ML20235N833

Safety Evaluation Supporting Conversion Order to Convert from High Enriched to Low Enriched U Fuel for License CX-22

ML20235N833
Person / Time
Site:	Rensselaer Polytechnic Institute
Issue date:	07/07/1987
From:	Office of Nuclear Reactor Regulation
To:
Shared Package
ML20235N799	List: ML20235N795 ML20235N818 ML20235N833 ML20235N857 ML20235N870
References
NUDOCS 8707200244
Download: ML20235N833 (18)

Text

-

, ,: s

_ pa ntouq E

1. k:g UNITED STATES NUCLEAR REGULATORY COMMISSION 5 .j WASHINGTON, D. C. 20555

\ ~

SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION.

SUPPORTING CONVERSION ORDER TO CONVERT FROM HIGH ENRICHED TO LOW ENRICHED URANIUM FUEL FACILITY OPERATING LICENSE NO. CX-22 l

RENSSELAER POLYTECHNIC INSTITUTE

1.0 INTRODUCTION

i In accordance'with.10 CFR 50.64, which requires that non-power reactors  ;

convert-toalow-enricheduranium(LEU) fuel,exceptundercertain-  ;

conditions, the Rensselaer Polytechnic Institute (RPI) has. proposed to '

convert the fuel in its critical facility from high-enriched uranium (HEU) to LEU. RFI has submitted a " Proposal for HEU/ LEU Core Conversion j Pursuant to 10 CFR 50;64" on October 3,1986, which included a Safety i Analysis Report and revised Technical Specifications. The staff's -)

safety review with respect to issuing an order for conversion from HEU l to LEU has been based on an analysis of this propesal and the Technical I Specifications, as well as on infonnation provided by RPI in response.to staff questions submitted by letter dated March 3,1987. Also, the Operator Requalification Program submitted by letter. dated April 5,1987 was reviewed. This material 15., available for review at the Commission's _

~

Public Document Room at 1717 H Street, N.W., Washington, D.C. This Safety Evaluation (SE) was prepared by T. S. Michaels, Project Manager, 1 Division of Reactor. Projects III, IV, V, and Special Projects. Office of 1 Nuclear Reactor Regulation, U.S. Nuclear Regulatory Commission. Major a contributors to the technical review include J. Buzy and J. Dosa of NRC ~

and W. R. Carpenter, C. H. Cooper and R. R. Hobbins-of EG&G, Idaho National Engineering Laboratory (INEL).

The purpose of this SE is to sumarize the results of the safety review ,

of the conversion from HEU to LEU fuel at the RPI critical facility.

The review has been summarized and edited to the extent practical to correlate with the Safety Evaluation Report issued in 1983 by the Commis-sion, NUREG-1023, which supported a renewal of the RPI license.

2.0 EVALUATION The section numbers below are not numbered in sequence because the numbers correspond to those in NUREG-1023. Any section numbers in NUREG-1023 which are not addressed below did not have to be revised for the conversion. -

4. REACTOR -

The Rensselaer Critical Facility (RCF) is a light-water moderated and reflected, heterogeneous reactor fueled with low-enriched UD, stainless-steel clad fuel pins. The reactor is controlled primarily by inserting and withdrawing neutron-absorbing control rods. . Additional reactivity control, used primarily during critical experiments, is' achieved by l changing the reactor tank water level.

e707200244 870707 DR ADOCK 0500 2 5

.- g 6 ..:

}

- 2^- 1 The 'RCF was constructed in 1956 by ALC0 prcducts, Inc. for the purpcse of 1 developing a series of power reactors for the-U.S. Army. In 1964, RPI assumed operation of the facility primarily for-instructing graduate ~*

students in the Institute's Department of Nuclear Engineering and l Science. In 1983, the Commission renewed the license for continued opera- l tion of the RPI facility with the original high-enriched core. In 1987,_

the high-enriched plate core will be removed and will be_ replaced with the -

LEU core.

The maximum authorized power level of the RPI. LEU reactor is 100 W, but actual operating powers are generally no higher than needed to accomplish the experimental objectives, normally below 20 W. Because of the low power levels and limited integrated energy generated, the RCF normally does not' l l contain a radiologically significant fission product inventory. l

.)

^

4.1 Reactor Core i The RCF LEU core and support structure are designed for critical experi-ments using variable arrays of fuel pins. Fuel, in the form of low-enriched U02 pellets in stainless steel cladding,-is arranged. in a generally cylindrical fashion with four control rods placed symmetrically about the core periphery. Fuel pins, with ari active length of 91.44 cm, are set on a- -l square pitch of 1.49 cm to yield an effective core radius of approximately 35 cm. This core rests on a fuel pin support plate and is immersed'in an open tank of water that serves as both moderator and reflector.- Although 3 the grid structure can accommodate approximately 550 fuel pins, initial l operation is expected to use only about 420 to 425 fuel pins. Two initial  !

fuel pin arrangements are proposed. The first, referred to as " Core A," .;

is a solid array of pins in the approximate shape of a right circular '

cylinder. The second, referred to as " Core B," is an annula'r array of pins _i with the central 5 x 5 square-of fuel pins removed. Calculations show that si Core A contains 421 pins during operation;, Core B is calculated to hold "'

424 pins during operation. Core B permits use of the interior region for ..i experimental purposes. Core A will be the initial LEU core to be loaded into the RPI facility and will contain 421 fuel Dins for a total U-235 loading of 14.82 kg.

4.1.1 Fuel Pins The RCF LEU core will employ 4.8 weight percent enriched SPERT(F-1) fuel pins. Although this fuel is unirradiated and has never been in a critical ,

reactor, a requalification program was conducted, primarily because of its age (N 20 years). Argonne National Laboratory requalified the SPERT fuel to verify that the pins have suffered no physical damage since fabrication. '

The pins were inspected under 6X magnification, and by X-radiographic, destructive, and metallographic examinations. Spectrographic and chemical analyses were performed on the U0 fuel. INEL formally evaluated the Argonne program (EGG-NTA-7660, R.2R. Hobbins) and concluded, on the basis of requalification studies and the nature of the intended use, that using'SPERT fuel pins in the RPI critical facility does not constitute an undue safety -

risk. -

l 1 -

7

Each SPERT LEU fuel pin contains 35.2 gm U-235, has an active length of 91.44 cm, and is made up of 60 sintered U02 pellets, encased in a --

Type 304 stainless steel tube, capped on both ends with a stainless steel cap and held in place with a chromium-nickel spring. An aluminum oxide (Al 0 ) insulator is installed between the fuel pellets and stainless ste$l 3caps on each end of the pin. Gas gaps to accommodate fuei expansion 1

-l are also provided at both the upper end and around the fuel pellets.

Figure 4.1 shows a single fuel pin, its pertinent dimensions, and its physical characteristics. The total amount of LEU fuel shipped to RPI is 595 fuel pins with a total U-235 inventory of 20.94 kg. Those not used in

'the reactor (*175 pins) will be kept in the fuel storage vault (discussed in Section 6.2).  !

4.1.2 Control Rods I

Control of the RPI LEU core is provided by .four new, nonfuel-follower-type  ;

control rods specially designed for use with the low-enriched SPERT fuel. l These rods are spaced 90 degrees apart at the core periphery as shown in )

Figure 4.2. Each control rod consists of a 6.99-cm-square stainless steel  !

tube,165.1 cm in length, which passes through the core and rests on a l hydraulic buffer on the bottom carrier plate of the support structure. -)

lioused in each of these " baskets" aro two neutron-absorber sections, one '

positioned above the other to provide for a total poison section length of )

91.44 cm. These absorber sections may contain either B-10 enriched baron i in iron, Eu03 in a stainless steel cermet, or an alloy of silver-cadium-indium. All absorber sections except the one containing silver-cadium-l indium are clad in stainless steel. All absorber sections are of the same dimensions, nominally 6.6 cm square, with their poisons uniformly distri-buted. Because of the symmetry of the LEU cores and the symmetric placement of the control rods, each of the four rods has approximately the same -; i reactivity effect. The initial absorber to be used at RPI in Core A is Oi B-10. Rod worths and shutdown margins discussed in Section 4.6.1 are based on the usage of B-10 absor'o er sections. ~

4.3 Support Structure The LEU core fuel pins are supported and positioned on a fuel pin support I

plate, drilled with 0.64 cm diameter holes to accept tips on the end of each pin. The support plate rests on a thick carrier plate that forms the base of a three-tier, overall core-support structure. An upper fuel pin lattice plate, shown in Figure 4.3, rests on the thick top plate and is drilled through with 1.27-cm-diameter holes on the prescribed pitch to secure the upper ends of the fuel pins. The carrier plate, the lower fuel pin support plate, a middle plate, the top plate, and the upper fuel pin lattice plate are secured with tie rods and bolts. The entire core structure is anchored by four posts set in the floor of the reactor tank.

Finally, in the event that the fuel pins are bowed but still satisfactory 4

en*+

e

.%+

. .: ' ~

4

.{

l

.1 ,

s

/ s Ible for Fuel 11arxil.12x 2bol

, Spritrg for 11oLiing 00 2 Pellets '

s-

~

-- Vold/ rep Decion which allcus for I the Dgansion of Pissica Gasses i p%:. 1

,e s,-

-l  :. .

/

c&

, j_  ;

~

i * - *

-LO2 Ftr.1 Pellet t _

L '

.;f 4 '

- Stainless Steel c'12Mirg Spert Fuel Ibd Characteristics 102 Ftul. Pellet D:rtaity

• 10.08 gm/m l U-235 Ddichment 4.81 1 0.15w/o U-235 Content per Ib:1 35.2 m .

. Pellet Outside Diancter 0.420 in Imx2th of Ftr1 Tbd 41.75 in clam LW Outside Dianeter 0.466 in c'laMLm '1hickness 0.020 in '

% s Active Fuel length . 36.00 in '

'N ' . .-

k c1aMhg Ianth 41.62 in

/ .\ - Ibd Overall Israth 41.75 in f

Figude 4 1 SPTRI TUD FIN

3

• %/

O_ :

9 ,

. f* e9 "

s 7 g y::

t 3"X [p

.i Y.  !,Ni. ...____.__ N sI ~

8  : -1 i i gg 3:- 5 -

el . a eo gI y

Eg G eserveess egv m

.- e gE 9 5) S 9"@ *%.)

\ X a O v.

tv X ;s2 v

o m l

x 1

vQ g y

% k u X X 8 3 l .1" x v

) X y ' l X >g .

$ l I '

l [ .-

_g- '

I @

OOQOOOOOOOG$MC'? OOOOO90 $

L'l

- @ oL ' _1

_ x. c O

• g o 1 l

0 X nEs %e

~ X6

^

g;<srs - -

-:. h , _-.

v x ,sO v  ;

y w i O>O'3 f_'O \

OO. ..

a .. ,

e .

a ..

u u i e H '

'o l l *=

5 O" ,

O msu un sie e 6880 l ~'Ai!

l -

i i +2 sieii+t #i nii nae + 4 .i le l T __

Figure 4. 2 UPPER RTEL PLATE LATIICE 1

i l

l l

. -4 '

w________-__--__..-- -- .-

, ^: ,

l g "

c= ins e e mc

(

[,11

/ .= Xl rr .

%ih-

-l' .,

ub . _

c

'A@ ~

A@ .'

(L

/  % .'f r!

( ,

! n:n vic -

I;

\

a 1 . +

1 .

~

., . ul f ,7 f}-PueL we.s Pl4 f, ,, ,. -

<}

q ,

\ -

.8 -

d(

p V  :

i N i, .i f ,

t ( , ~. v i ..

_A m j .-

i 4 -

y i serern i

t ,

, j, S m -

'i w+ AVv4

, N ";,,_,A w .

0 f% fyf I

i NI

{h

'f

' rg

?

7

[ .

.M ll . m. u rt ~,

\

[ ,

F L ..

. s\ .

. i .. ,p - rc e s w as ,, 1 i '

f f -

N .\ .

jl.' S i'.

f j,11.

\. .\

g

. '..'.',s,. \.\ l\ . V. g  :

s[

. ; 7 \bg\ ' .\\ -

nl

[i  !

f r, , '\ y

• @c?

r 4

.\

• I k .\'\\, '

.{y 7_ "

'g orx7tyi w

\ '. l

{' g l

N ] IEo l

t.'

l Tigure 4.3 CORE SUFFORT STRUCTUP2 i

~

l 4

+=

. i

?

.**j

~7-for use in the core, a plastic spacer plate may be installed on the middle _,

plate.

4.6 Dynamic Desion Evaluation The RCF LEU reactor is operated safely by using and manipulating a

~

reactivity control system that includes changes in the moderator water level and changes in position of poison-bearing control rods. The reactor instru-mentation monitors and displays changes in reactor parametdrs such as operating power and reactor period, thereby also providing information and signals for appropriate protective response. In addition, interlocks prevent inadvertent reactivity addition, and a scram system initiates rapid, automatic shutdown when safety settings are exceeded.

To ensure further stability and safety, the Techr.ical Specifications I require that all critical LEU cores mast have a predicted or measured negative isothermal moderator temperature coefficient of reactivity at temperatures above 100 F with no more than 0.00115 A k/k (0.155) integrated reactivity added.between ambient ( m50*F) to 100*F. This requirement is slightly different from the RCF HEU requirements of a y

) negative temperature coefficient above 90 F, with no more than 0.11$ total positive reactivity added. Both Core A and Core B satisfy the 100*F requirement on the isothermal moderator coefficient with Core A adding. .

0.000060 A k/k (0.008$) prior to turning negative and Core B adding 0.00103 6 k/k (0.13$) prior to turning negative.

An additional inherent safety feature of the RPI LEU tuel, since it is '

more than 95% U-238, is a strong negative temperature coefficient caused by Doppler broadening of the U-238 resonances. For the RPI LEU cores,- this -

Doppler coefficient reduces the core reactivity 0.12$ during the 50*F 7 temperature rise (between ambient and 100 F) where the moderator l temperature coefficient is required to become negative. For Core A, this --

means the total temperature coefficient is essentially negative from

! ambient. For Core 8, the 0.12$ negative Doppler essentially neutralizes the 0.13$ positive moderator frcm ambient to 100*F and, thereafter, the total temperature coefficient is negative. Therefore, in the unlikely event of inadvertent high-power operation leading to high temperatures, the total negative temperature coefficient of reactivity will tend to limit the reactor power.

eDe O

4 q

e I

- _ _ _ _ - - _ ______-l--------------- --- - - _ - - - _

-- a j

4.6.1 Excess Reactivity and Shutdown Margin The RCF Technical Specifications limit the maximum core excess reactivity-to .50.60$ above cold, clean critical . ([ effective = 0.00765).. Reactivity, worths of unsecured experiments are liinited tos0.60$ and worths of movable experiments (planned to be moved during reactor operation) are. limited to - ,

40.40$. In addition, the maximum reactivity insertion rate is limited to 50.05$ per second at source count rate multiplication levels greater than 10 and 50.12$ per second at lower count rate multiplication levels.

The RCF LEU Technical Specifications further require that all four control rod assemblies be operable and that the reactor shall' be subcritical by.  !

more than 0.70$ with the most reactive control rod fully withdrawn. j q

The RCF LEU cores may have up to 0.60$ excess reactivity above cold, clean cri tical . Cores A and B contain four B-10. control rods worth a total of about 3.60$ when removed as a bank. As'a result of core and control rod synnetry, the reactivity worth of all four control rods Lis essentially 'f equal. However if it is conservatively assumed that the first single rod to be fully withdrawn is worth one-half of the total and this rod becomes stuck upon scram, then the shutdown margin is + 0.60$ - 1.80$ = 1.20$,

meeting the Technical Specifications. The total expected shutdown of an RPI LEU core with the full 0.60$ excess allowed by the Technical Specif1-cations would be about 3.00$ with all four B-10 control rods seated._

4.6.1 Assessment The Technical Specifications require that at least four control. rods be l operable and that the reactor can be brought to a safely subcritical condition even if the highest worth control rod is totally removed from the core. These requirements ensure an adequate shutdown margin and provide sufficient redundancy in the unlikely event a control rod fails' to -

scram. Limiting the reactivity insertion rate to. values _less than 0.12$

per second provides that in the unlikely event of an unexpected and unabated reactivity insertion, the reactor operator and/or automatic scram signal will he.ve adequate time to safely shut the reactor down.

Limiting the reactivity worth of. movable and/or unsecured experiments to 50.40$ and 5 0.60$, respectively, prevents a prompt excursion caused by .

an experiment malfunction.

i On the basis of the above considerations, the staff concludes that  :

reactivity addition will be sufficiently limited and that adequate i redundant shutdown capability is provided to ensure safely controlled l operation of the RCF LEU core. ,

4.7 F!snctional Desian of Reactivity Control System' ]

Reactivity changes may be made in the RCF LEU core by changing the ,

reactor tank water level or by manipulating' control rod assemblies. An '

electric interlock is provided that prevents reactivity insertion from the control console by more than one means at a time.

- - ---- _____..__._._~___-_____________________J

4.7.2 Control Rod Drives ,

Primary reactor control is achieved with four control rod assemblies that are positioned vertically within the fuel lattice by remotely operated electromechanical drives. The control assemblies, which are described in Section 4.1.2,' are coupled to the drives by a shaft-turnbuckle connection. -

The four control rod drives are supported at the end of rigid cantilevers mounted on the reactor tank, protruding over the core. Structurally, the drives consist of a 1/20 horsepower motor, gear box, magnetic clu_tch, drive shaft, pinion gear, and control rod rack. Electrically, the control rods operate on demand from the control room and power is supplied to the I l

magnetic clutches from the safety amplifiers. Under nonnal operating I conditions, power to the magnetic clutches secures the control assembly to ,

i the rod drive. Any scra= condition or loss of power will deenergize and .

release the clutch, causing the poison section of the control assembly to drop by gravity into the core, thereby shutting down the reactor. A minimum holding current is adjusted for each drive individually to mini-mize magnet release time and rod drop time. The Technical Specifications rt. quire that the total control rod drop time from its fully withdrawn to its fully inserted position shall be less than 900 msec, including a ~

maximum magnet release time of 50 msec.

The control rods are operated remotely by spring-loaded switches located ~

on the reactor console. Position of each control rod over its full 91.44 cm of travel for the LEU core is monitored by a pair of geared anti-backlash synchrometers, with coarse and fine readouts displayed on _

the console. Full-in and full-out limit indications also are shown at the console. An interlock system prevents the withdrawal of control rods while the tank is being filled with water.

4.7.4 Assessment ,,

The staff agrees with the assessment of the safety and contrt0 systems as presented in NUREG-1023 and concludes that the reactivity control system of the RPI LEU reactor is designed in accordance with good engineering judgment and accepted safety practices and should function to adequately ensure safe operation and safe shutdowa of the reactor under all operating conditions.

I

~

1

[ _

1

-- i

. ,1 '

I 4.9 Initial Core load To safely implement the LEU conversion and measure nuclear characteristics of the LEU core, RPI will conduct a series of tests and evaluations.  !

Written procedures for this series will be evaluated by the Nuclear l Safety Review Board. Special emphasis will be placed on the careful and -

safe application of the extrapolation techniques used in the inverse ,

multiplication methods of core loading. In addition to formal written j restrictions regarding the number and placement of fuel pins during each l loading step, careful analysis of the multiplication data will be performed by senior RPI reactor personnel to be sure the " control rods in

and control rods out" multiplication is changing in a predicted and under- 4 standable manner. Should any anomalies appear during any phase of the i loading sequence, all further loading will be terminated until the l cncmaly is satisfactorily resolved and it is determined by the Senior l Reactor Operator that safe loading can continue. Should the operational '.i core vary significantly in either total fuel pin loading or geometry from (

the predictive calculations, these varidtions will be aaalyzed and under-

, stood by RPI before initiating the zero power core characterization test {

( series.

9 4.10 Conclusions -

L The staff has reviewed the information regarding the specific conversion of the RPI facility from high-enriched to low-enriched fuel, and the )

previous information regarding the safety analysis and license renewal of I the facility in 1983. The staff concludes that RPI adequately determined the necessary changes to the reactor facility caused by the conversion regarding the reactor fuel, core arrangement, control rods, structure, and instrumentation, and accurately assessed the impact of these changes from 11 both an engineering and a safety point of view. The staff further con- "

cludes there is reasonable assurance that the EU L hardware modifications mandated by the conversion are designed and will be built in accordance with good industrial practices, and that the design and performance capability of the components is acceptable to ensure the safe operation of the reactor.

The staff's review of the RCF has included studying its specific design and installation, its control and safety instrumentation, and its operating procedures. These features are similar to those found at other non-power reactor facilities. Furthermore, for the past 27 years, the RCF has operated safely and reliably with a high-enriched core. The staff believes this reflects both good engineering practices and sound judgment at RPI, and, can find no reason to believe this will not continue with the LEU core, which is actually an inherently safer core than the HEU core. On the basis of its review of the RCF, the staff concludes there is reasonable assurance that the RCF is capable of continued safe operatier, as limited by its Technical Specifications. ~

e.-

?

,y

~

6. ENGINEERED SAFETY FEATURES The engineered safety features for the RPI critical facility include the reactor room (sometimes called the cell in RPI's documentation) and the fuel storage vault. The reactor room is. unaffected by tho HEU/ LEU -

conversion; however the fuel storage vault was modified to accommodate the low-enriched SPERT-type fuel pins.

! I 6.2 Fuel Storage Vault j l

A 2.44 m x 3.1 m Fuel Storage Vault in one corner of the reactor room is provided with walls, ceiling, and floor made of 0.3-m-thick reinforced concrete. A storage. rack constructed of unistrut is mounted against one ,j wall of the vault opposite the vault access door. Forty-two stainless ,

steel tubes, 12.7 cm in diameter and wrapped with 0.038-cm-thick cadmium-sheaths, are bolted to the unistrut frame in parallel rows. These tubes and their cadmium sheaths are 107 cm long to ensure the 106-cm-long SPERT LEU fuel pirs are completely encased when stored in the vault. _These 42 tubes can safely accommodate, in physical weigilt and reactivity, all of the 595 LEU pins that have been identified for shipment to RPI. This -

constitutes a total U-235 inventory of 20.94 kg. Conservative --

calculations of the infinite multiplication factor for the vault, when housing the total LEU inventory in the 42 storage tubes under completely flooded conditions, yields a value less than 0.90.

~

6.3 Conclusions .

When NUREG-1023 was published for tho HEU core, the fuel storage vault contained eighty-one 12.7-cm diameter, cadmium-sheathed stainless steel ._J-storage tubes 71 cm in length. When the vault was modified for the SPERT -

LEU fuel, 42 of these tubes along with their cadmium sheaths were .

lengthened to 107 cm and the remaining 39 tubes were removed. The staff concludes this modification, consistent with the calculated safe infinite multiplication factor of less tha: 0.9, provides adequate assurance the RPI fuel storage vault is acceptable to ensure the protection of the  ;

health and safety of the public when SPERT LEU fuel is stored there.

~

l l

13. CONDUCT OF OPERATIONS 13.2 Training f The operator requalification program submitted by RPI on April 15, 1987,  !

l was reviewed by the staff and it was concluded that it meets the condi- l l tions of Appendix A of 10 CFR 55 and criteria contained in ANSI  ;

15.4/ANS-1977 in effect at the time of review. However, since revisions ,

to 10 CFR 55 became effective on May 26, 1987, the licensee is required to ~!

reevaluate the operator requalification program in light of the new ,

10 CFR 55 requirements and submit changes as appropriate. Generic Letter 87-07 provides information concerning revisions to operator licensing and - J NUREG-1262 provides additional clarification for implementing revised  !

2 10 CFR 55.

i 1

?

-i

_ _ _ _ _ _ - . ._ __ - _ __:O

o. ,

12 -

A special inspection was conducted by Region I to determine whether . ..

operator licenses issued to persons at RPI could be reactivated for use on the new LEU core under provisions of 10 CFR 55.31(e). The inspection is documented in Region I Inspection Report, Number 50-225/87-01, May 18, 1987. The conclusions in the report are that the licenses for two of the three corrently licensed operators will be reactivated upon proper appli- -

cation etJ subsequently renewed for use with the new core configuration.  ;

The facility licensee should determine the need for the remaining licensed i operator t7 hold a license valid for the. LEU core. If this.need is determined to exist, then the operator should participate in the requali-

. fication program and document this participation in an application to the NRC for reactivation of the license on the new core.

14. ACCIDENT ANALYSIS . .

The consequences of potential accidents in the RCF are minimal because of the low power levels at which the reactor is operated. In general, the operating power level is kept as low as practicable, consistent with ,

experimental and educational requirements, and is normally below 20 W. '

The Technical Specifications limit the integrated power in Eny 1-year ,

interval to 200 kW hours, which is equivalent to a 23% load factor at the maximum licensed power. Actual annual integrated energy production ~has, in practice, been considerably less (about 100 W hours) and is r.ot expected to change because the facility is being converted from HFU to LEU.

I l

The low power levels and low usage of the RCF result in a very small -

accumulation of decay heat during normal operations. The staff has estimated the initial decay heat power generation immediately following __

several hours' operation at 100 W to be approximately 8 W. Within 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> e" r atter shutdown, the decay heat level will be below 2 W and slowly decreasing. At 2 W, the adiabatic heatup ra te of the fuel is about -.

0.1*C/hr(0.2*F/hr). Therefore, a loss-of-coolant accident, which is normally postulated for non-power reactors, does not constitute any hazard l at the RCF LEU reactor. In fact, the rapid removal of moderator water is a l

design shutdown mechanism in this ' reactor.

Several potentially serious credible accident scenarios were hypothesized and the potential radiological ceasequences to the public were analyzed and evaluated. Even in the worst event considered in the following, no release of I

a significant quantity of radioactive fission products to the reactor cell would occur.

l 1. mechanical rearrangement of fuel l 2. reactivity insertion l 3. fuel handling accident e

o i

i

'I e

6

__-_a__. _ _ _ _ _ . _ __ _ _ - . _ _ _ _ _ _ _ _ - _ . _ a -

.c .- :

-14.1 Mechanical Rearrangement of the Fuel .,

This type of' accident would involve an event that causes rearrangement of the fuel to a supercritical configuration.. Mechanical. rearrangement of the fuel to obtain a supercritical configuration, inadvertently or with intent, is an extremely unlikely occurrence. Core design objectives -

included minimizing the critical mass and, therefore, other configurations of the fuel would be less reactive and not likely to support a large reactivity excursion. Physical damage to fuel pins is precluded both by each element's cladding and the stainless steel' support plates / stanchions that compose the basic core structure. In the unlikely event that suffi-cient force to break one or more of.the fuel pins was developed, the very low fission-product inwntory accumulated in the fuel pins would not cause a significant offsite hazard. On this basis, INEL concludes that no .

credible inadvertent mechanical rearrangement of the fuel would lead to a -

radioactivity release with more severe offsite consequences than those evaluated in Section 14.3 or lead to an inadvertent reactivity addition i with more severe offsite consequences than those evaluated in Section 14.2.

14.2 Reactivity Insertion ,,

Reactivity is normally inserted into the RCF LEU reactor by either adding d water moderator or withdrawing the control rods. The design of the reactor control system, incorporation of safety circuits, requirements of the Technical Specifications, and imposition of administrative controls all serve to limit the rate and magnitude of operational and accidental .

reactivity insertions. In Sections 14.1 and 14.3, the hypothetical l

accidents are assumed to occur with the reactor shut down. In this hypothetical event, it is initially assumed that the reactivity excursion -f might cause abnormally high power levels and a resultant abnormally large inventory of fission products. Therefore, loss of fuel integrity might -.

lead to release of sufficient radioactivity to cause significant offsite radiological consequences. Thus, the first question to be answered is whether a credible nuclear excursion would be likely to cause fuel damage.

If not, the fission products are contained in the fuel, and no further analysis is required to assess radiological risks to the public.

14.2.1 Scenario For purposes of evaluating potential impacts on the public, RPI hypothesized the inadvertent addition of excess reactivity, coupled with-the worst single credible failure in the safety system. Because the Technical Specifications limit both the excess reactivity above cold, clean critical and the maximum reactivity worth of an unsecured experiment to 0.60$, the RPI staff assumed an acc1 dental instantaneous insertion of that amount of excess reactivity while the reactor was operating at an _

abnormally high power level. The RPI staff has determined that this event would cause the largest credible reactivity excursion in the RCF. ;l e

m

4 The RPI staff made the'following additional assumptions, each of which adds conservatism to the accident scenario: Jl

1. The initial power level is 200 watts, based upon the Technical Specifications limit of 100. watts and incorporating a factor of _)

2 to account for the cumulative uncertainties associated with '

instrument calibration.

2. For analytical purposes, the reactivity feedback effects of I temperature and void formation were neglect 2d r,o that control rod insertion was the only mechanism for tenninating the event.

3. The ionization chamber feeding the logarithmic power level and reactor period channel feils simultaneously with the insertion of - I l the excess reactivity. This disables both the high. log power

• scram (135 W) and the short period (5 seconds) scram. f l 1

4. The two linear safety channels do not prevent the 200-W operation, but at least one channel is capable of initiating a scram signal at 90% of range after the nuclear excursion has {

progressed to this point. _

5. At the initiation of the 0.60$ step reactivity insertion, the linear power channel which will cause the scram is at 10% of range, i.e., the scram will occur when the reactor power has reached 200 W x 9 = 1800 W. 3 i

With the reactor operating initially at 200 watts, the insertion of 0.60$

positive reactivity causes power to promptly jump to 600 watts and then  : d increase on a period of 3.0 seconds to 1800 watts, at which point LP1 C_

and/or LP2 generate a scram signal. Allowing 1.5 seconds thereafter for the rods to be bottomed (Technical Specification is 900 msec), analysic ~

conservatively assumed the instantaneous insertion of 1.00$ negative reactivity (less than the core shutdown margin) at 5 seconds after the excursion began.

Maximum power reached during the transient is slightly below 3050 watts, l

depositir,g about 10 kJ of energy in the core and inducing a fuel temperature rise of less than 0.1*C (0.2*F) above an initjal value of 68'F (20*C). This energy depositign is roughly a factor of 10 less than the core safety limit of 3.3 x 10 kJ or 230 cal /g (the failure threshold for unirradiated SPERT fuel) identified in the RPI LEU Technical Specifications.

This increase in core energy and temperature is_nealigible compared to the melting point of stainless steel, U09 , and the boiling point of water, so i no damage to fuel or cladding would Pesult, and no fission products would be released as a result of-the maximum reactivity accident at the RCF LEU reactor. -,

M e.

=

l *-4 .

L___.__._____________.___________________._ _

l 19 g 14.2.2 Assessment The staff considers that the reactivity insertion accident postulated by il the RPI staff is representative of the most severe transient that can f credibly occur at the RCF LEU reactor. The staff has reviewed RPI's '

accident assumptions and calculations and finds them very conservative, but' reasonable and acceptable. Therefore, the staff concludes.that.it is unlikely that a credible nuclear excursion in the RCF would lead to fuel melting or cladding failure. Furthermore, the fission product inventory produced during the transient analyzed corresponds to less than-3 minutes of full-power operation. Thus, there is reasonable assurance that significant fission-product activity would not be released to the environ-I ment as a result of a reactor reactivity transient, even if there were j some damage to the fuel. Therefore, the staff concludes that the maximum postulated reactivity transient at the RCF does not pose a significant.

hazard to the public.

14.3 Fuel Handling Accident The fuel handling accident, which is designated'as the maximum hypothe-tical accident (MHA) for the RCF LEU core, includes various incidents involving one or more irradiated fuel pins in which the fuel cladding.

might be breached or ruptured. Because of the typicilly low power levels and the low usage ( m 100 W hr/yr), the fission-product inventory and l buildup is extremely small and a significant portion of the core would j have to be totally destroyed before release of the volatile fission pro- )

ducts would constitute a safety hazard. However, for completeness, the fuel handling accident is discussed.

I For the MHA, the staff did not try to develop a detailed scenario of how l the accident occurs, but rather assumed that the cladding _of one irradi-ated fuel pin fails and this occurs outside the reactor pool air, instantly releasing all of the available volatile fission-products that have accumulated in the free volume (gap) between the fuel and the .

cladding, Furthermore, the scenario conservatively assumes that the )

accident occurs following an extended run at full licensed power, i.e., i one day (24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />) at 100 W which results in a total fuel burnup more than )

20 times higher than the actual average yearly RPI fuel burnup of 100 W -  :

hours. Of the total gaseous fission-product inventory of about 23 milli--  !'

curies for the hot fuel pin, it has been estimated that less than 0.2% will escape from the fuel into the gap between the fuel pellet and the cladding Upon the cladding failure, all of this gap inventory is assumed to er. cape  !

from the fuel pin. If this inventory is assumed to all remain in the  !

reactor room volume, the I-131 radiation level will reach about 0.1 ,

maximum permissible concentration {MPC) for a restricted area with the ]

noble gas MPC at least a factor of 10 lower. If all of the gaseous fission- l products from the gap of the maximally irradiated fuel pin are assumed -l to be released from the 15-m stack, the maxir dose commitment offsite is less than 3 x 10~mrem. gum total thyroid inhalation l

On the basis of the above discussions and analysis, the staff concludes _.

that if the maximally irradiated fuel pin from the RCF LEU core were to release all of the noble gaseous and halogen fission products accumulated in its fuel cladding gap, radiation doses to occupational personnel and to the public in unrestricted areas would be significantly below the guideline values'of 10 CFR Part 20, Appendix 8. Accordingly, the staff concludes that _

there is reasonable assurance that the postulated accident poses no signi-ficant radiological risk to the health and safety of the public or to the j operational staff.

14.4 Conclusion I

The staff has reviewed the credible potential transients and accidents for the RPI LEU reactor. On the basis of this review, the postulated accident -

with the greatest potential effect on the environment is the luss of e

, cladding integrity of an irradiated fuel pin in air in the reactor room.

The analysis of this accident indicated that even if the cladding of {

several fuel pins failed simultaneously, the potential dose equivalents in l restricted and unrestricted areas still would be significantly below the l guideline values of 10 CFR Part 20. .

In NUREG-1023 the consequences of three accident scenarios were evaluated and found to be acceptable. The MHA for the LEU core, i.e., loss of cladding integrity of an irradiated fuel pin in air was not considered a credible event for the HEU core. The three accidents considered were natural phenomena, mechanical rearrangement of the fuel, and the reactivity insertion. An analysis of these same three accident scenarios for the RPI -

LEU core lead to the following observations. Because conversion to LEU does not affect the natural phenomenon accident scenario, the conclusions of ,

NUREG-1023 regarding this event are unchanged for the LEU core and this

r i accident was not discussed here. With regard to the mechanical rearrange-ment of fuel, although the mechanistic accident scenario is somewhat --

altered because of the physical differences between LEU fuel pins and HEU fuel elements, the conclusions regarding the probability of occurrence and  ;

radiological consequences are basically unaltered from those of NUREG-1023. '

Additionally, as was shown~in Section 14.1, the radiological consequences for the mechanical rearrangement of the LEU fuel is enveloped by the LEU fuel handling accident (Section 14.3) and the reactivity addition possible from mechanical rearrangement of the LEU fuel is enveloped by the maximum reactivity accident for the LEU fuel (Section 14.2), both of which have been shown to have acceptable consequences. Finally, with regard to the-reactivity accident scenario for the HEU core versus the LEU core, the LEU consequences are not totally enveloped by those of the HEU. However, the conclusions of NUREG-1023 are still valid. For the HEU core, the peak power reached during the transient was 2 kW with a total energy release of 8 kJ which resulted in a maximum fuel temperature increase of 16'C. For the LEU core, the peak power reached 3 kW with a total energy release of -

10 kJ. Hovcver, because of the much greater heat capacity of the LEU e

l W

~

, .. _1 w , .

,l l

core, the maximum fuel temperature increase.was less than 1*C. In both ~

1 cases the consequences of the maximum reactivity accident are negligible.

On the basis of the above transient.and accident review, and the favorable assessment in EGG-NTA-7660 regarding the suitability of using of the SPERT F-type fuel pins in the RPI critical facility, the -,

staff concludes that reasonable assurance exists that the RPI LEU reactor can be operated with a low probability of accidents and that even the maximum hypothetical accident will pose no significant risk to the health ,

and safety of the public. l

15. TECHNICAL SPECIFICATIONS i

The licensee's Technical-Specifications evaluated in this licensing action l define certain features, characteristics, and conditions . governing the -

continued operation of this facility. The Technical Specifications, which "

have been revised, are included in the attachment to the order modifying the

)

license (Amendment D ). Replacement pages for the revised Technical Specifi- j cations have been developed and will be provided to the licensee. Formats 1 and contents acceptable to the NRC have been used in the development of these Technical Specifications, and the staff has reviewed them using the l

)

I standard ANSI /ANS 15.1-1982 as a guide. -l On the basis of its review, the staff concludes that normal reactor operation within the limits of the Technical Specifications will not result in offsite -

radiation exposures in excess of 10 CFR 20 limits. Furthermore, the limiting

' conditions for operation, surveillance requirements, and engineered safety l features will limit the likelihood of malfunctions and mitigate the conse- _

l qu6nces to the public of offnonnal or accident events. j

18. CONCLUSIONS }

.. j i

Considering its evaluation of the application as set forth above, the "

staff has determined that:

1) The application for conversion from HEU to LEU fuel for its critical l facility reactor filed by the Rensselaer Polytechnic Institute dated October 6,1986, as supplemented, complies' with the requirements of the Atomic Energy Act of 1954, as amended (the Act), and the Comission's regulations set forth in 10 CFR Chapter 1.

I

2) The facility will operate in conformity with the application as amended, the provisions of the Act, and the rules and re.gulations of l the Comission.

3) There is reasonable assurance (a) that the activities authorized by the operating license can be conducted without endangering the health and safety of the public, and (b) that such activities will'be -

conducted in compliance with the regulations of the Comission set forth in 10 CFR Chapter 1.

T

18 -

4) The licensee is technically and financially qualified to engage in the activities authorized by the license in accordance with the regulations of the Commission set forth in 10 CFR Chapter 1.

5) The conversion from HEU to LEU fuel will not be inimical to the -

common defense and security or to the health and safety of the i

public. '

l l 19. REFERENCES Hobbins, R. R., " Technical Evaluation Report for the Requalification of SPERT Fuel for Use in Non-Power Reactors," Idaho National Engineering Laboratory, EGG-NTA-7708, May 1987 .

Carpenter, W. R. , and Cooper, C. H., "Rensselaer Polytechnic Institute, High Enriched to Low Enriched Fuel Conversion," June 1987, EGG-NTA-7722 Miller, H. J., and Buzy, J., "Rensselaer Polytechnic Institute Technical Conference Request of April 9,1987," May 22,1987 _

Dated: July 7,1987 4 's e_-.y 1

.e

+

9 9

w*N

_ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ . -