Human performance in the nuclear industry

15-25 ©2015 Vietnam Atomic Energy Society and Vietnam Atomic Energy Institute Calculation of excore detector weighting functions for a sodium-cooled TRU burner mockup using MCNP5 Pham

Nuclear Science and Technology, Vol.5, No 2 (2015), pp 01-06

©2015 Vietnam Atomic Energy Society and Vietnam Atomic Energy Institute

Human Performance in the Nuclear Industry

Steven M Koncz

Human Performance Engineering Pty Ltd,

PO Box 5036, Falcon, Western Australia Email: skoncz@hpeng.com.au

(Received 10 September 2015, accepted 25 October 2015)

Abstract: Management of employees human performance in the Nuclear Industry is endemic to their

safety when working In the United Kingdom it has been a key focus since 2003 Employees were made aware through a detailed program of workshops, of the error prevention methods and how to apply them The use of effective incident barriers became embedded in the safety culture The methodology implemented was personal ownership, to enable self assessment of behaviors, attitudes and beliefs When put in place, there are many specific barriers, which can reduce the chances of an error occurring They come under the headings of organisational, procedural and physical barriers All

of these were used in some way and continue to be reinforced on a daily basis Specific barriers are applied in specific situations However, some general ones are also effective In common use are the Take 2 or Take 5 Minutes, point of work risk assessments Applying the human performance barrier Independent Verification (I.V.) would result in 'Take 3 and I.V.' This would independently double check the risk assessment New ways of thinking are required to continuously improve and evolve Results of the error reduction process included; reduced workload, increased plant reliability, efficiencies and productivity

Keywords: Error, Human, Performance, Work, Prevention, Nuclear, Barriers, Safety, Process,

Behaviour

I INTRODUCTION

This paper describes the history of

human performance error prevention, as used in

the nuclear industry How error prevention tools

are used and how we could improve on the

ways in which they are employed on a daily

basis In the 13 years of use at nuclear facilities,

it is suggested the error prevention tools have

the error prevention tools applied to themselves

and review their application to promote

continuous improvement

'Complacency' is recognised as one of the

error enablers Being comfortable with the way

in which the human performance error

prevention methodologies are used, is itself an

error precursor If we think we have got it right

and don't need to change or improve, then we

are not applying the tools correctly

The 'Norms' is another recognised error enabler, "it's always been done like that," is a reply when asked why a particular action lead

to an event of some kind If we fall into the same trap and don't review how we employ the prevention methodologies, we again are not applying the tools correctly

II HISTORY OF HUMAN PERFORMANCE

IN THE NUCLEAR INDUSTRY

The nuclear event on April 26th 1986

at the Chernobyl-4 plant in the then Soviet Union, led to changes in the approach to process safety in nuclear plants the world over The World Association of Nuclear Operators (WANO) was formed on 15th May

1989, under a banner of international cooperation Through open exchange of

HUMAN PERFORMANCE IN THE NUCLEAR INDUSTRY

2

operating experience, all members could then

work together to achieve the highest possible

standards of nuclear safety

The Institute of Nuclear Power

Operations (INPO), founded in December

1979, established a Special Review

Committee on Human Performance in late

1993 This committee, along with several

working groups, was asked to identify actions

to bring about continued improvement in

human performance within the commercial

nuclear power industry [1] It was this

document, which was adopted and reviewed

by WANO to form the basis, in 2002, for

improving human performance [2]

III HUMAN PERFORMANCE

IMPROVEMENT

There is now good evidence through

human performance improvement to

demonstrate the benefits to safety, production

and output

In the UK over a 2-year period, the

performance of key performance indicators

(KPI's) were ahead of WANO “Best in Class”

targets for 2004/05 This was attributed to the

business improvements at that time

Implementing and reinforcing the Human

Performance error prevention process had a

bearing on these results, Non-outage defects

backlog reduced by 55%, Accident frequency

rate reduced by 40%, Unplanned automatic trip

rate reduced by 30%, Work schedule adherence

was 28% better [3]

Human error contributes to around 80%

of nuclear events in the industry, the remaining

20% attributable to equipment / plant failures

This not only has a bearing on the performance

of the facilities themselves, but the overall

public perception of the nuclear industry Of the

identified human errors, 30% of the mistakes

were down to the individuals and 70% due to the organisations failing to prevent the errors This is shown in Fig 1 [4]

Fig 1 Contribution of human error to the

“People know the right thing to do for any situation in three ways.” First, instinct triggers automatic responses This is a fixed reaction ’hard wired’ in the human mind that elicits a special response, such as the dilation of the eyes as one walks into bright sunlight No learning is required Second, a suitable response

is determined by learning either by education,

by trial and error, or from others' experiences Examples include reading a book on finances, learning to ride a bicycle, reading operating experience reports, or learning the expectations

of a new employer or work group Finally, thinking is a process of building idea upon idea

to make sense of a situation Thinking gathers data to generate cues that may help a person recognize a familiar pattern about what to do Thinking generates new ideas coupled with new knowledge leads to better understanding [5] The skills, knowledge and attitudes of individuals take time to change It is for this

STEVEN M KONCZ

3

reason that effective barriers must be put in

place Managers implement and strengthen

defenses, they reinforce error-prevention

techniques and maintain the standards and

expectations for staff

All WANO member nuclear plants must

aspire to the following human performance

objective;

"The behaviors of all personnel result in

safe and reliable station operation Behaviors

that contribute to excellence in human

performance are reinforced to continuously

strive for event-free station operations" [2]

The criteria contained within this

performance objective are assessed during peer

reviews and its effectiveness reported There

are two Nuclear Plant Event (NPE) definitions

associated with human performance

- NPE08, “Human error which degraded

nuclear safety related systems”

- NPE09, “Human error which could have

degraded nuclear safety related systems”

If you look at the timeframe of when human

performance error prevention was introduced and

concentrate on the years 1992 to 2006, it is

interesting to see the reduction in events at U.S

nuclear plants This is shown in Fig 2 [6]

Fig 2 Significant Events at U.S Nuclear Plants:

Annual Industry Average, Fiscal Year 1992-2006 [6] Significant Events are events that meet specific NRC criteria, including degradation of safety equipment, a reactor scram with complications, an unexpected response to a transient, or degradation of a fuel or pressure boundary Significant events are identified by NRC staff through detailed screening and evaluation of operating experience

V ERROR PREVENTION TECHNIQUES

& BARRIERS

In order to understand which error prevention techniques are most applicable, one must first understand what enablers can contribute to errors

12 main error enablers were identified and focused on as shown in Table I [7]

Table I The Error Enablers

Time Pressure Distractions/Interruptions Fatigue/High

workload

Inexperience/Lack of knowledge

Complacency Poor communication Stress Lack of assertiveness Resource planning Lack of Teamwork Lack of awareness Norms

Plant trip risk procedures were assessed and each error enabler considered for the current task Suitable barriers were then applied and reviewed in action

Barriers

There are many barriers to prevent things from going wrong, they can be Organisational, Procedural and Physical The most important aspect is all barriers set by management are reinforced at every opportunity It would be their expectation for

HUMAN PERFORMANCE IN THE NUCLEAR INDUSTRY

4

staff to adhere to procedural usage,

encouraged to have a questioning attitude and

to stop when they are unsure

Organisational

The organisational barriers are the ones

embedded within the company’s systems This

makes it less likely that a plant modification

occurs without drawing changes being in place

coupled with operational and maintenance

procedures There are many interconnected

systems that will not allow the next step to take

place until it is satisfied all the key elements to

a successful outcome are met This cascades

down to the competency levels of the person

writing the work order instruction

The organisational barriers can contain

latent errors These are hidden deficiencies in

the process or values that provoke an error or

cause the defense to break down The

organisation also influences the culture at its

locations through the reinforcement of its

standards and expectations People are

encouraged to work in a blame free culture but

not to the extent where they are unaccountable

for their actions One of the main organisational

barriers which sets the benchmark for all

expectations is training Shortfalls in training or

a lack of training reduces the effectiveness of

the understanding of what is required

Procedural

There are many procedural barriers in

common use across industry They hold the

individual responsible for their use The

following typical work task and barriers used

will highlight possible areas for concern

A work task can be broken down into 3

areas; Pre-work, Work and Post work

Pre-work – The barrier used at this point

is the Pre-Job Briefing Pre-work discussions

are carried out when there is potential to impact

on safety Everyone associated with the work is

involved The roles and responsibilities are

defined The critical tasks and each step identified The work instructions and procedures are verified and common understanding checked This barrier use may be mandatory depending on the task

Using prior knowledge, operational or maintenance can be utilised at this point It demonstrates we are prepared to learn from past experience and use it effectively Prior knowledge can be in database format or personal experience Whatever method is used,

it should capture previous incidents and near misses

Stop, Think, Act, Review (S.T.A.R.) or Take 2 / 5 minutes to assess the work area are part of the self checking barrier This can be formalised by filling in a check sheet to demonstrate its use Confirmed communications is essential use at this point, to ensure the correct plant item is worked on

It is evident the individual plays a major part in effectively utilising the barriers If they have not taken personal ownership of the process and endeavor to use it, there is scope for errors occurring When people work around these barriers there is scope for error

Work – The barriers used at this point can contain mandatory actions, depending on the work instruction Mandatory actions typically occur during the verification practices such as Peer Checking, Independent Verification or Concurrent / Simultaneous Verification Confirmed communications is also crucial during the work to exchange the right information at the right time Place keeping is another specific barrier employed during critical tasks to ensure the correct action is made at the right step Task Observations are carried when work is taking place This is an opportunity to carry out a formal or informal review of the complete scope of works It is a business improvement tool, used to capture the safety culture

STEVEN M KONCZ

5

surrounding the task A formal study of the

work process also checks the standards &

expectations are being met

Post work – This is an area where a

Post-Job Review takes place to determine if there are

any areas for improvement or worthy of note

for the next time Using this barrier enhances

the operating / maintenance experience data

gathering and can lead to further training,

where appropriate It is also a documented

opportunity to facilitate continuous

improvement processes

Physical

Physical barriers are the ones which

prevent entry to areas that require specific access

permissions The permit for work system is the

procedural aspect that controls this type of

barrier Boundary enclosures and containment

buildings fall into this category also

All of the barriers discussed were utilised

in specific ways in the British Energy, Human

Performance Awareness Workshops Similar

barriers are used in WANO member nuclear

power facilities, they are shown in Table II [7]

Table II Error Prevention Tools

Pre-Job Briefing Use of Operating

Experience Procedural Use and

Adherence

Self checking (S.T.A.R.) Questioning Attitude

(Stop When Unsure)

Peer-Checking

Independent

Verification

Clear Communication Techniques

Post-Job Brief Task Observation

VI ERROR PREVENTION THE NEXT

STEP

It is well recognised that human

performance error prevention hinges on the

behaviour of individuals It is this behaviour which drives them to implement the error prevention tools or choose not to utilise them Self ownership of the processes and methodologies employed to prevent error are essential Observing these behaviors can take place at the point of work or checked remotely through documented evidence of the barrier being used

If we look at the point of work risk assessment Take 2, which encourages the person

to take two minutes and review the potentials for error, the documented evidence can take the form of a tick sheet This barrier is open to any one of the error precursors stopping it from taking place, such as time pressure, complacency

or high workload If no one double checks it took place, it could lead to an event Adding in

an error prevention tool such as Independent Verification (I.V.), would make this process more robust It would only lengthen the risk assessment time slightly and possibly take three minutes with independent verification taking place or Take 3 and I.V Although this could depend upon the working party numbers, it could

be planned into the work pack This is an example of behaviour being observed and an additional barrier put in place

Since people choose their behaviour at any given time, it is perhaps worth using the questioning attitude barrier but applying it to oneself prior to engagement with the task A prompt to make the person think how their behaviour will affect the task A very simple example is will I rush this job if I start it 30 minutes from meal time or end of shift? If a behaviour check is covered before a critical task, it may lead to the understanding that they could be distracted due to a personal issue playing on their mind Carrying out a formal self behaviour check is another way to enhance the error prevention process

In the age of personal data devices and WiFi interconnectivity, there is now scope for

HUMAN PERFORMANCE IN THE NUCLEAR INDUSTRY

6

central databases with operating experience and

error prevention tools to be available at the

point of work, hazardous areas obviously

excluded

VII CONCLUSION

Management commitment to focus on

human performance, in particular error

prevention and effective incident barriers, were

the catalysts to improvements in this area

Through external peer reviews and

benchmarking current best practices, the UK

nuclear industry took a collaborative approach

to bring their power stations up to the expected

standards They continue to maintain those

standards and strive to exceed expectations

There are select businesses which invest

directly in their staff by focusing on their innate

human ability to make mistakes and how to

take steps to prevent them from occurring

Within a rational, unified, goal-seeking

organisation, business improvement must have

an understanding of human performance It is

this understanding that can lead to improved

business operations Trending of human

performance errors should form part of the key

performance indicators (KPI’s) This data can

be derived from a robust route cause analysis

process, which is performed by suitable

qualified experienced persons

Refreshing and repackaging the use of

the error prevention tools, is essential for the

success of the process and also facilitates

continuous improvement Readdressing how the

barriers are used in particular situations can

contribute to the As Low As Reasonably

Practicable (ALARP), process

A formal behaviour self check, will make

people think of additional barriers to use

dependent upon how they may feel on the day

Only they truly know what is going on in their

own mind

To avoid complacency with the known error prevention tools in use, revisiting all methodologies used and looking for ways to improve are advised Reviewing when things go right as well as wrong should also be trended to capture good practices for replication

ACKNOWLEGEMENTS

The author would like to thank Ms Liesa Platten, of Synergy, Perth and Mr Joe Wade Human Performance Engineering Pty Ltd, Mandurah for independent verification of the readability of this document

REFERENCES

[1] Institute of Nuclear Power Operations, Excellence in Human Performance, INPO, Atlanta, 1997

[2] World Association of Nuclear Operators, Principles for Excellence in Human Performance, WANO-GL 2002-02, 2002 [3] The ARUP Journal 1/2006 Table 2, pg 15,

2006

[4] International Atomic Energy Agency, Managing Human Performance to Improve Nuclear Facility Operation, No NG-T-2.7, pg 1,

Nuclear Science and Technology, Vol.5, No 2 (2015), pp.07-14

©2015 Vietnam Atomic Energy Society and Vietnam Atomic Energy Institute

Codes for NPP severe accident simulation:

development, validation and applications

ARKADIY E KISELEV

Nuclear Safety Institute of the Russian Academy of Sciences, B Tulskaya 52, 115191 Moscow, Russia

E-mail: ksv@ibrae.ac.ru

(Received 05 October 2015, accepted 25 October 2015)

Abstract: The software tools that describe various safety aspects of NPP with VVER reactor have been

developed at the Nuclear Safety Institute of the Russian Academy of Sciences (IBRAE RAN) Functionally, the codes can be divided into two groups: the calculation codes that describe separate elements of NPP equipment and/or a group of processes and integrated software systems that allow solving the tasks of the NPP safety assessment in coupled formulation In particular, IBRAE RAN in cooperation with the nuclear industry organizations has developed the integrated software package SOCRAT designed to analyze the behavior of NPP with VVER at various stages of beyond-design- basis accidents, including the stages of reactor core degradation and long-term melt retention in a core catcher The general information about development, validation and applications of SOCRAT code is presented and discussed in the paper

Keywords: nuclear power plant, safety, calculation codes, severe accident

A long-term IBRAE RAN experience in

developing software has allowed formulating

the methodological approach that includes the

following basic directions:

- Working out the models based on the

equations of mathematical physics and modern

knowledge of processes and the phenomena

that occur at various operating modes of reactor

installations;

- Based on these models, development

and validation of the deterministic computer

codes for nuclear power plants safety

assessment;

- Calculation-based and theoretical works

to support the experimental programs;

- Application of the developed software

complexes for safety analysis of nuclear power

plants

Using the physical approaches to develop

models allows considerable improving of the

process and phenomenon modelling quality and

reducing uncertainty of calculation results The software efficiency is tested through validation against experimental data While doing this, the assessment of the existing knowledge base on physical processes and phenomena is being conducted that allows formulating the tasks for experimental studies more accurately Participation of IBRAE RAS in the integral experiments is of a special significance since it allows verifying the new physical and numerical models in self-consistent way While developing models and codes for severe accident analysis, the basic uncertainties of the used physical models have been revealed; estimations of applicability of the existing codes to the safety analysis of NPP with various reactors have been made

Most of IBRAE RAS codes are developed within the frameworks of joint projects with the Russian and international nuclear stakeholders

CODES FOR NPP SEVERE ACCIDENT SIMULATION…

8

Table I Codes developed at IBRAE RIAN in cooperation with Russian and foreign institutions

Early stage of reactor core degradation SVECHA, QUENCH,

MFPR

NRC/IPSN/EC FZK/RIAR Late stage of reactor core degradation CONV2D&3D LOHEY OECD/RRC KI/ IPSN Interaction of melt with concrete and

Thus, in the late 1990s, the works on

development of the Russian code for safety

analysis of new designs of NPP with VVER in

conditions of severe accidents were started

upon the initiative of JSC "SPbAEP" in

cooperation of expert teams from IBRAE RAN,

Russian Federal Nuclear Center “All-Russian

Research Institute of Experimental Physics”

(FSUE RFNC VNIIEF) and National Research

Center «Kurchatov Institute» (NRC KI) Later,

this code, which received the name SOCRAT,

started to be applied also for safety assessment

of the VVER projects operated or constructed

in Russia In 2010, the basic version of the code

SOCRAT was certified by the Russian

regulator (Rostehnadzor) Since 2011, the work

has been conducted on developing and

validation of the advanced version of the code

that allows assessments of the radiological

consequences of severe accidents The quality

of the models and validation allow considering

the SOCRAT as a best-estimate code

Integration of numerous physical models into

one code provides end-to-end modelling of all

essential stages of severe accidents and

obtaining of the entire picture of the accident

evolution from a moment of its occurrence

(initiating event) up to release of radioactive

fission products out of the NPP containment

into the environment

Thermohydraulic models of the integrated code SOCRAT describe the behavior

of the two-phase coolant with non-condensable gases in the core, primary and secondary circuits of a reactor installation at all stages of severe accident including stage of total core un-cover and stage of in-vessel melt retention They include the various modes of coolant flow, interphase interactions, various modes of heat exchange with walls of hydraulic channels, friction at channel walls, presence of non-condensable gases, coolant ejection under containment Also, the models of the SOCRAT code allow describing the operation of pumps, valves, hydraulic reservoirs and other elements

of reactor installation equipment The set of the basic elements used to model the input deck of the primary and secondary circuits, allows describing the tracing of any hydraulic loops with the accuracy that is sufficient for modern calculations of severe accidents

Thermohydraulic processes in a system

of communicating containment rooms in are modelled self-consistently using the integrated

in SOCRAT containment codes KUPOL-M and ANGAR, representing the certified codes with lumped parameters

Physical mutually-consistent models describing the processes of fuel cladding

ARKADIY E KISELEV

9

oxidation by steam, thermomechanical behavior

of fuel rods and absorbers, melting of reactor

core and other in-vessel materials, melt

relocation are used for numerical analysis of

severe accidents at a stage of reactor core

degradation While doing this, the real material

composition of the reactor core is being taken

into account

Code SOCRAT allows modelling the

processes of melt interaction with water at a

stage of melt retention in the lower plenum,

formation and distribution of a corium liquid

phase, stratification of metal and oxide

components, reactor pressure vessel

degradation and melt release into containment

The basic NPP objects that are modelled

by the code SOCRAT in the advanced version are presented in Fig 1 They are as follows:

- Fuel;

- Fuel assemblies;

- Reactor core and in-vessel structures;

- Reactor coolant system including safety systems;

- Steam generator and main steam line;

- Containment

Fig 1 Phenomena modeled in SOCRAT

CODES FOR NPP SEVERE ACCIDENT SIMULATION…

10

Fig 2 Main processes simulated in the layer of homogeneous melt

The basic physical models of the

integrated code SOCRAT at in-vessel stage of

accident are presented below:

The advanced version of the integrated

code SOCRAT allows carrying out calculations

of parameters required for assessing the

radiological consequences of severe beyond

design-basis accidents at NPP with VVER

reactors and, in addition to the basic version,

describes in details the following processes:

- Buid-up of radioactive fission products

(FP) in fuel and their release into the fuel rod’s

gas gap;

- Transport and sedimentation of

radioactive fission products in various physical

and chemical forms in the reactor primary

circuit and in the containment;

- Release of radioactive fission products

into environment

Permanent validation of the SOCRAT

code as well as of its physical models is one of

the most important stages of the development

and application Models and algorithms of the

SOCRAT code have passed all-round

assessment against large data set, received in

separate effect tests and integral experiments

performed in Russia and abroad

The experimental programs that were used for the code validation are as follows: CORA, QUENCH (Germany), PHEBUS (France), RASPLAV, MASCA (Russia - OECD), ISTC/PARAMETER, ERCOSAM-SAMARA (joint Rosatom-Euroatom project), LOFT, PBF, international standard problem ICSP MASLWR, international benchmark BSAF (analysis of the accident at the Fukushima Daiichi NPP)

Fig 3 shows the calculated and measured temperatures of the surface of the fuel assembly simulator in the PARAMETER/SF1 experiment The PARAMETER program investigates phenomena associated with re-flooding of a degrading VVER like core under postulated severe accident conditions, in an early phase when the geometry is still mainly intact The figure confirms that the SOCRAT code correctly reflects the dynamics of the fuel assembly temperature behavior at all stages of the experiment (heating up, oxidation, and overheated core re-flood) under conditions of the presence of chemical power sources and convection and radiation heat exchange This results from a sufficiently large set of models of SOCRAT code and their validation in a wide range of initial data

ARKADIY E KISELEV

11

Fig 3 Modeling of fuel assembly temperature behavior in the PARAMETER/SF1 experiment /2/

Other example of SOCRAT validation is

participation in cooperation with JSC «OKB

“GIDROPRESS”» at all stages of the

international standard problem (ISP)

«Evaluation of Advanced Thermohydraulic

System Codes for Design and Safety Analysis

of Integral Type Reactors» In this exercise, an

accident with a feed-water loss in the secondary

circuit (test SP2) and a maneuvering mode of

the reactor operation (test SP3) were

investigated in a series of two integrated

experiments on scale model of perspective

reactor MASLWR with passive safety systems

Comparison of the calculated and measured

pressures of the primary circuit and

temperatures in containment for the SP2 test is

demonstrated in Fig 4а The close agreement

between the experimental and calculated data testifies the correct and consistent work of models of coolant flow and heat exchange in the presence of non-condensable gases that is of special importance for a reactor installation with passive safety systems Fig 4b presents the coolant temperature at the entrance to reactor core and its flow rate for the SP3 test Modelling of this test resulted in a good agreement with the results of measurements of not only temperature, but also flow rate parameters of two-phase coolant of the primary circuit in the natural circulation mode As a whole, the SOCRAT code is capable to simulate the thermohydraulic behavior of a reactor installation even prior to the beginning

of essential reactor core degradation

Fig 4 Modelling of the primary circuit pressure and containment temperature in the SP-2 test (a) and the

primary circuit coolant temperature and its flow rate in the SP-3 test (b)

Time, s 0

400 800 1200 1600 2000 2400 2800

CODES FOR NPP SEVERE ACCIDENT SIMULATION…

12

The code went through the practical test

and approbation in 2011, when a severe

accident stroke the Japanese NPP

“Fukushima-Daiichi” on March 11 The express analysis

was conducted at IBRAE RAN using the

integral SOCRAT code /3/ The possible

consequences of the accident, the forecast and characteristic times of emergency process development in reactor cores and SNF pools for the power units 1-4 were estimated /Fig.5/

2 (peak of pressure in containment after water ingression

in the core)

15.03 05:45 15.03 06:14

Fig 5 Estimated amount of hydrogen generated at Unit 3 by the time of explosion

The following sequence of processes was

analyzed:

- Decrease of the coolant level in reactor

core;

- Increase of containment pressure;

- Temperature increase and hydrogen

generation;

- Release of hydrogen and fission

products;

- Further degradation of reactor core

Fig 5 shows the calculated and real moments of hydrogen explosion at different units Comparison of the calculated and measured data for the mass level of coolant in the reactor core of the Unit 2 and for the pressure in the primary circuit for the Unit 3 is shown in Fig 6 The figure demonstrates that the SOCRAT code qualitatively describes the processes of heating, degradation and reflooding of reactor core The calculations were based on the assumptions that the power

of decay heat release corresponded to the typical BWR project, and data on water

ARKADIY E KISELEV

13

injection in the reactor coolant system and on

safety system operation corresponded to

TEPCO evidences that were available at the time of the accident

Fig 6 Comparison of the operational characteristics of the BWR-4 reactor installation measured during the

accident at the NPP Fukushima Daiichi with those calculated using the SOCRAT code: (a) Change of a water level in the reactor core of 3rd power unit; (b) Change of pressure in the primary circuit of 2nd power unit IBRAE RAN has continued this work by

joining the OECD-NEA Benchmark Study of

the Accident at the Fukushima Daiichi Nuclear

Power Station (BSAF) Project conducted by

Tokyo Engineering Power Company (TEPCO)

and the Nuclear Energy Agency of the

Organization for Economic Co-operation and

Development (NEA/OECD)

Today SOCRAT code is widely used

by the leading Russian design and scientific

organizations for analysis of beyond

design-basis severe accidents at NPP with reactors

on thermal neutrons with water coolant, for

assessment of hydrogen safety, efficiency of

melt retention systems, and for analysis of

efficiency of NPP passive safety systems

The typical thermohydraulic model of the

primary circuit of VVER-1000/В-320 reactor is

presented in Fig 7 It allows a quite detailed

modeling of beyond design-basis accidents with

loss of coolant in the primary circuit in a wide

range of locations and diameters of leaks

The full list of the SOCRAT code

applications is quite wide It can be noted that it

is used for the following units with VVER reactor: VVER-440/230 (Kola NPP), VVER-1000/В320 (the Balakovo NPP), VVER-1000/В428 (China), VVER-1000/В412 (India), VVER-1500/В448, VVER-1200/В392м (NVNPP-2), VVER-1200/В491 (LNPP-2) Presently, the code is used at SPbAEP, AEP, OKB GP, NRC KI, IPPE, and is transferred to MPEI as a tool of training of students and post-graduate students

In 2012, IBRAE RAN experts prepared and conducted a course of lectures for the Vietnamese specialists that were trained at the Central Institute for Advanced Training (TsIPK) Obninsk, Russia: Training course:

“Application of computer codes for safety analysis of NPPs Deterministic Safety Analysis and code SOCRAT” This course included 2 weeks of 96 hrs training Of them, the lectures took about 55 hrs, practical work - 41 hrs, and one day was devoted to testing

CODES FOR NPP SEVERE ACCIDENT SIMULATION…

14

Fig 7 Typical nodalization scheme of VVER-1000 reactor installation with passive safety systems used in the

SOCRAT code The further development of the SOCRAT

code includes the following:

1 Improvement of the current version of

the integrated code SOCRAT, participation in

international benchmarks in order to verify the

code, adaptation of physical models and

computing algorithms for various designs of

reactors with thermal neutrons and water

coolant, preparation and training of new users

2 Development of the new version of the

integrated code SOCRAT-BN for modelling of

physical processes in reactors with fast neutrons

and sodium coolant, that is being done based

upon practical experience received by the

cladding alloys // PROGRESS IN NUCLEAR ENERGY Volume: 52 Issue: 1 Pages: 19-36

Published: JAN 2010

[3] Dolganov K.S., Kapustin A V., Kisselev A E., Tomashchik D Yu., Tsaun S V., Yudina T A., Real-Time Calculation of the Accident at the Fukushima-1 NPP (Japan) Using the Sokrat

Code//ATOMIC ENERGY Volume: 114 Issue: 3

Pages: 161-168 Published: JUL 2013

Nuclear Science and Technology, Vol.5, No 2 (2015), pp 15-25

©2015 Vietnam Atomic Energy Society and Vietnam Atomic Energy Institute

Calculation of excore detector weighting functions for a

sodium-cooled TRU burner mockup using MCNP5

Pham Nhu Viet Ha*, Min Jae Lee, Sunghwan Yun, and Sang Ji Kim

Korea Atomic Energy Research Institute

1045 Daedeok-daero, Yuseong-ku, Daejeon, 305-353, Korea

*Email: phamha@kaeri.re.kr

(Received 23 September 2015 , accepted 23 October 2015)

Abstract: Power regulation systems of fast reactors are based on the signals of excore detectors The

excore detector weighting functions, which establish correspondence between the core power distribution and detector signal, are very useful for detector response analyses, e.g., in rod drop experiments This paper presents the calculation of the weighting functions for a TRU burner mockup

of the Korean Prototype Generation-IV Sodium-cooled Fast Reactor (named BFS-76-1A) using the MCNP5 multi-group adjoint capability For generation of the weighting functions, all fuel assemblies were considered and each of them was divided into ten horizontal layers Then the weighting functions for individual fuel assembly horizontal layers, the assembly weighting functions, and the shape annealing functions at RCP (Reactor Critical Point) and at conditions under which a control rod group was fully inserted into the core while other control rods at RCP were determined and evaluated The results indicate that the weighting functions can be considered relatively insensitive to the control rods position during the rod drop experiments and therefore those weighting values at RCP can be applied to the dynamic rod worth simulation for the BFS-76-1A

Keywords: SFR, TRU burner, BFS-76-1A, excore detector, weighting function, MCNP5

I INTRODUCTION

Sodium-cooled Fast Reactor (SFR) has

been widely recognized as one of the most

promising and attractive energy sources for

future generations since it can help efficiently

utilize the uranium resources and drastically

minimize the burden of nuclear waste from

nuclear power plants by closing the fuel cycle

In response to this recognition, the Korea

Atomic Energy Research Institute (KAERI)

had elaborated an advanced SFR concept for

transuranics (TRU) burning in the conceptual

design phase (2007-2011) of the long-term

advanced SFR R&D plan towards the

construction of an advanced SFR

demonstration plant by 2028 [1][2] Recently,

KAERI has been collaborating with the US

Department of Energy’s Argonne National

Laboratory to develop the 150 MWe Prototype

Generation-IV Sodium-cooled Fast Reactor (PGSFR) for testing and demonstrating the performance of TRU bearing metal fuel for commercial SFRs and the TRU transmutation capability of a burner reactor as a part of an advanced fuel cycle system [3][4]

For the demonstration of the metal fueled TRU burner core concept and securing

of the reactor physics database for design code validation, KAERI has been also collaborating with the Institute of Physics and Power Engineering (IPPE) in Russia for conducting reactor physics experiments [3][4] Correspondingly, four critical assemblies were constructed in the IPPE BFS-1 or BFS-2 facilities (called BFS-73-1, BFS-75-1, BFS-76-1A, and BFS-109-2A), representing either the metal uranium fuel (U-10Zr) loaded SFR concept developed in Korea in the late 1990’s

CALCULATION OF EXCORE DETECTOR WEIGHTING FUNCTIONS…

[1] or the current PGSFR design [3][4]

Especially, the BFS-76-1A, which stands for

the current PGSFR core, is a mockup of 300

MWe class TRU burner design without a

blanket, simultaneously loaded with uranium

and U-Pu metal fuels, and characterized by a

low conversion ratio, a high burnup reactivity

swing, and the consequent deep insertion of the

primary control rods at the beginning of the

equilibrium cycle Reactor physics experiments

in the BFS-76-1A were aimed to obtain

measured data on critical mass, spectral indices,

fission rate distribution, sodium void and axial

expansion effects, and control rod mockup

worth In particular, the information on control

rod mockup worth is very important and

requires careful evaluation because of its safety

implications

For that reason, a dynamic rod worth

simulation method applicable to SFRs needs to

be developed and then applied to the

BFS-76-1A for validating the measured control rod

mockup worths To simulate the pseudo excore

detector signals needed for inferring the

dynamic worth of control rods during the rod

drop experiments, the excore detector spatial

weighting functions which represent individual

contributions from specific core locations, i.e.,

fuel assemblies, fuel rods or portions of rods,

to the detector signal are required in advance

[5-8] It should be noted that the power

regulation system of a fast reactor is based on

the signals of excore neutron detectors The

detector signal contribution from each fuel

assembly depends not only on the power of the

fuel assembly but also on its position in the

core The excore detector spatial weighting

functions establish correspondence between the

spatial core power distribution and the signal

of excore detectors

In this paper, the excore detector spatial

weighting functions for the BFS-76-1A were

calculated and evaluated for further use in the dynamic rod worth simulation For generation

of the spatial weighting functions, all fuel assemblies were considered and each of them was divided into ten horizontal layers Then the spatial weighting functions for individual fuel assembly horizontal layers at RCP (Reactor Critical Point) and at the condition under which one control rod group was fully inserted into the core while other control rods at RCP were determined using the MCNP5 150-group adjoint calculations and inter-compared The results show that the spatial weighting functions were relatively insensitive to the control rods position during the rod drop experiments and therefore those weighting values at RCP can be applied in the dynamic rod worth simulation for the BFS-76-1A The calculation methodology is presented in Section II The results are provided and discussed in Section III Finally, concluding remarks are drawn in Section IV

II CALCULATION METHODOLOGY

The BFS-76-1A mockup consists of 326 LEZ-Pu assemblies, 488 LEZ-U assemblies,

322 HEZ-Pu assemblies, 648 HEZ-U assemblies, and the outer layers of relector, B4C shield, and radial shield assemblies as shown in Fig 1, where two excore neutron detectors were located outside the radial shield and symmetrically in the radial direction for this study (In Fig 1: 101= LEZ-Pu; 201= LEZ-U; 301= HEZ-Pu; 401= HEZ-U; 501, 601= primary, secondary control rods; 701= reflector; 801= radial shield; 901= B4C shield; 10= void; LEZ and HEZ= Low and High Enrichment Zones) In the vertical direction, each detector is located ~10 cm above the bottom of the active core The detectors are the

PHAM NHU VIET HA, MIN JAE LEE, SUNGHWAN YUN, SANG JI KIM

17

BF3 proportional counters They are cylinders

of BF3 with a radius of 2.5 cm and a height of

40 cm The cylinders are covered by a

polyethylene moderator layer with a thickness

of 5.0 cm to enhance the detector sensitivity

The excore detector response at arbitrary

time t is defined by [6]

where is the core power at position r

and time t; the spatial weighting function

at postion r; V the total core volume; it should

be noted that the unit of is arbitrary

In practice, the spatial weighting

functions for the excore detectors can be

generated using either the point kernel method

[5], the discrete ordinate transport method [6],

or the Monte Carlo method [7][8] It is noted

that an advantage of the Monte Carlo method is

the capability of modeling reactor

configurations with arbitrary geometrical

complexity With the Monte Carlo method, one

can also choose either the forward method or

the adjoint method The Monte Carlo forward

method allows the calculation of the weighting

function value of a given point in the reactor

and therefore gives more detailed results than

the adjoint method Additionally, the forward

method makes it possible to avoid the

approximations which stem from the

homogenization of the cross sections of the

assembly material and from the use of

group-wise data Nevertheless, since the calculation

of the weighting function is a fixed-source

neutron transport problem, the adjoint method

is much faster than the forward method

Especially, it will be very time-consuming to

generate the weighting functions using the

forward method if a large number of the

specific core locations are taken into account

Because of a much longer mean free path of neutrons in fast systems (~10 cm as compared to ~1 cm in PWRs), the neutrons from both the innermost fuel assemblies and the distant ones have higher possibility to leak out of the core and be “seen” by the excore detector Thus, all fuel assemblies of the BFS-76-1A (1784 assemblies) were taken into account for calculating their contributions to the detector response; whereas only the contributions from the peripheral fuel assemblies located close to the detector are considered significant for PWRs Therefore, the Monte Carlo adjoint method, which is much faster than the forward method as discussed above, will be applied in the calculation of the weighting functions for the BFS-76-1A using the well known MCNP5 Monte Carlo N-Particle Transport Code [9][10] Based on the adjoint method, the spatial weighting function is given by [6] ∫ (2) where is the spatial weighting factor at

position r i, the fission energy spectrum, and the adjoint flux at position r i and

neutron energy E

For the calculation of the weighting functions, each fuel assembly (FA) of the

BFS-76-1A (indexed by (i,j)) was divided

into 10 horizontal layers (each layer was

indexed by k, k = 1, 2, …, 10) Based on Eq

(2), the three-dimensional spatial weighting

functions of each FA layer (i,j,k) for each

detector at RCP (Reactor Critical Point- at which all secondary control rods were withdrawn out of the core and all primary control rods inserted into the core ~42% of the core height) and at the condition under which one control rod group (Group 1, 2, or

CALCULATION OF EXCORE DETECTOR WEIGHTING FUNCTIONS…

3; see Fig 1) was fully inserted into the core

while other control rods at RCP (hereafter

called the case G1IN, G2IN, or G3IN

respectively) were generated (using the

MCNP5 150-group adjoint calculations) and

normalized over the whole core by

∫

∑ ∫

∑

∑ ∑

(3) where is the fission spectrum at energy group g and the adjoint flux at the FA layer (i,j,k) at energy group g Thereafter, these weighting functions were averaged over the two symmetric detectors to relieve the effect of core radial position on the detector response In the MCNP5 150-group adjoint calculations, the neutron microscopic cross-sections for 150 neutron energy groups from the ENDF/B-VII.0 library were used To simulate the rod drop experiments, it is expected that a set of the spatial weighting functions insensitive to the control rods position can be generated On that account, the Assembly Weighting Functions (AWFs) and Axial Weighting Functions (also called the Shape Annealing Functions or SAFs) at RCP and at G1IN, G2IN, or G3IN were determined and inter-compared so as to select an appropriate set of the spatial weighting functions for the dynamic rod worth simulation The reason for the evaluation of the AWFs and SAFs instead of the three-dimensional weighting functions generated using Eq (3) is explained as follows Because the three-dimensional spatial weighting functions were calculated using MCNP5 and a very large number of FA layers were considered herein (1784 x 10 = 17840 layers), it is not intuitive and extremely time-consuming to compare these weighting functions (17840 values for each set of weighting functions) at different control rod positions, such as at RCP and G1IN Instead, the AWFs and SAFs at RCP and at G1IN, G2IN, or G3IN, were determined and inter-compared The AWF for the FA (i,j) which represents the detector response contributions from individual FAs is calculated by Eq (4) ∑ (4)

The SAF for the core layer (k) which represents the relative importance of core axial position to the detector response is calculated by Eq (5) ∑ (5)

III RESULTS AND DISCUSSION

The AWFs for the excore detector at RCP were illustrated in Fig 2 The relative differences of AWFs at RCP and at G1IN, G2IN, or G3IN were provided in Figs 3-5 The SAFs at RCP and G1IN, G2IN, or G3IN were shown and compared in Figs 6-8 It is noted that all the spatial weighting functions were obtained, in this study, with a relative error (fractional standard deviation) of less than

~0.035 (3.5%), provided the number of histories to be run in the MCNP5 calculations

of a billion

Fig 2 signifies that the contributions from the internal fuel assemblies or distant ones must be taken into account for an accurate prediction of the detector response It can be seen that the weighting function decreased from the outermost fuel assemblies close to the detector towards the innermost fuel assemblies

or those located further from the detector; for instance, it was reduced about one order after

~10 layers of fuel assemblies

PHAM NHU VIET HA, MIN JAE LEE, SUNGHWAN YUN, SANG JI KIM

19

From Figs 3-5, it can be found that

the relative difference between AWFs at

RCP and at G1IN, G2IN, or G3IN was on

average less than ~2.5% for the outer fuel

assemblies or those close to the detector

whereas it could reach up to ~22/39/49% for

a few inner assemblies located near the

dropped control rods (G1IN/G2IN/G3IN,

respectively) However, such difference of

at most ~22/39/49% can be practically

neglected in the calculation of the detector

response because the detector response

contributions from these inner assemblies

near the dropped control rods were at least

about one order smaller than those from the

assemblies located near the excore detector

(see Fig 2)

Figs 6-8 show that the SAFs have a

bottom-peaked shape because the two

symmetric detectors were axially located just

~10 cm above the active core bottom (the length of excore detector is 40 cm whereas the active core height is ~82.144 cm) As is seen in those figures, the SAF at RCP slightly overestimates that at G1IN/G2IN/G3IN for the core axial position below RCP and vice versa for the core axial position above RCP Generally, the relative difference of SAFs at RCP and at G1IN, G2IN, or G3IN was within

at most 1.8% and can be neglected

Hence, it was practically considered that the spatial weighting functions are relatively insensitive to the control rods position during the rod drop experiments and those values at RCP can be applied in the dynamic rod worth

simulation for the BFS-76-1A

Fig 1 BFS-76-1A radial core layout

CALCULATION OF EXCORE DETECTOR WEIGHTING FUNCTIONS…

20

Fig 2 AWFs at RCP (up) and a partial zoom-in (down), x10-2

PHAM NHU VIET HA, MIN JAE LEE, SUNGHWAN YUN, SANG JI KIM

21

Fig 3 Relative difference of AWFs at RCP and G1IN (up) and a partial zoom-in (down), %

CALCULATION OF EXCORE DETECTOR WEIGHTING FUNCTIONS…

Fig 4 Relative difference of AWFs at RCP and G2IN (up) and a partial zoom-in (down), %

PHAM NHU VIET HA, MIN JAE LEE, SUNGHWAN YUN, SANG JI KIM

23

Fig 5 Relative difference of AWFs at RCP and G3IN (up) and a partial zoom-in (down), %

CALCULATION OF EXCORE DETECTOR WEIGHTING FUNCTIONS…

The excore detector spatial weighting

functions for the BFS-76-1A were generated

using the MCNP5 150-group adjoint

calculations and evaluated in this study For

generation of the weighting functions, all fuel

assemblies were taken into account and each of

them was divided into ten horizontal layers To choose an appropriate set of the spatial weighting functions for further use in the dynamic rod worth simulation for the BFS-76-1A, the assembly weighting functions and the shape annealing functions at RCP (Reactor Critical Point) and at the condition under which one control rod group was fully inserted into the core while other control rods at RCP were determined and inter-compared instead of extremely large numbers of the calculated three-dimensional weighting functions The results indicate that the weighting functions were relatively insensitive to the control rods position during the rod drop experiments and consequently those weighting values at RCP can be applied in the dynamic rod worth simulation and evaluation for the BFS-76-1A

In future work, a dynamic rod worth simulation study based on those spatial weighting functions will be performed for validating the measured rod worths of the BFS-76-1A Finally, this work provides a basis for generation and evaluation of the excore detector spatial weighting functions for a SFR and will be applied for further analysis of the detector response aimed at evaluating the worth of control rods for safety design of the PGSFR and at designing a robust neutron flux/power monitoring system for the PGSFR

ACKNOWLEGEMENTS

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIP) (No NRF-2012M2A8A2025622)

-0.6 -0.4 -0.2 0.0 0.2 0.4

-2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5

PHAM NHU VIET HA, MIN JAE LEE, SUNGHWAN YUN, SANG JI KIM

25

REFERENCES

[1] D H Hahn et al., “Advanced SFR Design

Concepts and R&D Activities,” Nuclear

Engineering and Technology, 41(4), 427-446,

2009

[2] Y I Kim et al., “Preliminary Conceptual

Design Report of Gen-IV SFR Demonstration

Plant,” KAERI/TR-4335/2011, Korea Atomic

Energy Research Institute, 2011

[3] J Chang, “Status of Fast Reactor Technology

Development in Korea,” The 45th IAEA

TWG-FR Meeting, Beijing, China, June 20-22,

2012

[4] H Joo, “Status of Fast Reactor Technology

Development in Korea,” The 46th IAEA

TWG-FR Meeting, Vienna, Austria, May

21-24, 2013

[5] M W Crump and J C Lee, "Calculation of

Spatial Weighting Functions for Ex-Core

Detectors," Nuclear Technology, 41, 1978,

87-96, 1978

[6] J G Ahn and N Z Cho, "Generation of Spatial Weighting Functions for Ex-core Detectors by Adjoint Transport Calculation,"

Nuclear Technology, 103, 114-121, 1993

[7] T Berki, "Calculation of Spatial Weighting Functions for Ex-core Detectors of VVER-440 Reactors by Monte Carlo Method," International Conference: Nuclear Energy for New Europe 2003, Portorož, Slovenia, September 8-11, 2003

[8] G Farkas et al., "Computation of Ex-core Detector Weighting Functions for VVER-440

Using MCNP5," Nuclear Engineering and Design, 261, 226-231, 2013

[9] X-5 Monte Carlo Team, "MCNP- A General N-Particle Transport Code, Ver 5 - Vol I: Overview and Theory," LA-UR-03-1987, Los Alamos National Laboratory, 2003

[10] J C Wagner et al., "MCNP: Multigroup/Adjoint Capabilities," LA-12704, Los Alamos National Laboratory, April 1994

Nuclear Science and Technology, Vol.5, No 2 (2015), pp 26-32

©2015 Vietnam Atomic Energy Society and Vietnam Atomic Energy Institute

The recovery of metallic cadmium from the cadmium containing

residue in a electrolytic zinc plant

Luong Manh Hung, Tran Ngoc Vuong

Institute for Technology of Radioactive and Rare Elements, Vietnam Atomic Energy Institute,

48 Lang-Ha str., Hanoi, Vietnam Email: luonghung2505@gmail.com

(Received 02 April 2015, accepted 23 September 2015)

Abstract: This report presents a process for recovery and purification of metallic cadmium from a

residue of the purification process for zinc sulphate solution in Thai Nguyen electrolytic zinc plant The cadmium containing residue was digested by sulfuric acid of 140 g/l at a temperature of 700C for 4h, the obtained solution will be purified for removal of some impurities such as iron, copper, etc The purified solution with concentration 50 g/l of Cd, 120 g/l of sulphuric acid and 0.1 g/l of gelatin as an additive will be subjected to an electrolysis process with current density of 50 A/m2 for recovery of metallic cadmium The temperature of electrolyte is lower 400C Overall recovery of cadmium is 90%, purity of the obtained metalic cadmium is up to 99.0%

Keywords: cadimium, electrowinning cadmium,…

I INTRODUCTION

Metallic cadmium has various

technological applications such as in nickel–

cadmium and silver–cadmium storage batteries,

functional alloys and coatings [1] Cadmium is

used for the synthesis of chalcogenide

compounds and the production of

semiconductor intermetallics and also in control

rods in nuclear power plants [2]

Cadmium does not form separate

deposits, but is an element associated with zinc

and complex ores Therefore, cadmium

production technologies are developing in step

with methods for production of zinc and lead

Almost all cadmium producing installations are

part of zinc and lead producing facilities [3]

The main starting materials for cadmium

are byproducts of zinc and lead metallurgical

processes (copper–cadmium cakes, dusts left

after lead blast smelting, etc.) Cadmium can be

extracted from these materials either by the

pyrometallurgical (fractional distillation) or

hydrometallurgical method or using a combination of these methods The most widespread technique is the hydrometallurgical method, which consists of the following operations: oxidation of cadmium; leaching; cleaning of the solution and precipitation of the cadmium sponge; oxidation of the sponge, its repeated dissolution and cleaning of the solution; electrowinning; smelting of cathodic cadmium [3, 4]

hydrometallurgy combined with electrolysis is commonly used for the recovery of cadmium

in the process of purifying zinc sulfate solution

in a electrolytic zinc plant [1,2,4] This method has advantages of simple equipment, low chemiacls consumption (using sulfuric acid as

a byproduct of zinc metal production) and high purity of metallic cadmium product (up

to 99%)

The process of producing cadmium by hydrometallurgy method combined with electrolysis can be discribed as follows:

LUONG MANH HUNG, TRAN NGOC VUONG

27

Cadmium dissolution: cadmium in the

cadmium containing residue is transferred to

solution by using dilute sulphuric acid

Cd + H2SO4 CdSO4 + H2 (1)

Other metallic impurities are dissolved in

the solution according to the reaction

Me + H2SO4 MeSO4 + H2 (2)

solution: Adjust the pH of the obtained solution

to remove iron, aluminum by hydrolysis Other

impurities such as Cu can be removed by

cementation using Cd powder:

Me2+ + Cd Me + Cd2+

solution to obtain metallic cadmium:

Generally, electrolysis reaction of

cadmium sulfate solution can be discribed as

follows:

CdS04 + H20 Cd  + H2S04 + 1/202 - Q

(5)

During the electrolysis of an aqueous

solution of cadmium sulphate, metals more

electropositive than cadmium (e.g Cu) will

plate at the cathode in addition to cadmium,

while zinc will not plate at the cathode due to

more electronegative (Eo = -0.76V) than

cadmium So the presence of zinc in the

solution has no significant effect on the quality

of cadmium obtained Cadmium metal

produced by this method has a high purity (Cd

> 99%)

II EXPERIMENTALS

A Preparation of cadmium sulphate solution

Cadmium residues composition is mainly

Zn 13%; Fe 0.85%; Pb 0.25% and other impurities such as Al, Ni, Cu, with very small amounts The residue will be dissolved by sulphuric acid Cadmium, zinc and some other metallic impurities will be together dissolved

by sulphuric acid The removal of Al and Fe from the solution is easier by using hydrolysis method, by adjusting pH of the solution to pH 5.2 - 5.4, Al and Fe precipitate as Al(OH)3 and Fe(OH)3 then will be removed from the solution Ni and Cu can be removed by cementation method Zn will remain in the solution The optimum conditions for cadmium dissolution are as follows:

B Cadmium electrolysis

The feed electrolyte was prepared as discribed above The cadmium Electrolysis was studied with the experimental conditions are as follows: The cathode current density 35-60 A/m2; Concentration of in electrolyte feed 30-

70 g/l of cadmium, 90-150 g/l of free H2SO4; Temperature of the electrolyte: 25-60 0C and gelatin concentration: 0 to 0.3 g/l

Bench scale electrolysis was carried out

in a cell of inert plastic construction with working volmes of 800 ml, using lead alloy (Pb/Ca/Sn) anodes Aluminum alloy (HS1A)

THE RECOVERY OF METALLIC CADMIUM FROM THE CADMIUM CONTAINING RESIDUE…

28

was used for the cathode The cathode current

density was 35- 60A/m2 Operating current was

calculated assuming a current efficiency of

80% Cell voltage was approximately 2.4V in

all tests The temperature during the tests

ranged from 25 to 60oC

Power to the cell was provided by a

constant current DC rectifier supply

Electrolyte was fed continuous in to the

cell, and allowed to overflow to maintain a set

cathode immersion level and the cell was

operated for 4h in batch mode to bring the cell

contents to the spent electrolyte conditions for

continuous mode

At the end of that time, the cathode was

removed, weighed and cleaned The plated

cadmium was dried to determined the weigh

and the actual current efficiency and with that

information, the flow of electrolyte for the

continuous cycle was corrected

The current efficiency was calculated by

using Faraday law of electrolysis Faraday's

laws can be summarized by

H = m r / m

where:

liberated at an electrode in grams

mr is the practically obtained mass of the substance at an electrode

substance (electrons transferred per ion);

applied;

 H is the current efficiency For cadmium electrolysis, M=112.41g; z=2 The plated cadmium at the cathode will

be analyzed by ICP-MS to determine the contents of cadmium and other impurity elements

III RESULTS AND DISCUSSION

A Effects of organic additives

The effect of an organic additive gelatin

on the electrolysis of cadmium from acidic sulfate solutions are studied

Experimental conditions:

- Current density 50 A/m2

- Cadmium concentration in the electrolyte: 50 g/l

- The concentration of free H2SO4 : 90 g/l

H

Fig.1: Effect of the concentration of gelatin to the current efficiency

LUONG MANH HUNG, TRAN NGOC VUONG

29

It is observed that addition of gelatin

increases the current efficiency and decreases

the energy consumption Gelatin when present

in the solution polarizes the cathode causing the

electroreduction of cadmium at more negative

potentials The presence of gelatin affects the

degree of crystallinity of the electrodeposits

indicating that the deposits are also more

ductile Scanning electron micrographs of

cadmium deposits obtained in the presence of

magnafloc show that compact deposits are

formed with an instantaneous nucleation and

growth mechanism It is evident that the

presence of gelatin decreases the number of

grains and increases the sizes of the crystallites

Since cadmium is very prone to dendritic

deposition The cadmium precipitate create

multiple spikes, thickness of the cadmium layer

are different To overcome this drawback, a

small amount of gelatin can be added as a

surface-active substances into electrolyte

solution From Figure 1, it is found that the concentration of gelatin 0.1 g/l to achieve the highest current efficiency When the gelatin concentration exceeds 0.1 g/l, the current efficiency decreases due to reducing of polarization

B Effect of cadmium concentration

Experimental conditions:

- Current density 50 A/m2

- Cadmiumconcentration in the electrolyte solution: 30 - 80 g/l

- The concentration of free H2SO4 : 90 g/l

- Electrolysis temperature : 35 0C

- The concentration of gelatin: 0.1 g/l

Experimental results are presented in Figure 2

82 83 84 85 86 87 88

H

Fig 2 The effect of cadmium concentration on the current efficiency

From Figure 2, we see that, when the

cadmium concentration in solution increased

from 30 to 60 g/l, the current efficiency

increases When cadmium concentration is

higher than 70 g/l, the current efficiency does

not increase but somewhat diminished

C Effect of free acid concentration

Experimental conditions:

- Current density 50 A/m2

- Cadmiumconcentration in the electrolyte solution: 50 g/l

- The concentration of free H2SO4 : 90 -

150 g/l

- Electrolysis temperature : 35 0C

- The concentration of gelatin: 0.1 g/l

Experimental results are presented in Figure 3