficient to lif t the The necessary condition for initiation of film sweepout is obtained from the ZPSS as follows:
                            #Y 9g 1 9                  (IV-8) gapL      !
where:
og = gas density in the tunnel Va = gas velocity in the tunnel g = acceleration due to gravity a = liquid surface tension o n = density of the molten core material Equation (IV-8) can be written to solve for Vg:
                        -          ~
1/2 9                1/2 Vg 1          -
goA n-Og                -
(IV-9) ature, Equation and        (IV-9) should be evaluated at the pressure, tem pe r-the experiment. species conditions existing in the tunnel during gas By way of example, it is assumed that nitrogen at  100 psig and 570K is a typical lower bound of the possible range.
gives:      Using Table 4 for the appropriate material properties Vg i 1.2 n/sec This value represents the gas velocity in the tunnel neces-sary to insure film          sweepout and can be used to determine a corresponding system pressure.                  Using conservation of mass and the dimensions of the system:
m            =
tunnel          mbreach 40
 
kunnel                            Dgr  a Vg ) tunnel
                                                                                                                                      ,,      v2 k-1                        (yy_1g) k reach = PA                      or              k RT
[2\
                                                                                                                          \k+1) i where:'
Po==              pressure in the vessel k                  ratio of specific heats R        =        universal gas constant Solving for the vessel pressure gives:
Po        =
2                          (IV-ll) k+1 k-1 k        !2 I b _5                        k+1                .
Using the resultant above for V                                                              g f rom above and values f rom Table 4 yields:
1:19 Scale P1 9.928 MPa =                                                                  4 psi 1:20 Scale P-1 0.997 MPa =                                                                1 psi M Pa)
Thus, all pressures included in the test range (1.4 - 17.0                                                                                            g criterion.
will cause the gas velocity to exceed the film sweepout                                                                                              .d
                                                                                                                                                                                                    'r; a
  ,                            IV.3.3                  Film Entrainment i
velocity      Waves                    will form on is sufficiently                              large. the liquid pool in the cavity if the gas Liquid entrainment into the gas results f rom wave crest erosion by the high-velocity gas stream.
According to the ZPSS,
  ;                          criterion is satisfied. film entrainment occurs when the followin9                                                                                                    n 4-)                                                                                    V    2                                                                                                      "'
gg
                                                                                                  -V2 1 13.7 l                                                                                  -                                                                  (IV-12) 90D L
    !                                                                                                                        41 i
I h
      - - . - . . . , _ . _ _ . _ _ _ . , _ _ - _ _ _ - , . _ _ _ _ . . _ , . _ , . - _ . . _ , - . _ _ _ _ . . ~ , . . . _ . . . . . _ _ _            . _ _ _ _ _ . . . _ _ _ _ _ _ _ - , _ _ . . -
 
where:
o=
surface tension of the melt Thisofdiffers constant                  from the film sweepout criterion only by            the
                    '  proportionality, so that equation (IV-9) becomes:
V2 ~ V2
                                  ^
13.7 goon Vg    1                -
                  ..        9                          (IV-13) a Solving equation (IV-13) 1.4 MPa (200 psig) gives:                for an initial nitrogen pressure of Vg 1 1.4 n/sec conditions under consideration (P 1This criteria is easily                  s em satisfi 1.4 MPa) .
IV.3.4 Particle Levitation Entrained particles can be tunnel                                        swept out of the instrument pa  r ti clif ehydrodynamic against gravity.drag forces are sufficient to lif t the criterion is satisfied:                    This occurs when the following 3      (AV  gg23 4  Q              11 gDo nj                        (IV-14) where:
cd = aerodynamic drag coefficient D = entrained particle size The entragned pressure,        pV            particle size is dependent lowing:          gg.
A reasonable approach is to assume the fol-on the dynamic D=      #
2                        (IV-15)
                          #Y 99 trained into the gas stream, but once particles                                en- are entT they are subj ect to hydrodynamic f ragmentation                          rained, The maximum 42
 
stable fragment sizes can be estimated with Equation IV-15 when I-            the constant C is about 12 (Ref.13).
Substituting Equation IV-15 into the particle levitation criterion yields the same variable grouping shown in the film sweepout and film entrainment criteria.
                                            ~
2' 2 99-2 4._( d/
[CC\                  (IV-16) 90 LD              3 Therefore, the gas velocity can be found f rom:
1/4 f4 Ci                  f980L )1/2
,                      Vg      1                                                                            (IV-17) p Evaluating the expression containing C can be done using values from the ZPSS.
1/4
                                    !_4  C[1/4      ,      [4_ 12                        = 2.34 (3 Cd /                i3          0.5 j Comparing this result to the constant in equation (IV-9) and (IV-13) shows that the particle levitation criterion will also be exceeded for experimental conditions.
IV.3.5      Splashout The ZPSS proposes that                        the dynamic configuration of the i
debris will cause a high-amplitude, small-wavelength wave to be
,            accelerated down the instrument tunnel.                                          Film mass is converted to wave mass which has an ever-increasing material velocity.
j              Splashout will occur when the wave impacts the far tunnel wall if the wave mass has sufficient kinetic energy to overcome the gravity potential associated with lif ting material up and out of the instrument tunnel.                  The necessary condition for sweepout is:
2 Vn 2gh 11                                        (IV-18) where Vn = the material velocity when the wave impacts the slanted keyway at the end of the tunnel i
1
;                                                                    43
    - _ _ . - _ _          .___ _ _          . _ . _          -. _ _ _ - - _ _ _ _              _ _ _ . _ _ _ _ . _ _ _______m
 
a h = height which material must be lif ted to remove it from the instrument tunnel.
In order to determine if the neces'sary criterion for mass removal is met in the tests, the dynamic gas pressure in the tunnel and material velocity for splashout must be related to known quantities.
The gas velocity in the tunnel can be written in terms of the mass flow rate (s) in the tunnel.
_2 AV gg 2=Pg                                            (IV-19)
Where, A is the cross-sectional area for gas flow in the tunnel.
The mass flow rate in the tunnel is equal to the choked-flow condition exiting the reactor vessel.
                                  ~                            '
1/2 k+1 k        l  2    i k-1 A=PA  or                                                    (  ~
RTo        (k + lj The gas density in the instrument tunnel is given by the ideal gas law:
Ag =                                                (IV-21) i where the terms (P,T) are obtained at a specified location in the tunnel.
Combining Equations IV-19 through IV-21 yields an expression f or the dynamic pressure in terms of known quantities in the reactor vessel and instrument tunnel.
                                                  ~                              ~
k+1 k-1 2          A~ r          Po  T        I  2 I
OV gg    =P O
                                  .A.              P    To        (k + 1/
44
 
g          The material velocity of the high-amplitude, small-wave-length wave considered in the splashout model can also be related                ,
to known quantities.            This is accomplished by applying Newton's      '
g    Law to the wave. The resulting differential equation describing j    the wave position in the tunnel is given in the ZPSS as:
1-2 dx    ,
2AP AP L                        (IV-23) dt2 where the A P term is the              driving        pressure (P o - P)
The ZPSS expression for the position-dependent driving pres-sure is simplified by using a constant pressure equal in value to the actual driving pressure half-way down the tunnel.
g2                  2g 2 6P    =                  =
(IV-24)
Vab                Pg hjW where:
A      impact area on the instrument tunnel floor of the 3 = gas jet emanating from the reactor vessel L = distance traveled by the wave at the point where it impacts the far tunnel wall
    ;          W = width of the tunnel The mass flow rate in the tunnel is given by Equation IV-29 and the gas density in the tunnel is given by Equation IV-21.
The material velocity for the wave impacting the far wall of 23.
the  instrument tunnel is obtained by integrating Equation IV-k+1                4 k-1 2-2 2        o        r      4      T V      ,
k PP                                                (IV-25)    j L      _Aj _ A W To            . k + l_
Combining equations (IV-18) and (IV-25) gives an expression for scaling based on known variables:
k+1 k-1 2
2-gh p gP                AW      To k
k+1 1    1  (IV-26)
                        , Aj ,                    ,      _
45
                                                                                  <t
 
All of the terms        in the above expression can be easily established,        except f or the wave thickness (A). Rearranging equation other        (IV-26) yields:
parameters    and solving f or A using typical values for the For    Po  =
1.4 MPa N  2 : A I 1.0 m For    Po  =
17.0 MPa N  2 : A i 127 m For the lowest pressure value, the required wavelength thickness would in order  betogreater  than the not satisfy  thelargest splashoutdimension    of the scaled cavity criterion. Therefore, the pressure range of the experiments will cause the splashout cri-terion to be exceeded.
IV.4    Scaling Thermodynamic Phenomena The previous Section considered only the hydrodynamic as-pects of the debris removal mechanisms involved in the accident and experimental sequences. Scaling of the thermodynamic crite-ria are evaluated in this section.
IV.4.1      Concrete Erosion Concrete erosion is thermal decomposition caused by contact with material at elevated temperature. Two types of interactions are defined for the high pressure discharge of material into a scaled reactor cavity:
(a)        Pressurized ej ection of the melt onto the concrete below the reactor vessel.
(b)
Flow of melt over the concrete in the tunnel and keyway.
The aspectsrateofand theextent  of erosion ex-vessel  debris in    the HIPS tests are important behavior.      Concrete erosion may influence      the  development    and processes being studied in the tests. magnitude    of the key hydrodynamic Melt streaming onto a concrete surface will impose a heat flux that may be estimated by a stagnation heat flux correlation (Ref.14) for a fluid impacting an ablating solid surface.
I      0.35 p                  f 2V0 3 )1/2 Q    =
0.55K/( C v I    s
                                        ~
c Ikd" m /
(    }
where:
Cp = heat capacity f
46 i
 
v = viscosity K = thermal conductivit 7 Ts = temperature of the melt stream Tc = melting point of concrete V = stream velocity dm = stream diameter All the terms in the above equation are roughly equivalent in both    the accident and experiment cases, except for the stream diam eter. The appearance of terms involving the stream geometry distort the rate of concrete erosion f rom that expected in the accident.      To a first approximation, the rate of concrete erosion (6) will be Equation          proportional (IV-27).                    to the imposed heat flux as given by
                                                                                              )
i IdA i 1/2 /D          1/2 cE a
k      ai                    g )i C                                  ia i                        (IV-28)
A 3        4k/          k kl where D is the breach diameter and the subscripts E and A refer to the experiment and accident, respectively. Equation IV-28 suggests        that concrete erosion is inversely proportional to the square root of the breach diameter.
Table 4:                                                    For the dimensions given in E
E      / 2.5 3 1/2                        / 2.5 11/2
                      =1                =  1.4    or eg        (1.2 j)                          I (2.5 j      = 1.0            <
These results are appropriate only for the initial rate,                            '
where RPV  andtheexperimental breach diameters            are established apparatus.
by the geometry of the  >
Ablation of the breach aperture may cause the scaling to change at later times. Erosion rate i
scaling taking during ejection isinto            account of the    form:the possible growth of the breach              3
                                                                                            )
f        dr \
1l2 C          ro+ -(t)
E                dt
(              )3 e, " r              a, 3 v2                        (IV-29) ro+ -(t) dt
(              )E 47
 
where:
ro
                            = initial breach radius dr/dt = breach growth rate t        = time the equation reduces to equation (IV-28).For t = o, the breach growth sequence, the breach growth becomes significant.              For later times in the Assuming the growth rates are nearly equilvalent f or the test and accident, then the of duration    scaling    is dependent on the initial breach radii and the the event.
period are estimated to be in the range 1/150Because                  tothe time scales 1/1400,  the for the second accident. term    in the  denominator        (dr/dt)t E is small relative to the function of t.The erosion rate scaling then becomes a non-linear Equation (IV-29) then becomes:
1/2 E
E      fo+        t AA A        [roE t
Thus, 3.8 and  the4.2ratio    of test toon depending          accident      erosion the duration    of rates varies between limiting case of the small-break LOCA t g                        the event. For the the equation yields:                                              7 seconds, so that C
E V2
                  ,      2.5 + 2.6(7)
O                                      =  3.8 A              (1.2) 1/2 where the parameter values are stated in cgs units.
The total extent of concrete erosion (c) is the product of the erosion rate and the duration of melt streaming:
        =        x          =
C A      A          D A          (1.4 to 5.1) x (1/150 to 1/1400) = (1/30 to 1/1000)
: Thus, the total concrete erosion by the j et in the tests will significantly underpredict the accident value.
48
 
The second area of consideration for scaling erosion is the effect  induced by melt flowing along the tunnel floor and keyway.
Previous melt / concrete interaction experiments (Ref s. 6,14) have shown that concrete erosion is dependent in a nearly linear manner on the depth or static head of the melt pool. The ratio of test to accident erosion in the tunnel location, is                                  then given by:
cE        hE      tE            M)                PLA)        t a        -
                                  =                  -              -
E C          DA      t l
M l              (IV-30)
A                  A                                        t PA/E g                    /A      A l
where M is the amount of melt deposited on the tunnel floor (area A) .
If the density of the melt in the accident and experiment is nearly the same, then:
3 W                Lb C                            L C
E a      /B a            /E    =    Lt) E      (IV-31) 3 A                          L t)              Lt  3 A/                g2)
For 1:10 linear scale:
cE    1      f 1          1    i        l        l cg    10    $ 150        1400/" 1500              14000 For 1:20 linear scale:
cE      1            1    j c3  a [\ 300      28000 /
The results of the above calculations indicate that the erosion      caused by the melt pool in the experiment will greatly underpredict the accident behavior.
I IV.4.2      Thermal Mass                                                                      1 The total thermal energy of the melt mass at the time of vessel f ailure can be            found For an assumed temperature ofby                        using values given in the JPSS.
2200  C and a mass of 8 x 10 k g, the thermal energy of the melt is approximately 1.02 x 105                          g g, 49
 
of 3.2The to 3.6thermal MJ/kg,    mass of thermite is estimated to be on the order depending on the degree of completion of the reaction and the heat losses in the melt generator apparatus.
Using kg      the or mass)      lower 32 MJ    value (le kggives  a total energy content of 256 MJ (80 mass).      The higher specific energy of the  thermite        compared higher reaction temperature.            to the    accident is due principally to its Because stored energy of a system is based upon the mass of the system, of the form:    then proper scaling for the 1:10 scale test should be EE mE          1 Eg            g        1000      (1:1 scale)        (IV-32)
Using values f rom above, the estimated scaling ratio is:
EE 256            1
'                                          a                  ,
EA                              400 1.02 x 105 This result suggests that the energy content of the test in both scales is roughly 2-1/2 times greater than desired. The test value can be altered by lowering the tem increasing heat losses prior to ejection.perature of the melt or due to  Degraded the decay        core    material of fission        will continue to heat after ejection products.      The decay heating curve can be used to estimate the amount of a,dditional energy that is contributed sequence.          by fission product heating during the accident This value can then be used to compare to the experi-ment to evaluate the adequacy of thermite as a debris simulant.
time The      expression is given    by (Ref 15):      for the decay power level at any point in P(t) = PgAt-a                          (IV-33) i where:
P(t) = power level Po = power at time of shutdown A, a = constants used to fit the data
'                  It is reasonable to use P                        2400 MW (80% of 3000 MW) by assuming that some of the fissio=n products are lost during the melting process priog to ejection. For the time interval f rom 150 seconds to 4 x 10                      seconds after shutdown, the constants for 50
 
the expression are:            A = 0.13 9, a = 0.2 83.                      The power generated during over history  an accident the duration  is then of the  obtained      event. by integrating the power F%jection =              (.139) (2400)t-0.283 dt                    (IV-34)
                                                      ,) t1 where:
ti, t2 = the start and end times for the discharge.
The integration yields:
t 2
                                                    't exp(1 - 9.283)                    '
                                  %jection = 312  -
1-8.283                                          W-35) ty The values f or ti and t2 are determined by estimating the time af ter the start oT the accident that the ejection occurs and the expected interval for thermal interactions. Most of the accident sequences in the ZPSS indicate vessel failure between 1 and 19 hours af ter shutdown.                      The melt ejection and gas blowdown processes have a combined duration on the order of 50 to 100 seconds, pressure. depending on the type of f ailure and initial system                                                  )
Evaluating equation (IV-35) for the two bounding times and using a conservative 199-second duration:
                                  %jection 1 hr " *            (    *
                                                                          ~
                                                                                        **'        "  ''
* 1 1
                                  %jection 10 hr= 435 (1852.4 - 1848.7)                              = 1.6 x 193 m These values represent the amount of decay heat generated during the ejection interval.                          Given the larger value at I hour af ter shutdown, the decay heating represents a contribution to the total thermal mass (1.92 x 18g.9%                                  MJ) .
additional The result above indicates that decay heat provides only a small additional contribution to the total energy content of the system, for the time interval representing the discharge of the vessel.      For the small-break LOCA and transient sequences where the  time  duration  is even    willshorter be lessthan              thanabove,
: 24. the added energy i
i due to decay heating additional heating when usin                                                    Thus, the lack of                  )
significant source of error. g a thermite melt simulant is not a                                                    3 51
 
IV.4.3  Gas Generation carbon      dioxide from within t,he matrix of the material.Th releases occur over distinct                                        These temperature ranges allowing the gas generation material.      to be related to the propagation of isotherms into the relationship (Ref 5):The velocity of propagation can be estimated from the V=                  ;
(IV-36)
PF3+      pg/E T1 where:
V = isotherm velocity 6
p== density heat fluxofincident concreteon the surface c  = specific heat F
T,T 1
                    = enthalpy of decomposition of jth species
                    = temperature limits of the decomposition process in the accident or the test,If the heat flux is suddenly imposed, as would be the isotherm properties        velocities of concrete.      are dependent only on the thermophysical constituents that contribute to gas generation and their asso-Table ciated decomposition enthalpy values (From Ref. 6).
TABLE 6 Thermal Processes in the Decomposition of Concrete
* Decomposition Weight Percent          Enthalpy Constituent              (%)                (J/gm)      Temperature Range (C)
Free Water              2.7                  81.6            30-230 Bound Water              2.0                  120              360-485 Carbon Dioxide        22.0                  960              600-1000
* Generic Limestone-Common Sand 52
 
      .      .              .                ..                    =. _-      -  _- ._  - -
i
^
Comparison of the data in Table 6 indicates that the two water loss mechanisms occur over different temperature ranges of nominally equivalent span. Thus, enthalpies for these events are
  'f    approximately the same. The velocities associated with these two mechanisms will be nearly equal for the same imposed heat I
l        flux. The isotherm associated with the decarboxylation process
(    will proceed at a slower rate because of its relatively larger temperature range and enthalpy of reaction.
g The above analysis implies that the quantity of gas gene-i rated f rom concrete is related to the applied heat flux and the duration of the event. Using equation (IV-27) gives:
            "9E a-E x-tE    AE        Id N A        tE    A E
4 x-      al        'x                (IV-37)
]
m 9A Qg      tg    Ag        (dE/      -
tg xAg where:
mg  =  mass of gas released A =    area covered by jet stream In order to evaluate equation (IV-37), the stream diameter t
must be known.        One approach is to assume that the stream dia-meter is equivalent to the breach diameter in the reactor vessel or melt generator.        Thus, for the 1:10 scale model:
                "9E        /53 1/2*    f i          1i    1  f 1          1  i Igg      k2.5j        \ 150      1400j
* 100$10600      99000j The gas released during the experiment will vastly under-predict that expected during an accident.
This result is also subject to the same reservations stated previously        concerning the dependence of the incident flux and time scales on the breach diameter. Variations in breach radius and    growth rates will cause the ratio to vary within the indi-cated range.
The gas generation f rom the floor and sides of the tunnel in contact with the pool can also be related to the                    heat flux from the melt. As the melt moves across the tunnel                        floor, the heat flux is assumed to be a combination of conduction and convection.
After    the initial mov em ent ceases, the heat transf er will be principally conduction with gas-injection induced turbulence in the pool.      In both cases, the heat flux is proportional to the temperature diff erence . (AT) between the melt and the concrete.
The scaling is then:
53 m
 
m        ATE 9E                  tE a                          AE mg              x            x ATg        tg            3A                          (IV-38)
In this case the time scaling is obtained f rom the mecha-nisms controlling melt dispersal (blowdown of the vessel). If the temperature of the pool in the accident and experiment are nominally equivalent for 1:10 scale, equa tion (IV-3 8) becomes:
                  "9E        1        1          1 15
* 100            1000 The result indicates that the amount of gas generated in the test due to contact by the melt pool will                        scale in the same ratio as that desired for the relative masses of the melt.
IV.5    Scaling Aerosols The aerosol / fission product source term to containment building      is defined by:
1.
Mass of aerosol input to containment (penetration)
: 2.      Size distribution of aerosol
: 3.      Siz e-depende nt composition of the fission products and bulk material of the aerosol The melt ej ection aerosol / fission product source term is in
,      addition to and distinct from source terms arising                    from in-vessel f uel disruption and .aelling (the so-called in-vessel source term) and  from vessel      ex-vessel source    term).melt-concrete            interaction (the so-called ex-The aarosol source term is important be-cause the      it presents containment          a potential response          to theradiological accident. h'azard and may affect Experiments (Appendix A aerosols      upon    high-pressure            have demonstrated e)jec. tion of a    the formation thermitically generated  of melt.
In the accident case, r.erosols ejected into the reactor cavity  must    traverse the tunnel and keyway bef ore they are re-leased to containment. Material that does not escape the keyway is ot little significance to che containment aerosol source term because it settles out onto the cavity floor or impacts on the bend of the keyway and is returned to the melt. Figure 7 illus-
    .trates the relationship between the melt ejection aerosol dis-tribution, the size-dependent keyway penetration function, and the resulting aerosol source term to containment.
54 4
 
I z
9 E
Oam iIt e6 zJ
: n. 4 5E 9
x 0
log AERODYNAMIC DIAMETER 1.0 2
9 4
x 0
log AERODYNAMIC DIAMETER
_.i o
m am Ow                                                                                      \
8
                      -O                                                                                        (
EN z<
gE                                                                                      .
b Z
N w
: 0.                                                                                -
                                                                                                              )
1 log AERODYNAMIC DIAMETER 1
re 7.                Penetration Aerosol Relationship to the Source Term 65
 
                            .  ..  - . = -                  - - . -                = - _    - - -          --.          -    -    . --        - . - .
L                                                                                                                                                                  l 4
4 I
An aerosol source term to containment has two aspects of i
_importance separate from- one                    radiological another.and physical - which are difficult to fission product                                                  The melt ejection aerosol has a
'                                  These products are associated with bulk particles whose phys properties in the containment        determine              transport charcteristics and residence time atmosphere.
It is therefore necessary to know the particle size distribution and associated fission product inventory i                              scenario.                  to assess the radiological hazard for the accident                                                  ,
effect on containment response. Physical properties of the aerosol ma                                                          i 3                                                                                                      At large concentrations, aero-sols of fan coolerscan        plug          filters  and      vents      and  interfere with the function ejection case,, pumps,                              and unoxidized metal heat    transf    er surf aces. In the melt ding an additional heat load. rapidly oxidize upon provi-                                                entering the c ignition source                          for the hydrogen The particles could act as an process.
In the event                                released during the blowdown aerosol would be the principal radiological hazard.of an earlythe                                              containmen inadequate above              issues.                  to fully address the potential consequence The Phase I SPIT test data are of a qualitative been considered in the ZPSS. nature that prove useful in suggesting phe The scaled                      cavity tests melt relocation phenomena                  in mind. mechanisms                        of the ZPSS and are sized with these It is possible, however, vations on the test scaling with respect to aerosol formation.to make some ob
: 1.              Aerosol generation due
,                                                        diffusion (1:10).              will scale with path length through the meltto bubble n
: 2.              Material aerosolized by pneumatic atomization is pro-i i
portional to the material remaining in the pressure functions is not scaled.are dependent upon the vessel geometry i
but can be assumed to be between 1:199 and 1:1999. Scali 1                                      3.                                                                                                                      .
Vaporized surf  ace area        mater (al from an undisrupted jet scales with
!                                                      tics of aeros(ol generation are dependentL                              The  kine-time of propa
                                                                                                                                  ) and on th i
scaled parameters of temperature and pressure. e un-                              The l                                                      time of propagation is dependent on jet velocity (un-                                                _
scaled) water pool        andsurface.
the height -of the vessel to the concrete or
* It is not certain  that the aero-s sol generation scales linearly with                          jet height  (1:18).
!                        .                                                                  56 i
P
    , , , . - , - . . . - - ,          -em-wn,y.--.m.m        ,v---.w,.                                                -
 
I 4.
The velocity of the particles in the tunnel is depen-dent on slip and residence time.          The sr. aller size of the experiments may reduce the extent of aerosol re-moved because of the shorter length scale.
These scaling observations lead to the conclusion that aero-sol  formation in these tests will not scale in a straightforward linear manner.            If aerosol generation scales with melt volume, the concentrations volume    is scaled by the  willsame be atfactor.
realistic levels because the cavity This does not consider the effect of the shorter jet height, which would act to decrease aerosol mass.
Another important issue to be addressed by scaling is the inertial impaction of particles at the end of the keyway. Aero-sol size-dependent transport through the keyway in the scaled tests    will not correspond directly to that expected for the accident case.            Iner tial impaction of particles at the transi-tion f rom the tunnel to the keyway is assumed to be the principal removal mechanism during transport through the keyway. The Stoken      Number (Stk) is the appropriate scaling parameter for im pa ction.
Physically, it is the ratio of particle stopping distance to some characteristic length of the impaction geometry and is detined as:
pD  2 Stk =
pp C(Dg)g gy                                    (IV-39) where:
pp = particle material density Dp = particle diameter C(Dp ) = Cunningham slip correction i
g = acceleration due to gravity V= absolute viscosity of gas l
l                L = characteristic linear dimension of impaction geometry For similiar geometries of unequal size, it is reasonable to assume that the penetration (fraction of material into contain-i ment) as a function of Stokes number will be nominally the same.
Theref ore, the scaling r ela tionship f rom equa tion (IV-3 9) be-comes:
1  =
StkE cx Dg2g  #
E StkA                                    (IV-40) l                                [Uh2 3-1 r      >A l
l                                            57 l
t                                                                                    ~
 
where been    the equivalent terms in the accident and experiment have deleted.
can be found: From equation (IV-40) the particle size dependency U
pE      ILE "I
hg        ALA /                              (IV-41)
For 1:10 scale:
1 g          3.2 For 1:20 scale:
pE        1 g        4.5 Based on the Stokes number equivalency, particle entering experiments    thantheincontainment the accident. region is much larger in thethe diameter of the in the reactor case.the cavity in the scaled tests will almost certainly a This behavior is illustrated in Figure 8, showing      the Stokes number. loss  of  particles  by impaction as a function of If accurate data are obtained on the aerosol distributions at equivalent article can be made. conditions, penetration estimates f or                the tes Stokes Number dependence      Theseofdata    will allow penetration  andestimation estimation of of the the aerosol sizes leaving the reactor keyway.
58
 
1 I
I e
o  .
8 Q,
a  >,
O  8 e
f  @
O
                                                                                                                                                                                                                                      .M A
                                                                                                                                                                                                                                      }
h.
Q.
ed O      t t
O I
i
* O l
(-) NOl1YB13N3d I
f 59 i
l              *
                                                                                                                            \
                    , _ , , , . _ _ _ _ . , _ _ _ _ _ . _ _ - , _ _ . , _ _ _ . _ _ . ~ , _ _ . . _ _ _ , _ , _ _ _ _ _ _ _ _ _ _ . _ . . _ _ . _ - . . . . . . _ _ _ _ _ _ . . . . _ _ _ _ _ _ _ _ _ . _ _ . _ _ . _ _ _ _ _ . _ _
 
r V. Experimental Programs V.1  Program Overview j: 'L                                The results f rom the Phase I SPIT tests (described in 4    ,              . Appendix A) have shown the behavior of a molten stream ejected at high pressure to be uncharacteristic of that hypothesized in the ZPSS.
)
  ,,                  dent on the degree of gas in solution,The SPIT jet geometry appears t stable to expanded and finely fragmented. ranging from coherent and The tests have demon-strated an unexpectad propagation,                aerosol source term accompanying the jet mechanisms.        assumed to be caused by three or more separate 1
The experiments:
following    Phase II SPIT test program is designed to perform the
: 1)  Perform    jet  and aerosol  characterization  tests.
: 2)  Study  the  interaction of the jet stream with a pool.                                                water
: 3)    Conduct 1/20th-scale concrete cavity tests.
These test results will provide the majority of data con-
        '            cerning the dynamics of high-pressure melt ejection.          The Phase
;                    II SPIT test matrix is outlined in the f ollowing section        (V.2).
identification of the independent system v.ariables and the de velopment design approach.of the characterization test matrix using a factorial The HIPS verification      of experiments the ZPSS' are intended to provide experimental realistic, scaled geometry. debris        relocation mechanisms in a The dimensions of the HIPS apparatus i                    and planttest    article represent geometry.                a 1/10th linear scaling of the Zion The test matrix presented in Section V.3 is designed to provide data over the range of possible accident conditions.      A limited number of characterization experiments are also included, to allow correlation to the more extensive results obtained f rom the Phase II SPIT tests.
The HIPS apparatus and equipment are discussed in Section VII along with the strategy used to develop the test matrix. The complexity and resources required for the HIPS vent using a large factorially designed test program. experiments pre-V.2      SPIT Test Matrix i
The SPIT characterization test matrix (Table 7) is based on variables identified during the Phase I SPIT tests.a              f actorial design The results of the matrix should demonstrate the influence of these variables 69
 
and their interactions on the behavior of the ejected melt stream and aerosol generation.
oped and verified.        The data will allow models to be devel-TABLE 7 SPIT Phase II Jet and Aerosol Characterization Matrix Pressure                  Approx. Temp. Jet Test          (MPa)        Gas            (OC)      Height (m) 1            6.3 N2 /CO2        2250            0.9 2          17.0            N2          3000            1.6 3            1.4          N2          3000            0.2 4          17.0          CO2            3000            0.2 5            1.4          CO2          3000            1.6 6            17.0          N2          1500            0.2 7            17.0          N2          1500            1.6 8            17.0        CO 2          1500            1.6 9            1.4        CO2            1500            0.2 10              6.3                        2250 N2/CO2                          0.9 The Phase II jet-water experiments are shown in Table 8.
The tests  are designed to identify the type of interactions and quantify their extent. The results are intended to be used      for the selection of instrumentation devices for subsequent SPIT and HIPS scaled cavity experiments. The specific design and number of additional tests involving water is not known due to the uncertainty in the response of the system.
I 61
 
TABLE 8 SPIT Jet / Water Interaction and 1/20th scale Cavity Test Matrix Test    Pressure          Configuration (MPa)                                          Purpose 1        6.3          Deep water  pool,        Determine steam directly contacting      bubble growth in exit    aperature        " solid" water cavity 2        6.3          Aluminum 1/20th          Study dispersal of scale cavity, pool      water pool and hy-contacting exit          draulic forces in a aperature                " rigid" cavity 3          6.3        Concrete 1/20th          Debris dispersal scale cavity              (dry cavity) 4          6.3        Concrete cavity          Debris dispersal with water pool          (water present) contacting exit          jet / water inter-aperature                action 5        6.3          Concrete 1/20th          Interaction in scale cavity shal-      partially filled low water pool            cavity The scaled concrete cavity test (Test 3) will provide infor-mation at 1/20th scale for use in subsequent modeling ef forts.
The experiment will also allow the verification of instrumenta-tion techniques to be used on the HIPS tests. The damage incur-red by the test article will determine if other tests on the same unit can be accomplished.
Additional not been  defined, test conditions for the matrix in Table 8 have pending analysis of the results of the jet and aerosol characterization test matrix.
desired melt temperature will be chosen The        type of gas and the to obtain the maximum system response i.e.
the most energetic jet / water interaction or the highest probability of debris dispersal. The results of either test matrix may dictate that additional experiments            be performed behavior. in order to reduce uncertainties or study unexpected 62
 
V.3  HIPS Test Matrix Scaling analyses of (Section IV) indicate that material dispersal isthe mechanisms proposed in by the initial vessel pressure and temperature. most influenced do not allow for the influence of gas dissolved in the m Analysis of these latter two effects suggests that they may have a significant Table  9      influence on the debris relocation phenomena.
dispersal and their expected range. identifies the critical variables influ TABLE 9 Critical Debris Dispersal Characteristics Characteristics                    Range (Extremes)
Pressure:
1.4 - 17.0 MPa Melt Temperature:                15000 C - 28000C Water in Cavity:                  dry - water filled Gas Solubility:                    low - high tics given in Table 9 would require 16 tests total,A                  simple tw more than could be handled in a convenient manner. A f ractional factorial and with no replicationscheme to find just main effects and not interactions o of data would require at least 8 tests.
This  number HIPS testing. is also  impractical  under the scope of the Phase I The HIPS of the existence      test thestrategy  is therefore based on establishing effects.                    dispersal  mechanisms by isolating main a logic decision circuit given in Section VIIThe          technique (Figure 2 8). for accomplis The intent of  testof the circuit is to provide a method of finding the range                )
exist      conditions  where  the ZPSS  material dispersal mechanisms (or  do not  exist)  .
circuit is given in Table 10.The test matrix based on the logic
{
63
 
TABLE 10 g
HIPS Test Matrix Expected Melt    Tempera-4 Mass    ture L
Test      (kg)                Pressure (OC)        (MPa)        Gas  Cavity            Remarks 1        80      2400        17.0        N2    No cavity      Jet and aerosol 2        80      1800 characterization 1.4      CO2  No cavity      Jet and aerosol 3        80      2400 characterization 17.0        N2    MgO Most probable de-bris    dispersal conditions,  MgO cavity 4        80      240E        17.0      N2    Concrete      Most probable dispersal condi-tions, concrete cavity 5          80      1800          1.4      N2    Concrete      Least probable dispersal condi-6 tions 80      1800        17.0        N2    Concrete      Determine effect 7
of temperature 80      2400          1.4      N2      Concrete Determine effect of pressure 8W        80      2400        17.0      N2    Concrete      Water-filled cavity 9W*        80      2400        17.0        N2    Concrete      Par tially-filled
* cavity rss Possible ults.        inclusion in matrix depending on Phase II SPIT t est tests that bound solubility          ranges.
the expected temperature,The                      matrix is initia characterization pressure, They are followed by a test using initialand gas conditions relocation mechanisms.
conditions        are assumed the  highestFor to bethe most purpose  favorable of            for these illustration,    the existancei bility values within the prescribedtemperature, range.pressure, and gas solu-            ,
Presumably, the SPIT jet the decharacterization tests will provide a basis for establishing tions, gree of gas solubility associated with a set of test condi-                          ;
and the influence of dissolved gas on the dispersal                            e 64                                              ^
l
 
mechanisms. The cavity will be constructed of magnesium oxide to eliminate energetic concrete decomposition phenomena to minimize disruption of the dynamic debris configuration. The presence of water network.in the cavity is considered in another portion of the logic The results f rom the HIPS tests numbers 3 through 7 will establish nisms. The  the principal main ef fect on debris dispersal mecha-Phase II SPIT data will be used in conj unction with these involved.results    to quantify the secondary effects that are also The " dry-cavity" tests described above represent a lower bound on the possibilicies regarding the presence of water. The opposite extreme is when the cavity is completely filled with water, 8W to the lower extremity of the pressure vessel. Test HIPS-is designed to study this condition where melt is ejected under high pressure into a " water-1ccked" cavity.            The objective of  the test is All dispersal.        to determine      the  influence  water  has previous paths in the logic circuit are  onsubse-debris quently comes. directed along this branch, regardless of previous out-The HIPS-8W test      is applicable only to a completely filled cavity, a from    eitherpartially-filled this or the dry cavity  may cavity    behave quite differently case.
shallow    waterII pool                                  The behavior of a of  the Phase        SPITwill  be addressed in the jet / water test matrix experiments.
The results of the SPIT tests    ;
may a      suggest distinct,      that third      a classpartially-filled of experiments.cavity must be considered as      (
9W                                                  In that event, HIPS Test pool. will be conducted to study the effect of a limited-water Included in the HIPS test matrix are jet and aerosol charac-terization experiments. experiments to provide correlation with similar SPIT Characterization tests are necessary to confirm the data for scaling SPIT existence    of the          mechanisms at a larger scale and to provide analyses.      If significant differences are noted in the larger 1/10th scale HIPS jet and aerosol tests, additional characterization tests may be required.                                        )
The test matrix is not intended to provide definitive infor-mation about all possible combinations of accident conditions,                ,
l only    those dispersal      factors that appear to be most critical to the debris mechanisms.
i of possibilities for these factors,By studying the extremes of the spectrum bounds of potential test outcomes can be made. statements concerning the l
65 l
L
 
l l
VI. System Pressure Ejection Tests (SPIT)
The results f rom the Phase I SPIT tests (described in Appen-dix A) stream stable              have shown          that the ejected as hypothisized                              melt jet is not a coherent, in the ZPSS.
jet  appears to be highly dynamic and dependent                                              The behavior in part on    of the    the initial              conditions within the melt generator. The results also suggest during            thea ej number ection of    aerosol generation mechanisms are active process.                    These aerosols may cause a signifi-cant    source melt / concrete mechanisms.
term      separate,                    and    distinct f rom the in-vessel or The second phase of the SPIT testing is designed to provide a
ej cost ection.        effective study of the behavior of high pressure melt in scale,                It is also a method for obtaining quantative results, tion.                  concerning jet / water and jet / concrete cavity interac-The following subsections describe the SPIT apparatus and instrumentation and the salient features of the test matrices.
VI.1            SPIT Apparatus Description This section associated                            describes the SPIT m el t generator and the facility components.
the melt composition and temperature.                                    It also includes a discussion of VI.l.1                Melt Generator Assembly and Testing perature  The SPIT      melt withinmelt generator a pressurized,              is designed to create a high-tem-confined volume. The re-quirement for the high strength associated with pressure vessels is generally high      temperature. not compatible with devices capable of withstanding To accomplish the design ob                                              it is necessary nent.
to  separate      the melt from the pressure jbearing                      ective,compo-The SPIT generator design is shown in Figure 9. The pres-steel pipe is f abricated f rom a 12.7-cm-OD by 11.4-cm-ID mild sure      vessel bolted covers.section              The capped on each end by welded flanges and rated maximum working pressure of the vessel is 17.0 MPa.
section (10.2-cm OD x 8.9-cm ID)The melt crucible consists of a mild steel pipe each end, placed inside the pressure vessel. capped by graphite plates on The annular volume between the melt crucible and pressure vessel wall is filled with a ref ractory, alumina-based, dry-ram powder. The lower vessel flange 3/  4 " N PT.cover is threaded to accept a fusible brass plug normally toelsurround m      t.                    the melt plug and expose only the top surface to theT into theThe        melttop          graphite plate is drilled to allow gas to enter crucible.
l 66
 
GAS LINE TOP FLANGE COVER o            e o                                  e 2                      \                    -
FLJ////////                f/fffffL r    --z      1 l            // // z i  >A f      s -mm.            /
4
                                                  %s
                                                      /            THERMITE POWDER d4                %s dh                %v
                              ,g                g,        STEEL CRUCIBLE H
                              #4              -#  8 VESSEL WALL
                              ,                  4  s
                              #  %              4 d
                                %            )4d q            l% $
l *-
1 %            ..
4 s
                                              %v i4    j          g,                  FUSABLE PLUG
                            #4 1          hs d
g,                      LOWER FLANGE p
                                                    ~                  COVER o
7C r111    e ,/ r E // // s a
: /,
o RAW 6T// / //ATV // /7/1                            1/'
EXIT APERATURE Figure 9.        SPIT Melt Generator 67
 
flange cover removed. Assembly of the melt generator is performed with the top Af ter the brass plug is installed and the lower    graphite    plate lowered atop the lower graphite    dropped            intodisk.
place, the steel inner liner is poured into the gap between the two vessels.TheThe                                dry-ram  powder thermite  powderis is to then  gradually improve    settling.poured into the inner liner and lightly tamped The capacity of the melt generator is approximately (nominal density 2.2 g/cc). using 10.5        kilograms                  an Fe3 40 /Al thermite mixture
)
When the thermite is in place, an igniter  wire depth of 1-2 cm.is  embedded            into          the  top surface of the powder to a The      igniter
(~20 cm) to allow attachment to electrical              leads        are kept suf ficiently long in the top flange cover.                                                    feed throughs placed in position,                              Af  ter        the    top    graphite    cover is placed vessel.            the  top        flange    cover          is      bolted  onto    the pressure The test sequence L    and melt generator to the desired pressure.is initiated by charging the accumu in the upper flange cover monitors the pressure                                A transducer in the located vessel with the output recorder,              of the device recorded on a transient digitizing
(    When the preset a high-speed FM recorder, and active readout panels.
pressure level is achieved, is terminated and the thermite reaction is initiated.the pressure                                source A 28-VDC battery is used to cause an intermetallic reaction of the igni-ter, vicinity  creating of thelocalized igniter.        temperatures                  (in excess of 2700 C) in the greater than 800 0C for the reaction to proceed.The thermite requires temperat of nominally 2.5 cm/sec, although this value can vary de partly on the extent operation.
A high degree of tamping apparently improves the heato transf er within the powder bed, propagate at a higher velocity. causing the reaction front to the bottom of the crucible, the melt                        When    plug thequickly reactionf front  reaches ails and  the molten material is forceably ejected f rom the vessel.
Post-test inspection of the melt rior to be virtually devoid of any melt remains. generator shows the inte-mates      indicate less than 5% of the melt lef t in theVisual Some additional esti-generator.
the top graphiteresidue is found in the expansion volume between cover and the upper flange. This material appears to be ash-like and is assumed to be made up                                          of contami-less than 0.5% of the total melt mass.nants from the melt or nonstoichiom usually      penetrated          in    several          locations,        The but inner the    steel dry  ramliner is  is an effective protection for the vessel wall. In general, the aper-ture the plinug,  theapproximately bottom flange2.5          corresponds cm.                        to the outer diameter of Slight erosion of the steel insert      normally threads in the aperature. occurs,            suf ficient            to obliterate the internal Small amounts of residual melt are deposited in this location, possibly                              by material dripping down into this area af ter the ejection process is completed.
68 t
 
VI.1.2    Interaction Chamber For the Phase II testing, the melt generator will be placed in a large steel chamber similiar to that shown in Figure 10 The  chamber ticularly        allowsof aerosols,    containing    and measuring the products, par-the experiment.      The melt generator is mounted the chamber. on three load cells that are supported by the " roof" of Thus, the top flange of the device is outside the chamber      for  ease  of assembly while the lower flange and exit aperature are within the structure.
The height of the generator also permits longer jet propagation distances, a desirable feature  in the study of aerosol generation (see discussion in Section VI.2.)
The chamber is constructed of 1/4" thick steel plate with 6" I-beams on 26" centers for reinforcement. The large round open-ing in the side of the structure is used to house a junction box for instrumentation and control wiring. The cables terminate in the apparatus.
the  control center building located approximately 20 meters f rom The large covers on either end of the chamber are bolted insure sealing. with into  place        gaskets between the mating surf aces to The floor of the structure is comprised of a reinf orced concrete pad, nominally 6" thick.            A relief valve is used to prevent internal pressures f rom exceeding 0.2 MPa. The maximum MPa.          working pressure of the chamber is estimated to be 0.3 For the 1/20th scale cavity tests, the melt generator will              i be removed article.
Thefrom  the test    roof of article    the will    chamber then  be    and bolted placed  near  to the one  endtest of the interaction chamber. The exit of the cavity will be directly away f rom expand    awaythe    near from  theend testso that the ej ected debris will be f ree to device.
VI.l.3    Pressurization System The SPIT pressurize    the apparatus    uses melt generator. either nitrogen or carbon dioxide to merical storage bottles feeding separate  The gases are supplied by com-of manifolds. The output each manif old passes through a manual shutof f valve to a common pressure regulator (see Figure 11). The regulator in turn feeds through  intoaan  accumulator remote    controlvolume shutoffand  then to the melt generator valve. The accumulator is included caused by liberated gas during the thermite reaction. increases, in  the  system  to mitigate  potential pressure Experiments requiring pressures higher than can be con-viently provided by the gas bottles will use a gas intensifier incorporated into the system. The pressure rating of the intensifier is much greater than the maximum required for the testing.
tor directly Thewhen device  will charge the accumulator and melt genera-test  pressures above 10 MPa are needed.
69
                                                                              -R
 
                                                          +
                                          =\ INSTRUMENTATION PORT REMOVABLE DOOR MELT GENERATOR BOTH ENDS)                                          '
INTERACTION i
                                                              / CHAMBER
  ^7    W        W        W      W        W      W      /W      V ,s N'    !
Eb53      ,I
:                            ,V              /
                          /                          '
t k
                            \
I I
b E      o          3a      sca l
l Figure 10. Schematic Showing Relationship of Melt Generator and interaction Chamber l                                    70                                                      ,
i
 
PRESSURE TRANSDUCERS
                                                                          ' GENERATOR            BURST O                                        m    DISK
                                            -4        ~                        FEED g      ;  I        ..
9    I VALVE m
C MAIN y
HSHUTOFF              H w
VALVE          ,
X  I      X HIGH PRESSURE LOW PRESSURE                                            E VALVE        MELT REGULATOR                            HIGH PRESSURE M REGULATOR                        -            GENERATOR
                      - :4 4<:                                            2 VENT H
VALVE hhh              hhh  hh            hh                                    ACCUMULATOR e
\
              .N2              ..  .      CO2    ..
* SYSTEM SOURCE    SYSTEM SOURCE                                                  .
Figure 11. Pressure System Schematic w                        ~    ~                                                      _ __ -
 
via remotely controlled valves.The pressurizing system is operated f rom th ing pressure levels are used at the            Remote      transmitters gas manifolds      andforat monitor-the melt generator.
VI.l.4      Melt Composition and Temperature lothermitic reaction of iron oxide and aluminum:The melt used in 3Fe304      +    8 Al  -
9Fe    +              +    heat 4 Al O23                        (VI-1) the energy mately    876 cal equivalence
                            /g.            of a stoichiometric reaction is approxi-are mixed just prior to the test.The reactants are in the form of powders that The temperature achieved by thethe of  mixture reactionis assumed productsto(Ref.be limited 16). byThe  the vaporization of one completeness of the and pressure imposed on the reactants. reaction depends on the detail ceed to less than 85% completion,                          For reactions that pro-aluminum controls the maximum temperature.the Reactions          behavior ofbeyond the excess this point liberate sufficient energy to vaporize the excess aluminum.
The controlreaction      must proceed to 94% for the boiling of the iron to the temperature.
At pressure, the boiling of iron is
. suppressed and temperatures as high as 3300 K may be achieved.
Considering the above reaction, the amount of heat liberated tants, used to determine the maximum temperatures of the reac-can  be 3Fe304      +    8Al  -
9Fe    +
4Al O23    +    H Rxn (298)      (VI-2) where H Rxn(29 8) is the enthalpy of reaction at 298 K.
The maximum temperature can be found by:                                                  I T                    T H            =
Rxn(298)        HRef + MFe      Cp dt + M            cpdt A1 023                    (VI-3)              ,
Tref                  TRef where                                                                                            [
1
                                        \
HRef = Mpg l                                    f                t i H(Tref) ~ 298 /iFe+ MAl O23 I (Tref) - H298 l              (VI-4) t
                                                                          /Al 23 O                  -
where:
M pe        =
moles of iron M            =
Al O23          moles of alumina k
72
 
Handbook values (Ref. 17) are used to give:
H            =
                                      -797900 cal Rxn(298) cp(N)    =
10.74 + 4 x 10-4 (T - Tg) cp(Al 23O)  =
34.63 M/mle K.
Assuming the reaction goes to completion, 797900      =  9 H(T g) - H298 pe + 4 f H(T g) - H298 Alg T
          +9                                                      T 10.74 + 4 x 10-4 (T - Tg))bdt+4              34.63 dt TRef TRef (VI-5)
If T Ref is selected to be 2400 K H2400    -
H 298 pe      =
23m M/mle &
H2400    -
H              =
298 gig          91929 cal / mole A124 0
and 797900  =    9                                                    2 23917 + 9.78 (T - 2400) + 2 x 10-4 (T2 - 2400      .
                                                                                )~
            +4      91929 + 34.63 (T - 2400)                    (VI-6)
If the process is assumed to be adiabatic, then the maximum released is 876 cal /g. This value could cause up to 5% o t e iron to be boiled, process.                            if the pressure does not suppress the l
Maximum melt temperature above the iron boiling point is a
!  ambient environment. concern if the iron " flashes" as the melt is discharged into the source of                        Iron flashing could then serve as potential discharge. non-prototypic aerosol-like material during the mel t Non-reactive materials can be added to the thermite
'i  boiling point. mix to insure that the maximum melt temperature is below the iron the                      Ten percent, by weight, i'
temperature by an additional 150 K,of iron powder will lower output sufficient to insure that flashing does            a not reduction occur. in heat li                                              73 i
__ - - -_ - - A
 
l                            VI.2 Jet and Aerosol Characterization Tests i                                          The jet and aerosol characterization test matrix given in Table 7 is developed to obtain the response of the system over the range of conditions considered possible in reactor accident scenarios.
The results will be used to develop and verify an behavior at largerofscales.
analytical                model        the jet behavior that can be used to predict                                            ,
VI.2.1                  Information Sought i                                        The purpose of the characterization test matrix is to pro-vide a correlation of the jet behavior to the initial conditions of pressure, temperature, and melt and gas composition. The results will be used to develop an analytical model to predict
;                          the jet behavior over the input range under consideration. The tests are also designed to determine the extent of aerosol gener-ation            during melt ej ection and the ability of these aerosols to transport fission products out of the melt.
3 i                        test matrix will determine the extent of aerosolThe characterization                  in the experi-ments and the type and quantity of fission products in the aero-sol, over the projected range of accident conditions.
jet and        Table          11 presents aerosol              the types of information sought from the characterization test matrix. The list is com-piled                by reviewing the variables involved in the analytical models describing melt removal mechanisms, containment, and the effect                                                        aerosol transport in features.                                                  of aerosols on engineered safety i
l TABLE 11                                                                        '
Information Sought from the SPIT Jet i
and Aerosol Characterization Test Matrix                                                          ,
variable Phenomena Affected                                                    j Gas Pressure in Melt Generator                                Gas solubility Jet velocity Gas velocity Gas Temperature in Melt                                        Gas solubility Generator                                            Gas velocity Melt Temperature                                                Gas solubility
'                                                                                                                                                      [
Melt density Incident heat flux                                                    ?
Gas Solubility
<                                                                                Aerosol source term, Dynamic j et configuration 74
 
TABLE 11 (Cont.)
Information Sought f rom the SPIT Jet and Aerosol Characterization Test Matrix Variable Phenomena Affected Jet Velocity Dynamic pressure Mass flow rate Debris configuration Incident heat flux Aerosol source term
* Penetration velocity Jet Configuration Incident heat flux Dynamic debris configuration
* Water interaction
* Fragmentation Incident Heat Flux Jet / concrete interaction Concrete decomposition / gas generation
* Jet / Water interaction Reaction Forces                      Mass discharge rate Jet density Gas Velocity Debris removal Aerosol transport
* Water purge Aerosol Composition, Size,          Fission product transport Mass, Concentration Aerosol source term
* Water pool scrubbing
* Applicable to situations involving a water pool.
VI.2.2    Instrumentation Melt-ej ection experiments involve a with time durations ranging from tens of seconds combination  of events (thermite  reac-tion) to submillisecond (flash x-ray exposure). This character-
: istic, stream,combined with the high temperature and pressure of the jet the required measurements. requires the use of a number of unique devices to perfor will be used and the paragraphsTable  12 summarizes below                the in describe them  devices detail.that 75 1
 
TABLE 12 INSTRUMENTATION FOR SPIT CilARACTERIZATION TESTS Measurement          Device                Range              Number-Location              Remark Gas Pressure        Pressure              2000 psi or        1 - gas line transducer            10,000 psi                                      Placed near accumulator 1 - vessel flange cover      Thermally Shielded Gas Temperature      Thermocouple          14000C            1 - Generator expansion      Shielded volume 1 - Gas Line Melt Temperature    Thermocouple          1400 C              6 - Crucible sidewall        Used to infer tempe rature via inverse heat-conductior analysis Jet Temperature      Pyrome te r            3300 C            2 - Exit aperature Focused on melt jet at aperature M  Jet Velocity        Framing Camera            -
3 - Outside chamber Jet Configuration    Flash x-ray              -
4 - Film cassettes near      70-nsec exposure time Framing Camera                                (1 m) apparatus 3 - Outside chamber          View obstructed by vapor and aerosol clouds Reaction Forces      Load cells            8896 Nt            3 - Supporting generator Incident heat flux  Calorimete r 1 - Target plane - directly under jet stream Multiple elements, Temperature data used to Ae rosol *                                              -
infer heat flux
* See Table 13
 
VI.2.2.1              Gas Pressure and Temperature i
A transducer generator (Figure 9)                    placed    in the top flange cover of the melt expansion volume above the melt. measures the pressure of the gas in the melt out of the vessel. device The record obtained from this gives a precise indication of the gas pressure f A second,      duplicate transducer is placed in the gas line near the accum is chosen for two reasons; one,ulator                              (Figure the device                  ll). This isolated is physically location from gaugethe      arehigh        temperature minimized        and two, source so that thermal effects on the vides diameter      an gas    indication line. of the flow restriction caused by the smallthe reco Appendix B contains an analysis that shows the gas line acts as tor.a low pass acoustic filter during blowdown of the melt genera-The results indicate that the accumulator volume does not rator for the period immediately following melt ejection. contr locations adjacentGas temperature is obtained by placing thermocouples in                                              I to the pressure transducers described above.
The thermocouple in the gas line is a standard type K with a 16 mm diameter stainless steel sheath. The hostile environment                                                  .
i existing            in the expansion chamber during the thermite reaction requires that the second thermocouple be within a 6 mm diameter stainless steel tube. shielded by placing it is sealed                and numerous 1-1/2 mm diameter holesThe end of the tube the    side of              the tube to permit gas flow to theare                                  drilledThe in sensor.
design caused is          byeffective        in mitigating molten thermite            particles.damage to the thermocouple combined with the small diameter of theThe                                    gasmass    holesofcauses the shield the shielded device. response of the device to be slower than the comparable,un-VI.2.2.2                  Melt Temperature i
tor because of the rapid evolution of very high temperatu the corrosive nature of the melt composition. Common thermo-                                                              .
.i
{!I          couple materials such as stainless steel cannot survive the high temperatures.
Sheath materials capable of resisting the chemical attack mechanically  are typically  during the    thermally          shock-sensitive and will f ail thermite reaction.
Limitations in access
'            the use of optical pyrometry.to the melt crucible and the presence of op The alternative to direct measure-
;          ment of theismelt. the use of some other technique to infer the temperature Thermite reactions have been observed to be highly agitated, d          probably th  e m el t because  pool. Thus,    of the rapid expansion of trapped gas through time is short relative to the time required for the melt to calnit is ass J
77
 
and develop large thermal gradients.
mixture tivity of          is the thensteel nearly isothermal. Because the thermal conduc-The tem thermite, inferring            themonitoring melt temperature. the crucible wall temperatureowmay al l
the crucible wall at several depths and at various elevatioThermocou                                                  ns, to ness of the steel between the melt and the therm quate to protect the sensor during the thermite reaction. The mass of the steel also causes the initial temperature transient
'.                                  imposed the                      on the steel to be delayed in time at the locati ons of thermocouples.
the cu          rve.sensor        location and reduces the slope of the e-time                                                temperaturT der with a time-varying heat flux imposed onandthe inne an adiabatic outer surface.
highly turbulent during the reaction, so that the heat transferIt i at the inner surface is dominated by conduction across the inter-face.
by the Applying        cylinder wall an energy gives: balance to the control volume defined in -
out * $ stored                          (IV-7) can be                  For  an adiabatic outer surface, $out = 0 and the equatio n written:
                                                      -k2nrL      =    r o - rt  L Ccp (VI-8)                                                                >
gives:Converting                          the differentials to a finite difference notation i
2k(T)      't      t'    '2 T          "        2'    " t+1          t' In (ro/ri)    .i~    0.    [0 ~ S    D 80      ~
o_
where:
k(T)
                                                                  = thermal temperaturconductivity e            of steel as a function of T
                                                                  = temperature of the shell r    = cylinder radius subscript o,i
                                                                  = outside, inside cylinder surfaces superscript t+1, t = time interval 9,  c p = density and specific heat of steel 78
 
The analysis considers the temperature at the inner surf ace of the liner is to be approximately the same as that of the melt and that melting of the steel does not occur. Placing thermo-couplesT(t) history      in the  steel shell at various        allows obtaining the temperature radial and axial locations.          These data can then be fit numerically to obtain Ti(t) to infer melt perature.                                                                tem-A numerical technique (Ref.18) tion (VI-9) by using a boundary condition of the form:is used to solve the equa-T(r o, t) = Y (t)          i = 1, 2 ,
i                      ... N where:
N = the number of thermocouples.
The above equation is solved numerically using finite control sizes and a fully implicit technique. When multiple thermocou-ples and/or f uture temperatures are considered, the heat flux that minimizes the following least-squares error. is determined:
n+4    n  -
E=
{ { T(ri, tj) pn i=1      '
Yi (tj)-
2 (VI-10) where:
n = the number of future times considered The output from the code is either the heat flux or tempera-ture at the inner surf ace of the crucible wall.
VI.2.2.3      Jet Stream Temperature The temperature of the jet stream and/or vapor cloud outside the vessel is difficult to measure because of the high tem-peratures involved and the rapidly changing nature of the stream.
Most direct contact thermometry techniques will not function at slowanticipated the                temperatures (>2500 K) and may be hampered by response times.        The response time requirement for the SPIT jet stream measurements can be established by considering the gage behavior to a step-change in temperature. Assuming an exponential form for the governing equation:
Tmeas                      -t'
                  =T act 1 - eXP T                        (VI-11) where:
Tmeas, Tact = measured and actual stream temperatures 79 m
 
t = point in time when T              is obtained meas T=
time constant of the device 0.1 T Thect (Itime  constant
                                & 0.9 isT act.
defined as the period corresponding to:
equation        T VI-3If"gives:                Substituting these values into 0 .1 Tact:    tl
                                          =  0.1 0.9 Tact:      t2  =    2.3 T  =  0.45  (t2 -t l) change Thein above T        analysis shows that in order to measure a step act, the time constant of the pyrometer must be less than half the risetime of the pulse to insure proper frequency response.
temperatIf all ure,the  melt passing the sensor location is at the same                      ,
time required to ej ect all the material.then the device            theneed  only period of respond in with  time within thesmaller.
be proportionally          stream will require that the time constantTemperatures varyin period is on the order of 50 ms.At the highest ej ection velocity, the total me the pyrometer should have response to changes in time intervalsFor illustr corresponding to 1% of the total ejection time:
t    =
0.01 (50 ms)        =
0.05 ms Thethis for    time    constantis and example          then: natural f requency (fn) (Ref.19) required T =    0.45 (t-0)    =
0.45 (5 x 19~4)  =  0.2 ms fn= 2nr      =
                                              =  144 bz 2 w(0.2 x 10-3) i has a 1 microsecondA commercial pyrometer
* selected for the SPIT experiments response time with a temperature range up to                f l
c
* Model 8000-1, Thermogage, Inc., Frostburg, MD.
80
 
33880C.      The device is limited, however, in that it assumes the source is radiating as a blackbody. The spectral emissivity of If-  the melt stream must be known to correct the measured result.
The    relationship between the measured and actual temperature measurement is given by:
Eb(T) actual "EA ( A , T) Eb(T              = c A ( A ,T)g T0 ms or meas      c x( A ,T)        act where:
Eb (T) = total emissive power o = Stephan-Boltzman constant T = absolute temperature ex ( A ,T) = spectral emissivity h
The spectral dependence of ex(A, T) for a molten thermite jet  must be known to evaluate the error associated with assuming a constant emissivity. If the pyrometer is sensitive to all wavelengths, then measured temperature is related to the true temperature by the expression:
1 Tmeas    = Tact (E A)                      (VI-12)                    l The uncertainty in the data is a function of the difference between the assumed and actual emissivity value.                      This is given by the expression:                                                                    ;
                                                  ~
Tact    = C (c)                        (VI-13) where:
rey \
C=    constant      =
g )
The error in temperature is then:                                              '
ETact
                    =
1
                        -          = -
1    IEact - cT}
Tact        4              4  . ET      .
(VI-14) 81 et
 
1 l                    where:
Ec = uncertainty in the emissivity value
!                                          Tact  = actual temperature i                                                                                                                                                      i I                                            cT = assumed emissivity c
i act = actual emissivity i                          Over the range of temperature 1500-3000 0 that 0.3 < c < 0.8,                                                                            K, high as 42%.                      the maximum temperature error (ET/T)andcan                                    assuming be as i
insure accurate temperature measurement.Thus a good estimate of the e 4
cated stream.          by the presence of the vapor cloud surrounding t i                tus,                If the device is placed at a distance f rom the appara-the view may be obscured by the vapor cloud accompanying the melt stream.
an unhindered                Placing observation  the device            of on thethe jetlower as itflange cover allows vessel.                                                                                        emerges from the vesselMeasuring                  the temperature of the jet as it emerges from the surem ents. should          provide data to correlate with the in-vessel mea-                                                      i radiation and convection to the environment may cause t ature to target        or be  cavitysignificantantly floor.                              reduced prior to impacting the a function dix    C.                  of various experimental conditions is given in Appen-The results of the analysis indicate that the energy loss via radiation and convection f rom the melt jet to the environment is insignificant ity-driven              m elt. for all but the limiting case of a purely gra v-aperature should                      Measuring the stream temperature at the exit then provide a good overall indication.
VI.2.2.4                Jet Velocity The velocity of the jet stream is important because it determines floor.        It alsothedetermines dynamic configuration          the            of the debris on the cavity occur.                                                                      time for aerosol evolution to mated ture.          The  byequation applying  is of the          Bernoulli's form:          equation across the vesse fP + 1/2PV2+              pY g    y      =
P + 1/2 PV2,pgy          2                  (VI-15)
(        where:
P = pressure P = density j
V = velocity Y = elevation g = acceleration of gravity 82
 
Subscripts 1 and 2 ref er to the inside and outside of the vessel, respectively If the change in elevation is neglected, and the density is assumed unchanged, the above equation reduces to:
1/2    -2P, - 1/2 V2 2(P1-P)  2                    1
      " Vjet "  ,
p p
(VI-16) where the velocity in the vessel is assumed to be negligible, and Pi is defined in gage pressure.
Using the two extremes of pressure (1.4 and 17.0 MPa) gives the f ollowing:
For 1.4 MPa:
            .                      1/2 6
2(1.4 x 10 )
Vjet"_          5900      _
For 17.0 MPa:
7 1/2 2(1.70 x 10 )
Vjet " .        5900 These results represent the upper and lower bound for the prescribed set of accident conditions.                  The ZPSS predicts that the calculated velocity at the breach remains unchanged until impact with the floor or water pool.
VI.2.2.4.1      Turbulent Jet Theory The analysis in the preceeding section does not consider the possibility of turbulent mixing of the jet and free-stream gas.
Relative movement between a liquid and a gas will induce tangential separation surfaces at the interface. Resulting in-stabilities in the surf ace cause exchange of matter, momentum, and thermal energy. As a result, a region of finite thickness with continuous distribution of velocity, temperature, and spe-cies concentration is formed along the boundary of the two mate-rials. The region is termed the turbulent jet boundary layer (Ref. 20).
A turbulent jet discharging into a stationary fluid is called a submerged jet.                  If the velocity field at the initial cross-section is unif orm, the boundaries of the mixing layers 83
_ _ _ _ _ <t
 
zform l e. diverging surfaces which intersect at the edge of the noz-The outside surface of all points,                              the boundary is defined where, at gation is zero. the velocity with        respect to the direction of propa-The inner boundary is given by a constant velo-city  core n oz zl e. that  eventually    disappears at some distance f rom the is a function of propagation distance.Beyond this point, the velocity everywh appearance of an idealized            submerged jet.
Figure 12 illustrates the For a submerged, axially-symmetric jet of a single consti-ness ofthe tuent,      thelength boundaryof the layerpolar (Y) (core) region (Xp) and the thick-are given by:
E o
c(.214)V2 (VI-17)                    i Y = cx      0.416 + 0.021
(                    r o  /
where:                                                                              :
Xp = length of polar region to = radius of nozzle c = empirical proportionality factor x = axial distance from nozzle for submerged liquid jetsThe            value of c is typically 0.2-0.3 with 0.22 common (Ref. 2 0) .
and    Considering, as an example, the calculation for 17.0 MPa, to = 12.7 mm, the length of the transition region is:
12.7 b " (0.22) 0.463                    "
eters.The    length of the core region is therefore roughly 10 diam-At this limiting value, the boundary thickness is:
                      ~
Y p= 0.22 (124.7) 0.416 +          0.021 (0.22124.7) 2 ~
2.7      _
                                                            = 45.6 m This result suggests that the turbulent jet expands several times its initial size within the transition region.                Beyond this 84
 
                                      ~
  .                                                                N O
I G
E
                              .,  ,-ll ;                          R 1      1 y
S S
E                          ~
N                                                  % O K                                                    I G
C                                                %
I                                                    E H                                              N    R T
                                  .}                  %
R E
Y        -  A.,- z:            -
                                                    'N A                                                :
L "E
N C
O Y
R A
D A(' c.=j;#                _
N Y
T I
C U
O B
G gw
* N O
I G
O              -                                E L                                                R E
V T
N A    L E
I
                      /cj_,,_-[N N
s T
I N
I L
A I
T    F S    O N    R                            N O
C P
Y T
I C
                          "7' ~
:      5 O    E                            W L    L                            \
E V
O N
Z Z
3_'                    W W
W E
G R
A
_        ~
W W
W Hl_
C S
I W
D    1'                      W
                                          '  l i                l. l lP
 
point, the centerline velocity (Um ) is given by the expression:
0.96 Un Um=      m                                        (VI-18) r l-                                            o where:
Ua
                              - initial velocity of jet a - constant such that axp/ro = 8.96 For the example above,
                                  ,,  _9 .96 (12.7)  =
124.7        0.998 Equation (VI-18) reduces to:
124.4 Uo U,  =
for x > 124.7 ret tion distance, for. positions beyond the initial region of theT flow.
diverge from its initialThese results indicate that the melt stream may rapidly The central,                        diameter upon entering the atmosphere.
distance beyond the exit aperature. constant velocity core will diminish within a short becomes a function of both axial and radial position.The velocity field then VI.2.2.4.2 Measurement Techniques If the melt stream behaves as a turbulent jet, measurements Such measurements    mustarebe complicated referenced to bythe thelocation of the aperaturethen velocity                                    .
cloud and a large radiant-heat energy accompaning the dischargerapidly-exp                                                      .
resolve the motion of high-velocity objects. Fast-framing motion pictur j et emanating f rom the melt generator,                      In the case of the quickly obscures visual access to all but the leading edge of thethe expanding va melt stream. It is not certain that the behavior of the remaining portions of the melt stream correlates with the leading surface.
configuration of the meltA flash x-ray technique is used primarily to obtain the stream. The flash x-ray technique sponding film cassettes. typically uses four separate heads in conj unction with corre                                          -
The firing of each unit is accomplished 86
 
by individual trigger delay units referenced to a common signal such as a probe under the generator. The pulse width of the exposure is on the order of 70 ns, adequate to "f reeze" the 1
motion of even the highest-velocity j et. Velocity can be found by obtaining the propagation distance on two consecutive x-ray image photographs and dividing the resultant by the difference in the delay times for the two units.
The flash x-ray method is most applicable to observing a discrete        portion of the melt, such as the leading edge of the jet.        Distinct geometric features, later in time, rapidly evolve
!      so that the resolution becomes increasingly more difficult to obtain.
l The technique also suffers in that only average velocity l      over the interval is obtained.                                If the jet is de-accelerating rapidly as turbulant j et theory predicts, then time-resolved I
velocity would be much more appropriate than average velocity.
VI.2.2.5          Jet Configuration
    '        The dynamic jet configuration is obtained by penetrating the melt vapor and aerosol cloud that accompanies the ejection pro-cess. As described above, flash x-ray generators and film cas-settes can be used for this purpose.
The resolution of an x-ray film image is based upon the geometric and film unsharpness.                                The formcr is caused by slight dislocation of the x-ray field near the edge of an object, such as the melt boundary.                            The latter effect is caused by movement of the obj ect during the exposure period and the "graniness" of the f ilm.
The geometric unsharpness, X12, is given in Reference 21 as:
                                                    /D 1)
                                              = F1      1 X12                                                    (VI-19)
                                                    \Do/
where:
F = X-ray source spot-size                                                        '
Dy = object to film distance Do = source to object distance For typical experiments:
f (0.559)
X12 = 1 x 10-3                          = 3.1 x 10-4 m 87 CL
 
The f.ilm employed in the tests is very fine grained so that the film movement  unsharpness of the melt.factor          is dominated by the expression for highest predicted velocity:            The      j et movement is greatest for the
                                            =    Vt x X13                                      (VI-20) where:
V    =
jet velocity tx    =
exposure time (x-ray pulse width)
X13 =      75.9 (70 x 10-9) = 5 . 3 x 10 -6        m The combined resolution of the x-ray image is then given by:
t    '
Xy1  =        X      2  +
f I2  V2 =                        i i
i 12>l        l  X13          3.1 x 10-4 m
(      /
The analysis suggests that geometric f eatures of the melt jet less Thus,    than 0.3 mm will not be detectable on the x-ray images detectable.small particles within the vapor cloud will not be the The calculation also shows that the short pulse width  of jet stream.the    x-ray  exposure is very ef f ective in "f reezing" the VI.2.2.6    Reaction Forces Expulsion of the melt f rom the vessel will cause a reaction force in the direction opposite the jet stream propagation. This force is monitored by three compression / tension load cells lo-cated ing    between plate.        the chamber structure and the melt generator mount-A simple f ree-body analysis shows that each cell is measuring one-third the diff erence between the weight of the vessel and vessel.        the thrust produced by the expulsion of mass f rom the l        The thrust momentum    to be:  on the vessel can be derived f rom conservation of du                  2 l                FT  = V              "  PA22V 2 &                                    (VI-21) where:
FT    =  thrust
                    =
V2        exit velocity
                    =
A2        Area of aperture l                                                                                          l 88
 
Knowing the quantities FT ' A,2 and V2 permits the above relationship to jet stream as              be used to infer that the density of the molten applied at the aperture, it exits the vessel. Because the equation is occurs,                                before significant expansion of the jet the    thrust    on  the density of the melt in the stream. vessel    should accurately determine the The duration of the thrust forces is roughly equivalent to the period of time required to discharge the melt mass. Some force will be developed during gas discharge following the melt, but the less  than gas  thedensity melt. (and        the f orce) is approximately 150 times be estimated by assuming        The time  a required to expel 10 kg of melt can strictly correct, because onlyconstant                    mass flow the volumetric    fl ow (this is not rate is con-stant for choked flow) through the aperture.
t                        t E          =
final  .
mdt    =    A      final tetal                                            V(t)dt    (VI-22) 0                        0 where:
m total = ejected mass during the interval 0 to t in = mass flow rate t = time bulk melt density:For the highest pressure case, assuming constant velocity an t
final mtotal t          =
dt  =
final                              p                    (VI-23) 0
              ,                    10 kg 5900 kg/M (5.12 x 10-4) (75.9)
              = 4.36 x 10-2 sec The2)thrust Eq. (VI-2      .      can be calculated f or the same conditions using FT=V          =
Vin = 75.9 (m/sec) 2.19 x 102l FT = 1.66 x 104 Nt          =  3737 lb 89
 
Each cell will measure approximately one-third of the quantity or about minus    the Nt.
4.8 x 103    weight of the pressure vessel (2.2.x 10gbove              i Nt)
VI.2.2.7    Heat Plux Calorimeter th e mInelthe  high pressure t ejected          accident f rom the  vesselsequence proposed in the ZPSS, concrete basemat.                          impinges directly upon the              i concrete is related to the extentThe heat flux transferred f rom the jet to the amount of gas released.                      of erosion and, hence, the              i The  resulting  decomposition of the basemat and the quantity of gas released will affect the subse-l quent behavior of melt in the cavity.
prevent    debris configurations posed mechanisms.        Likewise, susceptable to removal by the pro-Extensive be hi tion.ghly    disruptive causing a very agitated, gas emerging f rom a molten pool The response of the concrete to the jet stream can then beunstable configura an  important of the            parameter in decermining the ultimate configuration core debris.
The incident heat flux will be measured during the SPIT tests using a slug-type        heat flux calorimeter of the type shown in Figure 13.
in a graphite body so that the f ront surf ace of each element isThe exposed.
A close fit simulate body to accurately          is maintained  between thebody.
a semi-infinite    elements and the A graphite f ace plate is attached to the body so that the f ront surf ace of each plate element is flush with the f ace plate. The sacrificial face each test. and elements undergo extensive damage and are replaced for Each element is instrumented with three Type-K thermocouples constructed radially-drilled  of AWG-3 holes.2 gauge  wire (0.18-mm dia) inserted into The thermocouples are b element with a high-temperature graphite coating,gnded                into the designed  to insure    good of the element    thermal  contact  between  the thermocouple  and  the body fine-grain ATJ,, graphite withAll of the graphite parts are f abricated f rom to    the  heat  flow.                  the grain oriented perpendicular sionalize the heatThe body and f ace plate serve to one-dimen-simplify the analysis.transf er patterns through the elements to Placing the thermocouples at three depths relative to the exposed surface gives the temperature distribution in the ele-ment as a function of depth and time. Multiple elements provide redundancy in the data, and allow determination of spatial varia-tions in the incident heat flux across the face of the
, Graphi-Bond 551 (TM) , Aremco Products, Inc., Ossining, NY.
(TM) Graphite Products Division, Union Carbide Corporation, Cleveland, OH.
90
 
l=!!!$!!
                                                                                  !            3
                                                                          <        !;;;=..:;-  Ar .
O bk.                  .
C      j;;;;;= :!:N!!
GRAPHITE          GRAPHITE                                  -
i!$        S..,.
FACE PLATE      CALORIMETER                  '
IEEE        --"-
(1 OF 4)                        -
iii      ?I f
4          15.g~~j.:' ;-
2.54 dia. P                                                                                Y.
3 s
                                        \
q                L    7 1
g                X                                          STEEL MOUNTING
                                '\                  '
PLATE
                        \
                    ;                          /                                              GRAPHITE BODY
{k              5.08 THERMOCOUPLE PLACEMENT Figure 13.
Slug-Type Heat Flux Calorimeter (Dimensions in cm) h                                .,. .                      ..
 
calorimeter.
recorder with a Thermocouple frequency response response is recorded on a FM tape subsequently stored on a floppy disk for analysis. The data are of 40 khz.
Analysis of the f requency response of the embeded sensor and an  estimate of the error in the recorded temperature values is given in Appendix D.
The method of determining incident heat flux from tempera-ture distribution within a body involves the solution of                  an inverse heat conduction problem to derive the boundary condition at the f ront surf ace of the element. Assuming the calorimeter              i body has a one-dimensional temperature distribution in the indi-vidual elements, the heat transfer can be represented as heat-flow in a semi-infinite slab. If the f ront surf ace is exposed to constantand location    heattime flux,isthen giventheby:temperature within the body at any (VI-24) 1/2 I -x2    i          /
T(x,t) - T                                    Ox          x      )
o = (20  k(T)    a(T)t/n) **k(T)  ~
k 4a(T)t k 2[a(T)t]I/2 where:
erf = Gaussian error function erfc (z) = 1 - erf(z)
Q = incident heat flux To = initial temperature of element a(T) = thermal dif fusivity = k (T)/pc p k(T) = thermal conductivity p = density c p = specific heat x = location from exposed face This equation can be solved if the heat flux is constant for the duration of the event and the functional relationship of the material properties with temperature is known. A solution is obtained numerically at discrete intervals in time because the incident heat flux f rom the impinging jet is not expected to be constant. The analysis uses an approach similar to that proposed by Beck (Ref.
: 22) by applying a least-squares method to damp the inherent instability of the numerical solution of the problem.
92
 
The quantity:
S2  =
T    -T
                                                $              (VI-25) i=1    j=1 represents the error between the observed T[      and calculated (Tjac temperatures where N denotes the number of embedded ther-mocouples ogtain  a      and Rheat useful      is the  number of future temperatures employed to flux.
S  term in Equation (VI-25) The              technique attempts to reduce the with respect to the calculated heat flux.
The calculated temperatures are found using a modified                            i Crank-Nicholson numerical routine that accounts for temperature-dependent thermophysical properties (Ref. 23). The procedure is applied piecewise over the data assuming the heat flux values to be constant over each specified time interval.
For the temperature range of interest (<l5000 C), the data of Ref. 24 were fit to obtain an analytical expression for k(T) of the form:
k(T) = 0.07 918 + 67 0.85/(T + 126.718)
                      - 0.5199 (T/100)-0.7 877 (cal /sec-cm-K)              (VI-26)
Over the same range, the density and specific heat are given 0 = 1.73 (g/cc) c p = 0 . 305 8 + 0 .15 8 9 x 10 -3T
              - 0.03107 x 10-6T 2 - 0.1684 x 105/T2 (cal /g)
Figure 14 gives the comparison between the values versus those recommended in Ref. 24. The greatest          calculated k(T) scatter occurs in the temperature interval above 2000K (0 .1                        to 0.06
                                                                                      ~
cal /cm-sec-K) extrapolated        where from  lowerthe    observed values temperature    data. have been estimated or The number of future temperatures (R) to be used in Equation                      I (VI-25) is determined by comparing the time associated with the                          ;
maximum change      temperature occurs.
in temperature        in the element to the time when the maximum              3 93 rL
 
5 3
0 4
2 l
F                                    0 E                                      3 R                              '
0                _.
M                                                        _
O R
F l                              S                                    5                e T                              '      2              i t
P                                    0 h
p A                                          )
a r
T                                                  G A                                          K -
J D                                            s-    T A
l 0      m      f
                                                                                '      2
                                                                                              /
c      o 0      l a      t y
(
c      i i
v t
D        c u
E      d I                                                                      5      V        n o
                                                                              '      1 R      C 0      E      l S      a B      m r
O      h e
K      T I
0                .
1
                                                                                        .          4 1
0 e
r u
g i
F I                                                                        5 0
                                                                    //
I 0
                                                                        's
                                              -          -    -    -              0 0                                                            0 5        0          5
* 3            2        2          1 0
1 5
0 0O 0            0        0                              0 0    0        O 2 4 Eoh $ owH' b<
e*
illll          111    111  l  l    11l          {'
 
Maximum Temperature:
                            =
x2 t1 g                        (VI-28)
Maximum Change in Temperature:
x2                t 1
t2
* 5 (1 - 2/3)
                                    =
p              (VI-29)
Using an average value f or a , the time intervals f or the three thermocouple locations used in the calorimeter elements are listed below.
Maximum              Maximum Depth        Temperature              Change xi =    5 mm      t1=    14.6 ms      t2=    4.9 ms x 2 = 10 mm      t1=    58.6        t2 = 19.5 x3 = 15 mm        t1 = 131.8          t2"    44 Based on these results, the calorimeter elements respond to the imposed flux in times comparable to the discharge of the genera-tor. Multidimensional effects such as the cooling f rom the sides of the block and the presence of a non-infinite dimension in the direction of propagation are theref ore minimized.            The results obtained f rom these data will be used to compare to those pro-posed in the ZPSS and to assess the attack on concrete by melt streams.
i VI.2.2.8    Jet Aerosol Instrumentation l
The SPIT free jet aerosol characterization tests provide an empirical basis for developing a model of aerosol generation by l melt ejection in the accident scenario. The size distribution l
and mass concentration measurements of aerosols generated by a l f ree jet or the debris exiting a scaled test article provide a i
basis to calculate the amount of aerosol material leaving the scale model's keyway (penetration).        It may provide the means to i back out the Stokes Number dependence of penetration to be ap-
; plied to the accident case.
Doping the melt with appropriate fission product mocks and analyzing the sampled aerosol for those mocks provides a basis f or assessing fission product aerosolization and transport to containment.
95
________ __.__ _ m
 
w Thechamber a closed    SPIT aerosol        characterization (approximately    43 m3 ).tests will be conducted in          i The controlled volume permits the assumption of an instantaneous source of well-mixed aerosol.
This assumption is facilitated by the less than 100-                        ,
bute the airborne material. msec ejection time and the use of a mixing f an to eve intervals will provide distribution decay information. SampleA collectionsummary over var of f romtheeach devices is given to beinused Tableand
: 13. the types of information obtained test. Six Anderson Mark III cascade impactors will be used on each These cascade as a f unction                    impactors of aerodynamic        give a particle mass distribution diameter.      The pre-separator col-lects particles above approximately le- micrometer aerodynamic diameter and will collect a relatively large mass of material to                        ;
prevent overloading the subsequent impactor stages. Four impac-                        ;
tors will over          be used sample    times in  of stationary positions sampling at 14 lit / min 0-1, 1-3, 3-7 and 15-20 minutes.
Two additional impactors will be placed on a rotary device                    !
designed to sample isokinetically, by matching the impactor's airstream moves.            velocity to the velocity of the inlet as the impactor The net effect is that large particles are obtained                        ,
without paths          having (Ref. 25). to significantly divert them f rom their natural I
the first two stationary impactor samples.These                    samples will Comparison      be taken con of the stationary of    sam      and rotary impactor samples, together with calculation theory,pling efficiency for the stationary impactors f rom sampling will provide information on sampling efficiency and the airborne size distribution.
provide size distributions, infThese                    instruments will directly ormation    on each of the modes, total concentration, and total mass aerosolized.
indirectly give the elemental distribution as a function                The data will of size when coupled with data f rom cascade cyclones.
Two Sierra These instrumentscascade            consist cyclones    will be used on each test.
of a series of cyclones to give an              !
aerosol as does mass    the cascade distribution  cs a function of aerodynamic diameter, impactor. The advantage of the cyclone is that larger amounts of material can be collected in each size class (each cyclone) for chemical analysis. One cascade cyclone i
will be operated f rom time 0-3 minutes.
erated for a longer duration to assure that enough material isThe other will be op-i          collected          for the analysis.
same information as the impactors.The first cyclone will provide the Size distribution information from the second cyclone will be distorted by the different decay times the            for concentration test.                        in each size class over the duration of j
i        second cyclone        The elemental composition of each size class from the 3
impactors and the other cyclone, coupled with the size distributions f rom          the i        sition distribution as a function of aerodynamic diameter.will give the elemental; co            ,
i                                                          96
 
TABLE 13 Aerosol Measurements and Instrumentation Instrument                  Measurement Filter Samples                  1,2,5 Photometer                        1 Cascade Impactor                  1,2,3,4,6 Cascade Cyclone                  1,2,3,4,5,6 Deposition Samples                125 Measurements:
: 1. Total aerosol mass concentration
: 2. Total fraction of melt aerosolized
: 3.      Size distribution of aerosol
: 4.      Mass, mean size, and spread of each mode
: 5.      Elemental composition of aerosol
: 6.      Size-dependent elemental composition Six Millipore 47-mm filter holders will be used on each test to collect  total mass samples for elemental analysis. The sample times of 0-1/2, 1/2-1, 1-2, 2-3, 3-7, and 15-20 minutes give time resolution of mass concentration.          They are taken concur-rently with other samples to permit inter-instrument comparisons.            3 A photometer is a light penetration device that continuously monitors tube.        the intensity of a light beam across an aerosol sampling It is strictly an empirical measurem ent device with in-          1 Jiitu calibration sured aerosol. Six    accomplished by f11ter collection of the mea-filters, operated concurrently with the six        i filter samples previously mentioned, are used f or calibration.
The device is placed in a large diameter (15 cm) duct utilizing a            t fan to draw in the airborne material. The duct serves to straighten the flow into the photometer so that the collected material 3
tion.
is representative of the entire aerosol size distribu-The photometer is used principally to give a continuous          ',
aerosol concentration measurement as a function of time.
Deposition samplers placed on the chamber floor and walls are used to give total and time resolved deposition.          The total deposition sample collection will be used to calculate            total 97
                                                                        -_m
 
,                              aerosolized rial.                                      mass and elemental composition of aerosolized mate-
'                              assess The                sampling          results  will be compared to the filter samples to efficiency.
exposes deposition surfacesThe                                                      at diff          time-resolved de erent times. position sampler are available with the sampling times: 0-1/ 2, 1/Five                                                                                                                            surfaces 2-1, 1-2, 2-3 minutes, and 3 minutes to test end.
:                                                                                                                                                      The time-resolved deposition used for elemental analysis. samples enhance the concentration decay inf VI.2.2.9
(
Pission Product Mocks and Elemental Analysis i
isotopes        Doping the melt with fission product mocks (nonradioactive purposes: or chemically similar simulants) serves the following 1.
j                                                              Correlating the concentration of fission products in t
the aerosol will determine if fission product concen-
!                                                              tration is enhanced or retarded in the aerosol,                                                                                                    and the particle sizes associated with the various fission products give mechanisms.                an indication of the aerosol formation
!                                          2.              The size and species i
;                                                            establishes a data base                                                    distribution for empirical estimates and of the aerosol i                                                            mechanistic the                      modeling melt ejection                process.              of fission product release f rom i
aerosol transport behavior, the                                                                            When used with keyway distribution yields a radiological particle size            source                  term                      to          the                containment,                          as well as information necessary to calculate residence time and transport.
j product          The      mocks  chemical              analysis can also  be used        techniques  t mea                                          used to me suce f                                    i composition of the aerosol.                                                                                                                                                          de    bu Po                                                          ""                              d (s important in assessing the potential of t e aeroso1 to act as a hydrogen                                                                      tal in theignition                                    aerosol      and the source.
Ideally, sented the          number            in the      of      melt product fission  and would                      mocks                    bethat              analyzed in the aerosol, how limited.                                                                                                                                                        can be included is If the amount of non-reactive material in a thermite mix exceeds 1 w/o, it will tend to retard the reaction, slowing 1                    the reaction rate Detectability                                  limitsand  (seelimiting Table the                16)maximum          dictate thatmelt temperature                                                      .
of material are required as dopants. The small minimum amounts amount of aero-techniques and, sol material typically collected limits the available analytical
;                    analyzed.                                              for some cases the number of elements that can be been done by taking the total dose ranking at 1 and le hou t
Ref. 26 and arbitrarily selecting the top 11 species f rom each t
)                                                                                                        98
 
time f rame. This procedure provides the field (see Table 14) from  which  to  select the important fission products for this type of accident sequence.
TABLE 14 Importance Ranking of Fission Products (Total Dose)
Rank              1 Hour                  10 Hours 1                    I                        I 2                  Np 3
Np Te                      Te 4                  Tc                      Mo 5                    Mo 6
Tc Sr                      Ce 7                    Ba                      Nb 8                    Cs                      Ba 9                    Y 10 Sr La                    Y or La 11                    Ru                      Ru Table 15 gives the simulant selection f or the fission pro-ducts obtained from Table 14.
because they are volatilized early    Iodineinand thecesium  are not included accident. Technetium (Tc) and Neptunium (Np) have no nonradioactive isotopes and are simulated by Manganese (Mn) and Cerium (Ce), respectively.
TABLE 15 Fission Product Simulants Fission Product                        Simulant Np                                Ce Te                                Te Tc                                Mn Mo                                                          i Mo Sr                                Sr 3
Ba                                Ba                        -
Y                            La or Y
'                  La                            La or Y Ce                                Ce Nb                                Nb 3                  Ru                                Ru 99 A
 
1 Identifying the fission products and their simulants allows                                  !
i              determining the analytical techniques to be employed. The three fluorescence (XRF), techniques considered for the SPIT aerosol analysis are coupled atomic absorptionneutron      activation (NA), and inductively-(ICP).
i
  ;          nondestructive elemental analysis method    The first  of these which      can be(XRF) usedisfor a j          aerosols    deposited particles (>10            on a filter or collected in bulk. Larger micrometer) i adverse affects on sensitivity and precision.in the distribution have potentially s am pl e, For the filter
!                      detection of square centimeter      limits filterare  on the surface      order of 1 microgram per area.
l 4
tion on the filter gives lower sensitivities. Lower mass concentra-i i
Neutron analysis        activation method.            is an additional      nondestructive      elemental ments of interest, Itsuch is limited    by  low    sensitivity    to  some      ele-i 4                                        as Te and A1.
Inductivelyanalysis tive elemental      coupledmethod.
plasma atomic absorption is a destruc-the sample material is consumed for each element analyzed.Its                Typi-primary lim cally,100 mg of sample dissolved in 10 ml of solvent provides enough solution for analysis of 3 to 5 elements. There are also
;            interference Z r.          problems among elements such as Mn interfering with i
the three latives  beinganalytical considered. techniques,              for the fission product simu-4 VI.2.3      Factorial Design Approach performed cal models.          to generate a data base for the developmen i                            The models can then be used to predict jet and aerocol to larger    behavior scale. for other system conditions and extrapolations The number of variables involved in the melt ejection and aerosol generation processes are large, complicating the correlation of input conditions to output r es ul ts. The f ollowing paragraphs discuss the methods employed to develop systematic                                                                                      a approach to obtaining the required test outcomes.
tical" approach is to identify the physicalThe first step in developing affect the process.                                                                                      ,
i From these mechanisms, mechanisms that may
:      f actors can be listed and their range of influence evaluated.the experimental I
Those variables that have a significant influence are carried
;      forward in the test matrix, while the less important are held constant.
mechanisms of the melt ejection process.The Phase I' SPIT tests served                              ,
j The SPIT are pressure,          test gas    results melt solubility,    determined temperature  that and the dominant the elapsedvariables I      time from ejection to contact with a structure or water pool.
199
 
TABLE 16 Aerosol Analysis Tcchniques Method (Sample Size)
XRF(I) 100-mg Sample on                                                      ICP Neutron Activation            100-mg Sample Dissolved 47-mm Filter                  200-mg Sample in 10-m1 Solvent Limit          Precision    Limit        Precision (2)      Limit          Precision Element          (ppm)            (%)      (ppm)            (%)              (ppm)              (%)
Ba
* 100          15                10 i 10 Sr            100          1 25          100        15                  10 La
* i 10 10        15                50            1 10 Y            100            1 25              t                            10 Ce
* 1 10 1
15                  50            i 10 Te
* t Mn
* 100            _+ 10 tt                            10 Ru 1 10 100          1 25              t                            50          + 10 Mo            100          1 25              1 15                10(3)            10 Np                        *                    -
Nb            100            1 25                                          10 Nd
* i 10 6
15              100 Zr            100            1 25          100 i5                10          + 10 Fe
* 100          15                10( )        i 10 Al                      **
tt                            50          i 10 U              100            1 25                                        100          i 10 o Particle-size distribution may adversely affect sensitivity.
Co Al not easily detected.
t Not sensitive to this element.                                                                          )
tt Involved procedure or Sanida Laboratories lacks facilities.
(1.) Particle size distribution affects sensitivity and precision.
percent is assumed.                                                  Precision of i 25 relative      -
(2.) Precision of i 50% at detection limits.                                                                c (3.) May have problems with interference.
(4.) Interference from Mn, Ga, Rh, and Tm.
101 m
 
Melt mass can also be considered, variables (test outcomes).of the SPIT apparatus cannot significa provide an eight-fold increase in the mass.The 1/19th scale test matrix does the                                                      Table 17 identifies thatrange  of each of these variables and the dependent they influence.                                                                  variables TABLE 17 Independent and Dependent Test Variables Independent Variable                  Range                          Influenced Dependent Variable                              i (Responses)
Pressure              1.4 to 17.9 MPa                Jet velocity Incident heat flux Gas Solubility          Low, High                      Aerosol generation Jet disruption
    . Temperature            15000 C to 39990C              Heat flux Aerosol generation Jet height              9 to 4.1 m                      Incident Heat Flux Jet configuration Aerosol generation The pressure and measured.          used for the experiments is easily controlled in the initial value induced during the thermite burn.The                                Gas principal diffi solu-bility is~ related to the types of materials involved, the pres-sure, and to a lesser extent, temperature (see Section III.2).
Establishing a value of gas solubility for each test is therefore complicated      by the f actors involved and by the uncertainty of their interrelationship.
crete values of gas solubility will be employed.For the                    Phase II SPIT tests, two Carbon dioxide represents a lower bound on the conditions expected for proto-typic situations (i.e., steam and hydrogen over molten corium)
Selecting' the expected  nitrogen  as the second gas provides an upper bound on behavior.
The
* temperatures achieved during the thermite reactions ap-pear to be a function of the constituents involved (particularly contaminants),
resides                pressure and the length of time the material
              ~in the vessel.
indicate that reducing the porosity by tampingThe                theobservations powder bed during the causes less heat  improved    heat transfer and hence, a faster burn rate and loss. Thus, f or highly-tamped charges containing only the stoichiometricthe high 192
 
material composition.        Conversely, the melt temperature can be lowered by adding iron powder to the thermite charge to cause energy to be absorbed in heating the non-reacting materials.
The aerosol generation appears to be a function, in part, of the length of time the jet is exposed to ambient atmospheric conditions.
The influence of this ef fect will be studied by experiments with widely varying distances from the exit aperature to the incident surface.
in jet velocity will provide a large spread in the totalThis variation, combined wi time of jet propagation.
allowsTheaexistence      of four f actors f actorial strategy            at two levels (low and high)
(Ref. 27)  to be implemented. The range of each of the independent variables defines its response surf ace i.e. the geometric plane over which all other variables are held constant.        The intersection of all the response surf aces represents the factor space, or all possible combinations of the f actors variables. For three independent factors, the response volume cube." is a cube, for four or more factors the shape is a " hyper-The intersections at the corners of the geometric volume represent their extremes.design points or conditions where the f actors are at This design pattern is called a "two-level" f actorial -- f our f actors at two levels each.
The two-level f actorial can determine the " main ef f ect" of each f actor, plus the test rial  " interactions" matrix would of the  factors require  24in combination. A full facto-tests (16 total) to yield both main effects and interactions of the f actors in combination.
Main effects are defined by considering the response of the system for factors X2 , two      diff erent X3 , &X          values of f actor X2 , while holding constant.
4                The estimate of the ef fect of X at these two points can then be considered with other estimates1 obtained at other values of X2, X3, & X4 The main ef f ect of X i is then the average of all the estimates of the ef fects of Xt.
In the same manner, the effect of X 2, X ,3 and X4 can be obtained.
The procedure above requires the use of all data to obtain an estimate of the main effect of a particular variable. This is called the " hidden replication" in a f actorial experiment, be-upon a each cause            f actor difference      ef fectaverages.
between    and each interaction ef f ect is based The improved precision of the factorial approach due to hidden (PR):
replication can be estimated from the precision ratio
                      =
oFE      2 FR          =
a          2              (VI-30) 103 1
 
where:
i U
FE = standard deviation of a factor effect                              i o = standard deviation of a single observation n = total number of observations = K2P K - number of replicates i
P - number of factors This expression applies to all balanced two-level factorial experiments; both multiple replicates (K = 1,2,3..) and balanced f ractional replicates (K = 1/ 2, 1/ 4 , 1/ 8, . . . ) .
The PR depends only on the value of n, whether it is r e pli ca te s.through the number of factors (P) or the number of achieved three parameters      Thethat PR canrequired    in a given experiment depends on be specified.
1.
The parameter a specifies the confidence level (1 - a) that will be used f or the t-statistic in testing the data for minimum significant factor effects.
: 2.      The parameter o is the size of the factor effect it is desired to detect.
3.
The parameter (1 - 8) is the desired probability of concluding that a factor has a significant effect when it has a true ef fect of size 6.
The probability level (1 - a) establishes how close the average value is to the real value, within a certain confidence.
givenf in The        actor Tableof ten
: 18. used f or this purpose is the "t-distribution" factor effects are statistically The t-factor can    be used to determine which significant.
tion of Por    thepoints) data      experiment and P  under
                                              = 4 consideration,    K = 1 (no replica-perature, jet height). Thus,                      (pressure, gas solubility, tem-l PR    =              =        =  0.50 (16).1/2        4 Perf orming the precision          ratioone    replication at any data point would improve markedly:
PR  =                  =  0.35 (2    24 [
104
 
TABLE  18 Probability Points if the t-Distribution Double-Sided Test df
        ~~
P
                  .995          .99        .95      .90 1    127 2
63.7        12.7      6.31 14.1        9.92        4.30 3    7.45        5.84                  2.92 4    5.60                    3.18      2.35 4.60        2.78      2.13 5      4.77        4.03        2.57      2.01 6    4.32        3.71        2.45 7    4.03        3.50                  1.94 8    3.83                    2.36      1.89 3.36        2.36      1.89 9    3.69        3.25        2.26 10      3.58        3.17                  1.83 2.23      1.81 11      3.50 12 3.11        2.20      1.80 3.43        3.05        2.18 13      3.37        3.01                  1.78 14      3.33                    2.16      1.77 15 2.98        2.14      1.76 3.29        2.95        2.13      1.75 16      3.25        2.92 17      3.22                    2.12      1.75 18 2.90        2.11      1.74 3.20        2.88        2.10 19      3.17        2.86                  1.73 f    20                                2,09      1.73 3.15        2.85        2.09      1.72 21      3.14        2.83 I    22      3.12                    2.08      1.72 2.82        2.07      1.72 23      3.10        2.81                          '
24      3.09                    2.07      1.71 25 2.80        2.06      1.71 3.08        2.79        2.06      1.71 26        3.07        2.78 27        3.06                    2.06      1.71 2.77        2.05      1.70 28      3.05        2.76 29      3.04                    2.05      1.70 2.76        2.05      1.70 30      3.03        2.75        2.04      1.70 40        2.97        2.70 60        2.91 2.02      1.68 120 2.66        2.00      1.67 2.86        2.62        1.98
    =
2.81                              1.66 2.58        1.96      1,64 105
 
Defining the response of the system using only the extreme              !
values  of a between the system    particularthe variable    does not indicate the behavior of two values.                                      j Obtaining the response for      i values intermediate to the two extremes allows estimating the deviation for          f rom linear (curvature) or obtaining a functional form the response.
i l
t In order to estimate the curvature of each response indivi-              !
didually  would require an increase in the number of tests by 59 to 199 percent.                                                                i ly by performingAn        estimate tests at theof middle curvature  can be points  ofmade  economical-all factors. The severity  of  curvature between the average of      can  then    be  estimated  by  the  difference and the average of the center  the design    points (corner intersections) points.
vere,  linear  model    assumptions      will    If be only  theaccurate curvature  nearis se-the corners of the cube.
Factor compare        effects can be calculated from response data and if th'      to  the " minimum significant factor effect" to determine can r      etor  is important to the response. The curvature effect
                  .culated and compared to the " minimum significant curva-ture emc.t" in a similar manner. The formulas for the minimum significant factor effect (MIN) and minimum significant curvature effect (MINC) are as follows:
MIN = T'(s) (2/mK)l/2                    (VI-31)
MINC = T' (s) (1/mK + 1/C)1/2            (VI-3 2) where:
T' = level t-distribution and degreestatistic of freedom forinthe the desired estimate probability s = pooled tion    standard deviation of a single response observa-m = 2P-1 where p is the number of degrees of freedom K = number of replicates of each trial C = number of center points Comparing the individual the relative importance of that factor. factor ef fects to MIN establishes ity (1 - 8) 1 9.95 is used.                        Typically, the probabil-value) system.
are the most important,The          largest factors (in absolute t elative to the response of the than MINC,Similarly, if the computed curvature effect is larger associated then with at it. least one variable has a non-zero curvature 106
 
I More sophisticated analyses recognize the need to consider the  linear, interaction the independent              and curvature effects with respect to all variables.
To estimate curvature, a full three-level factorial is required to. provide orthogonal estimates. For minimum of 64 tests.four independent variables, a three-level f actorial requires total A  f ractional number          factorial of tests      test matrixeight by performing  can be  used tests  withtoparameter reduce the vectors comprised of high and low values (corner points) and two replicants of a vector of intermediate values (center point).                By reducing the number of tests by nearly a factor of two, the results    will not be suitable to test for significant three-factor interactions.
characteristic of fractional factorial designs and isSome
      " conf ounding." Confounded factor effect conf usion called where the two interactions are defined identically.are termed  This " aliases" method gives    a  main  effects response surface of the form:test program  which permits    the  fitting  of a L
R  =
bi xi + bo                (VI-33) i=1 where:
R = response such as jet velocity, aerosol size,        etc.
th xi = i      variable bi = the coefficient of the i th      variable bo = intercept The choice of the f ractional factorial matrix is dictated by a  compromise quantity of resources between maximizing information and reducing the required for the matrix.
all 16 corner points and two replicants of the center point, Eighteen tests; would be required to fit the interaction terms xix4 where i =                j.
It should deemed          be noted that the additional 8 tests mty be run, necessary.                                                        if Pe rf orm ing the tests and analyzing the significant factor effects will aid in the develoresults for the to describe the jet and aerosol characteristics. pment            of a model The models con 1
identified in the error analysis of the data.then be applied over the en VI.3 Jet / Water Interaction Tests 4
A jet / water interaction is a creditable possibility during a reactor accident because of the presence of water in the reactor 107
 
cavity due to mitigation efforts or natural plant response. Only a(Ref.
limited 2 8) , amount of experimental data exist f or this regime mostly for low temperature molten materials.                      ,
The SPIT jet / water interaction tests are directed towards providing the design and instrumentation r equirements for ex-                {
periments involving a high pressure melt ej ected into a scaled reactor cavity containing a water pool. The data will also be                      ,
i used to verify the ZPSS hypotheses concerning jet / water interac-tions.
f rom a dry cavity to one that is f ully filled with water.The            The ZPSS extent and efforts  of water the    in the cavity natural  plant  is dependent on the mitigation response.
possibilities in the SPIT                                  To address all the matrix is beyond the scope of the program.
The test strategy is simplified by assuming that the behavior of all jet / water interactions can be identified with one of two catagories: 1) f ully-filled cavities or 2 partial water pool.
Quantitative differences are) cavities with a latter case depending on the depth of the pool, expected in the should not affect the specifying of instrumentation.but the range VI.3.1    Water-Filled Cavity Analysis The ZPSS analysis for a fully-water-filled cavity concludes that the process. water pool will have no effect on the debris dispersal steam bubble will be created at the breach in the RPV that subse quently forces all of the water out of the cavity.
VI.3.1.1      ZPSS Analysis The ZPSS estimates bubble growth by assuming that all the core material ously      quenched. coming out of the pressure vessel is instantane-cess creates a bubble,        The with steam generated during the quenching pro-the pressure and hence the size of the bubble to increase. additional      The steam ge in the ZPSS is: rate of bubble growth as derived from the basic equations given dfb                                                  ;1/2' dt "Ib 3      RT              .          2/4-P 2  b. ~ "C ' ~ b[ 3        /'
                      . 4 w rb                  '      Pg J
whereb  b = bubble radius R = gas constant 108
 
T = absolute temperature t = time sq = steam formation rate s c = condensation rate P
b = pressure in the bubble (instantaneous)
Pa, = atmospheric pressure P
g = pressure in the water away from the bubble The steam formation and condensation rates are given by:
sq = Af[2pp(P - P )]
o    b      cp (Tp-Tsat)/hfg    (VI-35) sc
                  = hb (2nrb )(Tsat - TB )/hfg                  (VI-36) where:
A p = area of breach in melt generator Pp = density of melt Po = primary system pressure cp = melt specific heat Tp = temperature of melt T
sat = saturation temperature in the bubble hfg = heat of vaporization h
b = heat transfer coefficient at the surface of the bubble T
B = temperature of the water away from the bubble The ZPSS basic equations    derivation mistakenlof the drb/dt                      the same first term in the equation.y omits the (1/2expressig)n  rb factor fin romthe latingThethe  pressure pressure in the in  water the    (Pg) must be known to allow calcu-bubble.
reactor,                                          For the geometry in the the expansion wave created by the steam bubble is trans-mitted to mately    twicethethe walls  of thepressure.
incident    cavity and reflected back at approxi-The velocity of the propaga-tion is assumed to be given by the compressive wave velocity.
109 1
 
The interactions of the reflected wave patterns are complicated by the geometry of the cavity; waves returned by the side walls will be out-of-phase instrumentation    tunnel.with those returned f rom the floor and the ZPSS analysis, only the wave propagation in the instrumentWav tunnel to the f ree surf ace at the containment floor. The calcu-lated bubble and cavity pressure histories are given in Figure 15 (Ref. ZPSS Figure 3.2.9-6) .
tunnel is predicted to cause theThe influence of the open instrement af ter the initial rise.                      cavity pressure to decrease VI.3.1.2      SPIT Experiment Design statedTwo    jet / water tests are proposed to satisfy the purposes above. The first test will use a large square box (0.6 m lateral dimension)        of melt plexiglass    to contain a water pool that is in    contact with the              generator.
The transparent structure allows hi-speed f raming cameras to be used to monitor the inter-action    of the jet and water and the growth of the steam bubble, if present.
The important result will be to determine if the steam bubble growth occurs in the manner postulated by the ZPSS.
The walls of the box are not prototypic of the reactor in composition, dimensions or strength.            Compressional waves emanat-box structure.ing f rom the interaction zone will cause prompt f ailure of the to be absorbed as movement of the water surf ace.The free water surfa ences will        cause the influence of water pressure        These dif f er-on bubble should be greater in the experiment than in the accident. grow i
within      a system can be evaluated by considering the mech impedances of the materials involved.
material is given by (Ref. 29) :                    The shock impedance of a where:
Z = impedance p = density p
o = longitudinal bulk sound speed cause  A compressional a " partitioning"  wave  in water of the        incident on an interface will incident relative                                              stress based upon the impedances of the materials. A simple approximation of the relative magnitudes can be found f rom the expressions:
2Z Bp =            A P
o ZA+ZB                      M@
110
 
8.0        '
i      '
i    '
i      -
TRANSIENT ACCIDENT 7.0  -
SEQUENCE                _
6.0 -  -
9c.
s 5.0 - .                                                        -
[                    BUBBLE PRESSURE m
4.0 -                                                        -
M n.
3.0 -                                                        -
2.0 -    .
CAVITY PRESSURE                        ~
        ,o-
                  ,.-. g                                            -
0.0        '    '      '      '    '      '      '
O.00        0.02            0.04      0.06            0.08 TIME (s) t Figure 15.
Bubble and Cavity Pressure Histories for a          '
Water-Filled Cavity l
I 111
                                                                          .ru
 
Pr  = Po (ZB~Z)A (gB+Z) A                M-38) where:
P o = incident stress level P
T = transmitted stress into material B
,            Pr = reflected stress Z
i = impedance of material i In equation (VI-3 8) P          represents reflected of  Po      back into the incident material (water).the pressure that is imposed  is on continuous, P          then the reflected pressure PIf the source propagated.            from the interface to the point wherer Pis super-  has how        these equations can be used to predict the w that will occur in the experiment and accident.
the calculations below:                  using equations (VI-37) and (VI-38)  The results are giof ven Interface Condition Frea Surface        Like Materials      High Impedance (ZB " 0)                (ZA"Z)B              (ZB >> ZI  A e
PT 2P o                      Po                                    i P
                                                                      ~9              !
r                  -P o                      9 P,
t gas exists above the water pool.The free surface condition represen                i sion (tensile) pulse will be reflected back into the water poolThe re A liquid can support a tensile wave only in absence of oth er                  .
t disturbances that cause shear stresses to destroy the wavefront Other wave periment      and interactions accident,      will undoubtably exist in both the ex-      .
to rapidly return to zero.therefore causing the reflected pressure                f cause a portion interface.            of  the water  The surf  transmitted ace  to      stress expand  away  (2Po ) will from    the The expansion will continue until P which off."    time the expanded portion of the poolo will        goes to zero, at be " spalled tinually to  exit the expanding cavity.            steam bubble will cause all of the 112
 
The second the interface.      case This      involves asimiliar represents            materials hypothetical      on both sides of interface,  reflec-tion of the  pressure pressure is unchanged. is predicted  to  be  zero and  the  transmitted The last exa m pl e involves a second material that is of significantly higher impedance than the incident material. This situation steel      is somewhat pressure  vessel. approximated      by water in contact with the For the limit Z >>Z        the transmitted pressure is essentially zero, but the reblecteh, pressure is equal to and superimposed on the incident pressure. The direction of the reflected pulse is away from the interf ace and towards the source.
source isThis  behavior will be maintained as long as the pressure constant.
regions encompassed by theThe pressure at the interf ace and in those double the initial pressure. propagation of P r will be at roughly poolsLoads by theplaced    on the structure used to contain the water wave interactions      are thus dependent upon basic material pr ope r ti es.
Four materials are of interest in the experiments and accident: water, plexiglass, steel, and concrete.
Standard 31  and arematerial given inproperties Table 19.are obtained f rom References 30 and TABLE  19 Material Properties for Impedance Calculations Densigy          Sound Speed Material            (kg/m )            (m/sec)            Impedance Water                  1000                1500            1.5 x 106 Plexiglass            1180                2680            3.2 x 106 Steel (Stainless)      7850                5960            46.8 x 106 Concrete              2340                4000              9.4 x 106 The ability of a plexiglass box to simulate the influence of a concrete    cavity can be estimated by comparing the reflected pressure levels for both situations.
For Plexiglass:
(3.2 - 1.5)
Pr=PO    (3.2 + 1.5)
O 113
 
For Concrete:
(9.4 - 1.5)
Pr=Po    (g 4 + l g)
                                        =  0.72Po The calculation shows that the plexiglass will reflect ap-proximately concrete.
50 percent less pressure f rom the interf ace than Because  the pressure in the water is inversely related to the bubble growth rate (Egn. VI-3 4) , the plexiglass wall will cause  the experiment to overpredict the bubble growth relative to the accident.      Thus, if the bubble is not observed in the experi-ment, it will likely not exist in the accident.
Pressure pulses propagated to a free surf ace interface will not reflect pressure into the pool.                This is contrasted to the reactor  case,  where return to the pool.
the  steel  RPV    will cause    a pressure pulse to i The reduction in the reflected pressure caused by the f ree surf ace in the experiment will also serve to overpredict the steam bubble growth. The presence or absence of trapped air pockets will af fect the magnitude of the reflected wave.
Forces acting on the initial bubble generation suggest the geometry to be hemispheric.          Compression waves emanating from the bubble will then be curvilinear as opposed to planar. Reflec-tions from plane interfaces will be returned as curvilinear waves so  thatare bubble    thestaggered.
return times of all portions of the wave f ront to the Likewise, waves incident at other than nbrmal to a plane surf ace will reflect at a complementary angle and not back towards the source location. Plane or curved-wave fronts incident on curved surfaces such as the cavity wall will be reflected in a distorted manner. Thus, the interactions of the various complex.        pressure pulses within any geometry will become very The time interval in the experiment prior to wave inter-actions can be estimated by considering wave transit times.
Considering the initial steam generation as a point source, the wave    transit time is given by the dimensions of the box and the sonic water velocity:
                =
2ax      2(30 cm) t1      C
                                            "    *  "8      *    ^
0.15 cWusec (VI-39) 2AY      2(110 cm) t2      C
                                            =  1.47 m    (V rtical) 0.15 cWusec 1
l 114
 
The dimension in t tanceThe box.      from times the center        of the stream to the nearest edge                of tht ini tial compres,s.ional    t 1 and  t wave f ront.2, represent the transit times of the e
As the bubble expands the the bubble and the motion of the side walls. distance will transit times generator,            are short compared to the discharge              Thetime calculated of the promptly distort and fail. suggesting that the walls of the structure will very expected during an accident whereThe results can be compared to that the transit time corresponds to the radius or depth of the reactor cavity.
t          2(260 cm) l                        "        ""
0.15 cny'psec        *
(VI-40)
* 2(450 cm) 2                              "8 0.15 cny'usec scaling ratio in the range of 1 to 10 to 1 to 4. Comparing the e effect of the interaction in a scaled cavity.The                                  second purp To do this, the the reactor cavity, wave interactions within the cavity must be properly              In    scaled from the boundary of            the compressional waves will propagate away contacted, surface      of theeither      the  the steam cavity    walls, bubble    until an interface is fl o o r, pool.                                      RPV, or the free complex cavity, floor,  due  and tovessel.
the differences in the geometry of the bubbleT cause instabilities in the bubble, Focusing of the reflected waves may, tation and dissipation of the bubble energy. inducing              Thedestructive waves may fragmen-also constructively interact at an interface pressure loadings on portions of the cavity. boundary to concentrate The second jet / water interaction test determine thewave interactions to occur in a prototypic manner.is designed to cause structural loads placed on the                  The test will hydraulic pressure in the water pool and the amountcavity water.                                                                              by the of expelled the correct A 1:20 scaled-cavity geometry will be used to maintain t                    wave interaction transit times.
be    used    because power of the distance          the traveled.
wave transit time is proportional to firstLinear scalin        -
susceptable to rapid decomposition andof aluminum to simula relationship f or reflected pressure (gas                      generation.      Uring the properties for aluminum from Ref. 30 gives:                  Egn. VI-3 8) and material p
r
                        ,    p    (17.3 - 1.5)  =
0                      0*84PO (17.3 + 1.5) 115
 
higher than that predicted for concrete o(0.72 of P ).The strength The influence the reflected wave on bubble growth may be                      slightly more detrimental compared to the reactor case.
by      The hydraulic loads on the cavity structure will be obtained pressure cavity.              transducers The devices      will  placed in the sidewall and floor of the location. Mapping the measured        measure      the value of (Pr+P) valuJs                    o at that allow inferring the total load placed upon the structure. should for all locations VI . 3 .1. 3    Predicted Loads The magnitude of the hydraulic pressure pulse can be esti-mated ciency. f  rom  a  presumed      thermal-to-mechanical conversion effi-interactions less (Ref. 3 2) . are estimated to be on the order of 1 percent or Using 1 percent as a basis and assuming that all of the melt contributes to the mechanical energy in the form of pressure / volume work, yields the following:
Total adiabatic thermal energy = 10 kg x 3.6 MJ/kg = 36 MJ Mechanical work = 0.01 x 36 MJ              = .36 MJ force For(F) an    expansion process, the work is equal to the applied integrated over the distance of displacement              (ds)
Work = /Fds = f(PdV + VdP) = /VdP (VI-41) where the PdV term volume does not change.is zero for a closed system in which the the cavity wall or floor,When the expanding pressure wave reaches the end of the keyway is not yet a f actor and the volume (V) canthe be approximated by the cylindrical cavity alone.                    Therefore:
work    =
V / dP  =
V(P2-P)  1
                                                          = VAP 6P    =          =
V 2.13 x 10-2 y
                                                  =
1.69 x 107b    = 16.9 MPa y
This value represents the incident pressure on the sidewall, P.
The
  + P. o pressure the cavity must withstand is given by the sum of li r
Pwall = Po+Pr = P o(1 + 0.84) = 16.9(1.84) = 31.1 MPa 116
 
c-This result establishes that the forces exerted through the water pool can be significant.
The apparatus will be designed where possible to withstand loadings of this order to permit accurate pressure readings to be obtained.
VI.3.1.4      Instrumentation For the two tests described above, the melt generator and aerosol instrumentation will be supplemented by high-speed motion picturecontainer.
water    cameras and multiple pressure transducers placed on the For the plexiglass box, two pressure trans-ducers base.      will be  used,  one in the sidewall and the other in the Wall motion will cause these devices to register propor-tionally lower values than actually applied at the interf ace.
The aluminum six pressure              scaled cavity will be monitored by a total of transducers.
the cavity, tunnel and keyway regions.The devices  will be placed The individual      in pairs pairs  will in be situated with one gauge mounted on a horizontal surf ace and the second on a vertical surf ace. Pairing the transducers pro-vides tion.
redundancy in the data and an indicator of spatial resolu-The motion picture cameras will be used to record the motion of  the water.
of the    jet in the plexiglass water box and the resulting behavior ble,                Visual access to the aluminum cavity is not possi-but  the  cameras f rom the exit of the keyway.will follow the movement of water and melt VI.3.2      Partially-Water-Filled Cavity Analysis The interactions between the jet and water in partially filled  reactor cavities differ from fully-filled in two signifi-cant ways:
: 1)    Aerosol generation and hydrodynamic forces are active during the jet propagation through the ambient atmosphere
: 2)      The  f ree upper surface allows unconfined expansion of the pool The first of these aspects is important in that the charac-ter  of  the jet may be altered depending on the distance between the RPV and pool surf ace. An unstable or divergent jet may be highly f ragmented before reaching the surf ace causing it to be more susceptable to melt / water mixing. Better mixing may promote a vigorous vapor explosion.
The presence of a free pool surface allows only partial confinement of the pool expansion caused by steam generation.
117
 
Structural loadin cavity    situation,g through the water is similiar to the filled significantly reduced.but the loadings above the pool surf ace are throughout                      The lack of a coherent water structure water into the containment region.the cavity may prevent aofcomplete    the    pur The principal interest in the test results is the amount of water aerosol  and  melt displaced from the cavity region and the extent of generation.
If the results are roughly equivalent        to those observed for the fully-water filled cavity, it will                be assumed that significant dif ferences are not detectable. This outcome will then dictate that only one type of jet / water inter-action test need be considered in the HIPS test matrix.
of jet / water testsIf detectable differences are observed between the two types  '
discern the source ,of the dif ferences.then additional tests may be required to consider an additional experiment with a transparentIt may bewaternecessary con- to tainer to permit visualizing the type of interaction that occurs.
The present state of knowledge of this interaction does not allow                    ,
predicting understandwhat  the type  of additional experiments may be required to interactions
(
The test matrix is developedtoinconsider  a partially-filled water cavity.
the possibility of one to three additional tests, ual experiments are not yet developed.although the specifics of the individ-VI.4      Scaled Cavity Tests study debris removal mechanisms in a realistic geom prototypic materials.
The test provides important information,            '
together with mechanisms          the HIPS tests, to determine the scaling of the to the reactor geometry. Quantative data on the also be obtained. aerosol source term and concrete decomposition processes will VI.4.1      Apparatus Description Figure  A 16.
schematic      drawing of the concrete test article is given in 1:29 the Zion  linear plant.scaling of the cavity, tunnel and keyway regions sand composition given in Table 29 (f rom Ref. 6).The                  concrete is the Exact repre-sentationinof included            the the    RPV and the instrumentation tube bundles are not model.
is cast into the concrete at the scaled height of the RPV (22.5A flat cm) and The    plateisalso used  as the bottom flange cover of the melt generator.
prevents                                                        ;
lating the insulating chield around the Zion reactor. upward gas flow from t
(
118
 
A
                    +                            +      e' r~~
                                                                                    \
                                                                                            +            ^
ir                                                                's
  )k                                                                                    N                    A n 5.7                        ~T~T~~~~~~~jj/                                \
                                                                                          \
      "                                                                  O 11.4                          _____,''$                \
_,'    j          61.0 s,          -
              /
                                                                                      ,e
                                                            %                    ',/
CAVITY                                                  N -a.
EXIT 1r MELT GENERATOR
                                    <          52.8-                      :        30.5 -+
                                        "  5
                                                                ;q 13.2      : (A                FUSABLE          s
                                                                                    + 10.2 +
                                \
PLUG                      ,
                      ''                                                ~-
R  CE                                            !f
                                                                  \~                          ,
3r
                ''        ~.*          ''
                                      '.( ;          $UMMi                                of f          2.5
                            !            h    o*f$tN                                        g 6.9
                ,                                                  _  ,                ,          7 6
                              *.                                                    I';46/o      n        55.9 l            15.6
              $h                    i[hbNh$$!!$Ijhh l)s2                                                  ,,
1    - 35.8 -        :    ;      > 12.7
              '                                                                  RAD
                                            -106.7                                          >
I l
Figure 16. SPIT 1/20th Scale Cavity 119
 
                                                    - -- ~ _ _ . - . . . _ _  . _ _
r 1
i TABLE    20                                        L l
I                                                                                        ,
t CHEMICAL COMPOSITIONS OF CONCRETE (Weight Percent)
[        Oxide                Limestone / Common                          Basaltic Sand Concrete                              Concrete b
Fe23 0                1.44 6.25  :
Cr 0                    0.014              '
23                                                                    ND  r Mn 0                    0.03                                                  t ND  )
TiO 2                  0.18                                                  K 1.05  >
K02                    1.22                                                    ?
5.38 Na2 0                  0.82                                                    -
1.8 Ca0                  31.2 8.8 M0 9
(
0.48 6.2    i SiO                  35.7 2
54.73    [
A123  0                3.6                                                    i 8.3 CO                    22 2
1.5 HO 2                    4.8 5.0 SO                                                                            k 2                  <0.2                                                    i
                                                                              <0.2    ;
i ND = not determined 120
 
i l
allows convenient access before and af ter the experim two mating surfaces are gasketed so that a good pressure se.alThe obtained halves are when  thetogether, bolted  long through-bolts are secured.                    is Af ter the two place and secured to the embedded steel plate.the melt generator      intois lowere The placed in a interactionexperiment  will be performed with the test fixture chamber      as depicted in Figure      17.
scribed in Section VI.1.1.The remaining assembly steps proceed              in the sa VI.4.2 Information Sought f rom the Cavity Tests In order to be useful in establishing scaling criteria, the results f rom phenomena      the SPIT cavity tests must identify and measure the involved in the debris di and concrete decomposition processes.spersal, aerosol generation, ma tional                                        Table 21 lists the infor-during the SPIT cavity tests. requirements and the dependentimportance phenomena of The term " debris" is used in the the jet contacts the cavity or water pool. table to refer to the po TABLE 21 SPIT Cavity Tests Informational Requirements Measurement Phenomena Pressure in Vessel Gas solubility Jet velocity Gas velocity Melt Temperature Gas solubility Melt density Incident heat flux
* Bubble growth Gas Solubility Aerosol source term Dynamic jet configuration Jet Velocity Dynamic pressure Mass flow rate Debris configuration Incident heat flux Aerosol source term
* Penetration velocity 121
 
TABLE 21 (Cont.)
SPIT Cavity Tests Informational Requirements Measurement Phenomena Jet Temperature Incident heat flux Aerosol generation Debris configuration Jet Configuration Aerosol generation Incident heat flux Dynamic  debris configuration
* Water interaction
* Fragmentation Incident Heat Flux                      Concrete erosion              t Gas generation
* Water interaction Concrete Erosion                        Gas release / composition Aerosol generation Debris configuration Dynamic Melt Configuration              Concrete decomposition Heat flux to surface, floors Removal mechanisms
* Water interaction Gas velocity                            Debris dispersal Aerosol transport, generation
* Water purge Cavity Pressure Debris dispersal
* Hydrostatic pressure Aerosol Composition, Size,              Fission product transport Mass, Concentration
* Water pool scrubbing Melt Disposition Removal mechanisms Concrete erosion
* Applicable for experiments involving water pool.                        i
[
Many of the informational needs arise f rom the discussion of debris dispersal and aerosol generation mechanisms given in Sec-tion IV.3 and IV.5. The following section discusses in more detail the instrumentation techniques that will be used.
122
 
I 243.00 e                      ~
                                                                                                                          ;I' n          v
                                                                                                        ,_    q n ;
I"A
                                &n]t) l                                                            =y' =                    l
:  _ /                                                                                :
x.
I Ta                    6                                                              "
                                                                        .i a
g 24.00,    ,          -48;O0 +            LD.
I+-- 52!OO                    O.D.                                  6.0 x 3.0 L.3 x 3 x                                                                                    26.435                        STL I BEAM
,    1/4 THK.        -
26.935 8 PLC'S.                                35 PLC'S.
na                                                              231.00 #'                    #
G                                                                                              112 523--+                                              114.76              I ii T                T 1.783 REF. : j:                - !=j: !=j                      j
                                                                                                                                -6.0 REF.          26.065          - -
l 3 50 T            T          1 Y            Y      Y
              'y                ::                                    i        j      ::
T_. u                    <
                                                                                                                                              '                          j, E                                                                                                                          ::*
5                          l 32.00                                          -
E mr : ,.
[              ]        "#*
o            ."        ""
:          l 1
so                    r                      i                                              '116.00      I i          }                          I                            l i
a 1                                    -q-+                      !
59.00                        5Ei              4
          /.                                                                      4 p
                                                                                                                                                          ;    ;        j i                      j                                                                        ..
18 PLC'S.
(,j                                  Q I          rh::,
a i
TYP.
U[ <\                  p      j l          '
y y            t_    -
r_2  ,  l      4 r l 1
                                                                                                                                                                        ]
                          \                          /            '
                                                                                                            ,  ,                  e                    ---- l
                                                                                                                                                                  ?    i GASKET            6.0 x 3.25 CHANNEL                                                    '
4 REQ'D.          2 PLC'S.                                                  4.0 x 8.0 RECTANGULAR TUBING Figure 17. SPIT Apparatus Placed in the Interaction Chamber
 
VI.4.3  Cavity Instrumentation to    Instrumentation for the SPIT cavity tests is more        difficult cle and the need for unobtrusive devices. implement because o the  instrumentation                            Consequently, inappropriate for the used cavityfor  the                            most of tests. characterization testing is given in Table 22 and are detailed in the paragraphs following.The TABLE 22 Instrumentation for the SPIT Cavity Tests Measurement                  Device                  Comment Melt Gen. Pressure          Pressure transducer          1- Expansion chamber
'                                                    1  gas line Melt Temperature              TC Placed in crucible sidewall, inverse heat    -
conduction analysis Gas Solubility              none Estimated Jet Temperature              none Estimated from charac-terization tests Jet Velocity                  none                  "    "    "    "    "
Jet Configuration            none Estimated f rom charac-terization results Incident Heat Flux          none                "      "    "    "    "
Concrete Erosion            Embedded sensors            Also post-test obser-vations Debris position /            Embedded velocity sensors            Optical detectors and electrical probes Gas Pressure and            Pitot-static Velocity tube / Pressure Placed in keyway and gage                tunnel regions Gas Temperature The rmocouple      Shicided Gas Composition              Sampling tube      Grab sample & post-test    gas chromoto-          ,
graphy 124
 
TABLE 22 (Cont.)
Instrumentation for the SPIT Cavity Tests Measurement                        Device                    Comment Cavity Act osol                  Sampling Tube          Filter sample Debris Disposition                Catch pans            Located    fixed dis-tances from apparatus Framing Camera        Film record X-ray                At exit aperture Chamber Aerosol                  Impactor, filter, Same as characteri-photometer, cy-      zation test clone, deposition surfaces Chamber Gas                      Time-phased composition                                          Poct-test gas grab samples          chromotography The    first    seven measurements given in          Table  22  through Incident same    data,Heat  Flux) as the  jet are  either done in characterization      the same manner or(une the experiments.
VI .4 .3 .1    Concrete Erosion Concrete    erosion by the jet or debris pool in inferred f rom the tionsoutput in theofconcrete.
thermocouples placed at various depths and Total crosion and erosion rate can be loca-obtained      using  the  time  of f ailure and knowing the position of the thermocouple with respect to the initial melt interf ace.
Figure 18 illustrates a typical installation pattern for the embedded thermocouples            placed devices are commercial Type K sonnoen        in the SPIT scale model.          The in    ungrounded 1.6 - m m-diameter stainican-steel cheaths.
The size of the aheath has been smaller selected    for  its  relatively  rugged    construction. Sheaths of diameter or bare thermocouples have been shown to be in-canting opewithstanding capable      of  rations.
the prenourca exerted during the concrete The array is formed using 0.6 mm thick stainless      steel  chim    stock  between    each welded. The arrays are positioned prior to the casting opera-unit and  then  spot tion by thin (0.4          mm) steel wiren stretched across the void formed by the concrete forms.                  This to give a positioning accuracy of i 1 mm technique    or better.
is estimated 125
 
V
                                              .,      L.
            - 10.2              - 30.s - -+
l f71              fil rLj iJ .,
1 (a) 2.5                          '
                                            -s    e 2.5
                                      --o-        += 7.6
                                  '                          Flush With Concrete Surface 1.0 -
ik G '6 JL 2d d0.?*j'hf.QX.Ypf tAS5            3:'dI@
u (b)          2.0                                3                                  ,
l 3.1      _h            '                                                      i V
l Dia. Type M b .6 mm
                                                        ,          Thermocouple        k
                                ,                                                        i (SS Sheath)        '
4 Figure 18. Embedded Thermocouple Detaas (a) Thermocouple Arrays Pieced                                  '
in Concrete Cavity i
(b) Detaas of the Array                                          l s
(oimeneaone h em) 128
 
ceilingThe of    arrays      are placed in numerous locations on the floor and the cavity.
array                                                      In all cases, the an accurate reference forsurf ace of the concretethe first sensor is    placed    flush          with          the so that in Each thermocouple                                                the other sensors is ing in a connector. is routed out of the concrete before achieved.                                                                                        terminat-The ability of the thermocouple devices to measure erosion is dependent upon the accuracy of the device's placement response of the measuring system (including the sensor) and the ability the behavior induced by the contact with the melt.of                                              The uncertainty the experimenter t in the il      mm,position of the probes prior to casting is estimated to be with              but the forces exerted by the concrete mixture combined shifting position less certain.of casting forms make the accuracy of                                                                                      the          final Experience with similar casting opera-tions    has      shown      that as 6 mm f rom their original position.        individual            thermocouples can shif t as much appears to be                on                                                  The mean displacement steel        shim insures the order of 1 mm. that the sensors The            in high-strength of the the          array will not be separated, displacements the sheath between                                            are therefore restricted to bending of spection via X-rays                            the is notsensor            tip because practical  and the steel strap. In-the technique is inadequate to image specimens of small size.                            the resolution of the uncertainty associated with usingEstimating                                thermocouples to monitor the expected eros concrete attack.
Reference 6 indicates that molten steel at-from about 12 cm/hr to 20 cm/hr. tacking limestone / common sand c sent a lower bound on the erosion caused by jetThese are representative of that values probably deposition,                                                      but    repre-the cavity in the form of a molten pool.to be expected if the melt remains in this analysis, a mean value of 16 cm/hr is used.For the purpose of Erosion couples is given by: rate determined by the failure of embedded thermo-Erosion rate (0) =
Distance between two sensors (D) + the time interval between failures (TF)
The uncertainty in the erosion rate (ED) is then given by the expression:
1/2 ED2              DrF        2 ED = D                          +
D2              77 2                            (VI-42)
The to approximately    time interval 0.063 hour.        between f ailures at 16 cm/hr                                  corresponds error in time measurement (accuracy of interval counter) isFor                                                                                                        verytime period small (0.01% of full scale) and is considered negligible compared 127
_                                                                                                                                                            n_
 
to the ability Typical          to discriminate the exact time the f ailure occurs.
values on the order of 10%
for the error in discriminating failure time are in  the                    if multiple sensors are employed. The error position of the sensor is estimated to average 1.5 mm.
Using these values in equation (VI-42), the uncertainty in the erosion rate becomes:                                                              {
l
                                                                                          \
1/2                                  i
,            Eb = 16        *
                                      +    (0.1) 2      =  2.88 This value represents an error as given by:
Error  =        =          =  18%
The  above    analysis        suggoats that cannot    give                                            ombedded  thermocouples accurate erosion rate data unless significant im-provement    can  be made in the detection of the f ailure time. The potentially high heat flux involved in the jet l
ejection could conceivably    make  the    thermocouple value used in the calculations.
failures  more certain than the    .
l Conversely, the f ailures may occur  so  quickly  that    the  response becomes the predominant cource of error.
of the individual detector      :
VI.4.3.2                                                                              i Debrin Position / Velocity                                                '
i The ZPSS analyala proposes that the molt jet f rom the vessel stagnatos that travela  at the cavity floor and forma a radially-expanding pcol sufficiently high, down the instrument tunnel. If the velocity la "splashout" of the melt will occur.              Instru-mentation capabic of monitoring the initial movcment of the                    melt would allow verifying the ZPSS hypothonia.
The  consor employed on the SPIT device relion on measuring the position of the melt with respect to time.
shown in Figure 19,                                                      The device, ceramic substrato imprinted with consist    of a thin strip of aluminum-oxido film conductivo and rociative paste.              a network compriced of thick-substrate has a continuous atri                            One sido of the coramic has a continuous recintor atrip. p of                              while the other At conductor, ap ciflod intervala, conduc-tivo " fingers" opposito              emanate f rom the rociativo atrip to locations the continuvua conductor.                Brid from the conductor            atrip to a " finger" ging        acrona will  give the  ceramic a apocified resistanco when measured            f rom the conductive taba.          An 01cetri-cally for    insulating small  portions paint ofintheused    to cover the conductive            entiro devico except taba.
placed into the concrete ao                                        The devicca are        I the continuous conductor                          that only the top portion of l
the    rociativo strip buried in the concreto.and opposing fingers are expocod with 128
 
r DIRECTION OF MELT FLOW CONDUCTIVE STRIP                        CONDUCTIVE FINGERS
                      ->    4- 0.51 cm (typ.)
SIDE A    IIIIIIIIIIIIIIII                    IIII G                                    RESISTIVE STRIP c
CONDUCTIVE STRIP SIDE B n
            <                  11.4 cm Figure 19. Melt Velocity Sensor
      ~
 
Each      device is connected to a DC power supply so that a circuit the of  the    type shown in Figure 20 is formed. The output of circuit is given by the ratio of the resistance of the el e-ment to the shunt resistance:                                                          i Rs E      =  E out          in n
                                              ))
( Rg + i=j    Ri  )
                                      \                /
where:
E        =
output voltage out Ein    "
power supply voltage R3      =
chunt resistance
                          =
Ri            resistance between element (1-1) and (1) j        =    element    number reprenonting the position        of melt f ront n        =
total number of elements                                    '
I When    the
( i. e. j=n) then melt        f ront causes the last finger to be the output voltage is equal to the input.bridged, n                                                                The quantity g{            Ri is on the order of 2000 ohma ao that choosing R a
= 100 ohma          gives E                              a nyatom consitivity of approximately 0.1 wkS1 bepor stop.            In this manner, multiplo discroto voltage stops acroan each  obtained finger  asintheturn.
melt pannes over the device and shorta finger position                                  Plotting the data in terms of velocity as                  an  a  function    of closure time will yield the slope of the line fit to the individual pointa.
The la    found error      associated with measuring velocity in this manner by considering the accuracy of the finger                                                          pooltion  of  each Appendixand  E. the timo measuromont.              This proceduro la detailed in by                  The    analysis    indicaton  that melt velocity datormined 1.6 % .
the  ceramic      nonnora    la  roanonably  accurato,  on the ordor well-behavedThis resultant la based on the assumption that the melt of              la conalatent manner.
and    that  shorting    of the clomonto  will occur    in a gosta that                        Observation of the free-jet experimento aug-along the cavity    the melt floor.willInteraction not behave an a fluid pool expanding betwoon the hot melt jot and algnificant quantition of acrosolo and ganon.concreto                        la most Those producta  are likely to not ahead  conductivo, ao that deposition of the materials on the connor of the melt may insulate the olomonta.                  The expanding cloud of vapor obnorved with the melt stream may                        also move through the cavity to cause premature ahorting                    of  the olomonta 130
 
e        e        e      u        .-
                ---~~
R,      R 2      R 3    R4      R3 Ein 1                                                      R. 3, o
E out Ein    - SOURCE VOLTAGE Eout - OUTPUT VOLTAGE Si
          - SWITCH REPRESENTING MELT SHORTING ACROSS FINGER i TO CONDUCTIVE STRIP R
        - RESISTANCE MEASURED FROM FINGER i TO CONDUCTIVE TAB R,    - SHUNT RESISTOR i
I 1
0 f
Figure 20.
Eloctrical Circuit for Molt Velocity Sensor 131
 
~
prior the to arrival of the melt.            Likewise,    melt sparged out of pool may precede the melt front and cause if depocited on the sensors.                                    erratic readings the      The melt  1/20th scale cavity provides the opportunity to evaluate expected invelocity the HIPSsensors cavities.under circumstances similar to that Failure of the devices or erratic the measurement or modification to the existing techniqu                            l VI.4.3.3      Gas Pressure and Velocity The                                                                          !
the blowdown city      and gases from the melt generator                  must which pass.
The velo-cavity density of the gases give the dynamic pressure term that is used in all of the hypothesized head  '
i mechanisms.          Thus, obtaining                            debria dispersal the gas velocity an a function of time is important in assessing the debria dispersal mechanisma, The 1/20th scale cavity will be instrumented with a pitot i
static ing        tube  to determine    the velocity of the gas stream. Apply-conservation of energy to a noction of the tunnel given equation for the velocity of the gas:                                            an I
(Pt  -
Pg Vg  =
g                j              (VI-44) 9 where:
C      correction factor p = friction and turbulancofor the pitot tubo to            account  for P
t = total or stagnation prosauro P
a = static pressure of gas og  = gas doneity The differential    quantity  (Pt-P) proasure      awill be obtained directly by using a gauge    betwoon the stagnation tic prosauro porta.                                                      and sta-unod in Equation (VI-44)          Thetodifferential prosauro data in then                !
particular location in the tunnelgive the local velocity at a alghly turbulant flow                                an a function of time. A to      be                      in noconaary to cause the velocity profile constant acrona the tunnel noction.
in expected pronauro          to be generated voanol,  plus ganos  principally by the gacon from theThe veloc:.ty field released f rom the concroto. Tho the high veloct ty    pattorn in  undoubtedly ofresulting    flow          highly turbulent because of the ejected gas and the many ubstruc-
{
tiona provided by the cavity geometry.                                                    t 132
 
The pilot-static device will be calibrated by performing a vessel brated blowdown with the melt generator in place.                        Using a cali-mass      flowmeter      on  the gas    line    to the  generator      will allow the or volumetric flowrate with time to be obtained. Knowing the density velocity        and the passing    over geometry the sensorof the      apparatus will then give the location.
done with nitrogen                                            The procedure will be pressures.                      and carbon dioxide over a range of system The tests with carbon dioxide may influence of gas condensation on the                                  illustrate the These results will be usef ul in the modeling              velocity      measurements.
process concerned with the blowdown of steam f rom                the reactor pressure vessel.
It    is wide              anticipated that the gas velocity will vary                    over    a value to  range the as  thecondition.
final    system pressure drops f rom the initially high the quantity (Pt-P)gwill require            The corredf>ondingly wide range in broad response.
a sensor with an Sever al pressure gauges in parallel mequally employed to improve resolution of the                                              ay  be tube also lends                                          data.      The    pilot-static An itself to measuring the static pressure alone.
absolute tube will yield the  pressure desired gauge value. connected to the P7 port of the Thus,    dynamic and static pressures can be obtained simultaneously using a single probe and properly selected pressure gauges.
VI.4.3.4        Gas Temperature in the Tunnel Measuring the temperature of the gas flow in the tunnel com plica ted by the        radiativo  heat    transf  er f rom    the  molten is and    cavity surfaces.          To minimize the ef fect,                            pool will be placed in a shield arrangement that allows                    the thermocouple flow f reely over the element.                                              the gas to prevent stagnation of the flow so            The    shields  will  be    designed    to that the proper convective heat transf er conditions are maintained.                        A mul tiple-shi eld, high-velocity          thermocouple      of a type similar Figure 21 will be used.                                              to  that shown in The    response of the probe to sudden changes in                    temperature can be approximated by the relationship (Ref. 33):
Tp=    Tg  =  (Tg  -
T) o    cY
                                              ~
T (VI-45) whore:
Tp  = temperature of the probo Tg = gas temperaturo To = initial temperature of the probo t = timo T=    timo constant of probo 133
 
TUBULAR SHIELDS THERMOCOUPLE                      -                l- -
ELEMENT -
HOUSING                '
THERMOCOUPLE ELEMENT E                        <                              4 cm                      Y I      1 W                                                                    .
j 1 cm GAS W                    - : .-
: - - : - . . :---=-
FLOW
                                                                          --~.-<...---~r-3 u
w v                                                                              I  <
4 I;      I Figure 21.
Multiple-Shiekt. High-Velocity Thermocouple Assembly
 
type of configuration. Reference 34 is used to estimate the time and    a thermocouple                                                        or this the time then            constant be capable element is  on the orderapproximately of 20 msec 1.6 mm in di rise-time as          fastof asresolving 50 msec.step  changes in temperature          withThe d configuration is                                Therefore the respond to a stop changeexpected            to be only marg,inally th e rmoco upl e discharge interval.                    in The device willtemperature        occuring  adequate not                during  the to shorter than 50 msec.ing rapid variations in the gas temperature th occur in times VI.4.3.5 Gas Composition necessary for the equations governingethe                                        meltThe tunnel  is      c
    " Grab" samples are typically used to obtain co removal process.
chromatography.
pally from the gas used to charge the melt generatorThe gases                                                                                        c Gassns    released released from the thermal decomposition of the
                                                                            ,  plus the f rom the concrete are                            concrete.
amount of noncondensable gas that ype accident.                                                                    canandbe exp a reactor sitionThe traditional method of gas sampling is to draw th by into a pre-evacuated              chamber that is subsequently  e compo-remotely-operated        values.                                        scaled The time required to equilibrate the dependent is    pressure in        the  sample    volume upon the flow through the with  the  experim The                                                                  ental apparatus interconnecting piping.
the f riction in the pipe (Ref. 35): flow can be estimated                  o obtain      using (f) J/2  =
2 log 10 lh        2 y -' + p.51            (VI-46) where:
f = friction factor 4
        @= relative pipe roughness D = pipo diameter R, a Reynolds number Eq ua tion velocity              (VI-46) can be sol ved connected by a length of pipeterm          (L):      G betwoon and sample  the  reservoir volume  (2)    (1)by ev 135
 
                          /
                                    .                    . 3    1/2 fP2)
PP11 1
                                          -        1 G = PV  =$-          ~                    ~
f P13            fL                        (VI-47) 2 logo            +      -
t                                I whero V is the velocity of the gas in the tube.                                  The maan veloc-ity can also be given in the f orm of the Reynolds number:
I D
C =
D                                          (VI-48) where p la the dynamic viscosity of the gas.
If P  >> P Equation                    then the first term in the denominator of and    (VI-48)(iVI-4 gives: 7) 2,can be noglected and equating Equations (VI 4 r            -
3                                      }1/2 D0P11                  f  P32          4 R [ =}
g                      1        1 (VI-49)
(            -
The maan velocit through the pipe.y can be unod to dotormine the masa flow rato (6) 2 s = ftGD (VI-50)
Solving    Equations              (VI-46) will give the masa flow into the camplo chamber.through (VI-50) nimultaneously The interval in timo at which the prosauro in the sample container larequir to the prosauro in the tunnel.                                                              equal the tont,                                                  For      the nominal    conditions  of the sampling accondo, or roughly        the same time required la on the order of 10 The response time oftimo                thinf technique rame as thew blowdown of the nyatom.
quate to renolvo details of gas roloaco procosa.ill It                                notmay; be ado-how-over, be capable of resolving the molt /concreto interactiona that occur over a longer period, doponding on the extent of dobria material removed f rom the cavity region.
A proach      continuous campling techniquo providos a alternativo ap-to filling gan bottloa.
tubo but the pressure differential would bo provided by aThis                                    method alao vacuum pump drawing on a manifold of parallel cample containora.                                  Valvon 136
 
placed time sequencer on the inlet to controlandvalveoutlet  of each All activation. bottle of the are connecte tainers areAtinitially open to allow gas to pass thro stricted.                                                                        con-seq uence,      the preselected valves          times during and af ter the ugh        unre-ej ection to trap the gas within. on the individual containers are closed variation delay of the  in composition sampling tubewith time to be assessed. Sequential                      valve same                                                                      The time time shif t to occur in each of the camples.4s a f actor but it sho VI.4.3.6 Debris Deposition Debris material      thatdeposition      or displacement refers to the        amount periment.                                                                            of quantity        The two questions that must be                      addressed  ex-
: 1) the the are cavity and 2) and    size the    ti distribution m e-f ram e      of material  ejected    out    of ejection, water pool contact, of melt dispersal relative to jet or gas blowdown.
The experiment.
first question can be answered by simply measuring                  the amount ofThis        material amount isthat  then is  outside compared      theoriginal to the  model    mass at ofthe end the thermite the                constituients melt generator.                and any residual material remai      n ng in i
placed melt, aerosols, The collected material will                  incl udo      dia-settled onto the chamber floorand possibly concreto particles that have to separate the smaller acros.                  Mechanical  acreens  canabe used distribution of                              olized  material,  giving          size done on the residue.the displaced melt. Chemical analyacs will be The ejected melt stream in typically accompanied by a lumi-nous cloud of vaporized melt that darkona as it cools and con denses.                                                                                -
cavity,        This cloud along      with thewillmolton undoubtably also be ejected f rom the graphic techniquen                                debrio.
debrin disporaal process.
will not penetrate          the Conventional cloud to show the photo-ing the position of the donne melt within the luminouX-ray detecti The (mol x-ray t) andrecord the other,should show good contraat betwoons th              cloud.
estimaton        of                lean-donne m ate r i al s.                e donco the amount        of                      Tim  e- r esol ved uncertain becauno the image melt material ejected will be three-dimensional obj e ct.                    is a two-dimensional view of a the      Post-test inopoction of the tout article will also indicate remainingextent of material removed from the cavity.                Molt concreto surface.in the cavity will probably be tightly adheredmaterial        to the then obtaining the mano of the entire toot article after                                                                              befIf the qua ore and remaining  the experiment in the cavity.will      be unod to cotimato the mann of debrin This method does not account for the maaninlost small            by concreto decomposition, comparison.                                  which in expected to be 137
 
l l
VII. HIPS Experimental Program mentalThe main objective of the HIPS tests is to provide experi-in the confirmation ZPSS. The of the debris removal mechanisms postulated aerosol generation mechanismsresults will also be used to describe the and fission product source term during melt ejection and vessel blowdown.
in developing scaling criteria to full-size The            data will be used reactor and dent a    model for predicting the ex-vessel portion of geometries process.                                                    the acci-The that          HIPS test matrix has been developed with the assumption the results from the SPIT acrosol and jet characterization tests can be scaled to larger geometries.
the larger, HIPS geometry will be limited to the    Similiar numbertests neces-in sary to confirm          the scaling hypothesis.
i assumes                                                    The program also that the SPIT 1/26th scale cavity experiment will demon-strate the type and location of the interactions, and the in-1 strumentation needed to diagnose the resulting behavior.
The l                    following sections cover the description of the experi-mental apparatus and instrumentation, the inf ormation sought from the tests and the test strategy employed.
VII.1      Experimental Apparatus The    HIPS apparatus consists of a melt generator, test    ar ticle,                                                    concrete the SPIT program. and the same interaction chamber as that used in scaling of the Zion reactor cavity.The test article represents a linear  1/10th VII.1.1      Melt Generator The the          HIPS melt generator incorporates improvements based on experience of the HIPS generator  gained during the SPIT program.          The dimensions able volume for gas expansion      are significantly larger,      with avail-over the SPIT equipm ent.                representing    an 18-fold increase A
larger diameter is that                second advantage of the relatively and vessel wall is increased.      the distance between the melt crucible refractory                                    The increased powder provides greater protection          to thethickness vessel in-of tegrity.
A schematic drawing of the preneure vennel to be used as the melt    generator      la shown in Pigure 22.
inch (41 cm) OD, Schedule 69, mild-steelT!uo  wall in made of 16 pipe wall      thickness)                                          caning (1.6 cm flange covers arewith          bolted flanges welded to each        end. The internal length ofacaled          with flat, reinforced gankets.
the device (1.34 m) given                      The volume ot 0.146 m3 .                                          an approximate allowable working pronoure    The ansembled apparatus han a rated maximum of 2500 paia (17 MPa),      sufficient to cover the expected range for the HIPS tent matrix.
138
 
e O O Oo    O O            o O                o 0        ,
c _ _
O                o O              o Oo O i
                        *--- 6 8.6 DI A.-->
11.0 -          ---> 41.1 DIA.1 ll                                        n t
l' 156.7 if Figure 22. Schematic of HFS Melt Generator (Dimensions in cm) 139
 
W +
Unlike            the    heavy pipe section used in the SPIT apparatus, the  HIPS      melt shown in Figure 23,          generator    will use a thin-wall melt cru cibl e.
As outer                                    the  crucible  is  constructed shells separated by a 2 cm thick layer of ref ractory wet ram material, of  inner  and heat flux. The designed      ramming to sinter when exposed to a high surface the molten thermite,                      material will form the prime barrier to f rom the melt will cause sintering to a depth of several centi-and meters.                                                                                                    i subsequentThe    testssintered          shell with the      will then be used as the crucible on apparatus.
Graphite plate, 12 mm thick, lower    ends of the melt crucible.                    is used to cap the upper              and 50    cm diameter to allow gas expansion volume above.
passage betweenThe upper    the plate has a central crucible and the    brass                                The lower plate is machined            to fuqable melt plug in the bottom flange cover. expose The bottom flange dover has been modified to accept a small-diameter
                    " insert" plate that is replaced for each test.
permits                                                                            The insert mizen damage  the size            of the  exit  aperature to the flange ccvor.          to  be    selected  and      mini-flange cover by an 0-ring and retaining bolts.The insert is sealed to                        the pattern  Assembly              of the melt generator procons in similar to                    the used for the SPIT device.                  c'i r s t, aporature plug            in determined                                            the  sizo and the insert machined to accept a melt of  the  exit of corresponding              dimonnion.
in the lower into            flange cover and the flange and incert ancemblyThe                        placed  insert in t the recepticlo located in the concreto toot a r ticl e.                              The ly bolted into position. melt generator in lowered onto the lower flango plato a in ponition so that the melt plug in exposed.The lower graphite plate in the loworod onto the graphite plato and held in The                                    crucible in extraneous material                                                          position. After mito powder in placed into the crucible.la removed f rom incido the shell, the th The the porosity and overall volume. thermite in lightly tamped                          to during reduce placement thermito to brought out through the upper graphito cover.a crimped wiron are  depth of no and            to the electrical                                            The loada are then the cover la bolted in place.foodthrougha in the vennel nido wall Attaching the gas fo and instrumentation loadr. completon the annombly procono.od line VII.1.2        Test Article Tho on regions.
a 1:10      linear acaling of the Zion cavity, internal Figuro dimonnions baned 24    chown details and dimensions of thetunnel, llIPS    and keyway tout      article.              The allow                                    external dimonnionn have boon nolected to ojection and vocool blowdown.the structure to withstand the atraina imposed du I
140
 
GAS INLET TOP FLANGE COVER O        O
                            /
O                        O
            !@IY[NfN D$2@f              k
      \Msy 2                      :
                                      ~)
      \;-                        u
                                          )
yPRESSURE VESSEL
                        ,                            SIDEWALL yCRUCIBLE CAP
                  * *J
                  ~Ti  '
f    MELT CRUCIBLE SHELL v
                              ,    -LOCATION OF THERMITE POWDER i
MELT PLUG i
C                  /O            STEEL PLATE J
(O)        prsf,/
j            /, LOWER FLANGE E
y          COVER O
f%Ihk Figuro 23. HIPS Molt Gonorator Assembly 141
 
A<                                                                    MELT GENERATOR MOUNTING RING
:. .....*.~. ...
c..
                                                                                .....                  .'. .; ... .... .;. . '.. . . ..6. ..
                            ....      ,.........., g . ... . .., .. ..... .. .
                    ;s.'?                                                                                                                                      ..        . . .
                                                                                                                . . g. ... .                            . :. . . . ..
                                                                                                                      ....0,e.,......
                                                                                                        . ..f.
                      . .. . .              4. y. . .. ...
                                                                                                                      .,....a..*.
p......=...,..........*.;......,.,        ..... . . .-
                                                                                                                                              ..                                          s
                        .:.y.*.....    ..;..;.....,9
                                                                                                                  ',.w . . ... . :: =... ,.. .. . . . . ,
                ,;.**....;                                                                                              ...,....,*..*yt-
                                                                                                                                    .l......'.'.t;."
                . :;. .: . c. .*. . .                            1
                ... .. .. ... .c.. . .
                                                                                                                                                      , . ;.. . ..v.
                                    ..                                                                                        1
                .n
:..f. n. .
                                                                                                                                            ~
              '..' .'.  .. ,* .*h. 'c.$. .                                                                                                                              .~.*
                                                                                                                                                                                    ~
                      ..... . . .                            A                                                              ,i              ,,            ..
              .'                                            4                                                                                . . .'.yW. '
                                                                                                                          /                .':s.          . . . .l:                  LIMESTONE / COMMON -
              @.R.'.~6.*l.7.
                . . . . N. ':
s~                                  -  ..i            .
                                                                                                                                  . . Q $....'.t'.'.Q.'
SAND CONCRETE c.
                                                                                                              ... 9..:.::;..<.:;
a..........r..
                                                                                                                .... <.. ... ;.... c.. .; . . .                                              .~....
                                                                                          ....w...............
                                                .                                                                                                                                                            i
            .............../......
                                                                                                              .. .......a...
                                . . . . . . .:............                                      . . . . . .. .. ... ..... .... .. ... .. .. .. . . . : . . e..
                                                                                              ......a...:.
                                                                                                  ..;..::...:.m.
                                        ..e. .. ..        .. ..:..;.....:.;                                                                . ..r;.e
::  .'...H.. .;'.i '::.. ..s                    v ..*:. .?                                                                    :
                                                                          ....            ,?...    .6..:'i:.'    . . ':;..      ..
                                                                                                                                                . .'. . . K EYWA Y O PENIN G
                                                                                                                                                          .4.;.'.; .'.
                                            ;. ... . ;,.      <. ...; .. .... ..;i.
                                                                                                    . .. .v .. ...... .;.:;
                                                                                                                                              .  . . , . ...a.....
                                                                                                              ". . .., ... ... .;..<.e.'..........
        . ..              ..                          ?                                                                                    .
          . . . . . . . . .o . ;.s. . .
                    . . ..:..  . . .. ........g....  ..                                                            ..                          .
                                                                                                                        .: ..:. : :. . ....                                  . . . . . :r.
        .. h..      . . .. :..            .::. .*..c.';,;.'.
                                                            . . . . . . .. . ; ? ." 2.:;. i.? : ... .. . . :
                                                                      ,  .  . v
                                                  ., .                  .    .y
                                                      . '. .'. ' ....,.*'..,..?*
        .    . . .; ;..,.;.''.Q;  . . .                      . .? :. *)
l
                                                      *                                                      <-.1 1.4 l
l l                                                            A M --
l    1
                                                                          - 61                                                                                    >
t Figure 24a.
,                                                  Top View of HIPS Test Article (Dimensions in cm) 142
 
s ow em oa j: 4 :': J
        .e e<g      _              -
hO          d f                    -,e    -
    @e                            To(Oa 6 ; '
                                              .N. If            'g{        ,I' O          l' QE me k%go      e. 'h(,F b,QS              * '9          -
G bio'."d
                                                =m0 g3                      s ?
                  -                    o                        o .O                                      S we            o                                  0        g906.:so              .-
5
    $E op              i L
                                              = %c o.c n ig=.7, C
u.
                                                                                                  *        =
                                                                          -Tgy ga                                                                                                      ?
C        8
  ;;j g        "                                                            y?.+ees t          m      .ti m
2            d  '                                    _-
                                                                              $i                        e Pg n.
j e                                                                                        g e
N:(o.G' oV /-d      o      i J' _
o8 Jv iD.gf
                                                -N0          n (M  3$                    E o
                                                                                                        =
5 a
                              ,$.0o..:              -                  -
e o              u                      0          3 oQ 0                              . .g . .          m w                            ,
0                                                        o
@                          h. O ' ' p 4
n o.
h F                            .
od        -
00        -
                                                                        'o, gf                      j o sh
* D.0 98                        9 '.p . 4 ' 9"    -
                                                                      - g,o    i,g (
* 39 f 8
Q u.
go                                                                    ;.
i 0
                          >  =0o                                    - %'.m o*' t s                        V                      .g 1}j.;g 5
it        tij
: y. egoe x    -
g a5            w?M o                                e i
9
* s ltw                                r m/[
a                                        +-- @                      >
l
                      ,                      g-                            >
143
 
i                                                                                                                                                                                                                                !
The line located            testalong    article            is formed in two sections with the parting the upper tunnel surface.
Each constructed using large steel channel members welded section                                                                                              together to is form the outer periphery.
4 the concrete structure and The                                          a steel provides added strength to convenient                        means for handling the apparatus.                      The        concrete is generic limestone / common-sand
;                                                                                                                                                                                                    (see
'                            Table              29)    similiar forcing steel is placed in the lower section to            that    used          at              the      Zion          plant.                      Some      rein-(approximately 8
'                            cm            from    tunnel            floor)                to  strengthen                              the        structure.                          The melt generator                    mounting ring in the upper section is placed                                                                                                    to locate the exit                              aperature at a scaled height equivalent to the bottom of the RPV.
Instrumentation                            in      the cavity is similiar to that                                                              of          the 1:20th                scale SPIT test fixture.                                            The larger dimensions of the HIPS              cavity permit the number of instruments to be with              three pitot-static tubes and temperature sensors increased, 3
and two additional gas and aerosol sample tubes.                                                                              The cavity also incor-
,                          porates additional diagnostics in the form of optical probes monitor initial movement of the melt in the cavity.                                                                                                                              to vices consist of a 6 mm diameter fiber optic probe placed                                                                                                  These          de-with the exposed concrete surface. The opposite end of the probe                                                                                                  flush is monitored                            with a photodector to give an electrical output i
when the melt passes over the exposed surf ace.                                                                                  Time-co r relating the position and signal from the various probes gives melt dis-placement and velocity within the cavity.
Figure 25 shows a conceptial drawing of the assembled                                                                                              HIPS apparatus consisting of the melt generator and test article. The melt The assembled    generator mounting prevents gas escaping out of the cavity.
apparatus placed in the interaction chamber is shown in Figure 26.                                          The device is placed on the concrete floor of    the          chamber, towards the opposite end.
near        one            end with the exit of the cavity directed Ej ected debris,                                  aerosols,                          and gases              will large f ans                    be retained and measured while in the chamber.
driven by                                                                                                                                  Two borne products of air motors are used to                                          stir the                          air-the                reaction to obtain a                                          homogenous ture              for sampling.                                                                                                                                      mix-                      ,
t
,                                                                                                                                                                                                                          i 1
VII.2                Test versus Accident Characteristics a
The ex-vessel behavior of the ejected core debris represents l
a complicated interaction of chemical, hydrodynamic, and thermal processes.
of              variation      Identifying is these phenomena and their possible range necessary to insure that the experiment accurately models the accident.                                                        Table 23 presents the correla-tion of the experimental and accident characteristics, and the effect              on the outcome of the test. The results indicate that the conditions present in the experiment                                                                            will tend to overpredict the          amount            of    debris          dispersed                      f rom                    the    cavity. The f ollowing 4
paragraphs                      present                    the          logic that was used                                            to assess the 4'
effect of test conditions.
144
  --      ,, , - - - .,erv    7.>,_.,,.--.    ,,-m  ,n    ,---w    -,    , , , , , - ,    ,w, , , , -    , , - , , , _ , , _ ,          . , , , ,        - , . - . - - . , - -            . - - . - , , . - - --
 
MELT GENERATOR 9      g KEYWAY                                                    d      d OPENING                                                  '4 s
s i
                                                                                          /
                                                                  /j:d/ /
An                h..o              ff        g                          /
p    h.            I w;, h    0, L.g    -
                                &      l,      f 3                      gy,.        /                        ,
F              ~
:5            lx                -      m '6              W / -
D. .
                                                                  \_ R.      p, c
l jl
                                  'f,'k..kdN.b.9' r
[~
_                            INSTRUMENTATION
                                                                              ;y
                            \          STEEL FORM 1/10th SCALE ZION CAVITY Figure 25. HPS Test Apparatus l
145
 
L 1    i I
                                                                                                                                            )
                                                                                                  '                                                I.
r    -      -----        -
i li                                                          -
i yy, ..;    ,  ...                          .+_        ei        .
I il                                            i 2.7 m }
I ll  i
                                                            . _ _ _ _ ..                                                                        I L            .      _ .  .
A e
i li.1 l l1 f
i I
I i
5.8 m
_AR            R      R        R        R                        R                                  R      Ti'i                A 1
7
                                                                                                            )\
                                    /                                  )        =          ';
[          .+                                          ,
2.7 m i
k a.
f l c3 I                                                        *p E                      E                ]                ];
Y c=i Figure 26.
HIPS Test Apparatus Placed in the Experiment Interaction Chamber 146
 
l TABLE 23 l
                !                                    CORRELATION OF ACCIDENT AND TEST PHENOMENA AND THE EFFECT ON TEST OtTICOME j        Phenomena        Accident              Expe rimen t i                                                                                          Effect on Test Conditions on Outcome f        Pressure          1.4 to 17.0 MPa I                                                1.4 to 17.0 MPa    None.
f      Cas Volume        Primary System      Melt Generator Linear scaling approximately 1:12; slightly less total energy in blowdown gas.
              !        Cas Composition    Steam and
              !                                                Nitrogen / Carbon    Higher dynamic head due to N 2 density (factor of 10); will enhance material reseval mech-Hydrogen            Dioxide anisms. Cas solubility effects in accident mknown. N2 and CO2 probably provide upper and lower bounds on solubility, respectively. CO2 may provide good simulation of steam
            !                                                                      condensation at blowdown.
l          Melt Composition  Corium f                                                  Iron and Alumina
                                        = 7.0 gm/cm3        = 5.9 /cm3            Hydrodynamic mechanisms only slightly affected unless metal / oxide stratification occurs i                                                                      before ejection. Cas solubility, aerosol production, and debris configuration unknown, j                                                                        Freezing point of altsnina only slightly lower than Zr-Zr0 -UO2 2    mixtures. Experiment tem-perature a fmetion of non-reactive mass and time in crucible. Too high a temperature may 5
overpredict decomposition and disruption of debris configurations.
Melt Temperature  1500-28000C          1800-25000C l                                                                        May overpredict concrete decomposition compared to lower temperature range. Density and viscosity may be too low--will enhance material removal.
Helt Vessel      RPV                  Melt Generator I
Isngth-to-dismeter the same              different (2 to .35), causing higher static head; melt depth nominally
          }        ~                                                                            "punchthru" of gas will not be overpredicted. Flat bottom may enhance atomiza-1        *-                                                              tion and jet breakup.
Breach Dimensions  4-40 cm            2.5 cm l                                                                        Exceeds 1/10th scale; will cause initially higher mass flow of melt and gas enhancing (Variable) material ture mayremoval.      Thickness be slightly            to aperature radius approximately 8:1, radial growth of apera-underpredicted.
Cavity Geometry  Zion Plant                -
1/10th linear scaling; geometric features not included (ladders, sump, instrument        tubes,
        '                                                                        etc.). Should allow more efficient material removal.
Concrete          Limestone-          Same J                                Common Sand                              Aggregate    site not ration and erosion    scaler may      ; relative amount of " surface" material greater. Initial gas gene-b i enhanced.
Containment      Containment          Interaction t
Structure        Building              Chamber          Free-expansion      out of keyway; no obstruction to material dispersal. Linear scaling approxi-mately  1:13.
J Containment      Possible Elevated  Ambient Atmosphere      Pressure (4 ata)                        Creater pressure differentiel will enhance melt transport out of cavity region.
Steam Hydrogen i
Water Pool        Dry to 5-m          Dry to 0.4-m Depth                Depth              Mixing length (time to initiate) and fragmentation not scaled. Rigid walls will simulate reactor to        cavity.
transit      Crowth timeg.      of steam bubble should be scaled if steam generation is rapid compared j
,3 u                          un                                                ..
 
l The geometry of the reactor pressure vessel is with is a flanged cylindrical and dished upper and lower heads; the melta generator    cylinder pipe section with flat flange covers on end. Based on                                                                      each the present design, have nominally equivalent melt depths.the experiment and accident head to be improperly scaled,                              This causes the static pressures above 1.4 MPa.                          but this is no consequence for experiment through the        could prevent the overlying gasBut relatively  from    deep melt pool in th melt. This effect                                  "p un chin g" generation would contribute to aerosol gas    passingbythrough causingthe  portions  of the melt to be entrained by the melt pool.
melt If      in this this      manner should cause large particles to be formed. Mechanical process also occur in a reactor        is observed accident. in the experiment it is likely to A second atomization          of effect of the experimental geometry is the possible the melt caused by stagnation of the          melt flow induced of ization        by the the flat-bottom jet            geometry of the melt crucible.        Atom-enhanced          aerosol        can    result  in  a less  coherent  stream    and The    curvature        of generation        via the production of fine droplets.
the dished head of the reactor pressure vessel The            is not significantly different from a flat-bottom vessel.
radius of the Zion RPV is approximately 2.2 meters. For breach tion            diameters up to le times the initial hole size, the devia-from a flat plate is less than 2% of the head radius.
Atomization the                    of the melt induced by the flat-bottom geometry of the experiment. vessel appears to be nearly as likely in the accident as                in of          The above both                illustrates the accident      and thea situation where the characteristics experiment are In contrast, some of the accident phenomena are not well known        well establihed.
or theaexperimental range of possibilitiesmatrix. must be considered in the design of the accident is not well established,As an example, melt temperature in the range    f rom 1500 C (steel solidification)    but is assumed to be in melting)        . Approaching          the lower bound,                to        0 2 800 C (Uo2 and resistive to displacement.                                melts are more viscous The reduced superheat causes the material energy loss      to be  f  more rom    theprone pool    to forming surf  ace. a solid crust layer        by melt displacement                                              At least three of        the surface.              The experimental meltmechanisms are dependent on an exposed liquid sured, but may be as high as 3000 0 C,                  temperature has not been mea-reaction and the heat losses involved.depending on the extent of accident melt temperature                                        The uncertainty of the lower temperature requi res and the possible consequence of a considered within                                  that a range of temperatures y behavior at the high-the              andexperimental low extremestest of matrix.      Com pa rivg permits            studying the                                  the temperature ranie debris. dispersal                and otherinfluence phenonena. of the  melt condition on tue i
148 m  .
 
VII.3    HIPS    Test Strategy The results of the analyses in the previous section suggest that accident come of the event. characteristics can vary and may          change the out-tions    can also be used toCorrcsponding changes      in the test condi-nomena.                              better  sim ula te The essential step in the test procedure is tothe accident phe-the event outcome,                                                          deter-mine the accident characteristics that are most in contribute significantly. while identifying those variables that do notn teristic and its range                  Addressing    each    accident require a large number of tests.of variation in the experimentcharac-        would A basic characteristicstwo-factorial given          test matrix of the critical input more than could be handled in a convenient manner.in                            Table 9 wo total, tional factorial scheme would require at least 8 tests test                                                                E ven  a  frac-ments strategy must therefore be based on performing the The HIPS    .
that will best satisfy the objective of the                      experi-verify and quantify the debris dispersal mechanism. program: to The scaling analysis pothesized by the            in the ZPSS          (Section IV) of the mechanisms hy-initial vessel indicates pressure that    they are most influenced and temperature.
analyses do influence            not account      for gas dissolved in the melt    Theo rZPSS effects show thatof  water    in  the cavity.                                    the the debris relocationthey          may also have a significant influence onAn phenomena.
Lacking trix, the        HIPSthetest resources    to consider a systematic test ma-ing    the existence              strategy  will concentrate on the main effects.                                                      establish-of Thethetechnique dispersal mechansims by isolating only decision circuit determine        the    given in Figure 27. is illustrated by the logic ZPSS                    range of test (and accident)The circuit is designed to material dispersal mechanisms exist (or do not exist) conditions w The                                                                  .
tions the    meltconsidered (HIPS-1            to be most probable for causingmatrix              -
conditions are ass).                                              reloca tion of umed to be:For the purpose of this illustration,        the of a non-reacting material.sure values within the prescribed                            range to                                        The test art                      ructed formed        described in Section VII.1.2 excep,1cle will be identical that using The                    a layer of magnesium oxide (minimum St the cm  cavity region is form cavity of the is  constructed cavity  dimensions.by casting the MgO around a Styrofoam  thick).
temperature                0 filled with of 400 C for ten hours.TheThe            material  is then remaining    baked form        at a is then limestone / common      sand  concrete.      The provides rigid structure.              supportheating Additional      to the (>l50 ceramic C) and strength to the ove concrete rall test to drive off the water absorbed in the MgO matiserial.          done just prior to the 149
 
    \
START SSESS DE8AIS DISPERSA AT MOST PROSASLE TEST CONDaTIONS HAPS t (w/O CONCRETE DECOMPOSITION-MgO CAVITT)
NsPS 2 (CONCRETE CAVITV)
P = 0 7 2 MPs T = 24o#C ph
                                                                                                                                  \g C
NO - SOTH TESTS                          VES  Of TESTS NO VES - MeO TEST a                                                                              ,
CONCRE TE                                                                                VES DESRIS DISPERSAL                DtCOMPOSITION                DISPERSAL MECHANISMS    DESRs3 DISPERSAL MECHANtSMS die EXIST            DISRUPTS REMOVAL                                        CONTROLLED SY            DESRBS DsSPERSAL EXIST AT UPPER LIMIT OF MECNAMISMS                                            PRESSURE OR          MECHAN 94MS EXIST OVER INPUT CONDITIONS TEMPERATURE              ENTtRE RANGE OF INPUT CONDtTIONS A                            REDUCE E AT FLUM                EIERM4NE ElHSTANCE A LOWER LIMIT OF      ASSESSsNFLUENCE ~
                                                                            $NPUT CONDtTION$      OF PRESSURE                      A I
HIPS S P = 1 F.2 M*s                      HIPS S P = 1.4 MPs            M8PS 4 T < soo#C                          7 < isorC            P = 17.2 MPs 7 < isovC 3,
VES 6                      .            NO CONCRETE                                                                                      YE8 HEAT FLUX                                                                            ir DECOMPOSITION TO CONCRETE PREVENTS DESRIS                                                                        PRESSURE LEVEL DISPERSAL                  DETERMINES DESRt3                                                                    TEMPERATURE OfSPERSAL                                          DETERMINES DESRIS DETERMeNES DESRIS OtSPERSAL                DISPERSAL A
A A
Figure 27a.
Logic Decisioh Network for the HIPS Test Strategy
 
l I
1 l(A))
i
!                                                                                y 6
q(ETERMINE EFFECT Oh WATER IN CAVITY J u                                      l i                                                                            HIPS 1W P = 17.2 MPa T = 2400*C 3                                                        WATER FILLED CAVITY U              ** WAS MATERIAL REMOVED FROM CAVITY?
YES - WATER ONLY              g i
V        NO YES - WATER AND DEBRIS P
P WATER PREVENTS OR d/n CAUSE                                                  WATER HAS NO l
!                                            DEBRIS REMOVAL                                    EFFECT ON DISPERSAL MECHANISMS Figure 27b.
Logic Decision Network for the HIPS Test Strategy        ,
v
 
The second
  ,          cavity is concrete.          test is identical to the first except that the ly                                  The two tests can then be compared direct-(                to dispersal. show    Itthe  influence is of concrete decomposition on debris                                                            !
the basemat should                expected      that reducing gas generation from disruption of the dynamic debris    inhance      material removal b configuration. y minimizing the                The comparison willthe by    also  show the concrete          extent of gas and aerosol generation produced decomposition.
Following        the second test, sion      block      to decide subsequent the        tests.            logic path leads to a deci-If  material was not removed then the removal mechanisms do not exist.f rom the cavity during both the placed in HIPS-1 and not in HIPS-2,                                                            If the melt was dis-crete                                                                  then the influence of con-wise, decomposition on removal mechanisms requires study. Like-if material was displaced in both tests, bounding condition for removal must be established. then a lower Test HIPS-3 is designed with minimum values for pressure and temperature to determine if material removal occurs for the lower bound of the accident conditions. If removal occurs at the these values of            pressure and temperature,                              then                the mechanisms assumed                                                                                                                            are If material tional is not displaced in this test, then one or twoto experiments be active o addi-are required to determine                                        if temperature is the dominating influence.                                                                  pressure                or The through 3)          results f  rom the HIPS tests numbered 1 through 5 (or will  establish the principal main influence on1 debris dispersal. The data f rom the tests will be used to model the processes port,        concreteof debris          dispersal, aerosol generation and trans-fission product source term. decomposition          and gas production, and the be    used in conjunction with these results to aidThe Phase                                              II SPIT data will also appropriate scaling criteria.                                                                            in developing The    second      portion of the decision circuit considers the effect      of  a    cavity  water    pool on debris dispersal.
SPIT matrix is planned to investigate the influenceTheofPhase                                                          II water using the SPIT    structures constructed of benign materials. The results of experiments will determine the differences, if any, between the resulting    fullyinteractions.
and partially-filled cavities and the magnitude of between          the two cases,              If significant differences are noted consider the two situations.thenThe                    the logic HIPScircuit                        test matrix incorporates  will also both classes of experiments, will be        based on the analysis of the SPIT tests.the decision to pursue both path Test HIPS-1W is designed to study the condition where melt is ejected under high pressure into a " water-locked" cavity. The objective of the test is to confirm the ZPSS hypothesis that melt will    initiate a steam bubble causing explusion of the water f rom the cavity.
the spectrumBecause                    this test represents a new condition within of possibilities,          all previous paths in the                                                  logic i
t 152
 
circuit of previous outcomes.are subsequently directed along this branch,regardless If necessary, ducted at identical conditions except for a shallow pool depthHIPS-2W    .
The    HIPS    test definitive                  program described above      will    not  provide dent            information about all possible combinations of conditions,                                                    acci-most    critical to the debrisbut onlydispersal those main f actors that appear to be tests                                            mechanism s.
factors will permitat the extremes of the spectrum of possibilities for theseConducting of potential test outcomes and modeling to be initiated. statements to 1
a 3
                                                                                        /
1 i
1 153
 
VIII. Summary The Zion Probabilistic Safety Study represents an extensive and innovative Zion plant.          analysis of the risks involved in operating the Within L
the document proposesthe context of the ex-vessel interactions, result in removal The resulting                of the iscore debris f rom the cavity regiona nu distribution ting plant mechanisms.                    considered coolable by the exis-        .
The to experimentally and analytically studyHigh Pressure Melt St the phenomena as-sociated with the ex-vessel accident sequences.                Tests are plan-the entire range expected in the accident.ned              The at two different form a data base for describing the ejected jet results will concrete interactions, debris removal processes, stream, generation.
and aerosoljet-The    System signed    to            Pressure Injection (SPIT) test matrix is de-during melt ej ection. the jet behavior and aerosol generation characterize generated      under pressures Thermitic      melts of 1.4 to 17.0    up to 10 kg will be MPa.
and nitrogen                                                    Carbon dioxide solubility on jet and aerosolwill be caed to determine the influence of gas is used to develop the                    beh avior. A statistical approach ence of                            test matrix the significant variables            to insure that is considered    overthe    influ-of conditions.                                                        the range investigaTheting SPITthematrix      will also    include  a limited    number testsof interaction    of the  melt jet and water      pools.
Theequipment and      results of for these  tests will be used to design instrumen t ation subsequent,    larger scale tests. The data will influence of water pools on the debris dispersal proces                    .
A 1/20th      linear tested using the SPIT apparatus. scale model of the Zion cavity will be sition is used to construct the test article.The prototypic concrete compo-performed        at initial                                    Thedispersal test will be the debris out        of theconditions    most likely to cause cavity region.                                    of The results from conducting the                    the SPIT testina provide the        basis matrix concentrates      larger scale HIPS experiments.                          for on verifying the existance The          HIPS    test dispersal mechanisms over a range of conditions. of the debris HIPS apparatus represents a 1/10th linear The geometry of thecavity.
Zion                                                                scale of the mitically 17.0      M Pa.
generated iron-alumina mixture,The melt generator contain the varibles ses.        Placing the  needed        to support verification and appa ratus                                  analy-modelingT allows                                      in a large process.
retaining and sampling the products interaction of the          chamber interaction 154
 
contributing to the debris dispersal mechanisms.The sion                                                      A logic deci-HIPS tes tions. circuit is used to determine the progression of test condi-within the logic circuit.The influence of water within the cavity is also included The results of the SPIT and HIPS tests are expected          to  be used in gation,    developing analytical tools for describing the jet propa-aerosol generation, and debris dispersal processes.          The polate the phenomena to reactor scales.              experiments will also pro Ultimately, w ill  be                                                      the models and differing system conditions.used to predict behavior for other cavity geom I
r
                                                                                    ?
0 t
155
 
t APPENDIX JL Phase I SPIT Test Program high pressure course  of the program,  melt generation and deliveryDuring            technique.Th the a
unexpected      phenomena requi red    the design of specialized instrumentation    were  observed and      that the    tests    to become                                                  caused progressively      more complex. This section principally concerningprovides a brief overview of the test results with th aerosol sa m pl es.                the nature of the jet stream and the given in Table 24. A listing of the pertinent test details is A.1  Jet Characteristics have  The a    nature of the jet emanating f rom the reactor vessel will within      direct  influence on the subsequent behavior of the melt the cavity.
forms the                    The initial shape and velocity of the promote in  the ZPSS  removal analysis ofisthe      material from the cavity.The dynamic to assumption
                                                                                            . deb formed so      that no that a stable,      single phase jet is the cavity floor.          Ageometric    expansion solid jet maximizes the  occurs during passage to loading that,        in turn,    causes the material              unit pressure ly across the cavity floor.                                  to traverse rapid-this report suggests that gas solubilityAn analysis given elsewhere in may            the material leaving the vesseland mechanical breakup causemixture.
two-phase                                                  to be an unstable, streamTheisPhase I SPIT tests show that the appearance of the                melt Figure 28 is a series of representativeradically different              thethan that presume ZPSS.
the    SPIT-3    test.
The times stated are referenced to thephotographs taken du appearance of the jet and are accurate to i 0.02 second. first                  The melt jet at 0.05 divergent              second hal cone (35-40      is characterized      as a highly  luminous The brightness of the cloudf-angle) emanating f rom the vessel.                  ,
other than the outer shape. prevents resolution of any details melt. The material within the cone ap-pears to be vaporized apparent,        but                          At 0.1 second, the cone is still condensed and darkened.the lower portion is masked by material that has deflected      off the brick bed      Some    of the condensed material has as indicated outward      direction.
sec), the aerosol cloud        Approximately 1 secondby later  its upward and (t = 1.15 The dark-brown appearance indicatescompletely envelops the apparatus.
significantly.          The behavior              that the material has cooled discernable.                                within the cloud interior is not Figure stream taken during two separate SPIT tests.29                        shows    two flash X-ra the melt The experiments were conducted at nominally the same conditions except for the 156
/
 
TABLE 24 Details of SPIT Tests Melt Test                        Mass Date          (kgs)      Pressure (MPa)
Initial        Final      Gas SPIT-1                                                                      Comment 25 May 82        2.15      Amb            0.69        Ar                                                            -
SPIT-2    26 May 82                                                        No melt thru, relief valve set at 0.69 MPa 2.37      Amb            0.67        Ar Relief valve opened prior to melt thru, gaseous melt.
SPIT-3      3 Jun 82      2.50        1.03          1.88 N2        Cas did not vent, SPIT-4                                                                      3 directions.        melt " sprayed" sequentially in 23 Jun 82      10.35      1.03          8.10        N2 No melt thru, relief opened at 7.24 MPa, 350 gm SPIT-5    24 Jun 82      10.35                                            La20 3 mixed in thermite.
0              ?
G                                                                        N2        350 gm La2
''                                                                                              3 Mgo brick bed.0    , no pressure record, instrumented SPIT-6    23 Jul 82      10.6        3.45          5.16        N2 SPIT-7    20 Sep 82                                                        Pressure vessel failure 10.4        2.99          4.77 N2        10 cm deep Al 0 SPIT-8                                                                      removed bed. 2 3 gravel bed, melt penetrated and 29 Sep 82      10.4        7.24          7.62        N2      30 cm deep Al 0 2 3 gravel, melt penetrated 20 cm into  bed,  no calorime te r.
displacement    of particles. Steel SPIT-9      3 Nov 82        5.0        3.45            ?          N2 SPIT-10    10 Nov 82      5.0                                            Real time x-ray of thermite reaction 2.52          3.16        N2 SPIT-11      7 Mar 83      10.0 Real time x-ray of thermite reaction 6.32          8.40 SPIT-12                                                            CO2      First CO 11 Mar 83      10.0                                                      2 test, no trigge_r on x-ray units 5.71          5.52 CO2 Coherent stream with ligament instabilities, max SPIT-13    17 Mar 83      10.0                                            heat flux 400 cal /cm2-sec 5.13        17.24        CO2 fragmented melt. Extreme pressure increase during o          -      -
 
\
TABLE 24 (CONT.)
Details of SPIT Tests Melt Mass    Pressure (MPa)
Test      Date    (kgs)  Initial      Final    Cas  Comment SPIT-14  6 Jun 83  10.0    1.59          1.68    CO2 SPIT-15 Lowest pressure value, large aerosol generation 27 Jun 83  10.0    6.87        11.67      N2 Solid water pool, jet penetrated water at approx.
16 m/sec, no vapor explosion, box destroyed SPIT-16  22 Jul 83  10.0  9.08        10.54    N2 1/20 scale aluminum cavity, water filled, melt failed to vent SPIT-17  25 Jul 83  10.0    8.76        10.21          Repeat test N2 16, over 200 MPa pressure spike in water pool (approx. 13 msec after jet entry),
cavity destroyed SPIT-18  9 Nov 83  10.3
  -                            10.62        12.26      N2
  $                                                        Alumina brick cavity, placed in 42 m3 chambe r.
approx. 58% of debris dispersed from cavity, several psi overpressure of recorded in chamber SPIT-19  16 Dec 83  10.3  10.76        12.77    N2 Scaled (1/20) Zion reactor cavity, limestone / common sand concrete, placed in 42 m3 chamber, significant overpressurization of chamber caused extensive damage. Approx. 95% of debris removed
 
d I
i m,                                                                                                  .
i
                                                                                          -                                                m-
;                                                                                                                                                      c r=-
                                                                                                      -                              .Y
[,).                                                                              ;
                                                                                                                          , , 3.,
i,    >
f,. ' I:
i                                              _-                            !!c 3
t = 0.05s t = 0.1 s MELT EJECTION BEGINS VAPOR CONDENSATION t
g                      ~~
                                        %91
                                                                ,g
(                                                                                                                                                  . J p.
7Jv.
                                        ~        :      ~-
                                ..;; g.
p "q, . . ,. +.-. x
                                                                                                                  & a? ; . ,
pc                                .<                --
i
    ~                                                                ,,                                        .
e                                                                    .
                                ^. g: q'W t = 1.15s t = 1.95a MAXIMUM AEROSOL CLOUD RESIDUE OF EJECTION l
Figure 28. Melt Ejection Sequence 159
> -                                                                    __ _ __ _ __ _ -                                            - ~ ~'--~~~- ~"~ ~
 
3 l
i
                                                                            ':j    '
                                                                                .j r                                                .,,
F
                                '                                                                                                                                            l
                                                                                ?)                                                                                          ;
e                                                                                          i 2-                                                    1 t
g                                                    .
(                        -
j g                                                                                                                                '
O                    '
                                                - .          ^n  .
                  *,,.4                            - '
_j C,t. r. ' '
                                                          ~.' ?I                                                                    ~
v.d,":: ..~';..;3  .
c-#. ..,. .                  .....4
                                                                        .a  .                                              ' -
                                    '5%3.i5. N ,
y.<,                .
1
                                            @J ,'YM                                                                                            .,
                              'wq . . .+. wa. ,                                                                                              ,
                                    - i>(Ps 3                                                                                  .
j TEST 8 - 7.6 MPa N 2                                                                                                                                          ii TEST 11 - 8.3 MPa CO 2 Figure 29.                          Flash X-ray Photographs of Jet Stream                                                !
1
_ __      ._                z                      ^*
w eensh-                    _--- -.
 
type of gas.
carbon dioxide.        SPIT-8  Thewas  charged with nitrogen while SPIT-ll used outward similar    to that seen in Figure 29.apparance of the jet was very is referenced to a signal                            The time given in each photo placed under the melt plug. generated by a break-wire trigger melt is estimated to be approximately 4.5The response          ms. of the trigger to the Four x-ray units are matelg typically 2 meters      used f rom onthe each melttest,    located stream  and at a radius of approxi-ly 25    apar t.
Film cassettes are placedpositioned            approximate-diametrically opposite the corresponding X-ray head, stream center.                                  approximately 0.46 meter f rom the break-wire signal.Both photos show the melt jet at 45 milliseconds The large nuts securing the lower meltafter the photo from SPIT-8 shows slightlyator flange are obvious The                                                                                  The in the u the                                                  over half of the melt stream.
boundary of the stream is clearly defined, suggesting that surrounding significantly less dense. cloud seen in the photographs of Figure 29 is on the order of 10-120                        The half-angle of the j et stream is and the time of exposure gives an estimated velocity of 20 m/s The      estimated velocity using      Bernoulli's                          can be compared to that obtained isentropic fluid flowing through an orifice. equation to determine          behaves if  asthe jet
                ~ 2(P  o - Pa ) .1/2        2Pog  .1/2 P
                .            F      .        .
P F-where:
Po    -P a      =  Pog a pressure in the melt generator (gage pressure) pp = density of melt ( fluid) 1/2 U=          *    *
                                                      =
                  ..      5900 kg/rd      -
16.1tt/sec For    this example, is assumed to be unity. The            the comparison discharge coefficient of the orifice the~ 1arge uncertainty in the experimental value.is resonable considering As stream        the jet expands and moves downward, appears                                                the density of to decrease,                                            the vcids within the structure.                      as evidenced by the appearance        of in the the        form of horizontal bands is apparent in the lower third ofA patter stream.
These bands may be caused by Helmholtz instabilities 161
 
i i
with the surrounding environment.that occur as the internal pressur
(
Helmholtz    instabilities flow relative                            arise when two immiscible fluids to each other alohg a surface of separation. There exists    a bance          maximum relative velocity above which a small distur-on the surface will grow and amplify.
can be approximated 'using a correlation (Ref. The SPIT geometry liquid jet and an upward vertical vapor blanket.36) The        for a downward velocity of propagation of surface wave is given by:
omgc                AA C2  ,            _
gv              2 P+A                          y t    v        (Ot+Av) 2 (V - Vg) where:
o = surface tension of liquid m = 2 /A wave number gc = proportionality constant A = wavelength of disturbance o,p g    y = density of liquid and vapor V,V g      y = velocity of liquid and vapor The condition for a stable jet requires that C 2          2.8, or 2oge          Pt Dv
                                  -                          2 A
( pg py)    (Yv~ Yt I    > 1 This criteria can tha t o = 1989 dyne /cm, Vtbe =evaluated            for SPIT-8 by 28 m/sec, Vy = 0, and        assuming that  the wavelength of the observed disturbances is on the order of 2 cm.
99
                                      -=    0.65 < 1 The result suggests that for the jet in the SPIT-8 test, the relative bility      velocity of the stream is high enough to induce insta-on theeven should become          surface    of the stream.        Instability ef fects due                          more pronounced at higher driving pressures term. to the dependence on the square of the relative velocity CO 2 -driven melt.The second radiograph in Figure 29 shows the appearance of a coherent                  In this case, the jet appears as a nearly the surface. stream with ligament-type instabilities emanating f rom No large voids or lower density areas are obvicus 162
 
within the stream, The 8  withestimated the high velocity is nearly equivalent to that                    seeas would system pressure.er density apparently of f setting the increased        n in SPIT-A.2 Aerosol Generation Cascade tests 3,    5, 6,impactors and 8      and filter samples were employed on so that they are enveloped byThe devices are placed near theSPIT                  apparatus controlled                                    the aerosol cloud.
at predetermined time intervals. valving permits samples to be drawn in activated                                          Ty pi cally, pates.          at ignitionimpactors Cascade          and closedallowafter the aerosol cloud dissi-the devices a according to aerodynamic                              the sample to be sized diameter based    on seven size ranges incorporated into each device.
mass of rate allowsaerosol over a fixed time period. Filters are used to obtain a Using a known flow rial from both types ofinferring a mean concentration.        The captured mate-to determine speciation. devices can be used in chemical analyses Cascade      impactors of the type used on the SPIT most    efficient    in than one-half micrometer  sampling      aerosols in the range of tests        areles slightly    s diameter).                              to about 15 micrometers carried      throughThe lowthe  mass  of smaller particles causes (aerodynamic surface.        Large              device without impacting on a them to be momentum      may        particles behave in an opposite manner;      collection cloud into the apparatus.be too great to be deflected outof                    their the large    size                          Filters are similarly affectedaerosol size range. range but are more efficient in capturing the                    in the lower Cascade Figurediameter.
particle      30 in the form of f requency                    of occurrence as a fim distinct        modes,            The curve fit                          unction of The smaller                  0.5 and 5 micrometer aerodynamicto the data suggests two the vaporized mode    species is assumed in diameter. of to be formed by the condensation the melt j et.
up of the melt, The            larger mode is attributed to mechanical brea of the melt at the exit aperature of the vessel.possibly by fragmentat As the jet enters the atmosphere, exposedf ragmenta dynamic      to a hightion. relativeThe velocity  gasthe  surface material is stream    that causes hydro breakup                                                                                          (
process until a stable fragment size isdroplets can this situation,                                                          multistage undergo a          -
achieved. For the fragment diameter size is given by (Ref. 13):
                                      "        -2
                                ,          V d  =  We c py2 1dV g,
163
 
150          .  .....y          . . , . ni          . .... sir 140 l,
TEST 3 130      -
N2 AT 1.90 MPa                    l                  -
120    -    NORTH OF NOZZLE                    !
8.4 mg i
i g
w 110      -
I 100    -
e                                                  8                  -
0    90    -
l o                                                  i                -
o                                                  i 80  -                                        i N                                                  i                  -
  ^                                                  I E    70  -                                        I 5    60  -                              ~
8 l                -
l 50 -                i i
                                  ;                l                  -
i 40 -              8 i
8 1                  -
I      i 30  -
i      i 1
I      i                i 20 -              !      l                                  -
i      i 10 -                                                        -
0          ''''''"'          '''''''''            ' ' ''''"
0.1                1.0              10.0                100.0 Dp ,, ( m)
Figure 30.
Size Distribution of Aerosol Produced During Pressurized Melt Ejection 164
 
where:
We c  =
critical Weber number d    =
maximum complete      fragment size when the processes          are V d=
fragment process      cloud velocity af ter the f ragmenti        ng o;V g    g=
o=
density and velocity of the gas stream droplet surface tension This particle    size equation accounts for Weber number reductio              n b plying typical values reduction      andthe f rom    relative  velocitytest SPIT-type        reduction.y both    Ap-atomization micrometer        process size range. can  yield  stable    particle s indicates that the pre-filter resul ts.                            Material recovered f roms      thein the impact10-orto 100-size material Microscopic examination of the particlesuse                        these of nominally was          65-micrometer aerodynamic shows          di a mean sampler    probably not      found because    on the impaction            ameter.
stages            This ciency ,for larger particles.of the device's low                            of collection    the  effi-Electron geometry of the aerosol particles. microscopicdetermine              examination is used the three particles of        sizes aggregates mentioned above (0.5-micrometer                    are n Fi shown i average        diameter) gure 31. R The smallest micrometer            of small par ticles        (<0.1-micrometer)        appear to be composed during                          probably    represent primaries.      The theaggr                0 .1 -
collisions,the  condensation while                process    with  the      material      formed lost during                                                    egates particles          the sampling.the individual particles settled        created out by appear                                                      or    were to be principallyThe singl5-micrometer e,        and 65-micrometer fconsistent ra gmenta tion.with    the proposed generation mech          spherical    shapes, anisms of aerodynamic smallerEnergy dispersive spectroscopy of th bearing. size ranges to be selectively either irone                  samples or alumin um-  twoshows the None    of  the  samples and iron coexisting in the same particle      that  were  analyzed iron-bearing particles were observed thaQuantita lack                                                    .          showed aluminum more may be due to the hiof aluminum composition                            Theparticles relative in the sm alumAnum (Figure 32).gher vapor pressure of iron compareder                                  size rang iron    and results of    aluminum        within The the  particle 65-micrometer samples    do show both to The th e m elsilica
: t.      and zirconium are assumed                      tthis typesam 65-micrometer        ofplanalysis e.
on the thermite charge prior to the testThe lanthanum .                  orm placed is  in derived fr 165 i
 
l l
j                                                                                                                                              i 1
                      ~
i
\
6
                                  . ,. m r
A    -c                          -
g                                  '
                                              .L T'
r                .n          b.                  -
t I          [''$ j          .
r r    %:,          '
                                                                ''                                                . h)t.'              -
DX%              '-
m.
1l'
                                                                                      ~                            ..
i
                                                                                                                            -"[
:    5 7 2s                seu i
i l                      Dp =65 MICRON                                  Op = 5 MICRON                              Dp = 0.5 MICRON            !
Figure 31.
Electron Micrographs of SPIT Aerosci Samples                                    ;
l
 
1
                                                                                                                                                                            )
l l
105          , ,, ,            , , ,,                        ,, ,
9      10 4            -
                        *a                                                                                  -
W                                Fe m                                                      Al2O3 M                3 m      10            -
E
                      <                                    VAPOR PRES SURE
                      >      10 2                            DATA FROM CRC HANDBOOK ON                                -
CHEMISTRY &
PHYSICS 49th EDITION 101
                                                        '''''''''                                    '2 2000            2500                  3000 TEMPERATURE (K)
Figure 32.
Vapor Pressure of iron and Aluminum Oxide i
as Functions of Temperature i                                                        167
 
1500                -
Fe 1125                -
en 750              -
0 Al Si 375              -
Fe O                  J " "'" W                      ~
                                                                -      ^
O                5                          10                            15          20 ENERGY (kev)
Figure 33. Elemental Analysis of 5-micrometer Particle 168
 
APPENDIY 3 Gas Blowdown of SPIT Apparatus The pressurized mel t generator, pl umbing.                      accumulatorcomponents of the SPIT apparatus            are gases                The                                reservoir and                          the liberated and heated during the thermiteaccumulator                                e      p burn. The gas volume the reservoir connecting          of pipin the volume    melt is      generator approximately          2  8 is x 10-estimated
                                                                                        . x 10-3  3  to bg 5 m and The m.
the two volumes g represents a neglible contribution inThevolinter-interconnectingshould        tubingnotmay  be be considered too to be additive um e simultaneous blowdown.                                              restrictive to because.
insure isentropic flow analysis of the discharge fThis point can                                      be illustrat and the flow through the interconnecting pipingrom the melt g                .
batic expansion    r                      through an orifice                      is35):
given reversible y (Ref.
bFor an adia-(Plt k-1/k f/(P 3 k-1/k l
2      k                                        ) 1/2 2D                                        1 ,
R ( k-1        h) h/
                                                                          ~
2 i 1/2 Z/k (A'2        (Di 1\-l \-/
( Al
                                          \
(P1
                                                    \
where:                                              ~
C    =
orifice discharge coefficient
                    =
A2 area of orifice P      =
2          ambient pressure R    =
gas constant T1      =
k    =
temperature of gas in the melt generator ratio of specific heats P      =
1 pressure in melt generator
                  =
A1 cross-sectional area of melt generator The discharge coefficient (C) final value depends on the ratio of Al/A2can vary from 01.0to, the through the orifice, and other factors                        .
type of gas, velocity i
169
 
I Flow pressure,      through and      elastic a pipe forces. is governed by the inertial, viscous, Assuming isothermal flow the mass I      flow rate through the interconnective piping is given by (Ref.
35):
(
2        2
                                                        \ 1l2 (Pg -P 1j i
D "-
2RT
                                                            -1/2 4f'RTL in          +
gA2      iP 1 gg2D where:
P,o  P1      =
pressure respectively in accumulator and melt generator, P,o  D1    =
density of gas in accumulator and generator D, L, A      =
diameter, length, and area of pipe, respec-tively g    =
acceleration due to gravity f'    =      pipe friction factor flow  Ifcan the Reynold's be number in the pipe exceeds 4,000, turbulent Colebrook equation (Ref. and assumed          35): the f riction factor is given by the Y2 f'    = -
1
                        -=
Y 2 logig i /c/D          +
2.51i 1
( 3.7          RY e /
where:
c  =
empirical value for surface roughness Re    =  Reynold's number                  =  y V  =  velocity of flow u =    viscosity of gas In tus,  the orderaboveto  obtainequations three      the blowdown    must be history of the SPIT appara-solved sim ultaneously.
170
 
The initial conditions of the problem are temperature, and type of gas in the pre given by the pressure, sure in the generator at any point in time        ssure is system.
det            The pr es-plug orifice                                                  ermined by the amount of          mass    that    has    flowed the interconnecting plumbing.minus the mass that out    of the rough      thevessel has    melt entered th th via the following values are selected foror thFor                  of the the    system, purpose e base case:
P1  =
15.2 MPa      T1  =
1670 K      T2    =
A2    =
5 x 10~4 m 2                                      330 K C    =
0.8 The    sensitivity the variables is illustrated by comparinof the calculationofto chang to the variation in the system paramete r s.
values      is assumed to be representative in the experiment.                                          ofThetheg      the base range          in    cas the                        In Figure 34a,                  possible variation blowdowneffective      flow  area are varied. the times are inversely                  As discharge shown, coefficient      and parameters.                            p                      the    calculated not                In all three cases,roportional to the change              in the tor.
resistivechange significantly during the blowdowneof                      t  thedoem lthe generas            ac The small flow area of the interconnecti the melt generator.to allow significant gas flow from the acng plumbing is cumula tor        to kPa in the accumulator pressure is predicted fA pressure                                    decrea Figure 34b provides a com                          or these cases.
tions in the perature    and gas s temperature,parison          to the base case for varia pressure, and species.                        -
of the system,pecies are significant contributorsBoth in    both    flow equations. basically due to the influence to the tem-response of the (RT)
The major effect of the                  product the mass in the system. system pressure is to cause                    decrease a longer in      disch down response of the system. accumulator volume does not To u e to the blow-of the system, a calculation was made length.sidered to have the same volume, n the where thfurthe response e piping was con-maintained                                  but of an infinitesimally small In this hypothetical situation, the sy tbut the discharge ,through illustrates                    the orifice in the melt          s em volume is the                                  genertor. vailable Figure for 35 and the base case of Figure 34.comrarison between            the hypothetical    situation Figure 35 significantly        to shows that equilibrate                the time required for thethe additional volume contributeso
!'                    with the ambient c                        melt generator i:  rate is the same for both cases,onditions.                Because    the  mass    flow to              u the interconnecting piping is functioninthe comparison illustrates that g as an acoustic low-pass 171
 
i i
i                                        15 I                i I              I 14 j                                                                                                                                  -
4 13                                      1-BASE CASE: C = 0.8 A2 = 5 x 10-4
!                                                                                                                          m2      -
12                                    2-C = 0.5 A2 = 5 x 10-4 m 2 I                                                                                                                                  -
11                                    3- C = 0.8      A2 = 1.2 x 10-4 m2 m                                                                                          -
j                                      g 10 i
j 2
v i
9    -
'                                    LU                                                                                        -
u) 8    -
4 M
y    7  -
:                                    E                                                                                        -
0-  6    -
3 i                                        5-j                                        4-3-
2-1  -    1 0                                      '                  '              '
O.0              0.1                0.2                0.3            0.4        0.5 TIME (s)
Figure 34a. Calculated SPIT Pressure Vessel Blowdown History for Variation in Orifice Size and Discharge coefficient
 
i K                                                                                4 0                                                                    I          2 7                                                                                0
                    ,                6 1
i n
                                      =                                                                    l n
2                                                                                                  i o
                  ,                T                                                                        ~      0                    i t
a r
2                                                                            2                      a N
V 0                    r
                ,            a    :                                                                                                    o P E                                                                            ~                            f y
MS                    K                                                        '                            r A                                                                                                      o 0 C                    0                                                                                    t s
                ,                                  5                                                                                  i 7                      5                                                              6                    H E
                            = S              2 =                                                      '        1 n
w 2 A do re 2                                                            0
              ,            P B                    T
                                                                                                                        )            wta    u
                            -    -          -      -                                                  I                  s        l o      r 1    2
(          Bep 4                                                                      E l
e m 2      M            s se
.-                                                                                                  I 1
I T          eT 0                  V d
          ,                                                                                                                        e na r
u s s
                                                                                                    '                              sie ec
                                                                                                      ~                          Pp e r
8                  TS 0                  I P e, 0                Sr u 4                                                        d s es are t
uP I                            l 3
l cs aa 4                CG I          0 0                  b i                                                                                                                        4 2                                                3 3
I                              e r
u g
                                          -      -    -                                                                    i
                                                              -  -      -                                0                F 5 4 3 2 1            1      1 1      0 g        8  7 6 0
1        1      1                      5 4      3 2    1      00 mg3 u$m@xo.J 4u
 
                                    . _ _ _  . - .- --                  - - - -      - - - --      ^ ^ "
s 15 i        ,            ,          ,        ,      ,            ,          ,    ,
                                                                                                                ~
13                                      1- GENERATOR AND ACCUMULATOR CONNECTED BY 12-mm dia. TUBING                        -
12    -
                                                                                                                ~
2- GENERATOR AND ACCUMULATOR 11    -
      ^                                                    CONNECTED BY INFINITESIMALLY                        -
g  10 -  -                                          SHORT PIPE 2                                                                                                        -
      -    g.  .
m                                                                                                        _
8-
      @  7-N
* E                                                                                                      -
1  6-                                2                                                                -
5-4-
3-2-
1  -
0          '        '            '        '          8      8            8 0.0    0.1      0.2      0.3        0.4            0.5    0.6          0.7    0.8    0.9      1.0 l
TIME (s) i Figure 35. Effect of System Volume on Blowdown History for Variation in the interconnecting Piping
 
filter, wise, cant    the large orifice in the melt generator                      .
preventpr Like-s signifi-from  additional the accumulator.pressurization af ter blowdown by theasflow atms    is implication  maintained    until the accumulatorAnisequilibrium pressure        2 of ju should of    these                                discharged.
results is that the accumulator                The the system volume.not be included in calculations involving a volume blowdown      of 1/40 linear scaling ratio to the Zion primary                      approximate syste means that the SPIT apparatus may underpredict                m volume.
the pheThis are dependent on the duration of the blowdown sequence              nomena that analysis40of the mately              primary seconds    are  system blowdown required  for        indicates that The ZPSS sient se            (P o =                    equilibration in a approxi-how ever, q uence              17.0 MPa) .                                  tran-and to      the composition maintain    the  proper    time of the gases (steam scaling,                    .
In and e r, hyd order last approximately 2 seconds.                    the SPIT blowdown should than  what underway to  is predicted    f  rom  the flow analysisThis    time interval is far Efforts are generator in order to overcome the present limitationim          .
e 175
 
APPENDIX E Jet Stream Heat Loss During the melt ejection process, the temperature of the stream is governed by the energy losses that occur during the propagation through the air.          Energy is lost from the jet stream by radiative and covective heat transfer to the environment.
The energy losses can be estimated by assuming the jet stream acts  like a small surface totally enclosed by the surroundings at ambient temperature.
expressed as:
The net rate of energy exchange (q) may be f4      4 i 9                        -
T        he Astream  (Tg C 4trean o (T3      sur[+                      -Tsur) where:
c = emissivity of melt Astream = surface area of jet a = Stefan-Boltzman constant T3  = jet surf ace temperature Tsur = temperature of surroundings hc = convective film coefficient Performing    an  energy balance on the      entire  jet  geometry shows in change    that thethe  energy stored is equivalent to the net energy system:
bin - bout = Estored The    sin term represents the mass influx into the system as jet ej ection    continues; the E        te and convection from the surYbce.rm is the loss due to radiation by defining a control volume represented    Calculations can be performed by the area of the jet of an incremental length. In thiscross-sectional manner,  the Ein term can be neglected by assuming no net heat or mass transfer from adjacent control volumes.              The energy loss then occurs at the outer boundaries exposed to                the ambient atmo-s pher e. Thus,
                                    $out " -b stored where    the minus sign represents a decrease in the initial stored energy.      This equation can be expressed in terms of the jet stream parameters as follows:
176 u.
 
4 cA30 Ts      -
Tsur              4 e
where:                      +hAs (Ts-Tsur) =            - pc.
j    (Acs dX)    dr E
dx = incremental length p = density c
p = specific heat t = time A
cs = cross-sectional area of the jet = Y Ag=
exposed surface area of the jet = WD dx                  ,
D = the average diameter Rearranging to remove common terms:
dr            4
                                  ~
4 E"        -
op          CU s
T 4
sur
                                                        +hc (T3 -Tsur) a This of time for an incremental control volume The equation sions    of assumes that,                                              as a functionwithiequa To  determine  thethe  jettemperature and the propertiesof the  str of the matat athe      given dimen-point in tim, erial  are the relationship of D, c , c p and h                                                  fixed.
e    eam  as    a  function    of  time The discussion of                              with time must be known,.
showed that the melt stream may divergethe any                                                              testsprevious (Appendix  SPIT
                                                              ,                                  A)
,i angle remains constant, then the dx                                      e. term        ipoint is a t f If the cone
;1    scheme tain  a    must decrease with propagation constant mass.
dista n the    control volume                r c
assuming constant density causes the area                                tTaking      thisand behaviornce into      account erms to take the form:
h " ~WD(x)4 2^                                                                          '
Ag    =
nD(x)dx the term D(x) is given as:
D(x)  =    Do    +    2(x tang) 177
 
where:
Do = diameter at aperture x = propagation distance 0 = melt stream half-angle For    the initial portion of the stream,              including the front face of the propagating jet, the exposed surface area becomes:
As  =
wD(x)dx +
one      The jet is then characterized using two different equations, jet.
f or the initial portion and another for the remainder of the The two equations can be solved numerically using a forward differencing representation of the time derivative of the form:
Initial Stream Portion:
At    !4 +
i+1                (        dx /        /14      4
                                                                            ~
Ts    "    -
i            i p
C"(Ts    Tsur + bc (Ts -Tsur)  +T s Main Stream Portion:
                              ~
i+1            4At          4 4
                                                                ~
CU/1 i
Ts    " -                T    -
T    jh                + T3 i
pb        \g          surf c (Tg -Tsur) the superscripts represent the time step interval as given by:
t = 16t: i = 1 to P at is established by:
A t = x/PV jet where:
V jet    =
velocity of the jet stream (assumed to be constant) x  =    total propagation distance
,                        P  =
total number of increments I
178 r
 
The T,    equations are then solved independentlyby s  T sur,c,P, c p,        he                                                    ini tializing and dx to give T s
* second                                                                          for each case. A dx value is then calculated using the volume size and an assumed expansionconstant                                            angle The (O)control process continues by setting Th equal to the just calc l above    equations        and                                                  u ated Tk+1 in the a new set of Tk+1 values are The procedure is continued until the                                                              obtained.
equal to the total propagation distancesum of the dxi increments is to fvalues assess rom the  SPIT the    extent of temperaturee loa experiments.                                                  aboveparametric procedure    stu ss that could for emissivity factor. coefficient and cone , angle to initial meltThe procedure involves varying thebe ex temperature, velocity, film The base  T i=0 case for the calculations
                                            =
:            isdetermine the g          3000 K Tsur  = 300 K c= 0.8 h=c    20,000 W/m 2K O= 120 given in Table 25.The range of values for the five vari b
.                                                                                          a les considered is TABLE 25                                                                  .
Parametric Values for Jet Temperatur e Calculations Variable Initial value                      Low i
Ts=1                                                                                              High l
3000 K Tsur                                                      2500 K 300 K                                                                  3500 K 300 K 0.8                                                                                        g I
Velocity                                                    0.5
'                                  45 m/sec                                                                  0.8 2.8 (Gravity)
Film Coefficient 20,000 W/m2K                                                                          76 1000 Cone Angle                                                                                        50,000 120 00 I                                                                                                            20 0 179
___                __ _ _-                  _ _ , - _ _ _ _ -        . . _ .    -        _      _J .
 
Calculations were made for separate situations wherein base                                                                                the range. case was modified to exercise one variable over its entire The results of the calculations are given in the form of non-dimensionalized          relations for graphical comparison:
Non-dimensional Temperature:
                            ,    Ts (t)    -
Tcur Ts(0)    -  Tsur Non-dimensional Distance:
vat x
where:
T s(t) =T s          i = 1 to P T s (0) = T s Figures 36 throu ature calculations. ghFigure    38 present      the the 36 shows  results      of theofjet-temper-influence tial melt temperature. The largest energy loss is associated                  the ini-with  the highest melt temperatures because of the larger thermal potential.
The spread between the lowest and highest cases is less than 0.5%. Figure 37 shows how the melt stream emissivity af fects the energy loss term. Changing the value f rom 0.8 (base-line) to 0.3 causes approximately a 0.5% increase in the calcu-lated temperature ratio (compare curves 1 and 3 in Figure 37).
Figure 37 also shows how the cone angle of the stream changes the ratio between surface area and volume. A larger angle will give a
lossproportionally terms. Curves  larger surface area and consequently enhance the 2 and 4 in Figure 37 demonstrate the calcu-lated response to the extremes in the cone angle selection. The fourfold difference in the cone angle in the calculated temperature response.gives            The  less than 0.5%
results        change of Figure  37 suggest that the emissivity and cone angle variations are also not signficant contributors to the temperature losses of the melt jet.
The velocity range considered in Figure 38                      represents  the expected      values extending from the lowest, the opposite situation where the generator gravity-driven                        case to is above the normal syctemthe tus,        operating range pressure.        For the parameters of the SPIT appara-only) to over 70 m/sec. of velocities      varies f rom 2.8 m/sec (g r av ity show that the gravity driven case has almost aThe calculated              5% temperature change in results give temperatur e.      This time required to traverse the distance. behavior is caused principally by the longer The lower bound velocity f or the pressure-driven melt in the SPIT tests, 21 m/sec, shows less than a 1% deviation in the non-dimensional temperature ratio.
180
: e.                                                      _  . _ _
 
1.000        i 0.995        -                                                            ,
0.990      -
O.985      -
3        2    1 E' 5 0.980        -
1 - Ts(0) = 2500 K                                                        _
T T                              2 - Ts(o) = 3000 K 3e $      O.975    -              ~3 - Ts(O)= 3500 K                                                        -
g;    e 3
n                                  V= 45 m/s g        0.970  -                    ( = 0.8                                                            _
6 = 12*
0.965 -                                                                                        _
O.960 -                                                                                        _
0.955 -                                                                                      -
0.950 '                '        '      '                                                  _
0.0                                        '          '
0.1      0.2                          8 0.3  0.4        0.5 8
8 8
0.6 0.7                        J 0.8      0.9        1.0 X. =- Vat X
Figure 36.
Calculated Jet Temperature as aa Function emperature of kiiti l T
 
1.000              ,
0.995      -
0.990    -
1 2          -
4    3 1-E = 0.3        0 = 120 i
0.985      -
2-E = 0.8          0= 5*                                                              ~
i
: m.      '
O.980    -
3-C = 0.8          0 = 120 E
I
                >a I
4 - E = 0.8        0 = 200                                                            -
O w      8 w
0.975    -
o      e
    'o g,g                                  V = 45 m/s 53 0.970    -
Ts(0) = 3000 K F                                                                                                            -
0.965    -
l 0.960  -
O.955  -
i 0.950              '        '        '        '        '      '        '          '      '
i 0.0        0.1      0.2      0.3      0.4      0.5    0.6 1
0.7      0.8    0.9    1.0 i
j                                                                        vat X. =
;                                                                          x Figure 37.
:                                            Calculated Melt Jet Temperature as a Function of Emissivity and Cone Angie 1 -
 
4 i
'1 l
l 1.000    -
I i
l            0.995    -
l              0.990    -                                                                                          I 3    2        -
i.
l 0.985    -
1 i            O.980    -
3    3                                                                                                              t F" s"                        1 -V = 70 m/s
    , 1    A  O.975    -
2-V = 45 m/s                                                                      ~
8 0    8                      3 - V = 21 m/s "F      O.970  -
4 - V = 2.8 m/s
        ,,                                                                                4                        -
F                              E = 0.8 l              0.965  -                6 = 12*                                                                      -
Ts(0)= 3000 K O.960  -
1              0.955  -
0.950                '        '        '                '      '        '    '          '    '
O.O            O.1      0.2      0.3          0.4        0.5    0.6  0.7        0.8  0.9  1.0 Vat l                                                                          X. =                                              ,
X I
Figure 38. Calculated Melt Jet Temperature as a Function of Jet Velocity l
 
2 Varying the convective film coefficient f rom 1000 to 50,000 w/ m  causes less than 0.5% change in the non-dimensional tempera-ture ratio.
The  results of the energy loss calculations indicate that the jet    stream temperature will not change significantly for distances up to 0.45 meter. This conclusion means that measuring the temperature at any point in the stream during the    experiment will give an accurate estimate of the jet temperature.
184                                  t
 
~--
APPENDIX D                                                                                                          ,
Frequency Response of Embedded Thermocouples The    response of embedded thermocouples is dependent on the ability of the devices to follow the rapidly changing temperature history. Ref. 37 suggests that the time response of an                                                intrin-sic thermocouple can be determined f rom the expression:
t*R2 t            a 95 where:
t 95
                    =    time required to reach 95% of the steady-state EMF R    =    radius of the thermocouple junction (assumed to be spherical) a=      thermal diffusivity of substrate (graphite) t* = dimensionless time The dimensionless time can be estimated by the relationships
                                                          ~*
* t*  =  5380 e the parameter a is a thermal property given by:
                                                                . -1 V2 f(kpc) wire              I a =      1 +1                              1
((kp c) substrate /              _
where:
k  =    thermal conductivity o  =    density c  =    specific heat The expression for t* is accurate to a f ew percent in the range:
9.3    i  a      1      9.7 For the case of a small diameter type k thermocouple placed in a graphite calorimeter elements a  =    9.89 185
 
                                                                                      /
r t*  =  2.7 t  =  0.004 see The result shows that the sensor will respond to a step-change in temperature in times less than the response of the recording instrumentation.
For thermocouples embedded in the concrete used to form the test cavity, the parameters are:
a  = 0.2 t*  =  934 t  =  11 sec The very low thermal conductivity of concrete increases the response of the sensor significantly. This behavior means that the thermal pulse is being conducted along the sheath of the thermocouple, away f rom the sensing location. The concrete also has the of f ect of slowing the isotherm velocity in depth, so that the seemingly long response time .nay be adequate for the deeper placed sensors.
l i
186
 
v                                                                                              ,
APPENDIX        E Error In Melt Velocity Measurement Melt velocity along the cavity floor and up the inclined key way will be measured using detectors constructed to give sequential resistance changes. Average velocity can be deter-mined using the equation:
                                        =-
xi Vi        t i
where:
Vi = average velocity during the ith interval xi = distance between two consecutive fingers ti = time int.erval between voltage steps The error in the velocity is then:
2 EVi          /Exi  1          f Etl i
                      =    1        l    +'          1 Vi            ( xi /            gt/i where EZ represents the uncertainty in parameter Z.
The    error        in xi is the variation in                  the  center-to-center spacing for an element based on the mean value (xi) and the total number of fingers on the element (n):
n Xj i
Xi              n n.
(*11 - Xi        + (X12 - xi)2 +              + (xin - xi)2 (Mi) 2  "
n-1 (xij - xi)2 n-1 j=1 1 87
 
The    term      (Exi)2 represents the variance for a single element;      the uncertainty for all sensors used in the apparatus can be based on the " pooled variance" given by the expression:
2            (n1 - 1)      (Ex 1)    + (n2-1) (Ex2) + '* + Ink-1) (Exk)
(Exi) pooled "                      (ni-1) + (ng-1) + . . . + (nk ~1) where ni,        n2,
                                , (nk_1), nk represent the number of fingers on each sensor.              If ail the sensors have the same number of elements (k), then the above equation reduces to:
(Exi)2 + (Ex2)          + '* + (Exk) 2          "
(Exi) pooled                              k Evaluating the above equations for an arbitrary selection of 6 sensors (k = 6) yields the following numerical values:
                          =    1.384 x 10-4 m (Exi)      ed 6
(xi) 2  =  0.5116 m (xi)mean =          11 and f Exi)        [Exi E  1*
j  =                        =    2.706 x 10-4 2
(Xi)mean The uncertainty in the time interval can be found as the absolute error for a given time increment or as a percentage of full ccale.        It is convenient to assume the latter, so that the uncertainty in a given velocity is applicable at all velocities.
For the device employed to record the velocity sensor output, the uncertainty          is      given as i 1 units for the entire sam pl e period.      Therefore,        for 4096        samples (number of time samples I for    the instrument) in a period:
188 a
 
I 2                2
                            =
f Eti l
                                /1    i
                                        "  0'  *
            \t$)                \ m96)
Substituting the results of the calculations above into the equation for the error in velocity:
Evi f
                                                              "N
              =      l  2.786 x 19~4    +    5.96 x 10-8 V$        .\                                      .
              =    1.64 x 18-2      =    1.64%
The  above analysis indicates that the melt velocity deter-mined by the sensors is reasonably accurate.                    The assumption must be made that the melt is well behaved and that shorting of the elements occurs in consistent, sequential manner.
189 r
 
4 4
REFERENCES                                    i
: 1. Rasmussen,    N. , et al. , Reactor Safety Studyr      AD Assessment    .
sf  Accident Risks in U.S.      Commercial Nuclear Power Plants,        i WASH-1400, NUREG-75/014, Washington, D.C. ,1975.                          ;
: 2. Lewis,  H. ,  et al.,    Risk Assessment Review Grouc Reoort 1g the U.S. Nuclear Regulatory Commission, NUREG/CR-s400', 1978.            f
: 3. Rogovin,      M.,    Three Mile Island:        A Report    tg    the  '
Commissioners And the Public, NUREG/CR-1250,1979.
t
: 4. Zion  Probabilistic Safety Study,      Commonwealth Edison      Co.,  i Chicago, IL. , 19 81.                                                    >
t'
: 5. Tarbell,  W.W., and Bradley, D.R., Sustained Concrete Attack hy Low-Temoerature.        Fragmented Core Debris,      S AND82-2 476 ,
NUREG/CR-3024, Sandia National Laboratories, Albuquerque, NM (in  press).
: 6. Powers,  D.,    and Arellano, F., Large-Scale, Transient Tests af the Interaction af Molten Steel with          Concrete,  S AND81-1753, NUREG/CR-2282, Sandia National Laboratories, Albuquer-que,  NM, January 19 82.
: 7. Perinic,  D.,    et al., Concrete EIucible Tests with Thermite Melts,    NRC Translation 669,          Nuclear Research Center, Karlsruhe, KFK 2572, July 1979.
: 8. Wooten,  R.O.,    and Avci,  H.I., MARCH      Enda Description And User's Manual,      NUREG/CR-1711 (BMI-206 4) ,    Battelle Columbus Laboratories, October, 1980.                                            t
: 9. Bordelon,    P.M.,    and Murphy,  E.T.,  WCAP-8327 Containment Pressure Analysis Code (COC01, July 1974.
: 10. Final Safety      Analysis Report Inx the      Zion  Nuclear    Power Station.
: 11. Catton,  I.,    Memorandum  to Jim Meyer,    USNRC,  February 2, 1982.
: 12. Blander,  M.,    et al., Solubility 91 Noble Gases ID Molten Flourides II,      In thg LiF-NaF-KF Eutetic Mixture,      J. Phy s .
Chem. El 1154 (1959).
: 13. Pilch,  M.,    et    al.,  Acceleration Induced Pragmentation of Liquid Drops, NUREG/CR-22 47, Univ. of Virginia,19 81.
: 14. Sitharamayya,      S. ,  and Subba Raj e, K. , Canad.. J, Chem. Eng2 11 365, 1969.
190 w
 
i 4
    .                                REFERENCES (Continued)
!4
  .      15. Knief,      R.A.,      Nuclear    Energy    Technology    Hemisphere Publishing Corp, McGraw-Hill Book Co,1981.
[      16. Powers, D. A. , and Arellano, F. E. , Direct obaarvation gf galt
  '            Behavior During Righ Tamnarature Malt / concrete          Interaction, SAND 81-1754,    NUREG/CR-22 83,    Sandia National Laboratories, Albuquerque, NM, January,1982.
: 17. Incropera,      F.,      and  DeWitt,  D.,  Fundamentals g Eggt Transfer, John Wiley & Sons, Inc., New York, NY,1981.
: 18. Blackwell,      F. B. ,    Umar's    Manual Inx .the Sandia        Dnt .
Dimensional Direct And Inverse Thermal fSonnIT) Eede, Sandia National Laboratories. Albuquerque, NM, (in press).
: 19. Doebelin, E.O. , Mammuramant Syntama r Application And Danign, McGraw-Hill Book Co. , New York, NY,1966.
28      Abramovich,      G.N.,.The Theory g Turbulent Jata, translation by Scripta Technica, The MIT press, Cambridge, MA,1963.
: 21. nadiography.in Modern Industry,          4 th ed. , Eastman-Kodak Co. ,
Rochester, NY,1988.
: 22. Beck,  J.V. , Non-Linear Estimation Applied .tg .the Non-Linear Inverse Ragt Conduction Problem, Int'l. Heat Mass Transfer (13) 783, 1978.
: 23. Powers,    D.A. and Arrellano, F. E. ,            EIsalgn af E1331 Structures by High-Temperatura. SAND 81-1755, NUREG/CR-2284, Sandia National Laboratories, Albuquerque, NM, June 1983.
: 24. Touloukian, Y. S. , et al., Thermophumical              Properties .gf Matter, IFI/ Plenum, Vol. 13, 1977.
: 25. Hameed,    R. et al.      " A New Rotating Course      Par ticle Sampler", Aerosol 101 and. Tach., Vol 2 (1983).
: 26. Powers, D. A. and Brockmann, J.E.,          Ralemme s.f 11331sn RIs:
ducts And Ganaration Af Aaromola Outside the Primary Syntam, Appendix C of BMI 2184, Vol I, Battelle Columbus Labora-tories, (In Press).
: 27. Stratany .gf Emparimentation : Revised Edition,            E.I. duPont de Nemours & Co. , Inc. , Wilmington , DE. , October,1975.
: 28. Bradley, R.H. , Witte, L. C.        " Explosive Interaction of Molten Metals    Injected Into Water" Nuclear Science and Engineering 48, 1972.
191
                                                                                          )
 
1 l
l REFERENCES (Continued)
: 29. Chou,  P. E. ,  and Hopkins,    A.K.,    Eds., Dynamic Responan DI Materials D Intenne Impulsive Loading,            published by Air Force  Materials    Laboratory,    Wright-Patterson AFB, Ohio, August    1972.
: 39. Handbook Di chemistry And Physics, 46th ed. , Chemical Rubber Co. , Cleveland, OH, 1965.
: 31. Sutherland,    H.J.,  and  Hagen,    S.,  Acoustically Measured Penetration        Profiles  far A      Molten  Netallic Pool into River-Stone      concrete,    SAND 8 2-067 6,  NUREG/CR-26 34,  Sandia National Laboratories, Albuquerque, NM, December 1982.
: 32. Buxton,    D.,    and Benedick, W.B. , Steam    Explosion Efficiency Studies,      SAND 79-1399,    NUREG/CR-0 9 47,    Sandia    National Laboratories, Albuquerque, NM, November, 1979.
: 33. Beckwith,    T.G.,    et al.,  Mechanical Measurements, 3rd ed.,
Addison-Wesley Publishing Co. , Inc. Reading, MA,1982.
: 34. Moffat,    R.J.,    "How to Specify Thermocouple Response",        IBh Journal, 293, June, 1955.
: 35. Mark's Standard Handbook far Mechanical Engineers,            8th ed.,
McGraw-Hill Book Co. ,1979.
: 36. Taylor, Sir Geof f rey, .ThR 1DRhnbillh2 D1 Liguld BMELADRR MhtD  AG221RIAhnd in h DiLeskiDD Et192ndisulA1 LD Their Planes, I, Proc. Royal Society, 201 (1950) 192-6.
: 37. Keltner,    N. R. and Bickle, L. W. , " Intrinsic Thermocouple Measurement Errors",      presented at ASME-AICHE Heat Transfer Conf erence, St. Louis, MO, Aug.1976.
t 1
192 n
 
DISTRIBUTION:
Division of Technical Information and Document Control NRC Distribution Contractor U. S. Nuclear Regulatory Commis ion 15700 Crabbs Branch Way Rockville, MD 20850 250 copies for R7 U.S. Nuclear Regulatory Commission (16)
Of fice of Nuclear Regulatory Research Washington, DC      20555 Attn:    O. E. B ssett B. S. Burson R. T. Curtis C. N. Kelber
[            J. Larkins T. Lee (5)
M. Silberberg R. W. Wright
'            T. Walker W. Pasedaq 4            M. Jankowski
-    U.S. Nuclear Regulatory Commission (4)
-    Office of Nuclear Regulatory Regulation Washington, DC
-                        20555 Attn      L. G. Ilulman P. Easky J. Rosenthal J. Mitchell U.S. Department of Energy (2)
Albuquerque Operations Of fice P. O. Box 5400 Albuquerque, NM      87185 Attn    J. R. Roeder, Director Operational Safety Division D. K. Nowlin, Director Special Programs Division Fort  C. B. Quinn D. Plymale U.S. Department of Energy Of fice of Nuclear Safety Coordination Washington, DC      20545 Attn    R. W. Barber Electric Power Research Institute 3412 Ilillview Avenue Palo Alto, CA 94303 Attna  R. Vogel R. Sehgal DIST-1
 
Prctossor a
T. Theoftnzuo urdue University School of Engineering West Lafayette, IN 47907 l    Dr. R. Henry Fauske & Associates i
16W070 West 83rd Street Burr-Ridge, IL    60521 M. L. Corradini Nuclear Engineering Dept.
University of Wisconsin j    Madison, WI 53706 I. Catton UCLA Nuclear Energy Laboratory 405 Hilgard Avenue Los Angeles, CA 90024 Brookhaven National Laboratory (4)
Department of Nuclear Energy l  Building 820 Upton, NY 11973 Attn:    R. A. Bari                l T. Pratt G. Greene T. Ginsberg Professor R. Seale Department of Nuclear Engineering University of Arizona Tucson, AZ 85721 Oak Ridge National Laboratory (2)    '
P. O. Box Y I  Oak Ridge, TN    37830                4 Attn:  T. Kress                      {
S. Hodge l
K. Holtzclaw                          i General Electric - San Jose            i Mail Code 682                          8 175 Kurtner Avenue                    -
San Jose, CA    95125                  8 i
l  Argonne National Laboratory            <
l  9700 S. Cass Avenue                    '
Argonne, IL    60439
,  Attn:  D. Pedersen                      '
Cathy Anderson l  Nuclear Safety Oversight Commission
;  1133 15th St., NW Room 307 Washington, DC 20005 DIST-2 i
f
 
7 I                                                  .
t i
g
          '        Battelle 505 King Columbus Avenue    Laboratory (3)
* Columbus, OH    43201
* Attn:  P. Cybulskis
* R. Denning
        .                  J. Gieseke
        .          J. E. Antill
        .          Berkeley Nuclear Laboratory Berkeley GL 139 PB Gluocestershire United Kingdom W. G. Cunliffe Bldg. 396 British Nuclear Fuels, Ltd.
* Springfields Works
* Salwick, Preston Lancs United Kingdom s
Reactor Development Division (4)
* UK AEA - Atomic Energy Establishment Winfrith,  Dorchester Dorset United Kingdom Attn:    R. G. Tyror, Head T. Briggs R. Potter A. Nichols Projekt Nucleare sicherheit (3)
Kerforschungszentrum Karlsruhe Postfach 3640 75 Karlsruhe Federal Republic of Germany Attn:    J. P. Hoseman Albrecht H. H. Rininsland Mr. G. Petrangeli Direzione Centrale della Srcurezza Entre Nazionnle Energie Alternative (ENEA) Nuclear Vrale Regina Margherita, 125 Casella Postale N. 2358 I-00100 Roma A.D. , ITALY Dr. K. J. Brinkman Reactor P.a. BoxCentrum 1
Nederland 1755 ZG Petten THE NETHERLANDS DIST-3              1
 
t-l Mr. N. Bairict, Chicf Department LNR Fuel Belgonucleaire Rue de Chanpde Mars. 25 l
B-1050 BRUSSELS, BELGIUM I    Dr. 8. Salto l    Japan Atomic Energy Research Institute
:    Takal Research Establishment
!    Tokai-Mura, Naku-Gun l    Ibaraki-ken l    JAPAN
!    Mang Lu TVA
!    400 Commerce, W9C157-CK Knoxville, TN 37902 l    M. Fontana l    Director, IDCOR Program l    Technology for Energy, Inc.
P. O. Box 22996                                                                                ]
10770 Dutchtown Rd.
Knoxville, TN    37922 H. J. Teague (3)
UKAEA Safety and Reliability Directorate                                                              g Wigshaw Lane                                                                                    I Culcheth Warrington, WA3 4NE United Kingdom Dr. Fran Reusenbach Gesellschaf t f ur Reaktorsicherheit (GR8 abH)
Postfach 101650                                                                                  ;
l    Glockengasse 2                                                                                    '
l    5000 Koeln 1 Federal Republic of Germany                                                                      }
l
,    8. J. Nienczyk                                                                                      :
(    Union of Concerned Scientists l    1346 Connecticut Avenue, N.W.                                                                      j i
S.1101                                                                                              4 Washington, DC    20036                                                                            '
t M. Jankowski                                                                                        '
IAEA
* Division of Nuclear Reactor Safety                                                                  '
Wagranerstrasse 5                                                                                  '
P. O. Box 100 A/1400 Vienna, Austria 3141  C. M. Ostrander (5) 3151  W. L. Garner (1) 6000  E. H. Backner DIST-4 A
 
I 6400 A. W. Snyder 6410 6420 J. W. Hickman
{        6420 J. V. Walker 6421 J. B. Rivard t
    '        6422 T. R. Schmidt
  '          6422 D. A. Powers (5)
  '          6422 F. E. Arellano 6422 J. E. Brockmann (5) i 6422 E. R. M. Elrick 5                    R. Copus 6422 J. E. Gronager 6422 T. M. Kerley 6422 6422 D. A. Lucero 6422 E. A. L. Ouellette, Jr.
,                      Randich 6422 A. R. Taig 6422 6423 W. W. Tarbell (5)
P. S. Pickard 6425 D. R. Bradley 6425 W. J . Camp 6425 6425 M. Pilch (5) 6425 A. Suo Anttila W. Frid 6427 M. Borman 6430 N. R. Ortiz 6440 D. A. Dahlgren 6442 6449 K. W. A. Von Riesemann 6450 D. Bergeron 6454 J. G.
A. Reuscher L. Cano 7530 T. D. Lane 7537 7537 T. N. R. Keltner Y. Chu 7537 R. U. Acton                  i 7537 8424 M. B. F. Blackwell A. Pound DIST-5
 
macpen. ass                                                    oe ove6ameansv6s. ton,coes.iss oes        e me o=1 %. eta n=swe e, reoc ene we u. ,,ea,s
      *                '3g' BIBLIO1RAPHIC DATA SHEET                                                    NUREG/CR-3025
        '            949 m5teuctiones os,ime ogvtast SAND 82-2477 6          ~g i.ggg .50 svetat6E                                                                                      3gg.vge6.gn High Pressure Melt Streaming (HIPS)
Program Plan
                                                                                                                                            . omitmE'oafco.htTED wo% t es                        vg.m
                    . .wv-oam                                                                                                      APRIL                            1984 W. W. Tarbell                                                                                                      . o n . .o v .sseio J. E. Brockmann                                                                                        "o*'"                            " ' "
M. Pilch                                                                                            JUNE                      l      1984
      .            , ... o...w o.s...z .v .o          .. . o ...t.=o .oo.u u,  i. c .                                        . ..oac,v.s. .oa.w a w . i-Sandia National Laboratories P. O. Box 5800                                                                                . . .~ o o .~1 % . .
      -                        Albuquerque, NM                      87185
)    -                                                                                                                                NRC Fin. No. Al218
                    ,  ,,0    so..w o.s      4.v.o ... .~o .. 6 .o .co i ss u . e, c ,                                        ii. ,,n o.an.oav Division of Accident Evaluation                                                                                        R7 Severe Accident Assessment Branch U. S. Nuclear Regulatory Commission                                                          '""'"*""'"~*~"*'
Washington, D. C.                        20555 I2 Sw*P64 4%T.av=0115 13 . elf m.cf f200 eerse or '.eu The Zion Probabilistic Safety Study (ZPSS) envisions accident sequences that could lead to failure of the reactor vessel while the primary system is pressurized. The resulting ejection of molten core material into the reactor cavity followed by the binwdown of steam and hydrogen is shown to cause the debris to enter into the containment region.
The High Pressure Melt Streaming (HIPS) program has been developed to provide an experimental and analytical investigation
    ,                          of the scenario described above. One-tenth linear scale models of the Zion cavity region will be used to investigate the debris dispersal phenomena. Smaller-scale experiments (SPIT-tests) are also used to study high-velocity jets, jet-water interactions,                                                                                        ,
and 1/20th scale cavity geometries.                                    Both matrices are developed using a factorial design approach.
The document describes certain aspects of the ZPSS ex-vessel phenomena, the experimental matrices, test equipment, and instrumentation, and the program's analytical efforts.                                                                    Preliminary data from SPIT testing are included.
: i. oocw. =v ...,vs        . . .  ..o os onc...rcas                                                                                      > > % . s. . s ,
High Pressure Ejection / Thermite / Jets / Aerosol Generation /                                                                  '''''''"
                              . Reactor Safety / Debris Dispersal /PRA unlimited as stcva.?v c6.ssaic.Tio%
                                                                                                                                                                                  ~
ora. o
                  .,oe.o..... ,o. .,i.e.oT..
uncTassified          ~
ara ee u unclassified
                                                                                                                                                            >> w. eta o nces is en.ca C U.S. GOVERNMENT PRINTl880 OFFICE.1964-676435 l 4300 M_      -~~r.,                      _                                                        -
                                                                                                                                              - - -}}

Revision as of 17:35, 2 September 2020

Redirect to:

Retrieved from "https://kanterella.com/w/index.php?title=ML20129J482&oldid=2940083"