ML20214L888
| ML20214L888 | |
| Person / Time | |
|---|---|
| Site: | Peach Bottom  |
| Issue date: | 05/27/1987 |
| From: | PECO ENERGY CO., (FORMERLY PHILADELPHIA ELECTRIC |
| To: | |
| Shared Package | |
| ML20214L881 | List: |
| References | |
| NUDOCS 8706010109 | |
| Download: ML20214L888 (64) | |
Text
{{#Wiki_filter:- Safety Evaluation / Analysis Supporting H Hrogen Water Chemistry For Peach Bottom Atomic Power Statien Units 2 and 3 p A P
I .
's MECHANICAL ENGINEERING DIVISION N2-1 2301 Market Street SAFETY EVALUATION FOR HYDROGEN AND OXYGEN STORAGE SYSTEMS, REV. 2 lk Peach Bottom Atcmic Power Station, Units 2 6 3 FILE: GOVT 1-1 (NRC)
RES 17-8-1 (HWC)
REFERENCE:
- 1) Bolling Water Reactor Owners Group (BWROG),
"GuldelInes for PerTnanent BoliIng Water Reactor Hydrogen Water Chemistry Installations",
1987 Revision
- 2) United Engineers C Constructors, Inc. " Safety Analysts for the Liquid Hydrogen Storage System", Rev. 2, May 1987
- 3) United Engineers & Constructors, Inc. " Safety Analysis for the Llauld Oxygen Storage System", Rev. 2, May 1987
SUBJECT:
This safety evaluation covers the cryogenic storage, vaporization, and gaseous storage of hydrogen and oxygen on-site at Peach Bottom, as required to supply gaseous hydrogen and oxygen to the station for the irrplementation of hydrogen water chemistry. CONCLUSIONS: The storage of hydrogen and oxygen on-site at Peach Bottom for use in Hydrogen Water Chemistry does not involve safety related equipment. The operation of the storage systems, including 11ould hydrogen and oxygen storage and vaporization and hydrogen gas storage, will not affect safety related equipment. The lo::ation and use of the hydrogen and oxygen storage systems on-site at Peach Bottcm does not constitute an unreviewed safety question. This modification maintains the capability to safely shut down the plant in the event of a fire. A change to the plant's Technical Specifications is not required. A change to the UFSAR is not required, however, it will be revised to discuss these systems. This modification does not result in a significant reduction in the margin of safety. A significant hazard consideration is not required. DISCUSSION: Cryogenic Ilquid hydrogen, liquid oxygen, and gaseous hydrogen storage facilities are required at Peach Bottcm to supply hydrogen and oxygen gas to a hydrogen water chemistry system.
s Oxygen Storage Systen: The oxygen supply system consists of a cryogenic storage tank, an ambient air vaporizer and associated controls. The systen will be Installed at Peach Bottcm, north of the Emergency Cooling Tower, outside of the plant security fence, on a concrete foundation pad. The , cryogenic storage tank consists of an inner stainless steel vessel and outer carbon steel vessel, with the annular space evacuated and filled with thermally non-conducting meterial to provide the required insulation. The Inner tank is conservatively designed and tested to ASME Code requirenents and is provided with a relief valve and a back-up rupture disc for overpressure s protection. The outer tank is provided for thernal insulation only and is not designed to ASME codes. The Ilquid oxygen flow is regulated through an ' arblent air vaporizer, with the resulting gas supplied to the station for use as part of a future hydrogen water chemistry systen. The systen also M incorporates provisions for temporary oxygen supply via trucks. i Oxygen is stored as a cryogenic ilquid at approximately 150 psig and - 300 F. Oxygen exists as a blue liquid or a colorless gas. Oxygen is nontoxic and non-flannable; however, it is a strong oxidizer. Materials which burn in air will combust more vigorously in oxygen. ___. . The primary potential hazard associated with the storage of liquid oxygen is the potential for adversely affecting the station's level of fire protection. The design of the liquid oxygen storage tank provides protection against a major vessel rupture and subsequent large oxygen release, but an unusual accident, such as a selsnic event rupturing the vessel, is credible. Cryogenic oxygen, both IIquid and gas, is heavier than air, so a large spill will disperse away fran the tank along the ground. The oxygen will begin mixing with the surrounding air as it warns. The increased flannability of combustibles in an enriched oxygen atmosphere is a concern. A conservative value of 30% by volone was chosen as the highest allowable oxygen concentration at any air intakes to safety related areas (r.ormal air is 21% oxygen). Liquid oxygen tanks are widely used, including providing the oxygen gas to most hospitals, and vessel failure has been shown to be extremely infrequent and always accompanied by an external cause. Nevertheless, the liquid oxygen tank at Peach Bottom has been located such that, in the unlikely event of a complete storage vessel failure, the oxygen concentration at safety related air intakes will not exceed the 30% by volume criteris. This detennination is based on Figure 4-9 of reference 1, which was developed using the industry recognized DEGADIS nodel for heavier than air gas behavicr. This nodel is appropriate for this analysts because cold oxygen gas is significantly denser than arblent air. This siting criteria will assure that the aunosphere within the station near safety related equipment will reneln well below the oxygen concentrations that would pose an increased flanmability concern. Additionally, since no ignition sources or combustibles are added, the probability of a fire near safety-related ' equipnent is essentially unchanged.
The oxygen storage tank is located outdoors, so any oxygen leakage, such as that from a pipebreak, will dissipate and is not a safety concern. Any failures of associated oxygen systen equipnent, such as piping or vaporizers, is enveloped by the rupture of the storage vessel. The foundation pad neterial is concrete, which does not burn and will not carbust even in the presence of pure oxygen. The storage system will be fenced and lighted, with barriers to prevent inadvertent contact with noving vehicles on-site. The location of the oxygen tank as vell as control and locatial of conbustible naterials is in accordance (and will be maintained in accordance) with NFPA 50, " Bulk Oxygen Systems at Consuner Sites". The oxygen storage system will be designed to reneln in place during the PBAPS design basis tornado wind (300 nph) to preclude the tank from falling nearer the station than t'he site analyzed in our evaluation. Based on this design, the tank was analyzed for a complete failure at its location (Ref. 2). Based on the ability to tolerate such a failure ard the fact that the BWROG guidelines (ref. 1) do not require the oxygen storage system to be designed for earthauakes, the liculd oxygen supply systen need not neet seismic design criterla. The referenced safety analysis (ref. 3) evaluates the siting of a large liquid oxygen tank at Peach Botton as compared to the guidelines developed by the BWROG Cref. 1). Based on this analysis, we have concluded that the siting of the I! quid oxygen tank does not involve a significant reduction in the margin of safety, and therefore does not constitute an unreviewed safety question. Based on the above discussion and on Reference 3, it is concluded that the location and operation of the liquid oxygen storage system as described is acceptable. The UFSAR need not be changed, however it will be revised to discuss this systen. The electrical power supply for the oxygen supply system will be covered separately. Hydrogen Storage System Hydrogen will be stored as a cryogenic liquid at approxinetely 100 psig and -425 F, in a cryogenic storage vessel consisting of an Inner aluninun vessel and an outer carbon steel shell . The annular space is evacuated and filled with thenna11y non-conducting neterial to provide the required insulation. The Inner liculd hydrogen tank is designed and tested to ASME code requirements and is provided with redundant overpressure protection devices. These requirements include pressure testing to 1.5 tines the Maxinom Allowable Working Pressure. A safety relief valve (set above the nornel working pressure) and a rupture disc (set above the relief valve setting) protect the vessel fran danege due to overpressure. The outer tank is provided for thernel Insulation only and is not designed to ASME codes. The IIquid hydrogen is pressurized to approximately 2500 psig by a cryogenic 11guld hydrogen, reciprocating type, punp. The liquid hydrogen is driven by the punp
_4 through an amblent vaporizer bank, which discharges the resulting gaseous hydrogen to high pressure storage tubes, which will provide the hydrogen gas to the station. The high pressure storage tubes are also equipped with a pressure relief device. The systan also incorporates provisions for temporary hydrogen supply via trucks. The hydrogen storage system, including the tank, purps, outdoor vaporizers, storage tubes, piping, and controls will be located outdoors approximately 1700 feet northwest of the Unit 3 Reactor Building,1180 feet frcm the offgas stack. The equipment will be mounted on a concrete foundation pad with the tank, tank supports, and Interconnecting piping frcm the tank to a flow Ilmiting device (valve or orifice), meeting PBAPS Design Basis Earthquake seismic design requirements as required by the BWROG guidelines. The storage system, including tank, vaporizers, gas storage tubes and associated controls, is designed to withstand the PBAPS design basis tornado wind which is 300 rrph. The storage system design and siting is in accordance with NFPA 50A,
" Standard for Bulk Hydrogen Systems at Consumer Sites", and NFPA 508, " Standard for Liqulfled Hydrogen Systems at Consuner Sites". The entire system is sited, designed and installed In accordance with reference 1.
The storage of hydrogen on-site is a concern due to the corrbustibility of hydrogen in air. The ccmbustion of hydrogen is exothennic, with a low activation energy required to initiate the reaction. Hydrogen, in air, has a lower flermability limit of 4% by volune and an upper flamnability limit of 74.2% by volune. Additionally, hydrogen is detonable in air, in confined spaces, at concentrations of from 18.2% to 58.9%. The quantity of hydrogen in the storage tank will be as much as 2.5 million standard cubic feet, which has an explosive equivalent (using industry accepted equivalence conversions which are utilized by the NRC in Regulatory Guide 1.91), if optimally vaporized and detonated, of approximately 14 tons of TNT. Additionally, this quantity of hydrogen, if burned rather than enploded, would form a fireball which would create significant thermal fluxes. The location for the hydrogen storage facility at Peach Bottcm, was chosen to provide adequate distance from the station so that, in the unlikely event wherein hydrogen is released, no resulting events would have a significant adverse effect on any safety related structures, including the offgas stack. Although the possibility of detonation was considered, an t.nconfined detonation is extremely unilkely. As detailed in the referenced safety analysis, all docunented, unconfined hydrogen sollis in the presence of an Ignition source, to date, have burned, but have not exploded. Additionally, for the analysis which considered large spills due to major vessel failure, it was determined (per Reference 1) that any carbustion, whether a fireball or an explosion, would take place at the tank site and would not drift near the plant or the offgas stack before ccrrbustion. This determination was made on the basis of the very low activation energy C.019 millIJoules) required to initiate the hydrogen / oxygen reaction. A review of the storage tank design and the tank's overpressure protection devices concluded that any accident which
l _5_ was capable of rupturing the vessel would be the result of external causes, such as aircraft crashes or tornado misslies. These events would be acconpanied by sparks, high tencerature meterial, or In the case of tornado misslies, static electricity, which would contain, considering the low activation energy of hydrogen conbustion, sufficient energy to ignite the release. In an unconfined area, such as the hllitop where the tank is located, a fireball is the most probable result if a large quantity of hydrogen was released. A fireball consists of a cloud of hydrogen vapor. The concentration of hydrogen in the cloud varies from the surface to the center. Sone volume of gas at the surface of the cloud is in the flanneble range. The Internal hydrogen thus feeds a large fire on the exterior of the volune. As detailed in the BWROG guidelines (ref. 1), the fireball resulting from such a large spill would last in the range of 3 to 9 seconds. The thennal flux at the station, or the offgas stack, would be less than half the heat input required to ignite wood; values this low would have little effect on reinforced concrete structures. As previously stated, the detonation of hydrogen gas in an unconfined, outdoor location is Improbable (no record of unconfined hydrogen / oxygen mixtures detonating is available); however, for conservatism this evaluation also considered the shock wave the station would receive due to the detonation of the entire tanks contents at the storage location. The calculations utilize an Industry accepted nethod for converting carbustible gas explosions to TNT equivalence, as recognized by the NRC in Reg. Guide 1.91. A neximun pressure pulse of approxinately 1 psig peak overpressure, which is below the 6 psig threshold value for partial denolition of brick buildings, would be received by the station. As detailed in reference 1, the 6 psig'value for partial denolition of brick buildings was detennined using a data base of buildings daneged during World War 11 from various bonb sizes at various distances. Overpressures of this negnitude would have little effect on reinforced concrete structures at Peach Bottom which are designed for 300 aph tornado wind loads. Additionally, an independent review using a method recognized by the NRC (NUREG/CR-2462) for calculating safe standoff distances fran explosions was perforned. This nethod is inherently nore conservative because the response of the Reactor Building to the pressure pulse is considered. The results of this evaluation, as detailed in Appendix A of Reference 2, require the tank to be sited approxinately twice the distance fron the station and offgas stack required by the initial analysis, or approximately 600 feet. As mentioned previously, the PBAPS tank siting neets this requirenent. Accidents which result in the failure of the liquid hydrogen tank will provide sufficient energy to ignite the release. The storage system gaseous piping, liquid piping after the flow limiting device, vaporizers, pumps and gas storage tubes, are non-seismic and could fall due to an earthquake, without an external Ignition source present. This would permit a non-Ignited cloud to drift towards the station. The storage system is sited considering the failure of any of these non-seismic conponents. Such failures are enveloped by the failure of
the largest pipe line containing liquid or gaseous hydrogen. Therefore, the system's location insures that the flarrmable hydrogen / air mixtures (hydrogen concentration between 4 and 74.2% by volune) created would not pose a hazard to safety related equignent due to the intake of an enriched hydrogen mixture. The referenced safety analysis (Reference 2), determined that, even in the worst case scenarlo where either a 2" high pressure gas line or a 3/8" 11 auld hydrogen line (IIquid hydrogen supply lines shall have flow Ilmiting devices that control hydrogen discharge to this equivalent flow), failed in the direction of the station, the concentration of hydrogen at any safety related air intake is well below 4% by volune which assures that hydrogen concentrations within the plant will remain well below hazardous levels. The storage system is selsmically designed up to the flow Ilmiting devices to assure that releases will be limited to only those analyzed. This determination was made using a jet dispersion model (for high pressure gas release) assuTilng the rrost conservative environmental conditions of a calm, still day with the light wind in the station direction (Gaussian dispersion, F weather stability). The quantity of hydrogen that would be released by the actuation of a pressure relief device is envelcped by the pipeline break analysis. This analysis of hydrogen concentration versus distance also shows that the combustion of such a gas or liquid pipeline release outdoors at the tank site is not a concern because the hydrogen released will not be in the corrbustible concentration range (greater than 4%) at the station. The carbustion of the dispersing hydrogen cloud due to a pipeline break, unilke the fireball caused by a large hydrogen release, is very short and produces insignificant thermal fluxes to surrounding areas. Since the burning cloud would not be located at the station or the offgas stack, no adverse impact on any plant structures would result. The referenced safety analysis (Ref. 2) evaluates the siting of the liquid hydrogen storage system at Peach Bottcm. The referenced analysis uses as its basis the guidelines (ref. 1) developed by the Bolling Water Reactor's Owners Group (BWROG). Based on this evaluation It has been determined that the siting of a liquid hydrogen storage facility does not involve a significant reduction in the margin of safety. This modification does not constitute an unreviewed safety question. Based on the above discussion and on Reference 2, it is concluded that the location and operation of the liquid hydrogen system as described is acceptable. A change to the UFSAR is not required, however, a revision will be made to duscribe this system. The electrical power supply to the hydrogen storage system will be evaluated separately. GENERAL: The transportation of hydrogen and oxygen on-site was also evaluated in accordance with the same criterla for the storage systems. Deliverles of hydrogen and oxygen will vary depending on usage but are expected to range frcm daily to tronthly. The supply tank trucks are of simliar design as the
pennanent vessels. The supply tank trucks also hold less hydrogen or oxygen than that analyzed for the storage systens and will be at the plant site only during deliveries. Deliveries will be scheduled such that at no time will the total quantity of hydrogen, including storage tank and delivery truck, be greater than that analyzed (see above discussion) for Peach Bottom. Additionally, the supply trucks will cone no closer to the station than the storage systen sites. Thus it is concluded that the potential hazards of on-site transportation are enveloped by the analyses for the permanent storage systems. The hydrogen and oxygen supply systens are each equipped with a low temperature switch downstream of the vaporizers. The switch will shutoff flow to prevent the Inadvertent Injection of cryogenic gas or IIquid which could danege station equipnent or materials. Due to the separation distance between the hydrogen and oxygen storage facilities (greater than 1000 feet) an evaluation of sinultaneous storage site accidents is not required. The following documents have been reviewed in the perfornence of this safety evaluation: PBAPS UFSAR Sect. 2, 12, 10CFR50, NRC Reg. Guide 1.91, and NUREG/CR-2462. 10CFR50.59 CHANGES. TESTS AND EXPERIMENTS:
- 1. The location of hydrogen and oxygen facilities at Peach Bottom does not involve an unreviewed safety question.
A. This modification does not involve a significant increase in the probability or consequences of an accident or malfunction of equipment important to safety previously evaluated In the safety analysis report. This nodification invokes the Installation of nea equipnent that will be physically located exterior to the station. Existing station equipnent or procedures are not affected by this nodification; therefore, the siting of oxygen and hydrogen storage systens at Peach Botton does not increase the probability or consequences of an accident previously evaluated. B. The nodification does not create the possibility of an accident or an1 function of a different type than any previously evaluated in the safety analysis report. This nodification involves the installation of new equipment exterior to the station. However, the hydrogen and oxygen storage facilities have been located at sufficient distances from the station such that no credible accidents could have adverse effects upon the station or any safety related equipnent in excess of any effects previously analyzed. This includes consideration of all accident scenarios included in the UFSAR.
C. The modification does not reduce the rnargin of safety as defined in the basis for any technical specification. The hydrogen and oxygen storage facilitles have been located at sufficient distances frcm the station so that no credible accidents could have adverse effects upon the station or any safety related equipnent in excess of any effects previously analyzed. This includes consideration of all accident scenarios included in the UFSAR. Additionally, there are no technical specifications concerning the hydrogen and oxygen storage facilities.
- 2. A change to the Peach Bottom Technical Specifications is not required. There are no Technical Specifications appilcable to these systems.
10CFR50.92 SIGNIFICANT HAZARDS DETERMINATION: A license ariendment is not required, therefore a significant hazards determination is not required.
_g_ Prepared by: h , [w Date: ! f s3!67 Reviewed by: 'c' /4- [ Ot[M Date: I//3/87 3 .$$ sfiblij f /,
" '* ' W Date $ Yf D PPSf (Section Fiead)
D (b ),' . ( % ~ Date: I N In erfacing Section (Civil Section Head)
'e Date: 8 87
' '/
Engine'er-in[hargeNES CMC /cnu/0t+308609 Copy to: L. B. Pyrlh M. B. Ryan l J. M. Madara, Jr. A. R. Lewis lh J. W. W. C. Austin Birely P. K. Pav11 des G. A. Hunger T. E. Shannon C. B. Patton C. Caprara ISEG Engineer (PB) D. R. Helwig R. J. Scholz D. Marano F. J. Coyle R. A. Mulford J. F. O'Rourke G. T. Brecht D. A. Anders C. M. Cooney L. C. Fletcher G. M. Leitch l[ W. M. Alden D. M. Smlth lh DAC (NG-8) DOCTYPE=565 l k F. W. Polaski F. J. Mascitelli b
~
i PHILADELPHIA ELECTRIC COMPANY t PEACH BOTTOM ATOMIC POWER STATION SAFETY ANALYSIS FOR THE LIQUID HYDROCEN STORAGE SYSTEM Revision 2 May 1987 A l Prepared by United Engineers and Constructors l l 4 (6202E)
TABLE OF CONTENTS I. INTRODUCTION II. DESCRIPTION OF LIQUID HYDROGEN STORAGE SYSTEM III. SCENARIOS FOR POTENTIAL HYDROGEN RELEASE IV. GENERAL BEHAVIOR OF HYDROGEN RELEASES V. DESIGN CRITERIA RELATED TO HYDROGEN RELEASES VI. HYDR 0 GEN RELEASE CONSEQUENCE ACCEPTABILITY VII. REFERENCES LIST OF FIGURES FIGURE 1 SITE PLOT PLAN FIGURE 2 HYDROGEN VS. METHANE IGNITION ENERGIES FIGURE 3 MINIMUM REQUIRED SEPARATION DISTANCE VS. HOLE SIZE AND ID OF PIPE FOR GASEOUS RELEASES FROM 150 PSI LIQUID HYDR 0 GEN STORAGE TANK FIGURE 4 MINIMUM REQUIRED SEPARATION DISTANCE VS. HOLE SIZE 'NDA DISCHARGE , RATE FOR LIQUID RELEASES FROM 150 PSI LIQUID HYDROGEN STORAGE TANK F WEATHER STABILITY, 1 M/S WIND VELOCITY. FIGURE 5 MINIMUM REQUIRED SEPARATION DISTANCE VS. LIQUID HYDROGEN STORAGE TANK SIZE FOR INSTANTANEOUS RELEASE OF ENTIRE TANK CONTENTS AND EXPLOSION AT TANK SITE F WEATHER STABILITY. FIGURE 6 THERMAL FLUX VS. DISTANCE FROM FIREBALL CENTER FOR LIQUID HYDROGEN STORAGE SYSTEM APPENDIX A CAPABILITY OF SAFETY RELATED STRUCTURES TO WITHSTAND BLAST LOADINGS APPENDIX B IMPACT ON TRANSMISSION EQUIPMENT IN VICINITY OF HYDROGEN STORAGE SYSTEM Page 2 of 24 (6202E)
I. INTRODUCTION ( This safety analysis of liquid hydrogen storage is intended to support the license amendment to permit liquid hydrogen to be stored on site. The hydrogen will be stored in a liquid hydrogen storage tank of 20,000 gallons or less. A. Purpose of this Report The purpose of this report is to establish the acceptability of the chosen site for the liquid hydrogen storage facility. The determination of the acceptability is based upon application of the criteria from " Guidelines for Permanent BWR Hydrogen Water Chemistry installations", by the BWR Owners Group for IGSCC Research, 1987 Revision (Reference 1). The Peach Bottom Atomic h Power Station (PBAPS) liquid hydrogen storage system will meet all appropriate guidelines provided by Reference 1. This report provides the basis for compliance with the liquid hydrogen storage guidance given in Reference 1. B. Structure of this Report
' ~
Section II of this report describes the liquid hydrogen storage system for PBAPS. The liquid hydrogen storage system will be provided by Air Products and Chemicals, a supplier with extensive experience in the design, operation, and maintenance of storage and supply systeus. This report (Section III) also develops possible event scenarios--both man-made and natural, as well as both onsite and offsite--that could damage the hydrogen storage system and lead to hydrogen releases. The general behavior of released hydrogen is described in Section IV. This behavior, along with a review of causal events, is used to develop general design criteria for the hydrogen storage system (Section V). These general design criteria are consistent with - Reference 1. l The general design criteria in Reference 1 are the result of a series of release event consequence analyses that are a function lh of distance from the storage system. These analyses were translated into a scries of curves for safe standoff distances as a function of certain storage system design parameters. This report evaluates the credible release scenarios and uses the applicable Reference 1 curves to assure that the storage system location is appropriate (Section VI). lh Page 3 of 24 (6202E)
C. Conclusions This report provides the basis for concluding that: (1) A safe location has been established for the liquid hydrogen storage system. The safety of this storage system site has been determined based on the relative location of other Peach Bottom equipment; and the effects of the normal, abnormal, and accident conditions which might occur at the storage location. The effects on the hydrogen storage system of the design basis external events used for safety related equipment (such as tornados and earthquakes) have also been considered. (2) The liquid hydrogen storage facility is designed and can operate safely without preventing safety-related structures and equipment from performing their design functions. Page 4 of 24 (6202E)
II. DESCRIPTION OF LIQUID HYDROGEN STORACE SYSTEM A. System Design The liquid hydrogen will"be stored in a liquid hydrogen storage tank. This tank consists of an inner vessel meeting Section VIII, Division 1 of the ASME Code for Unfired Pressure Vessels; and an outer " vacuum jacket" vessel, with the space between filled with insulation and evacuated. Liquid hydrogen will be pumped from the tank to ambient vaporizers to produce gaseous hydrogen for plant use. Reference 1. Section 3.2.2 provides additional details regarding ljdhg the design of liquid hydrogen storage systems. The PBAPS system will meet all of the design requirements of Reference 1, Section 3.2. l[ B. Siting The storage system location is shown on Figure 1. The storage system is within the site boundary, with access controlled by a locked gate. It is within its own locked fence as well. The route to be taken by liquid hydrogen delivery trucks is also shown on Figure 1. Delivery trucks will never approach closer to safety-related equipment than the storage site. The liquid hydrogen storage system will be located 1180 feet away from the nearest safety related structure, which is the plant stack. The next nearest safety related structure is the Unit 3 reactor building which is 1700 feet away. The nearest safety related air intake is for dilution air at the plant stack which is 1180 feet from the liquid hydrogen storage system. Page 5 of 24 (6202E)
III. SCENARIOS FOR POTENTIAL HYDROGEN RELEASE A. Hydrogen Storage System Specific Events Reference 1 includes several special requirements above and beyond l normal industry practice and ASME code requirements. These requirements deal with longitudinal weld radiography and overpressure protection. These added requirements were instituted based upon fracture mechanics analyses and overpressure protection reliability analyses, respectively, which were done in support of the Dresden HWC design (Reference 2). These added features are considered sufficient to render the possibility of vessel failure by detected or undetected crack growth, or by overpressure protection system failure, to be of such a low probability as to be negligible. The only normal operation of the storage system where hydrogen can be released to atmosphere would be when the station is not using hydrogen. Then, after a few days, the normal heat transfer from the ambient air will vaporize enough liquid to require controlled venting of the tank. The vent rate would be on the order of 0.002 lbs./sec., intermittently. Abnormal occurrences resulting in hydrogen releases, which are unlikely but might occur over the life of the storage system, are related to the operation of overpressure protection devicer.. The peak vent rates associated with the worst case failure would be about 8 lbs./sec. These normal and abnormal hydrogen release events will not impact the hydrogen storage system, or other Peach Bottom safety related or non-safety related equipment. This is the case whether or not the release were to be ignited. As shown in Reference 3, these events are less limiting in determining storage system siting than those associated with the failures discussed in Section III due to severe natural phenomena. Therefore, they are not evaluated further. B. Site Characteristics and Site-Related Events The site characteristics and site-related events important to the safety of the liquid hydrogen storage system are: (1) the location of the hydrogen storage facility relative to safety-related structures and equipment; Page 6 of 24 (6202E)
(2) the site-related characteristics that protect the storage area, and; (3) the control of bydrogen deliveries onsite.
- 1. Location Figure 1 is the site plot plan showing the location of the hydrogen storage area. The liquid hydrogen storage system will be located 1180 feet away from the nearest safety-related structure, which is the plant stack. The next nearest safety related structure is the Unit 3 reactor building, which is 1700 feet away. The nearest safety-related air intake is at the plant stack.
In order to protect the hydrogen storage area f rom any potential vehicle accidents, truck barriers are installed. 1 The approach road will be such that the delivery trucks will approach along the side of the storage area. The approach road is not a public road and is a route for only minimal plant traffic.
- 2. Security for Storage Area The storage area is inside the site boundary. The road leading to the approach road has a locked gate. An additional locked fence also surrounds the hydrogen storage area, as required by NFPA SOB.
- 3. Fire Protection There are no flammable materials stored near the hydrogen storage site. The system and its siting will meet NFPA requirements. The design basis for the overpressure protection systems for the liquid hydrogen storage tank assumes a hydrocarbon fire beneath these vessels. Thus external fires do not cause vessel failures. The hydrogen release through the relief devices may or may not be consumed in the fire, but in any case, would not impact other Peach Bottom safety related or non-safety related equipment. An external fire cannot cause a fire or detonation within the tank because of the absence of oxygen in the tank.
Ignition of relief valve discharges is possible, though not generally expected except in the case of an external fire. The preferred and most practical practice for extinguishing hydrogen fires, as stated in NFPA 50A and 50B, is simply to isolate the hydrogen supply. Even if the hydrogen supply Page 7 of 24 (6202E)
cannot be isolated, the fire will simply burn until the hydrogen is exhausted. This will have no impact on plant safety-related or non-safety related equipment.
- 4. Control of Construction Activities No construction activity of any kind is presently contemplated in the vicinity of the hydrogen storage system.
Any such construction would undergo a safety review to confirm that the hydrogen storage system would be unaffected.
- 5. Hydrogen Delivery and Offsite Hydrogen Transportation Figure 1 shows the route taken by the liquid hydrogen trucks through onsite areas. The hydrogen delivery trucks approach no closer than the storage area to safety-related equipment, and contain less hydrogen than the maximum sized storage
-tank. Therefore, the analyses of tank failures control the location of the storage area.
The liquid hydrogen delivery containers are of similar construction, and have similar overpressure protection systems, to that of the storage tank, except that the contents of the truck is less than that of the storage tank. One hydrogen supplier (Reference 4) reported that in more than 51,000,000 miles of delivery truck mileage a total of 6 serious accidents (truck roll-overs) had occurred; and that none of these accidents resulted in a liquid hydrogen spill, or a loss of vacuum between the inner and outer shells. Based upon the above discussion, the likelihood of a delivery-related accident is very small.
- 6. Conclusions About Site Related Events No site-related releases other than relief device action due to potential fires are postulated. This type of fire is considered a system-specific event, the effect of which is enveloped by more serious events discussed below.
Based on the excellent history of liquid hydrogen transportation, the possibility of a transportation related accident is very remote. Because the hydrogen delivery trucks approach no closer to the plant than the storage area and contain less hydrogen than the maximum sized storage tank, the analyses of tank failure are considered limiting on the location of the storage area. For these reasons, no transportation-related hydrogen release is postulated. Page 8 of 24 (6202E)
M C. External Events A set of external events was used as the basis for licensing the original plant and is dercribed in the PBAPS FSAR. These events have been considered to determine if any of them have the potential for causing a release of liquid hydrogen f rom the storage tank. These events include natural phenomena and offsite
~
industria/ hazards. For comparison, the way each event was treated in the FSAR is given first, followed (if applicable) by the evaluation of the event in light of the hydrogen storage system. .
- 1. j Earthquake
- a. Plant Design Basis The seismic design for safety-related structures and equipmend for PBAPS is based on dynamic analyses using accelera, tion or velocity response spectrum curves which are based on a horizontal ground motion of 0.12g; a 62 .
vertical acceleration equal to 0.08g is. assumed to occur ' 7, simultaneously.
'F.
f b. Hydrogen Storage System Design The liquid hydrogen storage tank and supporting structures will be analyzed for seismic loads using a static equivalent method. The liquid hydrogen storage tank and supporting structures will be capable of withstanding a seismic event of the magnitude used in plant safety-related system design. Therefore, the safe shutdown earthquake will, not cause f ailure either of the liquid hydrogen tank or 'of its supporting structures, and a seismic event .will not cause a large release of liquid hydrogen. 5 k' Piping containing liquid hydrogen' may not all be designed to withstand 'the safe shutdown' earthquake for PBAPS. For
,y non-seismic perticns,of, such piping, a break caused by an '
e earthquake would bejpoitulated. The amount of hydrogen
' I released due to such events will be restricted to an
( ,' acceptable level if n'ecessary by the use of h t seismically-supported excess flow check valves or by other flow-limiting devices. , l p t '
\ (j f .
j - l l
]
r, >A t . v p s e , a t
, ' s, e
- o Page 9 of 24
/
(6202E) s i
,A l
e
*? Y - e 5 ,* ./ __ ,,;_. __
2.- High Winds. Tornados.'and Tornado Missiles
- a. Plant Design Basis The design basis for PBAPS safety-related structures is a tornado with 300 miles per hour winds. Design basis tornado missiles are a 12 foot x 4 inch x 12 inch board traveling at 300 miles per hour striking end on, and a 4000 pound automobile propelled at 50 miles per hour.
The safety related structures are also designed for a 3 psi pressure differential caused by a tornado. In the case of the plant stack and reactor building above the refuel floor, these structures and equipment therein are not required in order to achieve or maintain safe l jf shutdown following a tornado. Therefore, they are not designed to withstand tornado effects.
- b. Hydrogen Storage System Design The liquid hydrogen storage tank and supporting structures will be designed to withstand the loads associated with tornado winds of 300 mph. Therefore, tornado winds will not cause any tank failures.
The tank is not capable of withstanding the design. basis tornado missiles. Therefore, a tornado missile could rupture small lines, or the storage tank and cause a large spill of liquid hydrogen. The tank failure is most limiting. The effects of a large spill of liquid hydrogen are discussed in Section IV.
- 3. Flood The hydrogen storage facility is 215 feet above the design basis flood level for the PBAPS Site, and flooding will have no effect on the storage area. Topography at the hydrogen storage site is such that local flooding is not a concern.
4 Offsite Industrial Hazards There are no offsite industrial hazards deemed of sufficient risk to require consideration in the design of the PBAPS. l The same conclusion is applicable to the hydrogen storage area. l l Page 10 of 24
'(6202E) l '^
- 5. Toxic Chemicals Toxic chemicals considered in PBAPS design basis will not affect the material strength of the liquid hydrogen storage facility. Toxic chemicals that would affect operator personnel would not affect the liquid hydrogen system, because no operator action is required for hydrogen storage system operation. Therefore, no failure of the hydrogen storage system due to toxic chemicals can occur that could lead to hydrogen releases.
- 6. Aircraft
- a. Plant Design Basis An evaluation of the potential hazard from aircraft in the vicinity of the PBAPS plant is contained in the FSAR. The conclusion of this evaluation is that the probability of an aircraft crashing into the PBAPS is sufficiently low that aircraft impact is not considered a design basis event.
- b. Hydrogen Storage System Design The evaluation of aircraft hazard for the plant is equally applicable to the hydrogen storage area, except that the hydrogen storage presents a smaller target area than the plant. Therefore, the probability of aircraft impact is even smaller for the hydrogen storage facility.
- 7. Lightning The hydrogen storage system will be protected from the effects of lightning per the requirements of NFPA 78. No hydrogen releases will be caused by lightning.
[ 8. Conclusions on Releases Caused by External Events Based on the above evaluations, the conclusions are: I ! a. No hazards associated with offsite industrial facilities or transportation routes will cause any hydrogen releases.
- b. No accidental hydrogen releases are expected due to onsite PECO activities. Nonetheless, the overpressure l protection system is designed to accommodate a fire
! directly under the storage vessels. The effects of any overpressure system related release would be less than those associated with those caused by severe natural phenomena. l Page 11 of 24 (6202E)
- c. The only natural phenomena with potential for causing a release of hydrogen are tornado missiles, which could rupture small lines containing hydrogen or create a hole in the tank, and.a seismic event, which could break small piping connected to the tank.
. d. The effects of system normal (very small) and abnormal operation releases of gaseous hydrogen are enveloped by releases caused by severe natural phenomena. 4 I i l~ l-l i- Page 12 of 24 (6202E) L
It GENERAL BEHAVIOR OF HYDROGEN RELEASES The conceivable hydrogen release events discussed in Section III are of three general types: (1) small releases of gaseous hydrogen from the liquid storage tank related piping or relief devices at a specific, determinable mass flow rate (2) small releases of liquid hydrogen from the liquid storage tank related piping at a specific, determinable mass flow l rate, and l (3) large releases of substantially all the liquid tank contents nearly at once. Hydrogen release may behave in any of three ways: . (1) dispersion duc to momentum, thermodynamics, and/or meteorology 1 (2) ignition resulting in a fireball type combustion, or ) l (3) ignition resulting in detonation. When mixed with air, hydrogen gas is flammable and potentially explosive. When hydrogen is released from the storage system, atmospheric processes and any fluid momentum contribute to the dilution and dispersion of the hydrogen. A hydrogen release will dilute first below the upper flammability limit (75%) and eventually below the lower flammability limit (4%). Between 75% and 4% there is a potential for ignition of the hydrogen air mixture, which would result in a fireball type combustion. Between 18.3% and 59% the potential for cloud detonation also exists under some conditions. This section of the report describes these types of behavior of releases of gaseous or liquid hydrogen. A logical result of this discussion is the identification of design criteria for locating a hydrogen storage system, such that the behaviors discussed in this section do not present unacceptable consequences to safety-related structures. Such criteria are given in Section V of this report. A. Dispersion Behavior of Hydrogen Releases Releases from the parts of the liquid hydrogen storage system l containing gaseous hydrogen have substantial momentum and relatively small volume such that they are fairly rapidly dispersed. Reference 1 uses a jet dispersion model to determine ljdhg the associated dispersion. This model is discussed in Reference l Page 13 of 24 l (6202E) t
- 3. Section V.B. Releases from the portions of the liquid hydrogen tank containing gaseous hydrogen are not as restrictive on the location of the storage area as other events.
Releases f rom the parts of the storage system containing liquid hydrogen will result in a liquid jet that will either flash to vapor due -to its superheated condition or evaporate rapidly upon hitting the ground. The resulting vapor has a low momentum. Gaussian dispersion models are considered the most appropriate way to model this type of release. The plume will be at or near neutral buoyancy and is assumed to disperse at or near ground level. B. Hydrogen Ignition Behavior Air Products and Chemicals, Inc. (APCI) has conducted a review of all known sources of liquid hydrogen spill events. This review is given in Reference 3. APCI's conclusions from these events regarding ignition were as follows: (1) vessel rupture was not shown to be a reliable source of ignition (only 3 out of 10 vessel ruptures resulted in ignition). (2) liquid hydrogen releases did not have a tendency to auto-ignite. There were identifiable sources of ignition for each case that ignited. (3) for the 79 tests where liquid hydrogen was released and ignition sources were intentionally provided, ignition always occurred without detonation. During spill tests in which ignition sources were not intentionally provided, spurious potential ignition sources were deliberately avoided. In normal industrial settings, this is likely not to be the case. In other words, ignition sources are more likely to exist. Potential ignition sources include flame, electrical, static or friction sparks, and anything above the auto-ignition temperature of hydrogen (932*F at I atm). A literature search located no studies of the likelihood of ignition in the vicinity of liquid hydrogen spills. However, hydrocarbon (e.g. methane) ignition data can be used as a conservative measure of this likelihood. James (Reference 5) examined records of the distance traveled by hydrocarbon vapor clouds in 81 rail car spills where ignition occurred. It was found that 58% ignited within 50 feet, 75% ignited within 100 feet, and all ignited within 300 feet. i I f Page 14 of 24 (6202E)
F When we compare hydrogen and methane ignition, we find that: (1) minimum energy required to ignite hydrogen is less than 7% of that required to ignite methane (2) the hydrogen auto-ignition temperature is 250*F lower than that of methane, and (3) hydrogen has a flammability range of 4 to 75% compared with 5 to 15% for methane. Minimum energies for spark ignition as a function of concentration of methane and of hydrogen in air are given in Figure 2 (taken from Reference 6). The curves clearly illustrate the much broader region for hydrogen ignition than that for methane. The minimum ignition energies for hydrogen and for methane are 0.019 and 0.28 millijoules, respectively. For comparison, 1.0 millijoule is the minimum spark energy that can be felt by humans (i.e., the human shock threshold). In other words, hydrogen can be ignited by a spark having less than one fiftieth of the energy of the static electricity one might feel when touching a door knob. C. Hydrogen Detonation Behavior The APCI review of hydrogen spill data in Reference 3 and the physics of detonation of flammable gases resulted in the following conclusions:
"As the test data have shown, no detonation of an unconfined liquid hydrogen spill has occurred. The unconfined conditions for these tests were mostly flat ground with only a shallow diked area and instrument towers as obstructions. It is well known that when flammable hydrogen-air mixtures are totally confined (e.g., a long pipe with closed ends) ignition can result in detonation. No quantitative data was found on the propensity of semi-confined liquid hydrogen spills to detonate or not detonate. A 1958 A. D.
Little report qualitatively discussed a test where liquid hydrogen spills were semi-confined by three walls of a test bay. The statement is made that 'even partial confinement can add substantially to the magnitude of the pressure wave generated by 4 the combustion of gases in free space'. It is also well known that objects immersed in a combustible vapor cloud can act as turbulence generators. These can cause acceleration of the ! burning velocity which could run up to a detonation. These considerations suggest that the possibility exists for the formation of a ' pressure wave' if ignition is delayed and the vapor cloud drifts in an environment that could provide accelerated combustion." l l l Page 15 of 24 (6202E) L
. - - .. . .~.
D. Hydrogen Release Behavior for the Hydrogen Storage System The facts given above (and elaborated in Reference 3) about the
- general-behavior of hydrogen releases lead to several logical observations.
(1) following any liquid spill resulting from violent disruption of the liquid hydrogen storage system, ignition near the site of the spill is considered very likely. (2) ignition of an unconfined cloud would be expected to result in a fireball (3) detonation may be possible in a semi-confined cloud, or a cloud in the vicinity of turbulence-generating structures or equipment that might be in relatively open surroundings (4) hydrogen vapor clouds drif ting between or being taken into structures by HVAC systems present some risk of detonation. (5) hydrogen releases through a pressure control valve, or relief device generally do not ignite, although there have been cases where they have. l l [ Page 16 of 24 (6202E) i
V. DESIGN CRITERIA RELATED TO HYDROGEN RELEASES A. Consideration of Release Events After reviewing the hydrogen behavior presented in Section IV, one can identify three scenarios following a release of hydrogen that could be unacceptable if not properly addressed using suitable design criteria. Those scenarios are: (1) if a hydrogen vapor cloud ignited where a fire could damage safety-related equipment (2) if a hydrogen vapor cloud ignited with a run-up to a detonation, where resulting blast overpressures could damage safety-related equipment, and (3) if a hydrogen vapor cloud could drif t to where it could be taken in by plant ventilation equipment or other equipment requiring an air supply B. Design Criteria The following design criteria provides adequate protection for safety-related structures from the three scenarios given above. (1) Prevention of Releases The potential for releases of liquid or gaseous hydrogen must be minimized to the extent practical by design of the hydrogen storage system and the equipment utilized for its protection. (2) Protection from Fireball The distance from the tank to safety-related structures must , be such that a fireball type of combustion at the spill site, ' which is the most likely consequence of a large spill, will not prevent safety-related structures from performing their design functions. (3) Protection from Detonation ! Even though detonations of unconfined hydrogen clouds have not occurred, the distance from the tank to safety-related structures must be sufficient to assure that complete tank failure and detonation of the tank contents in the detonable 1 i l E# l (6202E) l
concentration range, at the spill site, will not result in peak overpressures that would prevent those structures from performing their design functions. (4) Prevention from Hydrogen Accumulation Near Safety-Related Equipment For any hydrogen release, either early ignition of the resulting cloud must be sufficiently likely that the cloud will not reach any plant air intake system, or such releases must be out of the flansnable range due to dilution in the distance from the point of release to safety-related structures or plant air intake system. C. Relationship of Design Criteria to Reference 1 Guidance l Criterion 1 -- Prevention of Releases: Reference 1 provides guidance for the design of liquid hydrogen l supplies which meets and exceeds industry standards for the associated equipment. Reference 1 recommendations for protection g of storage equipment and safe handling of liquid hydrogen by experienced suppliers make abnormal hydrogen releases very unlikely. In fact, only some of the most severe natural phenomena used as the design basis for safety related equipment are expected to have the potential for causing significant hydrogen releases. Therefore, Criterion 1 is met by use of the Reference 1 guidance. lh Criterion 2 -- Protection from Fireball: Reference 1 provides analyses of fireball durations and heat fluxes assuming that the entire tank contents are involved. l[ These results can be used to establish that the distance from-the hydrogen storage area to the safety-related structures is sufficient to protect safety-related equipment, j The details of this type of analysis are discussed in Reference 3. The Reference 1 analysis will be used as the basis for determining l l that the PBAPS HWC System meets Criterion 2. l Page 18 of 24 (6202E)
Criterion 3 -- Protection from Postulated Detonation: Reference.1 providet, conservative analyses of TNT equivalency for l a hydrogen release associated with tank failure. Blast overpressures and impulses can then be calculated using the U.S. Army Technical Manual TM5-1300. Reference'l provides minimum separation distances vs. tank size l assuming that safety-related, reinforced concrete structures can withstand an incident blast overpressure of 6 psi. Reference 4 and Appendix A provide additional discussion of historical data on the response of reinforced concrete structures to blast loading. These discussions support the recommendations - in Reference 1. l The hydrogen storage system location will meet, with considerable - margin, the Reference 1 recommended separation distances. lh The calculated incident overpressure at the plant stack and Unit 3 reactor building are 1.21 and 0.76 psi, respectively. Criterion 4 -- Prevention of Hydrogen Accumulation Near Safety-Related Equipment: Reference 1 is based on the fact that violent storage tank damage such as that caused by an aircraft impact or tornado missiles will lh provide ample ignition sources such that early ignition is assured.
.In Reference 1 early ignition is not considered to be assured for l
events such as earthquakes that only result in pipe breaks. - Therefore, dispersion analyses are done and appropriate separation distances vs. break size are calculated. The Reference 1 guidance was used to establish the adequacy of l separation distances for pipe breaks. Page 19 of 24 (6202E)
VI. HYDROGEN RELEASE CONSEQUENCE ACCEPTABILITY Reference 1 provides the consequence analyses which are used in the l jf assessments given below of the acceptability of the proposed hydrogen storage system location. Based upon the previous sections, we have identified events that result in either: (1) no release, (2) a small gas release from the liquid. storage tank, (3) a small liquid release, or (4) a large liquid release. A summary of the events, analyses of the consequences of the releases using the Reference 1 guidance, and assessment of storage l site location acceptability are presented in this section. A. No-Release Events Either the facility design or site characteristics prevent certain events from causing hydrogen releases. The events are listed below. They include PBAPS safety-related structure design basis events and events specifically related to the hydrogen storage system. (1) Floods (storage area above design basis flood level); (2) High winds including 300 mph tornado winds (storage vessel design basis); (3) Toxic chemical releases (no impact); (4) Onsite and offsite industrial hazards (no PBAPS design basis events); (5) Manufacturing defect (prevented by quality control and application of appropriate code requirements); (6) Delivery or other vehicle impact with hydrogen storage facility equipment (area protected by barriers); (7) Construction activity onsite (none planned near storage site); (8) Lightning strikes (facility grounded); (9) Aircraft impact (not a PBAPS design basis); (10) Hydrogen delivery truck accidents (very unlikely and enveloped by other analyses); B. Small Gaseous Hydrogen Releases
- 1. Summary of Events The events that could cause small gaseous releases f rom the liquid storage portion of the hydrogen storage system are:
(1) Failure of small piping caused by earthquakes; (2) System abnormal operations leading to discharge through relief devices. (3) Extremely small and inconsequential pressure control valve releases. Page 20 of 24 (6202E)
r l l l l l l The worst case event is the seismically-induced rupture of relief piping upstream of the relief devices. This piping will be 2 inches or less in diameter.
- 2. Dispersion, Blast-Overpressure, and Fireball Effects Figure 3 (from Reference 1, Figure 4-6) shows minimum acceptable separation distances from release point to jj safety-related structures (based on blast overpressure calculations) and safety-related air intakes (based upon distance required to be below the lower flammability limit) vs. hole size in the storage system. For a 2 inch hole, 460 feet of separation is required from safety related air intakes. For blast protection, the (a) curve is considered 2 conservatively applicable for all PBAPS safety related structures. The required separation distances from this curve is 193 feet.
The minimum separation distance provided at PBAPS is 1180 feet for safety related structures and safety related air intakes, and is, therefore, acceptable. Therefore, Criterion 4 for preventing the intake of flammable concentrations of hydrogen into safety-related air intakes and Criterion 3 for blast ef fects are met. These events are not the limiting case for protection from fireball effects (Criterion 2). C. Small Liquid Release
- 1. Summary of Events The events that could cause a small release of liquid hydrogen are:
(1) Earthquake induced line break; (2) Tornado missile induced line break: The effects of tornado missile induced line break are enveloped by the impact of tornado missile induced vessel rupture, which is discussed under large liquid releases.
- 2. Dispersion, Blast-Overpressure, and Fireball Effects The consequences of a earthquake-induced line break are acceptable if the hydrogen is out of the flammable range before reaching safety-related air intake structures, and if detonation of the vaporized liquid in the explosive concentration range will not impact safety-related structures.
l Page 21 of 24 (6202E)
Figure 4 (from Reference 1 Figure 4-7) indicates that, for the 1180 feet of separation from safety-related intakes lh provided at PBAPS, the mass flow rate should be limited to leu than approximately 1.1 kg/sec. This corresponds to a j bra .4 in the liquid-line equivalent to an orifice of about 0.44 inches diameter. Air intake related separation l requirements are more limiting than blast overpressures. Based on the above, piping containing liquid hydrogen will be either (1) totally seismically supported; (2) seismically
. designed and supported up to and including a structural anchor downstream of excess flow check valves: or (3) seismically supported up to and including a structural anchor downstream of devices restricting mass flow in the event of a break to l'.1 kg/sec. or less. lh Based on the above, Criterion 4 for preventing the intake of flammable concentrations of hydrogen into safety-related air intakes and Criteria 3 for blast ef fects will be met. These events are not the limiting case for protection from fireball effects (Criterion 2).
D. Large Liquid Release
- 1. Summary of Events The only event that could cause large liquid hydrogen releases from the liquid hydrogen storage tank at the PBAPS site is a direct hit by a design basis tornado missile.
- 2. Dispersion. Blast-Overpressure, and Fireball Effects The PBAPS separation distances are 1180 feet to the nearest safety related structure (plant stack); and 1700 feet from the nearest safety related structure (reactor building) which is needed for safe shutdown in the event of a design basis tornado. Appendix A to this report provides detailed discussion on the capability of PBAPS safety-related structures to withstand postulated blast loadings. The conclusion of Appendix A, which is based on identification of loads, a review of historical data on blast effects, and Reference 1 requirements, is that the hydrogen supply system is safely separated from PBAPS safety-related structures.
Figure 5 (f roin Reference 1, Figure 4-4) shows the results of analyses of fireball heat fluxes and durations. This clearly l k shows that, at the separation distances provided at PBAPS, no unacceptable effects would result. Thus Criterion 2 is met for the worst case fireball. Page 22 of 24 (6202E)
A comparison of ignition
- sources caused by a tornado and associated missile strikes at the hydrogen storage system location with those associated with hydrocarbon spills caused by rail car accidents.(see Section IV.B) is difficult. Both events are destructive to industrial equipment on a significant scale and both would have the potential to disrupt electrical equipment, leading to arcing and other ignition sources. Moreover, the thunderstorms that frequently accompany tornados and high winds are likely to be sources of static charge -- the extreme case being lightning. Early ignition following a tornado missile is considered likely.
Any ignition would lead to a fireball type combustion and prevent any hydrogen cloud from reaching any plant air intake. Thus Criterion 4 is met and no dispersion analysis is considered necessary. It is also expected that dispersion of a tornado induced release will be rapid and that explosive and flammable concentrations would only occur in the immediate vicinity of the spill site and would be short lived. l l f l Page 23 of 24 (6202E) L- -
VII. REFERENCES
- 1. " Guidelines for Permanent BWR Hydrogen Water Chemistry Installations" 1987 Revision, by BWR Owners Group for ICSCC Research, Hydrogen Installation Subcommittee. f
- 2. Linney, R. E., " Hazard Analysis for Liquid Hydrogen Customer Station Located at the Commonwealth Edison Dresden Nuclear Power Plant", Air Products and Chemicals, Inc., October 22, 1984
- 3. ~Camilli, L. L.: Linney, R. E. & Doelp, L. C. , " Liquid Hydrogen Storage System -- Hazardous Consequence Analysis", Air Products and Chemicals, Inc., October 1, 1985.
- 4. Reichert, P. T.; Seeley, G. T., " Safety Analysis for the Liquid Hydrogen Storage System -- Dresden Nuclear Power Station", Stearns Catalytic Corporation, October 31, 1985.
- 5. James, G. B., " Fire Prevention in the Chemical Industry", NFPA Quarterly, Vol. 41, pg. 256.
- 6. Lewis, V. and Von Elbe, G., Combustion, Flames and Explosions of Cases, 1951.
- 7. Department of the Army, Structures to Resist the Effects of Accidental Explosion, TM5-1300, June 1969.
2 Page 24 of 24 (6202E)
WM23W##MZ:sais=
/ a~v m <,p m a 7.
) l p '
./} n ') yj-,i ,
l. l
>l f -
j fr ' - e;7,lf/ M+it s w < / y.T . * . w--- .% ( / ,a .._ /, M, y y1 :,
,3 N s: . s WP-5?
s-
-{ M s
% %/ ,d L
a 4 _
% \ / .j [: k._ I
_1 g{ c t% (D 2) il j Mk i. I ~~ (
..j
! g) * , . . . .
e 3r , - tj A ' s
~
'N N ,- 9 b N '*h cj?/dV(d mO!!rt c.. x..'
f//t ti 6 3 ., .Os 3!di
;p.
W '.'J!' .r Jhg hf
- g ;- . y
- v. ,.
/;c
\,
til gIh;.,.[' i y- - -
- N s ./ I L .
'/ < .. jl li wm;;
' h,j' &\
no spL, Y$' h ; b (f. r aa ,U gkfW -... g.
- s
';;Wb y, ,,.
ns...]a!;hf .
.r ,
~
yk3iq ;S / ', g- N@r \ n! - ! as (
'b
. .e !. 4.
9, Er MN\ %C . . . . , l(' Q hh t - q , de # =. , M- .i d. , a., _
~ .x.
FIGURE 2 HYDROGEN VS. METHANE IGNITION ENERGIE5* _ _ _ ,..._.. _. J 10 , ._ ,_ _ . . _ _ _ . . . . . . . . . .
. . _ . - -3.. .__ .. _ . . _ , ._.4..._......... .
s.
. . . .s
...a,.
a..s.- .
. . . . . . . . . . . . . . . . * . .a . . . ..
.... ..i_. ..-. - . .. . -
. ... 4 .
- 4. -
4 .
...t...HE.THANt.h... ..
e... _ ,.
. ._.i .
. _ . . . _ _ _ . ..i
.....i......_-
_.._._..1,. ._. . . _. . . .. _ . _. _. . ._ . . .. _
. . . .. . .. _ _..a._.. .. __....i._._...... .... .
- 1. 0 _ ._ ... _.. __ . ...
. . . . . . . _ . ._ _ . . . . . , ..r.. _. ..__ .. . . ..
. _ . . . . . . . . . . . = . .
MINIMUM . IGNITI0N . . . ENERGY .- .... . -... . .. .. . - - (MILLIJOULES) _ _ _ . .. = .
._- HYDROGEN 0 .1 ._ _. . . . . . . . . _ . . . . . . . . . . . . . . . .
s....... . . . . . . . . _ _ .
..__;.. ..;._._,.......3.. _ _ . . _ . , . . . . _ ..
__ ..- ...... __ _a....- .. .... . . .
...._._....__......_..?._...._. . . _
_ . . . . _ . . . . . _ . . . . . _. .. . o. i .._._t....-... g __i....._ -
.4.
...r. . s .. .
__ ;. _ _ _ . ._ . ..._s... ........ - - - - - . . . . -
...r
. _ -s . ..
i . : .. . . . . . - 0.01 . .
... .. . . . . .. .. .u-. .
0 10 20 30 40 __50 60 , CONCENTRATIONINAIR(%)
- FROM LEWIS 8.' AND YON ELBE C. COM.BUSTION, FLAMES AND EXPLOSION OF GASES,1951
FIGURE 3 Distances to safety related structures 10 3 '- (a) a 8-in. reinforced wall 8 - (b) = 18 in.; P. = 1.5 psi; # = 0.12 ksi 100 %. 6 - (c) a 18 in.: P, a 3.0 psi: = 0.30 ksi 4 D 460 f t (b) l u (a) (c) 5 Minimum required separation e distance to air pathways into E 2~ safety related structures $ lil/t T 10 2 - I34/I k ~5 8 - E c: 6 - E E 5 4 - 2 2 -
' I I ' '
I I 10 O.1 2 4 6 8 1.0 2 4 6 8 10 Hole Size or inside Diameter of Pipe (in.) From Reference 1: Figure 4 6. Minimum required separation distance versus hole size and ID of pice for gaseous releases from 150 psig liquid hydrogen storage tank.
l FIGURE 4 10' 8 - Distances to safety.related structures 6 (a) a 8 in. reinforced wall - - 4 - p'q (b) a 18 in.: P a 1.5 psi; a 0.12 ksi 1 e 1700 f t : 000A:
~
Ig - (c)' a 18 in.; P a 3.0 psi #' a 0.30 ksi
,,10004, 3 .$ 10 - * /t8ott (a)
C 8 - Minimum required separation W
.5 6 - eistance to air patnways into lii safety.related structures (C) lii 4 _
V2 E 2 - A
- l. I A /=c 3
- !6 - o.vuJL 10 2
4 - 1 I I I 14 12 1.0 2 Hole size or ID of pipe (in.) 10 I I II! I I ' '! I I I ' O.01 2 4 680.1 2 4 6 81.0 2 4 6 8to Discharge Rate (kgls) Frori Reference 1: Figure 4 7. Minimum required separation distance versus hole size and discharge rate for liquid releases from 150 psig liquid hydrogen storage tank F weather stability,1 m/s wind velocity.
FIGURE 5 I Key Tank Size (gal) Fireball Duration (sec) _ o 20.000 8.18 0 18.000 7.90 a 9.000 6.27 7 6.000 5.48 2 0 3.000 4.35 10 - e 1,500 3.45 g 5 - Charring of [ wood surfaces ? E 10 - s T
' I I I '
1 I l I I I O 200 400 600 800 1000 1200 1400 Distance from Fireball Center (ft) FiguSe 4dfYhermal flux versus distance from fireball center for fiquid hydrogen storage system. b
APPENDIX A TO THE HYDROGEN SAFETY ANALYSIS CAPABILITY OF SAFETY RELATED STRUCTURES TO WITHSTAND BLAST LOADINGS Al. INTRODUCTION A2. CALCULATED BLAST OVERPRESSURES A3. HISTORICAL DATA ON BLAST EFFECTS A3.1 JARRETT FORMULATION -- USED IN REFERENCE A5 jf A3.2 EQUIVALENT REINFORCED CONCRETE DATA A3.3 CONSIDERATION OF ACCEPTABLE DUCTILITY FACTORS l j hs A3.4 BLAST EFFECTS ON REINFORCED CONCRETE STACKS A4. EVALUATION OF PEACH BOTTOM CRITICAL STRUCTURE BLAST RESPONSE CAPABILITIES A4.1 PLANT STACK A4.2 REACTOR BUILDINGS A4.2.1 CONSIDERATION OF HISTORICAL DATA l2 i A4.2.2 CONSIDERATION OF " GUIDELINES FOR PERMANENT BWR HYDROGEN WATER CHEMISTRY INSTALLATIONS" j(( A4.2.3 USE OF NUREG/CR-2462 METHODOLOGY (6203E)
A1. INTRODUCTION This appendix provides justification for the determination that the l jdhg separation between the postulated hydrogen explosions and Peach Bottom safety related structures is satisfactory. Identification of critical structures and the blast parameters from the limiting postulated explosion is provided in Section A2. l The historical data on blast effects which provide a basis for expected blast response capabilities are discussed in Section A3. I A Section 44 provides an evaluation of blast response capabilities of critical structures considering both historical data and analytical l /h methods. The conclusion from these evaluations is that postulated explosions resulting from very unlikely, large releases from the hydrogen storage system will have no impact on the safety related equipment required to achieve safe shutdown following the release event. A-1 (6203E)
A2. CALCULATED BLAST OVERPRESSURES The two nearest safety related structures to the liquid hydrogen storage site are the Plant Stack and the Unit 3 reactor building. These are located at 1180 ft. and 1700 feet, respectively, from the hydrogen storage site. Safety related structures beyond these distances are considered either to not be threatened by the l postulated explosions due to the expected very low overpressures at these locations, or the response is enveloped by the results l presented herein. I L The worst case postulated explosion of released hydrogen for these two structures results from a tornado missile striking and piercing the liquid hydrogen tank, releasing its entire contents, followed by detonation of the resulting vapor cloud. The conversion factor from Reference A5 yields a 27,400 lb. TNT equivalence for the 20,000 lj~ gallon tank. Table 1 provides the relevant blast parameters. TABLE 1 BLAST PARAMETERS AT CRITICAL LOCATIONS (for 27,400 lb. TNT equivalent explosion) (Calculated Per Reference A2, Page 122) UNIT 3 PARAMETER PLANT STACK REACTOR BUILDING Distance (ft.) 1180 1700 Scaled Distance (ft./lb. 1/3) 39.14 56.39 Peak Side-On Overpressure (psi) 1.211 0.761 Peak Reflected Overpressure (psi) 2.506 1.555 Shock Wave Velocity (ft./sec.) 1155 1141 Particle or Wind Velocity (ft./sec.) 63.48 40.40 Dynamic or Wind Pressure (psi) 0.0352 0.0140 A-2 (6203E)
A3. HISTORICAL DATA ON BLAST EFFECTS Historical data on the effect of blasts on structures is principally from World War II. The effects of both conventional and nuclear explosives have been used to determine thresholds for various levels of damage to structures. A3.1 Jarrett Formulation -- Used In Reference A5 lh Reference Al provides a range of damage thresholds based upon a review of well documented blast effects on brick construction dwelling units. The threshold equation and constants are restated below: //3 l + (( Where R = Distance from Explosion in Feet W = TNT Weight Equivalent in Pounds DAMAGE EXPECTED (Incident Overpressure for Large Explosions) K = 9.5 - Almost complete demolition. (10.6 psi) K= 14 - 50 to 75 percent external brickwork destroyed or rendered unsafe and requiring demolition. (5.2 psi) K= 24 - houses uninhabitable - partial or total collapse of roof, partial demolition of one to two external walls, severe - damage to load bearing partitions requiring replacement. (2.3 psi) K= 70 - not exceeding minor structural damage, and partitions and joinery wrenched from fixings (0.57 psi) K = 140 - remaining inhabitable after repair - some damage to ceilings and tiling, more than 10 percent window glass broken. (0.22 psi) A3.2 Reinforced Concrete Data Data and conclusions in Reference A2 and A3, dealing with nuclear lk explosions and seismically designed reinforced concrete structures suggest the following K values: DAMAGE EXPECTED (Incident Overpressure for Large Explosions) K = 5.7 - SEVERE DAMAGE -- Walls shattered, severe frame distortion, incipient collapse. (35 psi) A-3 (6203E)
K = 6.5 - MODERATE DAMAGE -- Walls breached or on the point of being so, frame distorted. Entranceways damaged, doors blown in or jammed, extensive spalling of concrete. (25 psi) K = 9.8 - SLIGHT OR LIGHT DAMAGE -- Some cracking of concrete walls and frame. (10 psi) References A2 and A3 can be used for brick construction to predict a l zd i K of 14 for severe damage and 21 for moderate damage. These are in reasonable agreement with Reference A1. As can be seen from l historical data, seismically designed, reinforced concrete structures have considerable blast resistance. This blast resistance is far in excess of brick construction. The K = 9.8 damage threshold is associated with a level of damage which might be considered acceptable for nuclear power plant structures, i.e., some cracking of structural concrete, but no spalling or loss of integrity. However, for several reasons, a value should be chosen which provides some additional margin. These jp reasons include (1) the limited amount of data, (2) there may be significant structural differences, e.g. window openings may have served to reduce the differential pressure durations (3) the nuclear blasts may have been above the structure, (reducing the reflected overpressure effects), (4) the likely variation in historical response. A value of K equal to 20 would appear conservative for nuclear power plant use given that safety-related structures are designed for jp seismic events and tornado wind pressures. This would result in the placement of the explosive hazard at twice the distance and resulting overpressures at 30 percent of those historically associated with only slight damage. Effectively, a still more conservative K of 45 is used in Regulatory Guide 1.91. This would result in the placement of the explosive hazard at 4.5 times the distance and resulting overpressures at 10 percent of those historically associated with only slight damage. A.3.3 Consideration of Acceptable Ductility Factors One method of considering the actual blast resistance capability of reinforced concrete structures is use of ductility factory acceptance criteria which are more consistent with analyzed results of /\ historical experience. References A6 and A7 provide overall system ductility limit recommendations, and estimates of effects of varying degrees of displacement. A-4 (6203E)
For the analysis of a simple span Reference A7 indicates that support rotations of up to 12 degrees might be required before structural collapse, but to eliminate the formation of spalled fragments, a limit of 2 degrees is considered potentially effective. A limit of up to 5 degrees is considered acceptable for personnel protection although spalled debris protection might be required. The corresponding ductility factors for these three cases are 50, 8 and 21 respectively. Since no spalling would be acceptable in the nuclear power plant environment, acceptable ductility factors should be below 8. Reference A6 suggests the use of a ductility factor acceptance criteria of 7 for analysis of flexure failure potential in reinforced concrete. Reference A4 suggests that a ductility factor of 3 be used for nuclear power plants to maintain a measure of conservatism. The reduction in required separation distances with higher factors is modest. This is the ductility factor which is used in Reference A5 and the analyses below. No direct correlation can be made between general K values and wall specific analyses using various ductility factors. In general, a one-way-action analyses of blast response capability using the above range of ductility factors would be conservative relative to conclusions based upon historical blast response behavior. A.3.4 Blast Effects On Reinforced Concrete Stacks Reinforced concrete stacks respond to blast effects differently than do reinforced concrete buildings. These structures do not respond to blast overpressures because the blast pressures are quickly equalized due to their small lateral size. For this reason, stacks will respond principally to the dynamic pressure (wind loading or drag forces) which accompany the blast wave. (See Reference A2 for lj additional discussion.) In Reference A2, two industrial stacks were observed which survived l j 15 psi blast overpressure (with resulting 4 psi dynamic pressure). One stack was observed to have failed, with a 30 psi blast overpressure (with a resulting 20 psi dynamic pressure). Stacks survive dynamic pressures which are greater than their static pressure design basis because of the very short duration of these loads relative to the stack period. A-5 (6203E)
A4. EVALUATION OF PEACH BOTTOM CRITICAL STRUCTURE BLAST RESPONSE CAPABILITIES A4.1 Plant Stack The Plant Stack is not required for safe shutdown of the Peach Bottom Plant. It is seismically designed but not designed for tornado winds. Its safety function is to provide release dilution in the event of a design basis accident. The stack is designed for wind loads varying from 25 psf at the base to 55 psf at the top (0.174 to 0.382 psi). The calculated blast overpressure, dynamic pressure, and particle (wind) speed are given in Table 1, above. The dynamic pressure is far below those which have been observed to cause stack failure. In addition, the short duration, dynamic pressure is below the stack static wind loading design basis. Therefore, the postulated explosion of release hydrogen will have no effect on the plant stack. A4.2 Reactor Buildings The Reactor Buildings are constructed of two foot thick concrete walls, with floor elevations relative to grade of 0, 30, 60, and 99 feet. The wall above the refuel floor are of a steel frame and siding construction not designed for tornado winds or missiles. The concrete walls are designed for 300 mph wind speeds and a 3 psi tornado induced differential pressure. The Reactor Buildings are 1700 feet from the hydrogen storage site. The conclusion from the following discussions is that, by any reasonable assessment, a postulated explosion of released hydrogen at the hydrogen storage site will not prevent the PBAPS reactor 2 buildings from performing their safety functions including achieving safe shutdown. A4.2.1 Consideration of Historical Data l/\ The calculated incident overpressure of 0.76 psi is far below the observed threshold for slight damage, (cracking in walls and frames) of approximately 10 psi overpressure. The incident overpressure of 0.76 psi is also below the 3.0 psi A associated with a K of 20 and the 1 psi associated with Regulatory QL Guide 1.91 guidance. A-6 (6203')
y v, , s > ( ,' f .
& w - 1
) ?
s . A4.2.2 Considerationpf"Guideli]nesforPermanentBWRHydrogenWater Chemistry d 9BP Revision" (Rtference A5)
,, 'k ReferenceAS,providestwkcriteriafortheseparationofhydrogen supply systems from safety-related structures which could be conservatively app'11ed to the PBAPS Reactor Building.
~
+
)
The first criterion is ,the,Jarrett formulation which is described in Section A3.1. Referen'Ce A5 uses a K of 56 which corresponds to an s' incident overpressure'of 0.77 psi and which has historically resulted in less than a 5% probability of serious structural damage to I). standard British unreinforced brick dwellings. The resulting
- .e' required separation distance would be 1670 feet. It is suggested in Reference A5 that K = 56 is suitable for any reinforced concrete or reinforced masonry wall which is greater than 8 inches in thickness.
'Oiven the historical data on reinforced concrete wall blast response 'r < in Section A3.2, this criterion is extremely conservative.
j ( ' Nonetheless, this criterion is met since the provided separation distance is 1700 feet. The second criterion which is conservatively applicable to the PBAPS Reactor Building is based upon a simple span analysis for an 18 inch thick wall having a static capacity of 3.0 psi, 0.2% tensile steel reinforcement and an allowable ductility factor of 3. The resulting required separation distance calculated from Reference AS, Appendix B, Equation 7 is 776 feet. This would correspond to an incident overpressure of 2.09 psi. Again, this criterion is met since the provided separation distance is 1700 feet. Reference A5 als.) suggests the use of NUREG-2462 " Capacity of Nuclear Power Plant Structures to Resist Blast Loadings" (Reference A4) where the actual plant design may not correspond to the pre-analyzed conditions in Reference AS. Analysis of the response of the Reactor Building is considered unnecessary given the very low overpressures expected from the postulated hydrogen release explosion. However, a illustrative calculation, using the methodology discussed in NUREC/CR-2462 is included below to assess the possible results of an - analytical approach. A4.2.3 Use of NURcG/CR-2462 Methodology The structure of the reactor building wall facing the hydrogen storage area was reviewed to identify a nominally weakest element for a dynamic response analysis using the NUREG/CR-2462 methodology. The exposed lowest wall was selected for this anlysis, and modeled as a single span. The ends were conservatively modeled as providing simple support at the base and fixed support at the top. The output 2 of this analysis is attached. The walls above this span have been determined to have a higher dynamic strength, but are strengthened by heavily reinforced concrete T's which cannot be simply analyzed using NUREG/CR-2462. A-7 (6203E)
The calculated minimum allowable separation distance from the 27.400 lb. TNT equivalent blast is 560 feet. Again, the provided distance of 1,700 feet is well in excess of this criteria. With the 1,700 lk feet separation distance, the panel response is entirely elastic. h
'A4.2.4 Refuel Floor Effects The ef fects on the reactor building refuel floor (of a hydrogen storage tank breach, caused by a tornado missile, and followed by content release and postulated detonation) are considered to be comparable to that of a tornado striking the reactor building, i.e.
the steel walls may fail but safe shutdown would not be prevented. I A-8 (6203E)
REFERENCES A1. Jarrett, D. E., " Derivation of the British Explosives Safety Distances", Prevention of and Protection Against Accidental Explosion Ih of Munitions, Fuels and Other Hazardous Mixtures, Annals of the New York Academy of Sciences, Volume 152, October 28. 1968. A2. Glasstone, Samual, "The Effects of Nuclear Weapons", 1962. lk A3. The Lovelace Foundation under contract to the U.S. Atomic Energy Commission, " Nuclear Bomb Effects Computer", 1962. lk A4 NUREG/CR-2462, " Capacity of Nuclear Power Plant Structures to Resist [ Blast Loadings", September 1983. AS. " Guidelines for Permanent BWR Hydrogen Water Chemistry Installations", 1987 Revision, by BWR Owners Group for IGSCC Research, Hydrogen 2. Installation Subcommittee. A6. ASCE Manuals and Reports on Engineering Practice No. 42, " Design of ( Structures to Resist Nuclear Weapons Effects", American Society of Civil Engineers, New York, 1985. A7. " Structures to Resist the Effects of Accidental Explosions", TM 5-1300, Department of the Army, June 1969. l i l A-9 (6203E)
PROGRAM BLAST-RD Thi') progree is for determining the capabilities of reinforced concrete to l tithstand blast loadince. The methods and assumptions are directly fron l NUREG/CR 2462, ' Capability of Nuclear Power Plants to Withstand Blast i L0rdings', Sandia National Laboratories, September, 1983. Equation and 1 Raf;rences are for this report unless otheretse noted. l On3 exception is that this program uses actual, normal incident, reflected l cv;rpressures as opposed to assuming that the reflected overpressure is Ju:t 2 times the incident overpressure. Bicot side-on overpressures and impulses are calculated using curvefits to d;tn from ' Air Blast Parameters Versus Distance for Hemispherical TNT
-Surface Bursts", Re 1344 Ballistic Research Laboratories, Aberdeen Prcving Ground, MD, port No. September 19k6.
PROGRAM LAST REVISED: MAY 7, 1987
** NUREG/CR 2462 Ref. **
INPUT WALL THICKNESS (inches) = 24.00 INPUT DEPTH FROM COMPRESSION SURFACE TO CENTROID OF TENSILE STEEL (inches) = 21.50 INPUT RATIO OF AREA 0F TENSILE STEEL TO TO EFFECTIVE AREA OF CONCRETE: 2.280E-003 INPUT MAXIMUM SPAN LENGTH (feet) = 30. 00 UALL MODELED AS A SINGLE SPAN SIMPLY SUPPORTED AT ONE END FIXED AT OTHER. MU = ULTIMATE STATIC MOMENT CAPACITY tin-lb/in width) = 5.$8E+004 ** From Equation (All. **
** From Equation (A9). *e Ks 1.74E-001 **
IG = GROSS MOMENT OF INERTIA (in.*4/in vidth) = 8.28E+002 es From Equation (A7). IC = CRACKED MOMENT OF INERTIA (in.*4/in vidth) = 1.42E*e02 es From E untion (A8). so IA s AVERAGE MOMENT OF INERTIA tin.*4/in vidth) = 4.85E+002 es From E untion (A6). se PS = CALCULATED STATIC CAPACITY (psi) = 3.44 en From Table A29. ** RM = EFFECTIVE YIELD POINT RESISTANCE OF STRUCTURE (psi) = 5.68 se From Table A29. ** es From Table A29. ** T a BUILDING PERIOD (milliseconds) = 93 HC = CLEARING LENGTH (ft) = 90.00 TC = CLEARING TIME (milliseconds) = 225 se From Equation (2.2). ** INPUT DISTANCE FROM EXPLOSION TO WALL (feet) = 1700 INPUT TNT EQUIVALENT (1bs.) = 2.740E+004 PS3 = PEAK INCIDENT OVERPRESSURE (psi) = 0.76 PR = NORMAL INCIDENT REFLECTED OVERPRESSURE (psi) = 1.56 ISO = INCIDENT IMPULSE (poi-es) = 4.77E*001 125 es From Equation (2.1). se TI = INCIDENT TRIANGULAR LOAD DURATION (ms) = DUCTILITY DEMAND AT SELECTED DISTANCE FROM POSTULATED EXPLOSION = 0.45 j (Conservatively ignoring any clearing time effects) l l WITH 27400 POUNDS OF TNT EQUIVALENT AT MINIMUM ALLOWABLE DISTANCE: l AND WITH AN ALLOWABLE DUCTILITY OF 3.00: PATH #1: CLEARING TIME IS GREATER THAN LOAD DURATION ALLOWABLE 2 (scaled distance) = 18.6 ALLOWABLE INCIDENT OVERPRESSURE (psi) = 3.32 ASSOCIATED REFLECTED OVERPRESSURE (psi) = 7.27 ALLOWABLE R = 560 (feet) IMPULSE AT ALLOWABLE 2 (psi-es) = 1.39E+002 TRIANGULAR DURATION AT ALLOWABLE Z (as) = 84 TD e DURATION OF LOADING (es) = 84 es From Equation (A19). es TD/T = 8.96E-001 es Taken from Figure A-4. ** TM/T = 6.87E-001 TM s TIME OF MAXIMUM RESPONSE (es) = 64 ** F1/RM = PEAK (REFLECTED) PRESSURE / YIELD POINT RESISTANCE = 1.28E+000 es From Figure A-4.
APPENDIX B TO THE HYDROGEN STORAGE SYSTEM SAFETY ANALYSIS IMPACT ON TRANSMISSION EQUIPMENT IN THE VICINITY OF THE HYDROGEN STORAGE SYSTEM Bl. INTRODUCTION B2. RELATIVE LOCATION OF NON-SAFETY RELATED EQUIPMENT B3. COMPLIANCE WITH NFPA SEPARATION REQUIREMENTS B4. NORMAL AND ABNORMAL HYDROGEN RELEASES
~ B5. IMPACT OF RELEASE sEVENTS B5.1 NORMAL OPERATION B5.2 ABNORMAL OPERATION B6 CONCLUSIONS l
(6204E)
Bl. INTRODUCTION The hydrogen storage faellity is located at a safe distance from safety related structures. This is discussed in the main body of this report. The goal for safety related structures is to assure that they are undamaged, such that safe shutdown can be accomplished for any release from the hydrogen storage facility which has a credible cause. The impacts of the full range of severe natural phenomena used for'the design of Peach Bottom safety related , equipment are considered for the hydrogen storage system. For non-safety related equipment the design goal is to assure that this equipment will not be impacted by normal operation of the hydrogen storage system, nor by abnormal operation events which might reasonably be expected over the life of the equipment. This is particularly the case for equipment which might lead to a Peach Bottom plant transient. It is expected that compliance with NFPA 50B separation requirements will provide adequate protection for normal and abnormal operational events. The equipment which was evaluated to assure that protection is adequate includes the nearest transmission line and the nearest 500 KV substation. The credible events which could cause larger hydrogen release are very infrequent, i.e. tornados and earthquakes, and have their own potential for disrupting non-safety related equipment, independent of what happens at the hydrogen storage site. Therefore, protection of non-safety related equipment from the releases caused by severe natural phenomena is not considered necessary. B.2 RELATIVE LOCATION OF NON-SAFETY RELATED EQUIPMENT Transmission lines and a 500 KV substation are the nearest PECO facilities, whose failure might contribute to a plant transient (i.e. loss of offsite power). The nearest transmission line is a 500 KV line which is about 300 feet from the hydrogen storage area. A 500KV substation is located over 600 feet from the hydrogen storage area. *l B.3 COMPLIANCE WITH NFPA SEPARATION REQUIREMENTS NFPA SOB, " Liquified Hydrogen Systems at Consumer Sites", has no separation requirements for transmission facilities, however, the most restrictive limitation for this size of facility is 100 feet from combustibles (NFPA 50B, Table 2). NFPA 50B includes some special requirements for electrical equipment which is within 25 feet of a liquified hydrogen system. B-1 (6204E) l
The provided separation between the liquid hydrogen storage area and ' the transmission facilities is well in excess of the above requirements. The NFPA requirements are sufficient to protect PBAPS facilities from the worst case vent stack release (Section B.4). l Section B.5 provides assessments of the worst potential consequences of vent stack releases. B.4 NORMAL AND ABNORMAL HYDROGEN RELEASES The only normal operation of the storage system where hydrogen can be released to atmosphere would be when the station is not using
. hydrogen. Then, after a few days, the normal heat transfer from the ambient air will vaporize enoegh liquid to require controlled venting of the tank. The vent rate would be on the order of 0.002 lbs./sec.,
intermittently. Abnormal occurrences resulting in hydrogen releases, which are unlikely but might occur over the life of the storage system, are related to the operation of overpressure protection devices (relief valve or tupture disk). The peak vent rates associated with the worst case failure would be about 8 lbs./sec. B.5 IMPACT OF RELEASE EVENTS B.5.1 Normal Operation The pressure control releases which are necessary after extended shutdown of the hydrogen supply system are at a very low rate, and are routed through a vent stack for elevated release. They would be rapidly dispersed to below flammable concentrations at the exit and j hg would have no impact on PBAPS or hydrogen storage system equipment. B.S.2 Abnormal Operation The abnormal releases from relief valve or rupture disk operation are also routed through a vent stack for elevated release. A high velocity vertical jet results and is dissipated to below flammable concentrations before significant lateral drift has occurred. Vent stack release are well away from ignitions sources and ignition is generally not expected. Because of the unconfined location of the hydrogen release, even in the event of ignition, only a fire at the top of the vent stack would be expected. Calculated thermal fluxes from such a vent stack fire fall below that necessary to ignite wood at a distance of 35 feet from the plume (Reference B1). Such a fire would have no impact on l /(( Peach Bottom site transmission facilities. (6204E)
-- --, e - _ _ - -.
. _. .____._,.m. . _ . , . , . _ _
B.6 Conclusions Neither normal operation of the hydrogen system nor abnormal operations which might be expected to occur over the life of the
- equipment will adversely impact any PBAPS transmission facilities.
REFERENCE Bl. Camilli, L. L.; Linney, R. E.; & Doelp, L. C., " Liquid Hydrogen Storage System -- Hazardous Consequence Analysis", Air Products and Chemicals. Inc., October 1, 1985. B-3 (6204E) l I L-
PHILADELPHIA ELECTRIC COMPANY PEACH BOTTOM ATOMIC POWER STATION SAFETY ANALYSIS FOR THE LIQUID OXYGEN STORAGE SYSTEM Revisidn 2 May 1987 Prepared by United Engineers and Constructors 4 I i 4 t-1 ,I h r I (6205E) o
- k. - - - . . _ _ - . . . . _ _ . _ _
[; L I l l l TABLE OF CONTENTS l-l. I. INTRODUCTION II. GENERAL DESCRIPTION OF LIQUID OXYGEN STORAGE SYSTEM (LOXSS) III. POTENTIAL FOR SIGNIFICANT ACCIDENTAL RELEASES FROM THE LOXSS ! IV. SAFETY EVALUATION OF LIQUID OXYGEN STORAGE SYSTEM LOCATION , V. REFERENCES 4 3' 1 1 + 1 e i i Page 2 of 7 (6205E)-
I. INTRODUCTION This safety analysis of liquid oxygen storage is intended to support the license amendment to permit liquid oxygen to be stored on site. The liquid oxygen will be stored in a liquid oxygen storage tank of 11,000 gallon or less. A. Purpose and Scope of this Report The purpose of this report is to establish the acceptability of the design and chosen site for the liquid oxygen storage system. The determination of this acceptability, is based upon application of the criteria from " Guidelines for Permanent BWR Hydrogen Water Chemistry Installations", 1987 Revision, by the BWR Owners Group 2 for IGSCC Research, (Reference 1). B. Structure of this Report Section II of this report describes the liquid oxygen storage system. The liquid oxygen storage system will be provided by Air Products and Chemicals, a supplier with extensive experience in the design, operation, and maintenance of storage and supply systems. Section III identifies the circumstances which might result in the release of liquid oxygen, up to and including release of the entire contents. Section IV presents the evaluation of the location of the liquid oxygen storage system using Reference 1 guidance. l C. Conclusion The conclusion of this report is that safe separation distances are provided between the LOXSS site and all PBAPS safety related equipment. Therefore, installation in accordance with the analysis in this report presents no safety hazard. Page 3 of 7 (6205E)
II. GENERAL DESCRIPTION OF LIQUID OXYGEN STORAGE SYSTEM (LOXSS) A. Equipment Description The LOXSS will consist of:
- a liquid oxygen storage tank, up-to 11,000 gallons in size
- (2) ambient air oxygen vaporizer banks a pressure / temperature control manifold
- related interconnecting piping, valves, control panel and controls The oxygen tank will consist of an inner vessel or " liquid container" and an outer vessel with the space between filled with insulation and evacuated. The inner vessel will be designed, fabricated, tested and stamped according to applicable ASME codes. The tank will have appropriate overpressure protection devices. The liquid oxygen storage system will meet all applicable design requirements of Reference 1. J j hs The design of the LOXSS system makes large unplanned releases of oxygen very unlikely. Potential site-related causes of tank failure are evaluated to determine what other, related design criteria should be used. The tank will be designed to remain in place under Peach Bottom Atomic Power Station (PBAPS) tornado wind load design basis conditions.
B. Location of LOXSS The location of the LOXSS relative to other PBAPS facilities is shown on Figure 1. Table 1 shows the distance from the liquid oxygen storage tank to all safety related air intakes and their height above grade. The LOXSS meets all separation requirements of NFPA 50, Bulk Oxygen Systems at Consumer Sites including required separation from liquid hydrogen storage facilitie2. Page 4 of 7 (6205E)
III. POTENTIAL FOR SIGNIFICANT ACCIDENTAL RELEASES FROM THE LOXSS The only possible releases from the LOXSS under normal operating l conditions are in the event of extended supply system shutdown where i the gradual heat input must be controlled by occasional, small volume, [ venting the tank. The only possible releases from the LOXSS under abnormal operating conditions would be associated with tank relief valve actions. Both normal and abnormal operational releases would be rapidly dispersed in the immediate vicinity of the LOXSS. Adequate protection from such releases is provided by compliance with NFPA 50 standards. It is not possible to protect the LOXSS from the full range of design basis natural phenomena used for the PBAPS safety related equipment. Therefore, low probability, accidental releases of the entire tank contents are postulated and evaluated in Section IV. Design provisions are made to assure that the release will originate at the LOXSS site. These include that the liquid oxygen tank is designed to remain in place during a PBAPS design basis tornado and
- design basis flood.
l I Page 5 of 7 (620$E) t
IV. SAFETE EVALUATION OF LIQUID OXYGEN STORAGE SYSTEM LOCATION The only hazard from a liquid oxygen spill is a possibly increased fire risk. Oxygen-enriched air can make combustible materials ignite at lower temperatures and accelerate combustion. However, it would not increase area combustible loadings nor would spontaneous ignition result. An ignition source is still required. The design goal is to prevent ingestion of excessively oxygen-enriched air into safety-related air intakes. The limiting hazard is associated with an assumed total, instantaneous release of the tank contents. For safety-related building HVAC air jntakes, for the emergency cooling tower air inlet, and for the diesel generator combustion air intakes, the guidance from Reference 1 for location of the liquid l 1 oxygen storage system was used. Reference 1 uses a dense gas dispersion model to predict the extent l and height above grade of the oxygen plume resulting from a liquid oxygen tank instantaneous breach. Reference 1 uses a conservative [ limitation of 30 percent oxygen at any safety related air intake. Table 1 shows the distance from the liquid oxygen storage tank to all safety related air intakes, their height above grade and the appropriate limitations from Figure 2 (from Reference 1, Figure 4-8). lh All safety related air intakes are at an acceptable distance and/or height. V. REFERENCES
- 1. " Guidelines for Permanent BWR Hydrogen Water Chemistry Installations", 1987 Revision, by BWR Owners Group for IGSCC Research, Hydrogen Installation Subcommittee.
Page 6 of 7 (6205E)
TABLE 1 SEPARATION OF SAFETY RELATED AIR INTAKES FROM OXYGEN STORAGE SYSTEM DISTANCE HEIGHT MINIMUM FROM OXYGEN OF INTAKE REQUIRED HEIGHT STORAGE SYSTEM ABOVE GRADE ABOVE GRADE REACTOR BUILDING HVAC 885 FEET 79 FEET 19 FEET CONTROL BUILDING HVAC 1055 FEET 43 FEET 19 FEET DIESEL GENERATOR BUILDING 1422 FEET 16 FEET 0 FEET HVAC AND COMBUSTION
-EMERGENCY SERVICE WATER 1065 FEET 25 FEET 19 FEET PUMP ROOM EMERGENCY COOLING TOWER ELECTR0 MECHANICAL 260 FEET 39 FEET 23 FEET }
jdL g EQUIPMENT ROOM AND COOLING TOWER AIR INTAKE i l Page 7 of 7 l (6205E) i
O ( m J =
-. r - - ---- Ik,- w.-._,s =
dU IN h RiL. . .m=..{ N 8 =
-l., jO{ vW 5 =
- g.^ ML.; j S
! 'y
[E r $ 1: ltF.a
"! r l c :!! .' "
,5 .
g [ , llll i1 oo z n gig eed -[ g , j 8 w"IT / ' jf89 8:I f' zuea
- fi,% o
, 23 '
Fgd
- waa n
. f/mj~...,..., j
,gi ,
54g i
/)
t.l f . . , - I4 b o c s / f y8 tEi j I+
=
2
;v o I
0096h
'E h / .
.M P$ d ~
' ' 4 2f=il swi w iny ,
- - ~ ~ - -
, , x m' 4l-glji.tgg--
~
pM4 [
;I 4
g$3
= . . . .
* -- _....o. ~
l
FIGURE 2 1 60 N 9 2 r 40 Acceptable location of 5 safety related air intakes I 20 - h 8 8 A A 10 - Unacceptable locatiorr of safety related air intakes E l; - 0 1 I I I I I ' I ' I ' 0 200 400 600 800 1000 1200 Distance from Liquid Oxygen Storage Tank (ft) From Reference 1: Figure 4 8. Acceptable locations of safety related air intakes for various sizes of liquid oxygen storage tanks. l i 4-20}}