On the dose fields due to activated cooling water in nuclear facilities

This paper focuses on activation of oxygen isotopes in water and decay of this activated isotopes, i.e. 16N, 17N and 19O. An analysis of activation of water in pressurized water reactors and in fusion reactors was performed. Different evaluated nuclear data libraries were used in activation calculations (ENDF/B-VIII.0, FENDL-3.1b, JEFF-3.2 and TENDL-2015).

Contents lists available atScienceDirect Progress in Nuclear Energy journal homepage:www.elsevier.com/locate/pnucene

Review

On the dose fields due to activated cooling water in nuclear facilities

Andrej Žohar, Luka Snoj∗

Jožef Stefan Institute, Jamova cesta 39, SI-1000, Ljubljana, Slovenia

A R T I C L E I N F O

Keywords:

Activated cooling water

PWR

Fusion reactor

MCNP

Cooling pipes

Steam generator

A B S T R A C T Activated cooling water in nuclear facilities can present a significant radiation source around primary cooling system causing radiation damage to electrical components, increasing doses to personnel and in the case of fusion facilities additional heating to superconducting coils This paper focuses on activation of oxygen isotopes

in water and decay of this activated isotopes, i.e.16N,17N and19O An analysis of activation of water in pres-surized water reactors and in fusion reactors was performed Different evaluated nuclear data libraries were used

in activation calculations (ENDF/B-VIII.0, FENDL-3.1b, JEFF-3.2 and TENDL-2015) The calculated activation rates with different nuclear data libraries agree well for the16O(n,p)16N reaction and significantly differ for17O (n,p)17N and18O(n,γ)19O reactions In fusion reactor the specific activity of activated water isotopes is in the order of 1013Bq/m3/MW, which is five orders of magnitude higher compared to specific activity in a typical fission pressurized water reactor, amounting to 109Bq/m3/MW The results of specific activity of cooling water were used to perform parametric analysis of dose rates around pipes of cooling system and dose field around a steam generator in a pressurized water reactor as a representative of heat exchangers The analysis of dose rates around pipes include pipes featuring 1 mm to 8 cm thick walls and from 0.5 cm to 60 cm water radius Results can be used to estimate dose rates for all studied isotopes, provided the specific activity is known For heat exchangers the decay of16N contributes majority to the dose rates in the air surrounding them while17N and19O decay contributes together less than 0.1% For a typical 2 GW thermal power two loop pressurized water reactor the dose rates in air surrounding the stream generator are in the order of several mSv/h

1 Introduction

Water is cooling fluid in many nuclear facilities, such as fission

nuclear reactors and some fusion reactors In fission reactors water is

activated when passing through the reactor core, in fusion reactors

however water is activated when cooling the blanket, or other

com-ponents of the reactor such as diagnostic equipment Activation of

water consist of activation of oxygen and hydrogen as primary

con-stituents of the H2O molecule, activation of dissolved gasses, corrosion

products and additions to water and fission products in fission reactor

As all the latter are case specific, in this paper we will focus on

acti-vation of pure H2O only After being irradiated and activated the

cooling water flows through the primary cooling circuit, commonly

outside the primary biological shielding surrounding the reactor vessel

There the activation products decay, emit radiation, which causes

ra-diation damage to electrical components, increasing doses to personnel

working around the cooling circuit and in the case of fusion facilities

causes nuclear heating of various cold components such as

super-conducting coils cooled by liquid helium (Iida et al., 1997)

Decay of activated cooling water can also be used to obtain

important parameters of the heat producing component In nuclear power plants the decay of activated water is used to detect leakage of primary cooling system in the secondary cooling system (IAEA, 2000) Activated water can also be used to determine water flow and power of the reactor (Tsypin et al., 2003) In the case of fusion reactors the neutron yield of the reactor can be measured with the use of activated water (Nishitani et al., 2003) There are several papers on measurement

of activation of cooling water in fission power plants and research re-actors (Guo et al., 2018;Stepišnik et al., 2009) where the measurement are performed regularly for education of university students For fusion reactor only one experiment on the activation of water was performed

at the JAERI FNS facility in Japan (Uno et al., 2001) Activated cooling water is also present in spallation source facilities (Santoro et al., 1999) The main contributors to the activity of clean cooling water are radioactive isotopes of oxygen and nitrogen produced by activation of oxygen isotopes in the cooling water, i.e.16N,17N and19O Majority of studies dealing with activation of cooling water and dose fields due to decay of activated water focuses on isotopes16N and17N (Blakeman

et al., 2007;Santoro et al., 1997) while the majority neglects the effects

of isotope19O due to lower energies of gamma radiation emitted in

https://doi.org/10.1016/j.pnucene.2019.103042

Received 19 December 2018; Received in revised form 19 April 2019; Accepted 25 April 2019

∗Corresponding author

E-mail addresses:andrej.zohar@ijs.si(A Žohar),luka.snoj@ijs.si(L Snoj)

Available online 16 May 2019

0149-1970/ © 2019 The Authors Published by Elsevier Ltd This is an open access article under the CC BY-NC-ND license

(http://creativecommons.org/licenses/BY-NC-ND/4.0/)

T

decay and negligible activity compared to isotope16N However in this

paper activation and dose rates of all three isotopes of activated oxygen

in cooling water are studied As hydrogen has two stable isotopes,3H is

also produced in cooling water from2H activation However, as3H

half-life is in long compared to other activated isotopes and practically no

gamma rays are emitted at decay, 3H dose rate contribution are

ne-glected in this paper as they are in other papers

The analysis presented in this paper focuses on calculating

activa-tion of oxygen isotopes and the dose fields around primary cooling

pipes and heat exchangers of nuclear devices by using Monte Carlo

particle transport code MCNP (Goorley et al., 2012) In the first part of

the paper the neutron activation data, comparison of cross-sections for

activation between different evaluated nuclear data libraries and results

of activation calculations for fission and fusion reactors are presented

The second part of the paper presents the parametric analysis of dose

rates in the air surrounding pipes with different wall thickness and

water diameter and equivalent biological dose rates in the air

sur-rounding a heat exchanger are presented For the model of a heat

ex-changer a vertical steam generator in a typical 2 GW thermal power

pressurized water reactor was used Dose rates presented in the paper

are the H*(10) ambient dose equivalent for biological dose rates and

dose rates in silicon for electronic components

2 Neutron activation of water

Oxygen and nitrogen activated isotopes in cooling water are

pro-duced from activation of oxygen isotopes via the 16O(n,p)16N, 17O

(n,p)17N and18O(n,γ)19O reactions Activated isotopes in cooling water

decay by emitting various decay products with different energies

Summarized data is presented inTable 1(Chadwick et al., 2011) The

marked energies in table present the dominant energies of decay

pro-ducts emitted at decay As16N decays via decay path16N 16O + γ,

high energy gamma rays are emitted (E = 6.13 MeV) with half-life of

7.13 s17N decays via decay path17N 16O + n + γ with half-life of

4.14 s

Emitted neutrons can activate components outside the primary

circuit and produce neutron induced gamma-rays Activated isotope

19O has a half-life of 26.9 s and decays via decay path19O 19F + γ.

Reactions16O(n,p)16N and17O(n,p)17N are threshold reactions with

energy threshold at 10 MeV and 8 MeV respectively Reaction 18O

(n,γ)19O already takes place at thermal energies as presented inFig 1

Due to threshold reactions the activation of water is expected to be

higher in fusion reactors like ITER compared to fission reactors due to

higher neutron energies (14 MeV neutrons from deuterium-tritium

fu-sion)

As there are many different evaluated cross-sections for the above mentioned water activation reactions in different evaluated nuclear data libraries, the cross-sections can be significantly different In this paper cross-sections from four different libraries were used: ENDF/B-VIII.0 (Brown et al., 2018), JEFF-3.2 (OECD/NEA Data Bank, 2014), FENDL-3.1b (Koning and Trkov, 2016) and TENDL-2015 (Koning et al.,

2015) Cross-section for activation of16O (Fig 2) in all studied libraries

is the same as it is derived from the same experimental data (Nelson and Michaudon, 1999)

For17O activation however the cross-sections differ between li-braries as presented inFig 3 The cross-section in JEFF-3.2 library is taken from TENDL-2012 and cross-section in FENDL-3.1b is taken from TENDL-2010, which are predecessors of TENDL-2015 library Eval-uated cross-sections in TENDL libraries are based on computations by software for simulation of nuclear reactions TALYS (Koning and Rochman, 2012) The TENDL-2010 library is based on TALYS 1.20 version, TENDL-2012 is based on TALYS 1.50 version and TENDL-2015

is based on the TALYS 1.74 version for computation of cross-sections Due to this the cross-sections between this three libraries are similar unlike the cross-section from ENDF/B-VIII.0 library, which is taken from ENDF/B-V library which was released in 1978 and is based on computations by MODNEW (Uhl, 1972) and measurements performed

by Menlove (Menlove et al., 1970)

InFig 4the cross-sections for reaction18O(n,γ)19O from all studied evaluated nuclear data libraries are presented With the release of the ENDF/B-VIII.0 evaluated nuclear data library the cross-section for ac-tivation of 18O was added The cross-section is based on the

Table 1

Summarized data of activated isotopes of cooling water obtained from ENDF/B-VII.1 data library (Chadwick et al., 2011) The marked energies present the dominant energies of decay products

Fig 1 Cross-section energy dependence for activation of oxygen nuclide taken

from the JEFF-3.2 data library (OECD/NEA Data Bank, 2014)

Moghabghab resonance parameters below 5 MeV (Mughabghab, 2006)

while above 5 MeV the cross section is based on J Kopecky and D

Nierop evaluation for EAF-3 library The cross-section in the JEFF-3.2

library is taken from the TENDL-2012 while the cross-section in the

FENDL-3.1b library is taken from the 2014 library The

TENDL-2012 library is based on TALYS 1.50 version, TENDL-2014 is based on

TALYS 1.64 version and TENDL-2015 is based on the TALYS 1.74 version for computation of cross-sections As in the case for activation

of17O the cross-section for activation18O taken from the ENDF/B-VIII.0 library differs significantly compared to cross-sections in other studied nuclear data libraries The difference in cross-section between other studied libraries is in the epithermal region at around 0.08 MeV In this region the cross-section for18O(n,γ)19O from FENDL-3.1b and

TENDL-2015 libraries exhibit a resonance peak, while the cross-section from JEFF-3.2 library has no peak Due to this the calculated reaction rates are expected to be higher with the use of FENDL-3.1b and TENDL-2015 library

3 Activation of cooling water

It is difficult to measure the absolute value of activity of activated isotopes in cooling water in nuclear facilities due to short decay times of isotopes and high energy radiation which can cause high dose rates Another difficulty is the placement of large detectors (e.g High Purity Germanium detector) close to the primary cooling system for accurate measurements as the systems are normally shielded The easiest way to determine the absolute value of isotope activity is by calculations from parameters of nuclear facility The change of specific activity of a

stu-died isotope in the cooling water a with time is described by:

=

where λ is a decay constant [s−1] and F is an average reaction rate in

region of interest over irradiation time which is described by:

=

F ( , , ) ( ) ( , )d d d ,r E t E nr t E V t

where the ( , , )r E t is the neutron flux at position r, i( )E is the mi-croscopic cross-section for studied reaction and n( , )r t the number

density of target atoms at position r.

In nuclear facilities cooling water circulates in the primary cooling system and is exposed to neutron flux for a short time Hence the change in specific activity of studied isotope is described using a system

of equations:

=

a a

e ,

(3) wherea o is the specific activation of coolant on the outlet of heat producing component,a i is the specific activation of coolant on the

inlet of heat producing components, t i is the exposure time and T is

circulation time the coolant needs from the outlet to the inlet of heat producing component

From the eq.(3)the equilibrium value of specific activity at the outlet of the heat producing component can be obtained:

a F 1 e

o

t

t T

( )

i

In eq.(4)the F is the average reaction rate over whole heat

pro-ducing component The intensity of neutron fluxes as well the energy spectrum can significantly change through the heat producing compo-nent This changes can be taken into account by dividing the heat producing components in smaller sections with similar neutron fluxes and energy spectrum in which the reaction rates are calculated In

general there are n equations for n regions in the heat producing

component plus one equation for the region outside the heat producing component The general form of the system of equations is:

=

+

+

e

1 1

(5) From eq.(5)the equilibrium of activity of isotope at the outlet of heat producing component (a n) can be calculated

Fig 2 Cross-sections for reaction16O(n,p)16N taken from ENDF/B-VIII.0,

JEFF-3.2, FENDL-3.1b, TENDL-2015 libraries and experimental results from EXFOR

database

Fig 3 Cross-sections for reaction17O(n,p)17N taken from ENDF/B-VIII.0,

JEFF-3.2, FENDL-3.1b, TENDL-2015 libraries and experimental results from EXFOR

database

Fig 4 Cross-sections for reaction18O(n,γ)19O taken from ENDF/B-VIII.0,

JEFF-3.2, FENDL-3.1b, TENDL-2015 libraries and experimental results from EXFOR

database The cross-sections from FENDL-3.1b and TENDL-2015 library are

similar except in the high energy region (above 30 MeV) and are due to this

overlapped in the above graph

As there are many different nuclear facilities where cooling water is

activated a general analysis of activation of water in each of them is

difficult to perform In this paper an analysis of activation of cooling

water in a typical PWR and fusion reactor is presented PWR reactor

was chosen as a representative of fission reactors as they are the most

common type of fission power reactors

3.1 Activation of cooling water in PWR

To obtain the absolute value of specific activity of activated isotopes

of cooling water at the outlet of heat producing component, which in

the case of PWR is the reactor vessel, from eq.(4)some parameters are

needed The parameters are the exposure times of cooling water to

neutron flux, circulation time and the reaction rates The exposure time

and circulation time can be determined from volume flow rate of

cooling water and volume of it in different regions In the studied case

the circulation time of cooling water was calculated to be around 8.1 s,

while the exposure time was calculated to be around 4.2 s of this 1.4 s to

high neutron flux in reactor core

The reaction rates were calculated by the Monte Carlo particle

transport using the MCNP code A geometrical model of a typical 2 GW

two loop PWR reactor vessel was constructed in MCNP The model is

presented inFig 5and neutron spectra for all studied regions is

pre-sented inFig 6 The reaction rates in the reactor vessel were calculated

as follows The reactor vessel was divided in four sections: downcomer,

lower plenum, core and upper plenum and then the section averaged

reaction rates were calculated by multiplying the neutron flux

calcu-lated by the track length estimator (F4 tally in MCNP) with the

corre-sponding cross-section The reaction rates were calculated using all

studied nuclear data libraries and density of water at 600 K Results of

average reaction rates in regions for some libraries are presented in

Table 2 The highest reaction rates are in the reactor core where the

neutron flux is highest Two orders of magnitude lower reaction rates are in the downcomer while in lower and upper plenum the reaction rates are five orders of magnitude lower as in the core due to support structures in lower plenum and control rods in upper plenum The uncertainties in the calculation results are due to systematic error of Monte Carlo calculation, uncertainty in the model and uncertainty of nuclear data However, only the Monte Carlo statistical uncertainties are presented inTable 2 For lower and upper plenum the greatest contribution to uncertainty is the Monte Carlo statistical uncertainty

To improve the statistic of simulation the calculation times would need

to be extended However, the contribution to the total reaction rate from this two regions is negligible compared to the contribution of the core and additional calculation time would not change the final result significantly

For calculation of specific activity the total value of reaction rates are needed but the spectral analysis of reaction rates were also per-formed The results of reaction rate spectra in core for some libraries are presented inFig 7 For activation of18O results of reaction rate spectra

in core from two different libraries are present to show the effect of additional resonance peak in cross-section in TENDL-2015 library From calculated reaction rates and exposure times the equilibrium specific activity at the outlet of the reactor vessel was calculated using

eq.(5)and the obtained values for all studied nuclear data libraries are presented inFig 8 The most activated isotope of cooling water is16N due to high natural abundance and higher cross-section for reaction The equilibrium specific activity is four orders of magnitude higher than equilibrium specific activity of17N and two orders of magnitude higher than equilibrium specific activity of19O Due to differences in cross-section for activation of17O the equilibrium specific activity of

17N obtained with the ENDF/B-VIII.0 library is by a factor of three higher compared to results obtained with other libraries The equili-brium specific activities of19O obtained with libraries TENDL-2015 and FENDL-3.1b are by a factor of three higher than equilibrium specific activity obtained with the use of JEFF-3.2 library due to resonance peak

in the cross-section at epithermal energy Despite significant differences

in cross-section for activation of 18O in ENDF/B-VIII.0 library the equilibrium specific activity is comparable to results obtained with the use of TENDL-2015 and FENDL-3.1b This is due to the resonance peaks

at fast neutron energies

3.1.1 Time dependence of specific activity

Results of specific activity presented inFig 8 present the equili-brium value However, during start-up, shutdown and power changes of reactor the value of specific activity changes The behaviour in specific activity can be simulated using the calculated reaction rates in specific areas This is described with a set of eq.(6)(Žohar and Snoj, 2016):

=

<

+ + +

=

A t

a t t t t t N T t

( )

(1 e ), (1 e )e (1 e ),

o

t

i

t

i

4

0

i i

4 3

4

(6)

where a is the saturated value of specific activity after one cycle at the

outlet of reactor vessel The first equation in eq.(6)presents the specific activity produced in region 4 (upper plenum) The second equation presents the saturated specific activity produced in region 4 plus spe-cific activity produced in region 3 (reactor core) The equations follow this order until the time the cooling water goes through the whole re-actor vessel After that the last equation describes the specific activity behaviour

The first analysis performed was the change of specific activity of all activated isotopes of cooling water during the start of the studied re-actor with no activated cooling water Special attention was given to

Fig 5 MCNP model of reactor vessel in a typical PWR with marked regions for

reaction rates calculation and marked direction of water flow in reactor vessel

the time and number of recirculation cycles it takes for the specific activity of a radionuclide to reach the saturated value and the results are presented inFig 9 The specific activity of all products is asymp-totically increasing in steps due to recirculation cycles The results show that it takes 4.3 min (21 cycles) for specific activity of16N to reach saturated value, 2.9 min (14 cycles) for specific activity of17N to reach saturated value and 14.3 min (70 cycles) for specific activity of19O to reach saturated value The specific activity changes in the last cycles are too small to be visible in the graph

The specific activity behaviour during power changes and after shutdown was also analysed for the studied reactor To simulate this the

Fig 6 Lethargy neutron flux spectra in all four studied regions of a 2 GW thermal pressurized water reactor.

Table 2

Reaction rates in studied regions of reactor vessel in a 2 GW thermal power

PWR

Region 16 O(n,p) 16 N ENDF/

B-VIII.0 [cm −3 s −1 ]

17 O(n,p) 17 N TENDL-2015 [cm −3 s −1 ]

18O(n,γ)19 O FENDL-3.1b [cm −3 s −1 ] Downcomer 1.39·10 5±6.98·10 3 9.85± 0.79 5.05·10 3±4.54·10 2

Lower plenum 6.71·10 2±1.61·10 2 0.037± 0.009 3.60± 0.38

Reactor core 1.11·10 7±5.56·10 5 780± 62 1.99·10 5±2.11·10 4

Upper plenum 1.53·10 2±4.15·10 1 0.022± 0.007 1.92± 0.21

Fig 7 Reaction rate per energy bin in the core of a 2 GW thermal power PWR for activation of all isotopes of cooling water obtained for some nuclear libraries For

activation of18O results from two different libraries are taken to present the effect of additional resonance peak in the cross-section

power level in simulation was changed from full power to 50% power

and kept constant till the specific activity of all isotopes reached new

saturated value Then the power was changed back to full power and

after the specific activity of all isotopes reached saturated value the

reactor was shut down The results of the simulation are presented in

Fig 10

The specific activity behaviour during power changes is similar to

the behaviour during the reactor start-up The times and numbers of

cycles needed to reach new saturated values are the same The last part

of the simulation presents specific activity behaviour after a rapid

re-actor shutdown 15 min after rere-actor shutdown the specific activity of

all isotopes falls below 0.001 Bq/m3

3.2 Activation of cooling water in fusion reactors

In fission reactors the energies of neutrons at birth are distributed

according to Maxwell spectrum with peak energy below 1 MeV and

average energy of neutrons at around 2 MeV On the other side in fusion

reactors fusing deuterium and tritium (D-T) the energies of neutrons at

birth are around 14.1 MeV, an order of magnitude higher Due to this

the neutron spectrum between fission and fusion reactors differ The

comparison of both neutron spectra in cooling water is presented in

Fig 11 The absorption lines in the fast neutron region of fission spectra are due to elastic scattering of neutrons on16O

As already mentioned in the beginning of the paper due to higher energies of neutrons in fusion reactors and threshold reactions for ac-tivation of16O and17O the activity of cooling water is expected to be higher in fusion reactors Due to this calculations of activation of cooling water in fusion reactor using MCNP were performed The methodology for calculation was same as for calculation in fission re-actor Reaction rates in water pipes 6 cm from first wall of reactor were calculated using the MCNP for all studied nuclear data libraries The neutron spectrum used was from a D-T plasma and the exposure time was estimated to be 1 s It was also assumed that the cooling water was not activated before the cooling of reactor as the circulation time of system is large enough for all activated isotopes to decay before new activation The results of activated cooling water for an ITER like re-actor with 500 MW thermal power is presented inFig 12 As predicted for ITER like fusion reactors the specific activity of cooling water for all isotopes is higher compared to fission reactors For isotope16N and17N the specific activity is four orders of magnitude higher while for isotope

19O the specific activity is one orders of magnitude higher Due to lower thermal neutron flux in fusion reactors compared to fission reactors the specific activity of17N is higher than specific activity of19O despite lower natural abundance

The specific activity for19O obtained with the ENDF/B-VIII.0 li-brary is higher compared to results obtained with TENDL-2015 and FENDL-3.1b library due to higher cross-section in the energy region of fast fusion neutrons

Fig 8 Equilibrium specific activities of activated isotopes in cooling water for

all studied nuclear data libraries for a typical 2 GW thermal power PWR

Fig 9 Time dependence of specific activity from the start of a reactor with no

activated cooling water to saturated value Steps in the graphs corresponds to

individual cycles of cooling water

Fig 10 Time dependence of specific activity during power changes and after

shutdown

Fig 11 Comparison of neutron flux energy spectrum in cooling water for

fis-sion and fufis-sion reactors

Fig 12 Equilibrium specific activities of activated isotopes in cooling water for

all studied nuclear data libraries for an ITER like fusion reactor of 500 MW thermal power

The results of activity were also normalized to 1 MW thermal power

of reactor for comparison of results between fission and fusion reactors

and are presented inTable 3 For isotope16N and17N the specific

ac-tivity is five orders of magnitude higher while for isotope 19O the

specific activity is two orders of magnitude higher

4 Dose field in nuclear facilities

Gamma rays and neutrons emitted at decay of isotopes of activated

cooling water cause radiation damage to electrical and structural

components in vicinity of primary cooling system and increased dose

rates to personnel performing tasks close to the cooling system Due to

this the determination of dose field is important for designing of

shielding for components and personnel As already stated in the paper

measurements of dose field due to decay of activated cooling water in

nuclear facilities is in most cases difficult if not impossible As a result

of that the dose field needs to be obtained using computational methods

like Monte Carlo or deterministic methods For complex nuclear

facil-ities the Monte Carlo method is the preferred method In this paper the

dose fields were calculated using the Monte Carlo code MCNP

In Monte Carlo calculations the dose rate at a specific location in

studied facility is calculated using the particle flux at studied location

and flux-to-dose conversion factors For effects on biological tissue (the

H*(10) ambient dose equivalent) the flux-to-dose factors from standard

ICRP-21 (ICRP, 1973) are used for gamma rays while for neutrons the

flux-to-dose factors from standard ICRP-74 (ICRP, 1996) are used which

have been independently validated by several different institutions

(Traub, 2010) For dose rates on the silicon components the flux-to-dose

factors from standard ASTME722-14 (ASTM International, 2014) for

neutrons are used while for gamma rays the dose rates are calculated

using energy deposited in studied material All flux-to-dose factors used

in this paper are presented inFig 13

All nuclear facilities have some common components in cooling

systems like pipes As there are many different sizes of pipes through

the cooling system of a facility a parametric analysis is needed to cover

all possible pipe sizes for general study On the other hand heat

exchangers wary significantly between different types of nuclear facil-ities and a general analysis of dose field surrounding them is difficult to perform Due to this an analysis of biological dose field in the air sur-rounding a vertical steam generator in a typical PWR was performed The results and analysis of both studies are presented in following chapters

4.1 Parametric analysis of pipes

Pipes are connectors between major components in the cooling loop and guide tubes for instruments measuring parameters of coolant Due

to this the diameter of cooling water and thickness of walls changes throughout nuclear facilities A parametric study of the dose filed in the air surrounding the pipes due to decay of activated cooling water was performed to include all possible sizes of pipes The aim of this para-metric study is to provide guidelines on expected dose rates around different pipes containing activated water The results of dose rates were calculated by using the MCNP with the ENDF/B-VIII.0 library at

50 cm distance from surface of pipe The simulated model was a two meters long pipe with reflecting surfaces (boundary conditions) at the end and surrounded by air The material used for the pipe wall was stainless steel (SS 304) as it is used as the main material in primary cooling circuit of PWR Despite the simulated pipes being a part of primary cooling system the thermal isolation was not modelled in the analysis The source for Monte Carlo calculations was a uniform and isotropic emission of decay particles in the whole water volume in pipes

The parametric analysis of dose rates included pipes with wall thickness from 0.1 to 8.1 cm and water radius from 0.5 to 60.5 cm This limits were chosen to include all possible pipe sizes in the primary cooling loops of nuclear facilities On the graphs of results the dimen-sions for pipes according to ANSI B36.19 schedules 160 for sizes 1.27 cm (1/2 inch) through 30.5 cm (12 inches) and according to ASME Boiler and Pressure Vessel Code, Section III, Class 1 components for larger pipes are given as red dots

Neutrons emitted in decay of isotope17N can activate components around primary cooling system Two types of gamma rays are produced

at activation: prompt and delayed In this paper only the study of prompt gamma dose rates due17N decay was performed as the acti-vation analysis of the pipes was not performed

4.1.1 Biological parametric analysis

Results of dose rates were normalized to one source particle to study the diffusion of prompt gamma rays due to decay of17N and results are presented inFig 14 They show that there is a region where the dose rates are at highest due to greater absorption of neutrons compared to thinner pipes and lower absorption of prompt gamma rays compared to

Table 3

Comparison of specific activity of activated isotopes of cooling water in fission

and fusion reactors normalized to 1 MW thermal power Only the Monte Carlo

statistical uncertainties are presented in table

Activated isotopes Fusion reactor specific

activity [Bq/m 3 /MW] Fission reactor specificactivity [Bq/m 3 /MW]

16 N 5.92 10 13(1 ± 0.0069) 9.60 10 8(1 ± 0.0069)

Fig 13 Flux to dose conversion factor using different standards, for both

neutrons and rays in terms of Sv/h and silicon equivalent Gy/h per particle flux

Fig 14 Parametric results of biological gamma dose rates due to decay of17N normalized to one source particle The red dots present the pipe parameters according to ANSI B36.19 schedules 160 and ASME BPVC section III

thicker pipes The maximum is a around 10 cm of water radius and 2 cm

of wall thickness From the results it is also visible that pipes according

to ANSI B36.19 standard 160 are in the maximum of dose rates For

other isotopes of activated cooling water the highest doses are in region

with small water radius (few cm) and thin walls (few mm)

The results of parametric analysis (d) obtained by MCNP were

re-normalized such a way that they can be multiplied with specific activity

of water (a) and divided by volume flow rate of activated water to

obtain the dose rates in Sv/h at distance 50 cm from pipe surface:

=

D d a

Such approach allows easy estimation of dose rates around pipes by

using pre-calculated values in this paper Results for all activated

iso-topes are presented inFig 15 For easier readout of physical quantities

from the figures the data for certain pipe diameters are provided in

tabular format in appendix A Unlike results normalized per source

particle the highest dose rates are for pipes with big water radius (over

50 cm) and thin wall (few mm) For prompt gamma rays due to decay of

17N the region with highest dose rates is not present due to small water

volumes for normalization

If the specific activity of all three isotopes of activated water would

be the same value the highest contribution to biological dose rates

would be due to neutrons from 17N and the lowest contributor are

gamma rays from 19O decay However, as already presented in the

paper the specific activities of isotopes are different due to differences

in activation The majority contribution to the total dose rate is thus

from16N decay while the dose rates from17N and19O contribute the

same order of magnitude

4.1.2 Silicon parametric analysis

Dose rates for electronic components were calculated in 1 cm thick silicon dummy model Neutron dose rates were calculated using ASTM standard For gamma rays the dose rates were obtained using tally multiplier which consisted of number density of silicon, total cross-sections for gamma interaction in silicon and gamma heating number of silicon As in the case for biological dose rates, the results were nor-malized to one source particle to study diffusion of prompt gamma rays due to decay of17N and are presented inFig 16 The highest dose rates are around 10 cm of water radius and 2 cm of wall thickness due to absorption of neutrons and low absorption of prompt gamma rays in

Fig 15 Parametric results of biological dose rates normalized to water volume The red dots present the pipe parameters according to ANSI B36.19 schedules 160

and ASME BPVC section III

Fig 16 Parametric results of gamma dose rates in silicon due to decay of17N normalized to one source particle The red dots present the pipe parameters according to ANSI B36.19 schedules 160 and ASME BPVC section III

Fig 17 Parametric results of dose rates in silicon normalized to water volume The red dots present the pipe parameters according to ANSI B36.19 schedules 160 and

ASME BPVC section III

Fig 18 Comparison between the CAD model and the constructed MCNP model of the steam generator Figures of the steam generator are mirrored for better

comparison

pipe walls as in the case of biological dose rates.

The parametric results of dose rates for silicon were renormalized in

a way they can be multiplied with specific activity of water and divided

by volume flow rate of activated water to obtain dose rates in Gy/h at

distance 50 cm from pipe surface Results for all activated isotopes are

presented inFig 17 For easier readout of physical quantities from the

figures the data for certain pipe diameters are provided in tabular

format in appendix B The highest dose rates are for pipes with big

water radius and thin wall similar to biological dose rates However,

unlike biological dose rates, the lowest contribution to the total dose

rate at same specific activity for all activated water isotopes is due to

neutrons from 17N decay while the highest contribution is due to

gamma rays from16N decay

4.2 Dose field around heat exchanger

Important components in cooling loops in nuclear facilities are hear

exchangers The decay of activated cooling water presents one of the

main radiation sources around heat exchangers as they are normally

positioned away from primary radiation source Steam generator is the

heat exchanger in nuclear power plants and is located together with the

primary coolant pump in an area separated from reactor core For

ef-ficiency of heat exchange the water is majority of flowing time in the

heat exchanger In the case of steam generators in PWR the water is

more than 70% of circulation time in steam generators Thus the

ma-jority of the radioactive isotopes of water decay in the heat exchangers

As the primary cooling pumps can be located next to the heat

ex-changers the decay causes increasing doses to workers performing

emergency repairs on the pumps during operation

4.2.1 Steam generator model

A detailed geometrical computational model of the steam generator

and the radiation source was constructed in MCNP The computational

model of the steam generator was based on the steam generators in a

two loop PWR and the resulting geometrical model is presented in

Fig 18 Material used in the model are low alloy steel SA 508 Cl 3a,

stainless steel SS 304, Inconel 690 TT, borated water for the primary

side of the steam generator with 1400 ppm boron concentration and

pure water for secondary part of the steam generator The model of the

steam generator was surrounded by air and concrete structure similar to

structures in the power plant By the construction of the computational

model it was taken care that the mass of the model was preserved The

final mass of the model deviated from the real mass by 1.38%, i.e 4.7

tons

To construct the MCNP model of the steam generator the U-tubes

(over 5000 U-tubes) have been defined using universes and repeated

structures in hexagonal lattices MCNP has several limitations for

de-finition of particle source especially if the source description depends

on defined cells in universes and lattices To use the axial function along

tubes a cylinder needs to be defined within the source definition To

properly define a cylinder in the source term two parameters are

needed Due to this it is not possible to define an axial function for each

U-tube as there are too many U-tubes If a larger cylinder would be used

to encompass all U-tubes the number of cell for the primary water

in-side U-tubes needs to be given In such a case MCNP distributes points

inside such cylinder for source locations thus creating point source Due

to this limitations the particle source was modelled as a set of discs In

the area of U-tubes the centres of the discs were placed in the middle of

the tube with the radius of the tube In the height the source discs were

placed in layers with 50 cm distance between each layer At the bottom

of the steam generator the source was defined as layers of discs on a 2

cm×2 cm grid with 30 cm distance between layers The source location

in the steam generator is presented inFig 19 The probability for

se-lection of a disc as the source was defined using the exponential decay

of the radioactive isotopes of activated cooling water As each

radio-active isotope has its own decay time and decay products with different

energies the sources were made separately for each studied isotope

4.2.2 Dose field results

As the model of steam generator was taken from a 2 GW thermal power PWR the results of dose fields were normalized to activity of activated water in such PWR which were presented earlier in the paper

Fig 19 MCNP source (marked with red dots) used to obtain dose field due to

decay of activated cooling water

Fig 20 The gamma dose field distribution outside the steam generator due to

decay of isotope16N The arrow presents the direction of primary cooling water flow