ML20199H928
| ML20199H928 | |
| Person / Time | |
|---|---|
| Site: | 07000036 |
| Issue date: | 01/30/1998 |
| From: | Sharkey R ABB COMBUSTION ENGINEERING NUCLEAR FUEL (FORMERLY, ASEA BROWN BOVERI, INC. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM), NRC OFFICE OF NUCLEAR MATERIAL SAFETY & SAFEGUARDS (NMSS) |
| References | |
| NUDOCS 9802050156 | |
| Download: ML20199H928 (9) | |
Text
- _. - _ _ - _ _ _ _ _ _ _ _
o January 30,199b Docket No. 70-36 License No. SNM-33 Director, Office of Nuclear Materials Safety and Safeguards U.S. Nuclear Regulatory Commission ATTN: Document Control Desk
'Vashington, DC 20555-0001
Subject:
SupplementalInformation Regarding Criticality Safety of the llF Absorber System
Dear Mr. Felsher:
O
^e we di8ce ed en aanearr 20. i998. nrevided hereie is edditienai detaii resardins the determination of the criticality safety of the IIF absorber system.
Enclosed. is a description of the process, the pertinent dimensional details, a description of the normal and accident conditions which were investigated, and a listing of the controls used to ensure criticality safety.
If there are questions regardmg this matter, please feel free to contact Robert S. Freeman 1
of my staff at (314) 937-4691 Ext. 425 or myself at (314) 937-4691 Ext. 399.
Sinceiely.
COMBUSTION ENGINEERING,INC.
i s'
1 i
Robert W. Sharkey Y
i
,d \\ () j Director, Regulatory Affairs 9802050156 980130 lll ll' ll11 ll Qi l l PDR ADOCK 07000036 Hl.II till
.I C
PDR RA98/680
'(
M " l
ABB CENO Fuel Operations n
combustion En;neenng Inc.
33m State Hoad P Te4* phone (314) 937-4691 Post Office bn 107 St Louis (314)2945640 Hnmtitite Mssoun 6.K)d7 Fax (314) 937-7955
Enclosure to 11A98/680 Itcsults of the Criticality Safety Analysis and Evaluation for the llF,.bsorber System O
m
-%c)
L........
===.
System Description===
p V
The purpose of this system is to capture the hydrogen nuoride (ifF) contained in the off-gas streams of tl oxide conversion process and, through a series of packed bed absorbing vessels, convert the llF gas into r. concentrated IIF solution using water as the dilutant.
The liF absorbing process consists primarily of a series of liquid absorbing / scrubbing vesels which allow conversion of the hydrogen nuoride off-gas stream from the oxide couversion process into a concentrated llF solution. The solution is then passed through a heat exchanger and stored withh an array of favorable geometry qualification tanks while the liF free vapors exit through a dedicated stack. Once the material within the qualification tank array is qualified for release, it is pumped into a 8,500 gallon storage tank. The entire system is positioned within a thirty five by thirty three and a half foot square retention dike for environmental protection. Under normal operation, the system is designed to operate unattended until reaching the fill limitations of the qualification or storage tanks.
The initial design contains a series of three packed bed absorber vessels with provisions for two additional vessels if praduction requires.
The three absorber vessets are designated as the first, second and third stages of the gas scrubbing process. As the reactor off-gas flows sequentially through the three stages, the concentration of hydrogen fluoride in the gas vapors is reduced.
During the process the hydrogen fluoride gV concentration in the scrubbing solution is highest in the first stage and decreases in each subsequent stage.
The concentrated ilF solution liquid will flow from the heat exchanger into one of the P
two groups of qualification tanks. The first group contains 6 tanks adjacent to one another with a combined capacity of 282 gallons. The liquid will normally be allowed to flow pass the first group of 6 tanks into the second group of 18 tanks. The combined capacity of the 18 tanks is 846 gallons.
The material in the second group of tanks can now be circulated with a transfer pump for sampling and qualification.
Following acceptable sampling results, the material is
~
pumped into the unfavorable geometry of the 8,500 gallon storage tank. While all process equipment is intended to operate at ambient pressure, the pressure rating for all piping types is significantly above the operating conditions.
Prior to transfer from the bulk storage tank into the tanker truck, an additional material sumpic will be taken following re-circulation of the bulk storage tank to ensure no accumulation of material has occurred over time within the bulk storage tank.
p)
I_
v
The follov,ing statement was provided in Reference 1, with additional clarification
- O
. provided below:
.v "In the unlikely event of a process upset which would raise the uranium concentration in the hold tank above the release limits, then prior to release from the qualification tanks to the hold tank the Fquid will either be diverted for later introduction into the system or placed into a container and neutralized."
The container mentioned in the above description represents a generic containing device, and is not specific to any actual container or tank. In the unlikely event of a process upset where the uranium concentration increases above the release limit based on multiple sampling, it will be transferred into a criticality safe container. The container sim, shape and dimensions will be evaluated prior to transfer and is dependent on the magnitude of the upse'. and the uranium concentration.
Computational Modelling Details The llF abscrber system consists of three vessels whose orientation is perpendicular to the surface of the concrete dike lowd 92 inches below the bottom flange of each vessel.
The vessels consist of concentricahy layered piping with the center pipe constructed out (o) of an liF resistant polymeric material with the following repeating chemical structure (-
Cil - CF -),,. The internal sleeve is tightly fit into the exterior carbon steel piping with an 2
2 r-irterference fit. The maximum internal diameter of the internal pipe is modelled as 7.68 inches with a thickness of 0.186 inches. The exterior carbon steel piping ht a wall thickness of 0.286 inches and an external diameter of 8.62.5 inches. The three vertical vessels are separated by a nominal center-to-center spacing of 27 inches. Two of the HF absorbers extend 289.125 inches in length, with the third vessel reaching ? 15.250 inches.
Tne IIF absorber system also contains a 6 x 4 array of tanks oriented with their longitudinal axis horizontal rdative to the concrete dike. The tanks are all dimensionally identical, and consist of on internal polypropylene sleeve surrounded by a carbon steel shell. Although the qualification tanks are manufactured in the same manner as the vertical vessels using the interference litting approach, the internal diameter difTers slightly between the two due to difTerent standard material thicknesses and tolerances.
The maximum internal diameter of the polypropylene was taken as 7.63 inches, with a thickness of 0.210 inches. The exterior carbon oteel piping has a wall thickness of 0.286 inches and an external diameter oi8.625 inches.
The edge flanges of the horhontal tanks are located approximately 27 inches from the center of the vertical vessels. The horizontal tanks are squarely arrayed such that the nominal vertical and horirmtal separations are 29 and 27 inches, respectively. The center n!j of the lowest qualification amk is 25 inches above the floor of the concrete dike.
A reCector region was modelled which surrounded the bottom portion of the array of
(~N V
horizontal tanks. The reflector region was established to evaluate the elTects of a snow load. The region below the lowest row of horizontal tanks, betuen the tanks and the floor of the concrete dike was reDected with full density water. In addition, a close lhting 12 inch thick reflector region was included on each of the focir sides. This reflector was modelled as approximately 4 feet high to simulate a buildup of snow. The reDector regions were filled with full density water ihr consenatism, even inaugh the density of snow is far less.
With the exception of the fbur inch piping between the vertical units, the connectin; piping between vessels within the system is less than 2 inches in diameter and was not modelled. Thc 4 inch diameter piping between the three absorber vessels was evaluated and found to have no significant effect on the system reactivity.
The 8,500 gallon storage tank was not modelled since controls are in place to preclude significant uramum from reaching this unfavorable geoiaetry vessel.
The sump consists of a 7.625 inch int rnal diameter pipe with a 0.5 inch v/all thickness. A two (bot by two (bot by one inch thick steel plate was imbedded below the ton surface of the concrete base of the dike surrounding the collection sump, to provide sufficient shielding between the geometry of the sump pipe and an accumulated slab of solution above the pipe.
pV Normal Operation Normal operating conditions for the liF absorber system do not involve fissile material. By nature of dr or,ide conversion process, the multiple system interlocks, and the redundant littering osvices, the uranium inventory is contained within the oxide conversion system. In addition. d"al sampling is perfonned prior to transfer from the favorable geometry to the unfavorable vessels.
p
i
~
4 I
Prbcess Upset Analyses
(~}
V F.act of the various components of the I F absorber system, as well as the system as a who, has substantial margin to the criticality safety limit of 0.95 under both normal and accia nt conditions. This is due primarily to the geometry of the individual components in the system and their separation. Nthough multiple process barriers exist to prevent fissile material from entering the liF absorber system, the system was analyzed with optimally moderated solutions of both uranyl fluoride and uranium dioxide in all components, excluding the 8,500 gallon storage tank, in order to determine the upper limit of the effective multiplication factor. This assumed condition would require approximately 13,000 kgs of UF,, (5. full 30B cylinders of material) homogeneously distributed in the system.
The dimensions modeled in the analysis are conservative. Specifically, the separation distances between units in the 6 x 4 horizontal tank array were reduced both horizontally and vertically by one inch from the nominal dimensions.
In all instances, as-built component separation distances were verified ibliowing installation and found to be greater than those modelled. For case in modelling, all three vertical vessels were conservatively assumed to be 315.25 inches in length. The packing bed material was conservatively om.We> &cm the model. This is conservative since the presence of the packing material si d ik u displaces volume which could be occupied by the assumed fissile solution.
t i
U Tit absorber system was analyzed for fbur conditions. The first two analyses assumed the nystem, excluding the 8,500 gallon tank, to be full of optimally moderated uranyl fluoride at both 5.0 and 6.0 weight percent "U. The second set of analyses assumed the system, excluding the 8,500 gallon tank, to be full of optimally moderated uranium dioxide at both 5.0 and 6.0 weight percent 2"U.
In each case, the optimum conditions were determined through a series of parametric calculations in which the hydrogen content and uranium density were varied in order to determine the most adverse conditions from a reactivity standpoint. The parametric studies considered the geometric model of the system companents to determine the optimum conditions. The results of the study were in agreement with his'orical data, intemal calculations, and industry published data, with the most adverse densities of 2.2 g/cc (H/U ratio of 11.50) and 2.0 g/:e (II/U atio of 12.25) fbr UO F or UO at 5.0 w/o "U, respectively 2
2 2 2
The calculated data show that approximately 2.5% reactivity is attributable to a close fitting bottom reflector,1.2% for side reflection up to 47 inches, and no added reactivity for end reflection due to the length of the piping. The data also clearly define the limiting geometry to be that of the 6 x 4 horizontal array of tanks and not the series of three vertical llF absorbers. The system was evaluated Ihr a variety of interstitial mistin; cenditions and it was determined that the most adverse mist density was 0.025 g/cc water.
n h
(a
Using the most adverse interstitial mist density between units, the most adverse uranium density under optimum moderating conditions, reduced separation distances, and the
- )_
reficctive boundaries mentioned above, the maximum system multiplication factor is v
0.89057 or 0.91701 with all applicable biases and uncertainties applied when the system f
is assumed filled with 5.0 w/o "U uranyl fluoride or uranium dioxide, respectively.
2 Criticality Controls (Primary)
The primary criticality controls which alone satisfy the requirements of the double contingency principle are:
Geometry / Interaction ControhThe as-built configuration of the system is sullicient to maintain criticality safety for material enriched up to 6.0 weight percent in the "U isotope in all system components 2
and vessels with the exceptico of the bulk storage tank.
Shape and dimensions of the individual components establish the basis for geometry control.
Mass / Concentration Control: Criticality safety in the bulk storage tank located downstream from the qualification tanks is maintained by mass / concentration control. Two samples are taken of the O
liquid be:ng transferred to the receiving tank. Therc is a time delay of approximately thirty minutes octween the samples.
The liquid is well mixed between samples and equipment used to test the liquids is checked by counting a standard to ensure that equipment mis-calibration does not introduce a common mode failure.
The controlled parameter from a criticality safety viewpoint is mass. Transfers to the 8,500 gallon receiving tank may not take place unless the concentration is less than 0.2 g U per liter. At this concentration, the total mass m the bulk storage tank can then not exceed 6.5 kg U.
In addition, the bulk storage tank is to be tested for accumulation of material periodically. This may be accomplished by circulating the contents of the tank, taking a sample, and testing for uranium content. Testing during the annual uranium physical inventory would be sufticient -
to satisfy this requirement.
k
, i I
QJ '
?
s' j
Criticality Controls (Secondary)
.f3 V
'the secondary criticality controls, in addition to the measurement controls listed above, provide auxiliary support and protection toward satisfying the requirements of the double contingency principle are:
Mass / Concentration Control: Criticality control during the transfer of material from the favorable geometry of the qualification tank array to unfavorable geometry of the bulk storage tank is also maintained by preventing uranium from initially entering the IIP absorber system.
Gaseous UF. is controlled from entering the liF absorber system by redundant active engineered steam and UF.110w interlocks. Steam and UF. Ilow are regulated to ensure that UF. is fully reacted and not released to the HF absorbers.
The physical barriers of the in-line secondary filtration units (sintered metal filters) prevent particulate SNM from initially reaching IIF absorber system. The filter units undergo periodic scheduled inspections.
OV Alternate Protective Aspects (Tiertiarv)
The following protective aspects of the system exist An initial redundant set of uranium particulat 'ilter units prevent particulate from entering the secondary filtration units.
Oxide Control room clarms are activated by high differential pressure in the secondary litter units. This may indicate a break in the upstream litters.
UF. and stemn pressure indicator interlocks.
A blind flange connection is in place prohibiting dual operation with Dry Scrubber System.
p J
.i
. O)
. Sutnmarv and Conclusions
(
Based on the documented analysis and evaluation, the liF absorber system will remain suberitical under both normal and credible accident conditions. Criticality control is provided throughout the process consistent with the double contingency principle.
References:
1)
Letter from CE to Mr. Weber," Amendment Request fbr Unrestricted Release of Ilydrofluoric Acid", August 12,1997
,nq v
i l
'8
)
(
- - - - - - -