preliminary hazard analysis groundwater treatment …

127
PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT PLANT ORICA AUSTRALIA PTY LTD BOTANY INDUSTRIAL PARK, NSW Prepared by: Dean Shewring 8 November 2004 Pinnacle Risk Management Pty Limited ABN 83 098 666 703 PO Box 5024 Elanora Heights NSW Australia 2101 Telephone: (02) 9913 7284 Facsimile: (02) 9913 7930

Upload: others

Post on 16-Oct-2021

3 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

PRELIMINARY HAZARD ANALYSIS

GROUNDWATER TREATMENT PLANT

ORICA AUSTRALIA PTY LTD

BOTANY INDUSTRIAL PARK, NSW

Prepared by: Dean Shewring

8 November 2004

Pinnacle Risk Management Pty LimitedABN 83 098 666 703

PO Box 5024 Elanora HeightsNSW Australia 2101

Telephone: (02) 9913 7284Facsimile: (02) 9913 7930

Page 2: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 2004

Disclaimer

This report was prepared by Pinnacle Risk Management Pty Limited (PinnacleRisk Management) as an account of work for Orica Australia Pty Ltd (Orica).The material in it reflects Pinnacle Risk Management’s best judgement in thelight of the information available to it at the time of preparation. However, asPinnacle Risk Management cannot control the conditions under which thisreport may be used, Pinnacle Risk Management will not be responsible fordamages of any nature resulting from use of or reliance upon this report.Pinnacle Risk Management’s responsibility for advice given is subject to theterms of engagement with Orica.

Preliminary Hazard Analysis,Groundwater Treatment Plant, Orica

Australia Pty Ltd, Botany Industrial Park

Rev Date Description Reviewed By

A 5/7/04 Draft for Comment Orica

B 4/10/04 Orica Comments Included Orica

C 8/10/04 Draft EIS Issue Orica

D 18/10/04 Revised Orica

E 28/10/04 Revised Orica

F 8/11/04 Revised Orica

Page 3: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 2004

CONTENTS

CONTENTS ............................................................................................................II

EXECUTIVE SUMMARY............................................................................................ I

GLOSSARY............................................................................................................II

1 INTRODUCTION .............................................................................................. 1

1.1 Background ..................................................................................... 1

1.2 Objectives ........................................................................................ 1

1.3 Scope................................................................................................ 2

1.4 Methodology .................................................................................... 3

1.5 Findings and Recommendations ................................................... 3

2 SITE DESCRIPTION......................................................................................... 5

3 PROCESS DESCRIPTION ................................................................................. 9

3.1 Background to the Groundwater Contamination.......................... 9

3.2 Plant Description........................................................................... 10

3.3 Groundwater Composition ........................................................... 14

3.4 SEPP 33 Evaluation Information .................................................. 15

4 HAZARD IDENTIFICATION.............................................................................. 17

4.1 Hazardous Materials...................................................................... 17

4.2 Potential Hazardous Incidents ..................................................... 20

4.3 Safety Management Systems ....................................................... 32

4.3.1 Safety Software in Risk Assessment ............................................... 35

5 CONSEQUENCE ANALYSIS............................................................................ 37

5.1 Scenarios Modelled....................................................................... 40

5.2 Release Sources............................................................................ 40

5.3 Release Rates ................................................................................ 40

5.4 Release Duration ........................................................................... 41

Page 4: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 2004

5.5 Meteorological Data ...................................................................... 41

5.6 Terrain Effects and Degree of Confinement................................ 41

5.7 Consequence Results ................................................................... 42

5.7.1 Pipes or Vessels Failures Leading to Gas Plumes.......................... 42

5.7.2 Recovered Waste EDC Liquid Isotainer Events .............................. 46

5.7.3 Natural Gas Line Failures................................................................ 51

5.7.4 Thermal Oxidiser Explosion............................................................. 53

5.7.5 Caustic Scrubber Failure ................................................................. 54

5.7.6 Thermal Oxidiser Feed / Product Exchanger Failure....................... 55

5.8 Calculation of Fatality Due to Fires, Explosions and ToxicReleases................................................................................................... 55

6 FREQUENCY / LIKELIHOOD ANALYSIS............................................................ 57

6.1 Generic Equipment Failure Frequencies..................................... 57

6.2 BLEVE Likelihood.......................................................................... 58

6.3 Thermal Oxidiser Internal Explosion Likelihood ........................ 58

6.4 Caustic Scrubber Failure Likelihood ........................................... 58

6.5 Domino Incidents .......................................................................... 59

7 RISK ANALYSIS ........................................................................................... 61

7.1 Risk Criteria ................................................................................... 61

7.2 Mitigating Features for Off-site Individuals................................. 62

7.3 Risk Results................................................................................... 62

7.3.1 Individual Fatality Risk..................................................................... 62

7.3.2 Injury Risk........................................................................................ 62

7.3.3 Irritation Risk ................................................................................... 63

7.3.4 Property Damage ............................................................................ 64

7.3.5 Cumulative Risk............................................................................... 64

7.4 Transport Risk ............................................................................... 65

Page 5: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 2004

7.5 Risk from Combustion Products.................................................. 65

8 RISK TO THE BIOPHYSICAL ENVIRONMENT..................................................... 67

8.1 Escape of Materials to Atmosphere............................................. 67

8.1.1 HCl and Chlorine ............................................................................. 67

8.1.2 Dioxins............................................................................................. 67

8.2 Escape of Materials To Soil, Waterways or Sewerage System.. 70

9 CONCLUSION AND RECOMMENDATIONS......................................................... 73

10 REFERENCES .............................................................................................. 74

LIST OF FIGURESFigure 1 – BIP Site............................................................................................ 7

Figure 2 – GTP Layout ..................................................................................... 8

Figure 3 – Proposed Groundwater Extraction Well Locations ................... 11

LIST OF TABLESTable 1 – Plants on the BIP Site......................................................................... 5

Table 2 – Material Summary – SEPP 33.......................................................... 15

Table 3 – Health Information for some of the Groundwater Contaminants....... 18

Table 4 – Hazard Identification Word Diagram................................................. 22

Table 5 - Summary of Safety Related Procedures ........................................... 34

Table 6 – GTP Neighbouring Industries ........................................................... 38

Table 7 – EFFECTS Release Cases................................................................ 39

Table 8 – Scenarios Modelled .......................................................................... 40

Table 9 – Thermal Oxidiser Feed Stream, Dispersion Modelling ..................... 43

Page 6: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 2004

Table 10 – Thermal Oxidiser Exit Stream, Dispersion Modelling...................... 46

Table 11 - Radiant Heat Impact........................................................................ 48

Table 12 – Layout Considerations – Tolerable Radiant Heat Levels................ 49

Table 13 – Radiant Heat vs Distance – Pool Fire Scenarios............................ 49

Table 14 – Large Recovered Waste EDC liquid Pool (9 m diameter), DispersionModelling .......................................................................................................... 50

Table 15 –Natural Gas Jet Fires....................................................................... 51

Table 16 – Effects of Explosion Overpressure ................................................. 52

Table 17 – Natural Gas Vapour Cloud Explosions and Flash Fires.................. 53

Table 18 – Distance to Specified Levels of Explosion Overpressure for PotentialInternal Thermal Oxidiser Explosion Scenario.................................................. 54

Table 19 - Generic Equipment Failure Frequencies ......................................... 57

Table 20 - Risk Criteria, New Plants................................................................. 61

Table 21 – Fire Plume Rise Modelling.............................................................. 66

Table 22 – Water Discharge Hazard Identification Word Diagram ................... 71

LIST OF APPENDICESAppendix 1 – GTP Process Flow Diagrams.

Appendix 2 – Typical Groundwater Composition.

Appendix 3 – Selected MSDS’s.

Appendix 4 – Description of EFFECTS

Appendix 5 – Meteorological Data

Page 7: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 2004i

EXECUTIVE SUMMARY

Orica Australia Pty Ltd (Orica) at the Botany Industrial Park (BIP) is proposingto build a Groundwater Treatment Plant (GTP). Some of the groundwater atand around the BIP is contaminated with chlorinated hydrocarbons (CHCs).The primary contaminant is ethylene dichloride (EDC or 1,2 dichloroethane). Itis proposed to pump contaminated groundwater to the BIP and treat thegroundwater at the proposed GTP on the BIP. The GTP will be designed toremove and destroy the CHCs such that all emissions and wastes from theplant will comply with the EPA requirements.

To assess the risk associated with the GTP and associated operations, apreliminary hazard analysis (PHA) has been performed. This report details theresults from the analysis.

The risk associated with the proposed GTP and associated operations at theBIP has been assessed and compared against the DIPNR (Department ofInfrastructure, Planning and Natural Resources) risk criteria.

The results of this PHA show that the risk associated with the proposed GTPcomplies with DIPNR guidelines for tolerable fatality, injury, irritation andsocietal risk. Also, transport risk, risks to biophysical environment, the risk ofpropagation and the impact on cumulative risk in the Port Botany / Randwickarea from releases from the GTP and associated operations are broadlyacceptable. These conclusions apply to both off-site (e.g. residential areas) andon-site (i.e. neighbouring industrial facilities) risk.

The primary reason for the low risk levels is that significant consequentialimpacts from potential hazardous events associated with the GTP andassociated operations do not reach the nearest site boundary or, for theneighbouring industrial facilities, their likelihood is acceptably low.

As with most PHA’s, limited detailed design information is currently available.Correspondingly, some of the analysis in this report is based on assumedconditions. The assumptions made in this analysis are to be reviewedthroughout the project design stage and updated in the Final Hazard Analysis.Therefore, no specific recommendations are made.

It is assumed that the GTP and associated operations will be reviewed via theHAZOP methodology.

Page 8: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 2004ii

GLOSSARY

AS Australian Standard

BIP Botany Industrial Park

BLEVE Boiling liquid expanding vapour explosion

CHC Chlorinated hydrocarbon

CTC Carbon tetrachloride

DG Dangerous Goods

DIPNR Department of Infrastructure, Planning and Natural Resources

DNAPL Dense non-aqueous phase liquids

DTL Dangerous toxic load

EDC Ethylene dichloride

EPA Environmental Protection Authority

ERPG Emergency Response Planning Guideline

FHA Final hazard analysis

FTA Fault tree analysis

GTP Groundwater Treatment Plant

HAZOP Hazard and operability study

HCB Hexachlorobenzene

HCl Hydrochloric acid

HIPAP Hazardous Industry Planning Advisory Paper

IDLH Immediately dangerous to life and health

LEL Lower explosion limit

ML megalitres (or 1,000,000 litres)

MSDS Material safety data sheet

NCUA Notice of Cleanup Action

ORP Oxidation reduction potential

PCA Primary containment area

Page 9: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 2004iii

PHA Preliminary hazard analysis

PIC Products of incomplete combustion

P&ID Piping and instrumentation drawing

PCE Tetrachloroethene

ppb Parts per billion (e.g. 0.001 mg/kg)

ppm Parts per million (e.g. 1 mg/kg)

PPE Personnel protective equipment

QRA Quantitative risk assessment

RO Reverse osmosis

SCA Secondary containment area

scm Standard cubic metres (corrected to 15°C and 1 atmosphere pressure)

SEPP State Environmental Planning Policy

SHE Safety, health and environment

SIF Safety Instrumented Function

SIL Safety Integrity Level

SLOT Specified level of toxicity

SSU Steam stripping unit

STEL Short term exposure limit

TCE Trichloroethene

TNO Dutch organisation specialising in risk software

UVCE Unconfined vapour cloud explosion

VC Vinyl chloride

VOC Volatile organic carbons

VRA Voluntary Remediation Agreement

WHB Waste heat boiler

wt weight

Page 10: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 20041

REPORT

1 INTRODUCTION

1.1 BACKGROUND

Orica Australia Pty Ltd (Orica) at the Botany Industrial Park (BIP) is proposingto build a Groundwater Treatment Plant (GTP). Some of the groundwater atand around the BIP is contaminated with chlorinated hydrocarbons (CHCs).The primary contaminant is ethylene dichloride (EDC or 1,2 dichloroethane). Itis proposed to pump contaminated groundwater to the BIP and treat thegroundwater at the proposed GTP on the BIP. The GTP will be designed toremove and destroy the CHCs such that all emissions and wastes from theplant will comply with the EPA requirements.

The groundwater to be treated is to be pumped from three locations:

From the Primary Containment Area (Blocks 1 and 2 of Southlands, tothe south west of the BIP);

From the Secondary Containment Area down gradient of the PrimaryContainment Area, along Foreshore Road; and

From the DNAPL (dense non-aqueous phase liquids) containment linealong the western perimeter of the BIP.

As part of the Environmental Impact Statement for this project, a preliminaryhazard analysis (PHA) is required in accordance with the guidelines publishedby the Department of Infrastructure, Planning and Natural Resources (DIPNR)NSW Hazardous Industry Planning Advisory Paper (HIPAP) No 6 (Ref 1). Theneed for a PHA to be prepared was found by application of SEPP 33 (Ref 2), aprocess for evaluating whether proposed developments will be potentiallyhazardous and hence whether some form of hazard analysis is required. Asummary of the SEPP 33 determination information is included in Section 3.4 ofthis report. The requirement for a SEPP 33 analysis was part of the Director-General’s Requirements for the EIS (Ref 3).

Orica has appointed Pinnacle Risk Management Pty Ltd (Pinnacle RiskManagement) to prepare this preliminary hazard analysis report.

1.2 OBJECTIVES

The main aims of this PHA study are to:

Address the Director-General’s requirements for the PHA;

Page 11: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 20042

Identify the credible, potential hazardous events associated with theproposed GTP and associated operations;

Evaluate the level of risk associated with the identified potentialhazardous events from the GTP and associated operations tosurrounding land users, including other BIP companies and theiroperations, and compare the calculated risk levels with the risk criteriapublished by DIPNR in HIPAP No 4 (Ref 4);

Review the adequacy of the existing safeguards; and

Where necessary, submit recommendations to Orica to ensure that theproposed GTP and associated operations are operated and maintainedat acceptable levels of safety and effective safety management systemsare used.

1.3 SCOPE

This PHA assesses the credible, potential hazardous events and correspondingrisks associated with the GTP and associated operations.

The assessment includes:

The GTP process operations;

Material storage;

Pumping and transport of groundwater through pipelines to the GTP;

Pumping and transport of treated water through pipelines from the GTP;

Road transport of recovered waste EDC liquid from storage at theTerminals Pty Ltd facility at Port Botany; and

Pumping and transport of salty waste water (water treatment byproduct)via an existing unused pipeline from the GTP to a stormwater channeldischarging to Brotherson Dock.

It does not include any aspects of the storage and handling of recovered wasteEDC liquid at the Terminals Pty Ltd facility at Port Botany, as this facility isowned by a third party and covered by its own planning approvals, whichinclude storage and handling of waste chlorinated hydrocarbons such as therecovered waste EDC liquid.

As with most PHA’s, limited detailed design information is currently available.Correspondingly, some of the analysis in this report is based on assumedconditions. The assumptions made in this analysis are to be reviewedthroughout the project design stage and updated in the Final Hazard Analysis.

Page 12: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 20043

1.4 METHODOLOGY

In accordance with the approach recommended by DIPNR in HIPAP 6 (Ref 1)the underlying methodology of the PHA is risk-based, that is, the risk of aparticular potentially hazardous event is assessed as the outcome of itsconsequences and likelihood.

The PHA has been conducted as follows:

The design and location of the GTP and associated operations werereviewed to identify credible, potential hazardous events;

The frequency and consequence of each significant potential hazardousevent were estimated;

The risk results have been quantified by combining the frequency andconsequence for each event and summing to give total (cumulative) riskas appropriate; and

The risks associated with the facility are compared to the criteria inHIPAP 4 (Ref 4).

1.5 FINDINGS AND RECOMMENDATIONS

The results of this PHA show that the risk associated with the proposed GTPand associated operations complies with DIPNR guidelines for tolerable fatality,injury, irritation and societal risk. Also, transport risk, risks to biophysicalenvironment, the risk of propagation and the impact on cumulative risk in thePort Botany / Randwick area from releases are broadly acceptable. Theseconclusions apply to both off-site (e.g. residential areas) and on-site (i.e.neighbouring industrial facilities) risk.

The primary reason for the low risk levels is that significant consequentialimpacts from potential hazardous events associated with the GTP operation donot reach the nearest site boundary or, for the neighbouring industrial facilities,their likelihood is acceptably low.

The significant hazardous events identified and quantitatively analysed are:

Pipes or vessels failures leading to gas plumes;

Recovered waste EDC liquid Isotainer events, e.g. BLEVE and toxicimpact from plumes emanating from large pools;

Natural gas line failures;

Thermal oxidiser explosion; and

Caustic scrubber failure.

Page 13: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 20044

As with most PHA’s, limited detailed design information is currently available.Correspondingly, some of the analysis in this report is based on assumedconditions. The assumptions made in this analysis are to be reviewedthroughout the project design stage and updated in the Final Hazard Analysis.Therefore, no specific recommendations are made.

It is assumed that the GTP and associated operations will be reviewed via theHAZOP methodology.

Page 14: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 20045

2 SITE DESCRIPTION

A map of the area showing the location of the Botany Industrial Park site andthe GTP in the context of its surroundings is presented in Figure 1.

The BIP site is bounded to the north by Corish Circle, the east by Denison St,the south by Beauchamp Rd and to the west by the Botany Goods railway lineeasement. The land around most of the BIP site perimeter is zoned commercialand industrial. The exception is land adjacent to part of the eastern boundary ofthe site, which has significant residential areas along Denison Street, andbeyond. The nearest public road to the proposed GTP is Denison St, adistance of approximately 325 m.

There have been some changes relating to ownership of the various chemicalplants operating on the BIP site. The BIP land was wholly owned by Orica(previously ICI Australia) until late 1998, when changes resulting in the sale ofsome Orica plants or formation of joint venture companies occurred. The sitewas subsequently subdivided (in 1999) to form the Botany Industrial Park (BIP).There are now six main industrial complexes on the site, operated by threedifferent companies, Orica, Qenos (a joint venture between Orica and ExxonMobil) and Huntsman as shown in Table 1 below.

Table 1 – Plants on the BIP Site

Plant Operator Description

Olefines Qenos Manufactures ethylene from ethane feedstock foruse in downstream plants

Alkathene Qenos Manufactures high density polyethylene plastics

Alkatuff Qenos Manufactures low density polyethylene plastics

Site Utilities Qenos Supplies steam, nitrogen, cooling water etc. tovarious plants at the site

Surfactants Huntsman Manufactures a range of materials such asdetergents etc largely based on ethylene oxide

Chlorine andDerivatives

Orica Manufactures chlorine, hydrochloric acid, causticsoda, ferric chloride and sodium hypochlorite

The location of the GTP is on Orica land on the plot formerly occupied by theSilicates Plant on 10th Avenue - see Figure 2 for details of the plant layout. Asthe GTP is part of the Orica ChlorAlkali facility, there are in effect twoboundaries to consider in this PHA. The first is the boundary with neighbouringindustrial land users, the nearest being the other (non-Orica) BIP sitecompanies, referred to as on-site. The second is the boundary with theresidential land users external to the BIP site, referred to as off-site.

Page 15: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 20046

Where associated operations required for the GTP Project such as wells,pipelines and road transport are located outside the BIP, then the relevantboundaries for risk assessment are the land users adjacent to, and nearby,these operations.

Page 16: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 20047

Figure 1 – BIP Site

Page 17: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 20048

Figure 2 – GTP Layout

Page 18: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 20049

3 PROCESS DESCRIPTION

3.1 BACKGROUND TO THE GROUNDWATER CONTAMINATION

From Ref 5, ICI Australia (now Orica) manufactured chlorinated hydrocarbons(CHCs) at its Botany Site from approximately 1945 to 1994. A range ofchlorinated solvents including carbon tetrachloride (CTC), tetrachloroethene(known as PCE) and trichlorethene (TCE) were produced until 1991. Theirmanufacturing processes involved other chlorinated hydrocarbon intermediates.Ethylene dichloride (EDC) was produced from 1966 to 1994 as an intermediatefor production of vinyl chloride monomer, subsequently polymerised to poly vinylchloride (PVC).

Leaks of chlorinated materials may have occurred over the years fromproduction, waste storages, effluent systems and other underground lines, drumstorage areas (once stored on unbunded bare ground) and storage tanksystems.

The Botany Industrial Park is constructed on the Botany Sands Aquifer, whichprovided the original potable water supply for Sydney and has been usedextensively as a source of industrial water. The Botany Industrial Park islocated close to Botany Bay, where the Botany Sands Aquifer discharges to thesea. There are peaty layers in the sands, with a general layer present at 8 to10m below ground surface (bgs) that divides what are referred to as the shallowand deep aquifers.

The chlorinated solvents are all heavier than water and thus sink ingroundwater. They are all only slightly soluble in water and as pure or mixedliquids are referred to as Dense Non-Aqueous Phase Liquids (DNAPL). DNAPLwill slowly dissolve and sink until they reach less permeable layers where theymay form pools and/or move following the topography of the low permeabilitylayer.

Some material will, however, remain trapped within soil pore spaces due tocapillary and surface tension effects; this form of DNAPL is termed "residualDNAPL". Groundwater flows through and past the DNAPL source zones, slowlydissolving CHCs into dissolved contaminant plumes. The CHCs from thedissolved plume can then sorb to soil as they move with groundwater. There isa wide variety of properties of the CHCs (key parameters include watersolubility, volatility, density, surface tension, soil adsorption) and aquifercharacteristics (permeability, geology, groundwater velocity etc) that affect themovement of contaminants in groundwater.

The CHCs are in general toxic chemicals to human health. In particular, EDC isone of the most soluble and least absorptive of the CHCs; it dissolves andmoves rapidly in groundwater compared with lower solubility, highly absorptivematerials such as PCE, TCE and CTC.

Orica has been investigating the groundwater issues since 1989 under variousregulatory regimes (most recently a Voluntary Remediation Agreement (VRA)

Page 19: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200410

under the Contaminated Land Management Act). Under the VRA, Orica’sproposed remediation approach is the widespread application of in situenhanced bioremediation and reactive iron barriers where bioremediation is notapplicable. Accordingly Orica are currently implementing field trials ofbioremediation and are progressing studies on reactive iron barrier installation.Should bioremediation trials prove ineffective, Orica's fallback was hydrauliccontainment and ex situ treatment.

Recent detection of higher than expected concentrations in an extraction boreclose to residential locations, combined with concerns about the rate ofmovement of the most concentrated contaminant plume, led to issue of a Noticeof Cleanup Action (NCUA) by the NSW EPA, (now part of the Department ofEnvironment and Conservation, DEC). As required, Orica has produced aGroundwater Cleanup Plan (GCP) for the EPA's consideration. The EPAsubsequently authorised Orica to implement the GCP.

The NCUA requires hydraulic containment and remediation of the extractedgroundwater, including the use of ex situ treatment technology. Orica’sbioremediation trials have proved less successful than hoped, and hydrauliccontainment and ex situ treatment is being used for all contaminant containmentduties.

3.2 PLANT DESCRIPTION

From Ref 6, the three areas from which the groundwater will be extracted are asfollows:

From the Primary Containment Area (PCA - Blocks 1 and 2 ofSouthlands, to the south west of the BIP);

From the Secondary Containment Area (SCA) down gradient of thePrimary Containment Area, along Foreshore Road; and

From the DNAPL (dense non-aqueous phase liquids) containment linealong the western perimeter of the BIP.

The locations of these three areas are shown in Figure 3.

The GTP will be designed to treat groundwater from all three extraction areasas a common feed.

Due to time constraints imposed on the GCP by the NCUA, some groundwaterwill be extracted and treated prior to the operation of the GTP. Thisgroundwater will be sourced either from the PCA, the SCA or a combination ofboth. The groundwater will be treated by the recommissioned Steam StrippingUnit (SSU) located on BIP. It is proposed that pipelines will be constructed totransfer the groundwater from Southlands and Foreshore Road to the SSU fortreatment.

Page 20: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200411

Figure 3 – Proposed Groundwater Extraction Well Locations

Page 21: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200412

Treated water from the SSU will be discharged to sewer and/or re-used on theBIP, while the recovered waste EDC liquid will be transported to Port Botanyand stored in existing tanks. Once the GTP is operational, this recovered wasteEDC liquid will be transported to the GTP site in Isotainers, for destruction in theGTP thermal oxidiser.

The GTP is being designed to treat 15 ML/day. The plant life will depend on therate of cleanup of the contaminated groundwater and source areas. At thisstage, the plant is expected to operate for up to 30 years.

Process flow diagrams showing the main items of equipment for the GTP andassociated operations are given in Appendix 1.

Feed Handling

Groundwater will be pumped via submersible pumps in bores and connectingpipework from the containment areas as noted above to a groundwater feedtank on the GTP plot. The pressure in the feed tank will be atmospheric.Hydrochloric acid (recovered from the GTP – refer to Air Treatment below) isdosed into the feed tank to reduce the pH to approximately 2.7. This reducesthe chance of fouling in the air strippers.

The feed tank is nitrogen padded to reduce the risk of explosions.

The acidified water is then pumped to the air stripping section.

Stored recovered waste EDC liquid will be transported by road in an Isotainerfrom Port Botany to the GTP. The expected average delivery frequency isapproximately once every two weeks. Recovered waste EDC liquid will beinjected at a low rate into the thermal oxidiser (see below).

Air Stripping

The air strippers are modular design tray towers specifically designed for easycleaning in fouling service. Air stripping is performed by drawing ambienttemperature air up through a falling column of water, transferring almost all thevolatile chlorinated hydrocarbons such as EDC from the water to the air. Thewater and air are then treated separately.

The air stripper is designed to achieve a very low EDC concentration of< 0.003 mg/L in the stripped water bottoms.

Components which are less volatile than EDC will be present in the strippedwater. These consist mainly of phenol and chlorinated phenols with a totalconcentration of approximately 200 ppb. Of these, some components (e.g.phenol) are present in excess of their identified removal target values and sorequire further treatment.

The stripped water is pumped to activated carbon adsorber beds which "polish"(remove) almost all the remaining hydrocarbons, so achieving treated water tothe required standards (see below).

Page 22: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200413

Air Treatment

Contaminated air from the air strippers is drawn by an induced draught fan to athermal oxidation unit where the stream is heated to a high temperature in thepresence of air to break down the contaminants to form carbon dioxide, watervapour, hydrochloric acid and chlorine.

Initially, the air from the air strippers is preheated by heat exchange with thethermal oxidation unit’s product gas stream. Further heating is provided byburning natural gas within the thermal oxidation unit, as well as heat releasedfrom the contaminant destruction. This heat input maintains the correcttemperatures for the oxidation process to work. Recovered waste EDC liquidwill be injected directly into the thermal oxidiser at low rates using a speciallydesigned nozzle that will atomise the liquid to ensure effective destruction. Therecovered waste EDC liquid will be added once the contaminant load in thegroundwater falls with time, so that the total EDC load on the thermal oxidiserdoes not exceed the original design value.

Thermal oxidation operates at a relatively high temperature (approx 1,000°C) toensure effective destruction of chlorinated hydrocarbons (>99.99%).

The product gas stream from the thermal oxidation unit is cooled in a wasteheat boiler (i.e. steam is raised) and then by heat exchange with the incomingair stream. After this heat exchanger, the product stream is quenched withweak hydrochloric acid (HCl) solution (5 wt%) to approximately 70oC. Thequenched stream then passes through the acid absorber where the hydrogenchloride (HCl) generated in the thermal oxidiser is recovered as a 5 wt% HClsolution. The recovered hydrochloric acid (5 wt%) will be used in the feedhandling unit for acidification of the feed.

The air stream then continues to the caustic scrubber for further treatment toremove other acid gases and chlorine to meet emission specifications. Air exitsthe plant via a 20 m high stack at about 67°C. The scrubber feed (46 wt%caustic soda solution) is pumped from the existing ChlorAlkali plant. Sodiummetabisulphite is also added to reduce the level of free chlorine in the scrubbereffluent.

Water Treatment

Stripped water from the air strippers is treated with caustic soda solution toraise the pH to approximately 8 to precipitate iron compounds. These will beremoved using a filter, dewatered and sent to landfill.

Filtered water from the iron removal stage passes through an ion exchangesystem to remove organic acids, an activated carbon filter to remove phenol,organochlorides and other residual matter not removed in the air stripper andthen through a breakpoint chlorination unit to reduce ammonia levels.

A portion of this water will then enter a Reverse Osmosis (RO) unit. The ROunit will remove dissolved solids (salts), producing water to the quality

Page 23: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200414

standards required. Treated water from the RO unit is recycled to the absorberor scrubber, or exits the GTP for distribution to users.

Treated Water Reuse

Treated water discharged from the GTP will be processed to meet the requiredstandards for re-use on BIP process plants and other industrial users in theBotany area. Initially, this water will be distributed throughout the BIP for use asprocess water, for example, as cooling tower make-up water and feed to thedemineralisation plant for steam generation. The treated water will first bedischarged to a buffer tank on the GTP site to allow approximately 2 hoursresidence time before distribution throughout BIP.

Most of the treated water will be consumed by process operations around theBIP. As additional users of recycled water are found, the flow to the RO plantwill increase to its full capacity of 15ML/d. Until that time, the balance of thewater not required for polishing by the RO unit for re-use will be discharged withthe salty waste water stream through an existing pipeline to an existing channelwhich discharges into Botany Bay at Brotherson Dock. The pipeline will beupgraded with an internal ‘plastic sleeve’ to minimise corrosion.

In the event that process operations on BIP and other users cannot use all theavailable treated water from the GTP, the unused treated water will bedischarged to Botany Bay via the above pipeline. Other potential uses for thetreated water are being investigated.

Operation

The GTP and most of the associated operations will be controlled by anautomated monitoring and control system in a dedicated control room. In theevent of an abnormal condition being detected, the plant will be automaticallyshut down, isolating all feeds and stopping all discharges. The GTP andassociated operations will be designed to be a robust and effective process,and the technical design specification includes a 95% availability with maximummaintenance shut-down period of a week, to ensure that control of thegroundwater movement and associated plume of EDC is maintained.

When groundwater treatment is no longer required, the plant will be shut-downand decommissioned.

3.3 GROUNDWATER COMPOSITION

A typical groundwater composition to the GTP is shown in Appendix 2. As canbe seen from this data:

EDC is the primary contaminant;

The concentration of the bulk of the contaminants is relatively low; and

The total concentration of contaminants is less than 1%.

This information confirms current site practice when dealing with thegroundwater via excavation work etc, i.e. releases of the groundwater will only

Page 24: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200415

impact those personnel near the point of release as the concentrations of thecontaminants are not high enough to generate modest to high concentrationplumes in the atmosphere. This was also a finding from the PHA conducted onthe steam stripping unit operation (Ref 7).

3.4 SEPP 33 EVALUATION INFORMATION

SEPP 33 (Ref 2) requires a summary of the potentially hazardous materials forthe proposed development to be compiled and evaluated. This materialsummary is shown in Table 2. Some non-hazardous materials are included forinformation.

Table 2 – Material Summary – SEPP 33

Material Class Quantity ExceedsSEPP 33Criterion

?

Location

5 wt% hydrochloric acid Corrosive - 8(PG II)

40 m3 Yes On the GTP plot

46 wt% caustic soda Corrosive - 8(PG II)

30 m3 Yes On the GTP plot

Recovered waste EDCliquid (Isotainer)

FlammableLiquid - 3 (Toxic

- 6.1) (PG II)

20 m3 Yes (toxiccriterion)

On the GTP plot

Groundwater - 15 ML/day NA Piped to and onthe GTP plot

Natural Gas Flammable gas- 2.1

Negligible No In piping to thethermal oxidiseronly

Nitrogen Non-flammable,non toxic gas -

2.2

Negligible NA In piping

Sodium metabisulphite(33 wt%)

Corrosive - 8 4.5 m3 No On the GTP plot

Sodium hypochlorite Corrosive - 8 5 m3 No On the GTP plot

Various water treatmentchemicals

Corrosive - 8 Low No On the GTP plotand/or broughton-site forequipmentcleaning and thenremoved

Flocculant - 6 m3 NA On the GTP plot

Page 25: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200416

Material Class Quantity ExceedsSEPP 33Criterion

?

Location

Salty waste water - Negligible NA Pipeline fromGTP toBrotherson Dock

Steam - Minor NA On the GTP plot

Given the above information, the GTP and its associated operations isconsidered a potentially hazardous facility by the SEPP 33 process andtherefore a PHA is required.

Page 26: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200417

4 HAZARD IDENTIFICATION

4.1 HAZARDOUS MATERIALS

Groundwater

The following MSDS’s (material safety datasheets) are given in Appendix 3:

Southlands groundwater;

Southlands groundwater – Central Plume;

GTP groundwater;

Recovered waste EDC liquid; and

EDC (the main contaminant).

The bulk of the groundwater is water; greater than 99 wt%.

Depending on the source of the groundwater:

EDC is less than 0.02%;

Trichloroethene (TCE) is less than 0.001%;

Carbon tetrachloride (CTC) is less than 0.002%;

Tetrachloroethene (PCE) is less than 0.002%; and

Vinyl chloride (VC or VCM – vinyl chloride monomer) is less than0.001%.

Some groundwater contaminants may cause cancer. Groundwater is notcombustible although some of the components in the pure form are flammable,e.g. EDC and VC.

Table 3 lists some of the exposure levels for the common groundwatercontaminants.

Page 27: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200418

Table 3 – Health Information for some of the Groundwater Contaminants

Component ShortTerm

ExposureLevel

(STEL),ppm

IDLH(ppm)

ERPG 1(ppm)

ERPG 2(ppm)

ERPG 3(ppm)

EDC - 50 50 200 300

TCE 200 1,000 100 500 5000

CTC 10 200 20 100 750

PCE 150 150 100 200 1000

VC - - 50 5,000 20,000

Chloroform - 500 notappropriate

50 5,000

Notes for Table 3: - = not recorded due to insufficient data

IDLH = Immediately Dangerous to Life and Health

The American Industrial Hygiene Association's Emergency Response PlanningGuidelines (Ref 8) provide data on "injury" toxic exposure levels for a fewindustrial chemicals for exposure periods of one hour. Three sub-fatal toxicexposure levels are defined by the American Industrial Hygiene Association(AIHA). The definitions are as follows.

Emergency Response Planning Guidelines (ERPGs) are values intended toprovide estimates of concentration ranges above which one could reasonablyanticipate observing adverse health effects; see ERPG-1; ERPG-2; ERPG-3.The term also refers to the documentation that summarises the basis for thosevalues. The documentation is contained in a series of guides produced by theEmergency Response Planning Committee of the American Industrial HygieneAssociation (Ref 8).

ERPG-1: The maximum airborne concentration below which nearly allindividuals could be exposed for up to 1 hour without experiencingmore than mild, transient adverse health effects or withoutperceiving a clearly defined objectionable odour.

ERPG-2: The maximum airborne concentration below which nearly allindividuals could be exposed for up to 1 hour without experiencingor developing irreversible or other serious health effects orsymptoms that could impair an individual's ability to takeprotective action.

ERPG-3: The maximum airborne concentration below which nearly allindividuals could be exposed for up to 1 hour without experiencingor developing life-threatening health effects.

Page 28: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200419

In terms of effects, the ERPG-1 and ERPG-2 definitions seem to approach mostclosely the "Irritation" and "Injury" guidelines suggested by DIPNR, howeverwith a 1 hour rather than “relatively short exposure period”. The use of thesevalues for this PHA are considered conservative as the release durations will berelatively short (see Section 5.4).

Whilst EDC is present in the groundwater, it is also present in the recoveredwaste EDC liquid stored in the Isotainer that is loaded at Port Botany. Thecomposition of this liquid is expected to typically be:

Component Percentage by mass

EDC 95.1Vinyl Chloride 1.6Water 0.3Benzene 0.1Other CHCs balance

Note that the flammability limits for EDC are 6.2 to 16 vol%.

Hydrogen Chloride

Hydrogen chloride is produced in the thermal oxidation unit and is present in theexit stream to the acid absorber. Hydrogen chloride is a corrosive gas whichcan impact people, property and the environment. The health impact data forhydrogen chloride is as follows:

ERPG 1 3 ppm

ERPG 2 20 ppm

ERPG 3 150 ppm

IDLH 50 ppm

Note that chlorine is also present in the thermal oxidiser off-gas but only atconcentrations much lower than hydrogen chloride.

Natural Gas

The thermal oxidation unit burns natural gas for additional heat input. This gasis the same as is used in domestic situations, e.g. household water heaters andstoves. Natural gas is a flammable gas. It is not toxic. Releases of natural gascan lead to fires and/or explosions if ignition occurs.

Corrosives

The two bulk corrosive fluids proposed to be handled and/or stored at the GTPare hydrochloric acid (5 wt%) and caustic soda (sodium hydroxide; 46 wt%).Also stored at the GTP and used are sodium metabisulphite solution (used as areducing agent in the caustic scrubber, 4.5m3 storage as 33% solution, 18 te as100% annual consumption) and sodium hypochlorite solution (for ammonia

Page 29: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200420

removal, 5m3 storage as 12% solution, 220 te as 100% annual consumption).Other minor quantities of corrosive materials may be stored on a permanent ortemporary basis to enable cleaning and maintenance of GTP equipment,especially the equipment in the treated water plant area. These are not yetspecified in detail.

Releases of these materials can affect those who are exposed to the liquidand/or vapour above the liquid, as well as affect the environment if allowed toflow to soil and/or waterways.

Nitrogen

Nitrogen is non-toxic (it is the major constituent of air) but it presents the hazardof asphyxiation to those on-site personnel who may enter a tank which containsnitrogen or a reduced oxygen content due to dilution by nitrogen.

Oxides of Sulphur and Nitrogen

Oxides of sulphur and nitrogen (SOx and NOx) are formed in small quantities inthe thermal oxidation unit. The impact of normal release of these materials isanalysed separately in the project EIS.

Salty Waste Water

This is the waste stream produced as part of the water treatment process. Itcontains the chlorides removed from the main treated water stream in order tomeet the specifications for supply of treated water to other users. It is not aDangerous Good or a Hazardous Material.

Steam

Medium pressure (around 1000kPag) steam is produced in the waste heatboiler which is part of the thermal oxidation unit. Its hazards of hightemperature and pressure could have local effects only.

4.2 POTENTIAL HAZARDOUS INCIDENTS

In accordance with the requirements of Guidelines for Hazard Analysis, (ref 1),it is necessary to identify hazardous events which could be caused by the GTPand associated operations. As recommended in HIPAP 6, the PHA focuses on“atypical and abnormal events and conditions. It is not intended to apply tocontinuous or normal operating emissions to air or water”. The latter arediscussed elsewhere in the EIS.

Potential hazardous events for the GTP have been identified by the followingmeans:

Hazard studies involving design and operating personnel;

Page 30: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200421

Operational experience;

Literature reviews;

Checklists; and

Historical incident reviews.

In keeping with the principles of PHAs, credible, hazardous events with thepotential for off-site effects have been identified. That is, “slips, trips and falls”type events are not included nor are non-credible situations such as an aircraftcrash occurring at the same time as an earthquake.

Where abnormal events are assessed as having potentially chronic effects,these are discussed elsewhere in the EIS. As stated above in accordance withthe required methodology, the PHA itself only discusses acute effects.

The large majority of the specific release scenarios are generic equipmentfailures, e.g. failures of vessels, pipes etc, from previous industrial events.These are supplemented by process incidents due to other abnormal modes ofoperation, control system failure and human error.

The credible, significant incidents identified are summarised in the HazardIdentification Word Diagram following (Table 4). The diagram presents thecauses and consequences of the events, together with major preventative andprotective features that are included as part of the design.

A column showing which scenarios are included in the quantified riskcalculations is included in the hazard identification table. Only the events withthe potential for significant consequences have been included in the quantifiedrisk assessment described in subsequent sections of the report (i.e. “yes” infinal column of Table 4).

Page 31: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinn

acle

Ris

k M

anag

emen

t

Oric

a G

TP

PH

A R

epor

t Rev

F.D

oc8

Nov

embe

r 20

0422

Tabl

e 4

– H

azar

d Id

entif

icat

ion

Wor

d D

iagr

am

Item

No.

Even

tC

ause

sPo

ssib

le C

onse

quen

ces

Prev

entio

n/Pr

otec

tion

Ris

kQ

uant

ified

?

1.

Rel

ease

of

vapo

urs

from

the

grou

ndw

ater

feed

tank

vap

our

spac

e

Indu

ced

drau

ght f

an s

huts

dow

n so

can

't dr

aw ta

nkov

erhe

ad v

apou

rs (t

anks

are

not p

ress

uris

ed)

Chl

orin

ated

hyd

roca

rbon

vap

our

from

sto

rage

tank

s es

cape

s to

atm

osph

ere

Loca

l im

pact

onl

y

No

haza

rdou

s co

nseq

uenc

es to

adja

cent

indu

stria

l or r

esid

entia

lar

eas

Filli

ng o

f the

gro

undw

ater

feed

tank

tobe

sto

pped

dur

ing

a sh

utdo

wn

Vap

ours

to b

e ve

nted

aw

ay fr

omop

erat

iona

l sta

ff. T

he fe

ed ta

nk w

illve

nt to

the

ther

mal

oxi

dise

r or b

e“b

oxed

in” i

f the

ther

mal

oxi

dise

r is

unav

aila

ble.

If th

ere

is a

larg

ete

mpe

ratu

re c

hang

e in

the

feed

tank

,so

me

vapo

ur m

ay b

e di

scha

rged

to a

sepa

rate

, act

ivat

ed c

arbo

n, v

apou

rco

llect

ion

syst

em. S

pent

car

bon

wou

ldbe

rege

nera

ted

or s

ent t

o ap

prov

eddi

spos

al.

No

2.

Dis

char

ge o

fav

aila

ble

chlo

rine

e.g.

sodi

umhy

poch

lorit

e(N

aOC

l) or

chlo

rine

(Cl 2)

Dis

char

ge fr

om c

aust

icsc

rubb

ing

tow

er (n

orm

al d

esig

nco

nditi

on)

The

disc

harg

e st

ream

join

s th

em

ixed

site

effl

uent

The

pote

ntia

l exi

sts

for r

eact

ions

tooc

cur i

n th

e m

ixed

effl

uent

from

site

(e.g

. chl

orof

orm

cou

ld b

e pr

oduc

ed)

and

henc

e di

scha

rge

limits

may

be

exce

eded

No

haza

rdou

s co

nseq

uenc

es to

adja

cent

indu

stria

l or r

esid

entia

lar

eas

Des

ign

of a

cid

abso

rber

and

cau

stic

scru

bber

sys

tem

to a

void

exc

essi

vefo

rmat

ion

of a

vaila

ble

chlo

rine.

Are

duci

ng a

gent

(sod

ium

met

abis

ulph

ite)

will

be

adde

d to

the

caus

tic s

crub

ber

Mon

itorin

g of

dis

char

ge a

nd s

iteef

fluen

t stre

ams

No

Page 32: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinn

acle

Ris

k M

anag

emen

t

Oric

a G

TP

PH

A R

epor

t Rev

F.D

oc8

Nov

embe

r 20

0423

Item

No.

Even

tC

ause

sPo

ssib

le C

onse

quen

ces

Prev

entio

n/Pr

otec

tion

Ris

kQ

uant

ified

?

3.

Non

-com

plia

nce

with

Syd

ney

Wat

ersu

spen

ded

solid

ssp

ecifi

catio

n

Bac

kwas

h of

san

d fil

ters

will

caus

e hi

gher

sus

pend

ed s

olid

sflo

ws

to p

lant

effl

uent

Sus

pend

ed s

olid

s le

vels

indi

scha

rge

vary

dep

endi

ng o

nba

ckw

ash

met

hods

, fre

quen

cyet

c

Exc

eed

perm

itted

leve

ls in

BIP

site

efflu

ent

No

haza

rdou

s co

nseq

uenc

es to

adja

cent

indu

stria

l or r

esid

entia

lar

eas

Sys

tem

to b

e de

sign

ed to

ens

ure

com

plia

nce

Mon

itorin

g of

dis

char

ge a

nd s

iteef

fluen

t stre

ams

No

4.

Emis

sion

s of

diox

in fr

om th

eth

erm

al o

xidi

ser

Ther

mal

oxi

dise

r doe

s no

tpe

rform

to s

peci

ficat

ion

unde

ral

l req

uire

d co

nditi

ons

Dio

xins

may

be

prod

uced

with

impa

ct to

peo

ple

and

faun

a

Chr

onic

impa

ct

Con

trol a

nd M

onito

ring

Sys

tem

will

shu

tdo

wn

the

plan

t und

er a

bnor

mal

cond

ition

s, i.

e. s

top

the

air s

tripp

er a

ndsh

ut d

own

the

feed

to th

e th

erm

alox

idis

er. R

egul

ar te

stin

g w

ill oc

cur t

opr

ove

good

ope

ratio

n

Yes

–ch

roni

cef

fect

sas

sess

ed in

hum

anhe

alth

risk

asse

ssm

ent

5.

Fire

or e

xplo

sion

from

ele

ctric

alfa

ults

Elec

tric

mot

ors

on fa

ns a

ndpu

mp

Pla

nt d

amag

e; p

ossi

ble

igni

tion

ofan

y fla

mm

able

vap

ours

Loca

l im

pact

onl

y

No

haza

rdou

s co

nseq

uenc

es to

adja

cent

indu

stria

l or r

esid

entia

lar

eas

Det

erm

ine

haza

rds

and

desi

gn to

the

appr

opria

te e

lect

rical

sta

ndar

dsN

o

Page 33: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinn

acle

Ris

k M

anag

emen

t

Oric

a G

TP

PH

A R

epor

t Rev

F.D

oc8

Nov

embe

r 20

0424

Item

No.

Even

tC

ause

sPo

ssib

le C

onse

quen

ces

Prev

entio

n/Pr

otec

tion

Ris

kQ

uant

ified

?

6.

Loss

of

cont

ainm

ent o

fgr

ound

wat

erfro

m th

e fe

edta

nk o

rco

nnec

ting

pipi

ng

Leve

l con

trol f

ailu

re le

adin

g to

an o

verfi

ll

Wat

er h

amm

er

Tank

or p

ipe

failu

res

Dra

in v

alve

s le

ft op

en

Rel

ease

of g

roun

dwat

er

Loca

l im

pact

onl

y

No

haza

rdou

s co

nseq

uenc

es to

adja

cent

indu

stria

l or r

esid

entia

lar

eas

Pre

vent

ativ

e m

aint

enan

ce o

fin

stru

men

tatio

n an

d ta

nk

App

ropr

iate

mat

eria

ls o

f con

stru

ctio

nfo

r the

tank

and

pip

ing

Bun

d fo

r con

tain

ing

any

spills

Slo

w c

losi

ng v

alve

s to

avo

id w

ater

ham

mer

Pro

cedu

res

and

oper

ator

trai

ning

No

7.

Aci

d ta

nk fa

ilsU

ndilu

ted

hydr

ochl

oric

aci

d(H

Cl)

adde

d in

to ta

nk w

ithin

suffi

cien

t wat

er

Aci

d sp

lash

es o

n ex

pose

d ta

nkco

mpo

nent

s ca

usin

g se

vere

corro

sion

Tank

dam

age;

tank

fails

Loca

l im

pact

onl

y

No

haza

rdou

s co

nseq

uenc

es to

adja

cent

indu

stria

l or r

esid

entia

lar

eas

Inte

rlock

s on

aci

d pu

mp

Bun

d fo

r con

tain

ing

any

spills

No

8.

Sam

pled

mat

eria

ls m

ayco

ntai

n ac

ids

and

chlo

rinat

edhy

droc

arbo

ns

Loss

of c

onta

inm

ent d

urin

gsa

mpl

ing

Hea

lth e

ffect

s on

ope

rato

rs

Cor

rosi

ve m

ater

ials

dam

age

drai

nsor

nea

rby

stru

ctur

es/e

quip

men

t

Loca

l im

pact

onl

y

No

haza

rdou

s co

nseq

uenc

es to

adja

cent

indu

stria

l or r

esid

entia

lar

eas

Des

ign

of s

ampl

ing

poin

ts to

min

imis

eth

e lik

elih

ood

of lo

sses

of c

onta

inm

ent

Com

plia

nce

with

Wor

kpla

ce H

azar

dous

Sub

stan

ces

cont

rols

PPE

Firs

t-aid

, eye

was

h, s

afet

y sh

ower

s

No

9.

Pip

ing

from

the

strip

per t

o th

eth

erm

al o

xidi

ser

fails

Pip

es c

an fa

il du

e to

a n

umbe

rof

reas

ons,

e.g

. des

ign

and

man

ufac

turin

g er

rors

, cor

rosi

onan

d im

pact

Loss

of c

onta

inm

ent o

f air

stre

amco

ntai

ning

the

grou

ndw

ater

impu

ritie

s. P

oten

tial t

o im

pact

peop

le a

nd th

e en

viro

nmen

t

Des

ign

and

prev

enta

tive

mai

nten

ance

proc

edur

es

Pip

e pr

otec

tion

from

traf

fic e

tc

Yes

Page 34: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinn

acle

Ris

k M

anag

emen

t

Oric

a G

TP

PH

A R

epor

t Rev

F.D

oc8

Nov

embe

r 20

0425

Item

No.

Even

tC

ause

sPo

ssib

le C

onse

quen

ces

Prev

entio

n/Pr

otec

tion

Ris

kQ

uant

ified

?

10.

N

atur

al g

as je

tfir

e im

ping

ing

onst

ored

reco

vere

dw

aste

ED

Cliq

uid,

oth

erpl

ant a

nd/o

rpe

ople

Rup

ture

of n

atur

al g

as li

ne o

rle

ak fr

om v

alve

s w

hich

isig

nite

d

Fire

; pos

sibl

e B

LEV

E, o

f the

reco

vere

d w

aste

ED

C li

quid

Isot

aine

r; pl

ant d

amag

e, in

jury

tope

ople

, env

ironm

enta

l im

pact

s

Des

ign

and

mai

nten

ance

pro

cedu

res

for t

he n

atur

al g

as s

uppl

y lin

e

Fire

pro

tect

ion

equi

pmen

t for

emer

genc

y re

spon

se

Isol

atio

n of

the

natu

ral g

as li

ne a

t the

plan

t bat

tery

lim

it

Yes

11.

N

atur

al g

assu

pply

line

failu

re

Mec

hani

cal i

mpa

ct

Cor

rosi

on

Wel

d de

fect

Gas

ket l

eak

Rel

ease

of n

atur

al g

as to

atm

osph

ere

- fire

, UVC

E if

igni

ted

EIV

(em

erge

ncy

isol

atio

n va

lve)

on

feed

line

to p

lant

plu

s m

anua

l iso

latio

ns

Des

ign

and

prev

enta

tive

mai

nten

ance

proc

edur

es fo

r the

nat

ural

gas

sup

ply

line

Fire

pro

tect

ion

equi

pmen

t for

emer

genc

y re

spon

se

Yes

12.

G

as b

urne

rm

anag

emen

tsy

stem

fails

unsa

fely

Non

-com

plia

nce

with

gas

supp

lier r

equi

rem

ents

Com

pone

nt fa

ilure

s

Exp

losi

on in

side

the

ther

mal

oxi

dise

r

Fata

litie

s or

inju

ries

to o

pera

tors

;pl

ant d

amag

e

The

ther

mal

oxi

dise

r bur

ner

man

agem

ent s

yste

m is

to c

ompl

y w

ithth

e A

GL

requ

irem

ents

Yes

13.

Th

e th

erm

alox

idis

er d

oes

not o

pera

teco

rrect

ly

Bur

ner p

robl

ems,

poo

r gas

qual

ity, b

urne

r man

agem

ent

syst

em fa

ulty

, air

flow

prob

lem

s, fe

ed fl

ow p

robl

ems

Off

gas

is n

ot in

spe

cific

atio

n

Pot

entia

l chr

onic

and

acu

te e

ffect

son

peo

ple

and

the

envi

ronm

ent

Stri

ppin

g sh

ould

not

occ

ur if

the

ther

mal

oxi

dise

r is

not o

pera

ting

corre

ctly

, e.g

. cor

rect

tem

pera

ture

or

fan

stop

ped.

Ala

rms

and

trips

are

prop

osed

to b

e us

ed to

avo

id in

corr

ect

oper

atio

n

Yes

–ch

roni

cef

fect

sas

sess

ed in

hum

anhe

alth

risk

asse

ssm

ent

Page 35: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinn

acle

Ris

k M

anag

emen

t

Oric

a G

TP

PH

A R

epor

t Rev

F.D

oc8

Nov

embe

r 20

0426

Item

No.

Even

tC

ause

sPo

ssib

le C

onse

quen

ces

Prev

entio

n/Pr

otec

tion

Ris

kQ

uant

ified

?

14.

P

lant

wat

erdi

scha

rges

not

in c

ompl

ianc

ew

ith E

PA

licen

ces

Inad

equa

te w

ater

was

te s

tream

mon

itorin

gW

ater

dis

char

ges

are

not i

nsp

ecifi

catio

n; E

PA is

sues

No

haza

rdou

s co

nseq

uenc

es to

adja

cent

indu

stria

l or r

esid

entia

lar

eas

Ala

rms

and

trips

are

pro

pose

d to

be

used

to a

void

inco

rrect

ope

ratio

nN

o

15.

U

ndet

ecte

dre

leas

e of

ED

Cet

c fro

m s

tack

Hol

e in

ther

mal

oxi

dise

r fee

d /

prod

uct h

eat e

xcha

nger

allo

win

g un

treat

ed v

apou

r int

oth

e th

erm

al o

xidi

ser e

xhau

st

Noz

zle

re-in

ject

ing

reco

vere

dw

aste

ED

C li

quid

mal

func

tions

and

reco

vere

d w

aste

ED

Cliq

uid

is n

ot fu

lly a

tom

ised

Env

ironm

enta

l im

pact

Effe

ct o

n pe

ople

If no

zzle

mal

func

tions

, lar

ger t

han

desi

gn li

quid

ED

C d

ropl

ets

will

ent

erth

e th

erm

al o

xidi

ser

Onl

ine

stac

k m

onito

ring

of E

DC

Noz

zle

is s

elf-c

lean

ing.

Noz

zle

back

pre

ssur

e w

ill be

mon

itore

dto

che

ck fo

r nor

mal

ope

ratio

n. R

egul

arvi

sual

obs

erva

tion

and

rout

ine

mai

nten

ance

Any

unb

urnt

ED

C w

ill be

con

dens

edpo

st th

e qu

ench

uni

t and

ther

efor

e no

tre

leas

ed to

atm

osph

ere.

The

cond

ense

d E

DC

will

be

pres

ent i

n th

eac

id a

bsor

ber a

nd c

aust

ic s

crub

ber

blee

d st

ream

s w

hich

are

man

ually

and

auto

mat

ical

ly m

onito

red

Yes

–ch

roni

cef

fect

sas

sess

ed in

hum

anhe

alth

risk

asse

ssm

ent

16.

In

adeq

uate

scru

bbin

gC

ircul

atio

n pu

mp

failu

re

Floo

ding

of a

bsor

ber o

rsc

rubb

er

Inad

equa

te le

vel o

f circ

ulat

ing

abso

rben

t che

mic

als

Loss

of l

iqui

d fro

m a

bsor

ber o

rsc

rubb

er

Atm

osph

eric

em

issi

ons

toen

viro

nmen

t exc

eed

licen

cere

quire

men

ts

Env

ironm

enta

l im

pact

, effe

ct o

npe

ople

Pla

nt s

hutd

own

due

to a

larm

s an

d tri

pson

the

abso

rbin

g an

d sc

rubb

ing

tow

ers,

e.g.

diff

eren

tial p

ress

ure

for f

lood

ing,

corre

ct re

flux

flow

rate

s, p

ump

oper

atio

n, s

ump

leve

l, ci

rcul

atin

g liq

uid

com

posi

tion

Onl

ine

HC

l mea

sure

men

t

Yes

Page 36: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinn

acle

Ris

k M

anag

emen

t

Oric

a G

TP

PH

A R

epor

t Rev

F.D

oc8

Nov

embe

r 20

0427

Item

No.

Even

tC

ause

sPo

ssib

le C

onse

quen

ces

Prev

entio

n/Pr

otec

tion

Ris

kQ

uant

ified

?

17.

Fa

ilure

of

abso

rber

Inad

equa

te o

pera

tion

ofqu

ench

resu

lting

in h

igh

tem

pera

ture

exi

t gas

to th

e ac

idab

sorb

er

Hig

h te

mpe

ratu

re le

ads

to fa

ilure

of

the

acid

abs

orbe

r res

ultin

g in

rele

ase

of to

xic

gase

s

Trip

pro

tect

ion

for q

uenc

h fa

ilure

Yes

18.

Lo

ss o

fco

ntai

nmen

tfro

m th

ere

cove

red

was

teE

DC

liqu

idst

orag

e or

deliv

ery

syst

emto

ther

mal

oxid

iser

Mec

hani

cal f

ailu

res

Hum

an e

rror

Bun

d fir

e if

igni

ted

lead

ing

to d

amag

eto

the

grou

ndw

ater

trea

tmen

t pla

nt,

inju

ry to

peo

ple,

toxi

c co

mbu

stio

npr

oduc

ts.

BLE

VE

of t

he Is

otai

ner i

sa

poss

ibili

ty

Isot

aine

rs a

re o

f rob

ust d

esig

n an

d ar

eus

ed fo

r tra

nspo

rting

milli

ons

of to

nnes

of D

ange

rous

Goo

ds e

ach

day

thro

ugho

ut th

e w

orld

Haz

ardo

us a

rea

desi

gn a

ndpr

ecau

tions

Fire

figh

ting

equi

pmen

t and

pro

cedu

res

Loca

tion

is a

dequ

atel

y pr

otec

ted

from

vehi

cle

impa

ct

Slo

ped

bund

floo

r (dr

ain

liqui

d aw

ayfro

m th

e Is

otai

ner)

Yes

19.

C

orro

sion

of

pipe

fitti

ngs

Wet

reco

vere

d w

aste

ED

Cliq

uid

hydr

olys

es in

sto

rage

,cr

eatin

g ac

idic

chl

orid

es

Cor

rosi

on o

f pip

es e

tc re

sults

inle

aks

or d

amag

e to

pla

nt o

r Iso

tain

erA

ppro

pria

te m

ater

ials

of c

onst

ruct

ion

Spi

ll co

ntai

nmen

tY

es –

par

tof

Item

No.

18

20.

B

LEV

EP

ool f

ire in

bun

d en

gulfs

reco

vere

d w

aste

ED

C li

quid

Isot

aine

r

Tran

spor

t tru

ck fi

re

Fire

ball

Pos

sibl

e fa

talit

ies

and

dam

age

tosu

rrou

ndin

g ar

ea

As

per I

tem

No.

18

Yes

Page 37: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinn

acle

Ris

k M

anag

emen

t

Oric

a G

TP

PH

A R

epor

t Rev

F.D

oc8

Nov

embe

r 20

0428

Item

No.

Even

tC

ause

sPo

ssib

le C

onse

quen

ces

Prev

entio

n/Pr

otec

tion

Ris

kQ

uant

ified

?

21.

E

xpos

ure

toch

lorin

ated

hydr

ocar

bon

mix

ture

in th

ere

cove

red

was

teE

DC

liqu

idIs

otai

ner

Rec

over

ed w

aste

ED

C li

quid

cont

aini

ng V

C le

aks

from

hos

ew

hen

disc

onne

cted

from

Isot

aine

r

Per

sonn

el in

jury

from

toxi

c fu

mes

;pr

olon

ged

/ fre

quen

t exp

osur

e co

uld

caus

e ch

roni

c pr

oble

ms,

e.g

. can

cer,

from

VC

Loca

l im

pact

s on

ly

No

haza

rdou

s co

nseq

uenc

es to

adja

cent

indu

stria

l or r

esid

entia

lar

eas

Com

plia

nce

with

Wor

kpla

ce H

azar

dous

Sub

stan

ces

cont

rols

Cor

rect

PPE

Cor

rect

pro

cedu

res

and

mai

nten

ance

Firs

t Aid

Dry

bre

ak c

oupl

ing

to b

e pr

ovid

ed a

tho

se c

onne

ctio

n fro

m Is

otai

ner t

oun

load

ing

pum

p

No

22.

S

tore

d sp

ent

carb

on s

tore

dca

tche

s fir

e

Spe

nt c

arbo

n be

com

espy

roph

oric

Pla

nt d

amag

e; to

xic

fum

es

Loca

l im

pact

onl

y

No

haza

rdou

s co

nseq

uenc

es to

adja

cent

indu

stria

l or r

esid

entia

lar

eas

Pro

cedu

res

for w

aste

han

dlin

g, s

tora

gean

d di

spos

al

Fire

figh

ting

No

23.

O

ff sp

ecifi

catio

ndi

scha

rge

from

strip

ped

wat

ertre

atm

ent u

nits

(e.g

. rev

erse

osm

osis

(RO

)un

it)

RO

failu

re a

nd b

reak

thro

ugh,

prob

lem

with

iron

rem

oval

uni

t,io

n-ex

chan

ge o

pera

tion,

activ

ated

car

bon

or b

reak

poin

tch

lorin

atio

n (e

.g. c

hlor

ides

,iro

n, a

mm

onia

etc

)

Dis

char

ge o

ff sp

ecifi

catio

n to

cust

omer

s, B

otan

y B

ay o

rst

orm

wat

er

Loca

l im

pact

onl

y

No

haza

rdou

s co

nseq

uenc

es to

adja

cent

indu

stria

l or r

esid

entia

lar

eas

Mai

nten

ance

of p

roce

ss u

nits

Onl

ine

mea

sure

men

t of k

ey p

aram

eter

s

Ala

rms

on a

bnor

mal

ope

ratio

n of

proc

ess

units

Buf

fer t

ank

for s

tora

ge c

apac

itanc

e

Pla

nt w

ill be

put

into

recy

cle

mod

eup

on d

etec

tion

of p

robl

em

No

24.

Tr

ansp

ort

Inci

dent

Bad

wea

ther

con

ditio

ns

Driv

er e

rror

Truc

k fa

ilure

Pot

entia

l for

loss

of c

onta

inm

ent o

fre

cove

red

was

te E

DC

liqu

id.

This

can

igni

te w

ith im

pact

to p

eopl

e an

dth

e en

viro

nmen

t

Rec

over

ed w

aste

ED

C li

quid

trans

porte

d in

inte

rnat

iona

lly u

sed

Isot

aine

rs

DG

regu

latio

ns fo

r driv

ers

and

truck

s

Em

erge

ncy

resp

onse

for a

ccid

ents

No

- see

Sec

tion

7.4

Page 38: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinn

acle

Ris

k M

anag

emen

t

Oric

a G

TP

PH

A R

epor

t Rev

F.D

oc8

Nov

embe

r 20

0429

Item

No.

Even

tC

ause

sPo

ssib

le C

onse

quen

ces

Prev

entio

n/Pr

otec

tion

Ris

kQ

uant

ified

?

25.

A

irpla

ne C

rash

Bad

wea

ther

con

ditio

ns

Pilo

t erro

r

Pla

ne fa

ilure

Des

truct

ion

of th

e gr

ound

wat

ertre

atm

ent p

lant

. Fi

re fr

om th

e pl

ane

cras

h w

ould

be

mor

e ha

zard

ous

than

prop

agat

ion

even

ts fr

om th

ede

stro

yed

grou

ndw

ater

trea

tmen

tpl

ant

Site

is n

ot u

nder

maj

or fl

ight

pat

hs

Air

indu

stry

regu

latio

ns re

gard

ing

pilo

ttra

inin

g an

d pl

ane

safe

ty

No

- see

Sec

tion

6.5

26.

S

abot

age

/Te

rror

ism

Dis

grun

tled

empl

oyee

or

intru

der

Pos

sibl

e re

leas

e of

gro

undw

ater

,re

cove

red

was

te E

DC

liqu

id, H

Cl,

caus

tic s

oda

or o

ther

Dan

gero

usG

oods

with

con

sequ

ence

s as

abo

ve

Rec

over

ed w

aste

ED

C li

quid

rele

ases

hav

e th

e po

tent

ial t

o im

pact

on a

djac

ent r

esid

entia

l are

as

The

grou

ndw

ater

trea

tmen

t pla

nt is

tobe

loca

ted

on th

e B

IP.

Exi

stin

gse

curit

y m

easu

res

are

deta

iled

inS

ectio

n 4.

3

Con

sequ

en-

ces

of w

aste

ED

C li

quid

rele

ases

anal

ysed

as

per I

tem

Nos

18

to21

27.

In

cide

nt o

nad

jace

nt p

lant

sE

mer

genc

y si

tuat

ions

resu

lting

in fi

res,

exp

losi

ons

or to

xic

rele

ases

Fire

s an

d ex

plos

ions

hav

e th

epo

tent

ial t

o pr

opag

ate

the

inci

dent

on

the

grou

ndw

ater

trea

tmen

t pla

nt.

Toxi

c ga

ses

may

be

draw

n in

to th

eai

r int

ake

to th

e st

rippe

rs

The

cons

eque

nces

of p

ropa

gatio

nev

ents

in th

e G

TP a

re a

s pe

r the

abov

e G

TP s

cena

rios

Dem

atch

ed s

ite re

duce

s ch

ance

s of

igni

tion

Cla

ssifi

ed h

azar

dous

are

as re

duce

chan

ces

of ig

nitio

n

Em

erge

ncy

resp

onse

act

ions

tosh

utdo

wn

the

grou

ndw

ater

trea

tmen

tpl

ant p

lus

deal

with

the

emer

genc

y at

the

adja

cent

pla

nt

As

abov

e

Page 39: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinn

acle

Ris

k M

anag

emen

t

Oric

a G

TP

PH

A R

epor

t Rev

F.D

oc8

Nov

embe

r 20

0430

Item

No.

Even

tC

ause

sPo

ssib

le C

onse

quen

ces

Prev

entio

n/Pr

otec

tion

Ris

kQ

uant

ified

?

28.

Lo

ss o

fco

ntai

nmen

t of

grou

ndw

ater

from

the

pipi

ngsy

stem

supp

lyin

g th

eG

TP, i

nclu

ding

DN

AP

L, p

rimar

yan

d se

cond

ary

cont

ainm

ent

pipe

lines

Third

par

ty a

ctiv

ity, e

.g.

exca

vato

r, es

peci

ally

in n

on-

BIP

are

as

Sab

otag

e

Pip

ing

and

equi

pmen

t fai

lure

s,e.

g. c

orro

sion

Loss

of c

onta

inm

ent o

f gro

undw

ater

at th

e le

ak p

oint

. Lo

cal i

mpa

ct o

nly.

Pot

entia

l to

affe

ct th

ose

who

are

expo

sed

to th

e gr

ound

wat

er.

Odo

urw

ill oc

cur l

ocal

ly

No

haza

rdou

s co

nseq

uenc

es to

adja

cent

indu

stria

l or r

esid

entia

lar

eas

Pip

ing

and

equi

pmen

t to

be p

ositi

oned

and

secu

red

to m

inim

ise

loss

es o

fco

ntai

nmen

t. S

igna

ge to

war

n of

unde

rgro

und

pipe

s an

d eq

uipm

ent

Sec

onda

ry c

onta

inm

ent w

ith le

akde

tect

ion

Incl

uded

in ‘D

ial B

efor

e Y

ou D

ig’

syst

em

No

29.

Lo

ss o

fco

ntai

nmen

tfro

m p

ipin

g an

dpu

mpi

ng o

f sal

tyw

aste

wat

er

Third

par

ty a

ctiv

ity, e

.g.

exca

vato

r, es

peci

ally

in n

on-

BIP

are

as

Sab

otag

e

Pip

ing

and

equi

pmen

t fai

lure

s,e.

g. c

orro

sion

As s

alty

was

te w

ater

mee

ts A

NZE

CC

disc

harg

e gu

idel

ines

, the

re w

ill be

min

imal

impa

ct to

wat

erw

ays

from

any

leak

age

No

haza

rdou

s co

nseq

uenc

es to

adja

cent

indu

stria

l or r

esid

entia

lar

eas

Pip

ing

and

equi

pmen

t to

be p

ositi

oned

and

secu

red

to m

inim

ise

loss

es o

fco

ntai

nmen

t.

Sig

nage

to w

arn

of u

nder

grou

nd p

ipes

and

equi

pmen

t. In

clud

ed in

‘Dia

l Bef

ore

You

Dig

’ sys

tem

Exi

stin

g pi

pelin

e is

in g

ood

cond

ition

.P

ipe

will

be p

last

ic li

ned

befo

re u

se to

incr

ease

inte

grity

for s

alty

was

te w

ater

No

30.

Lo

ss o

fco

ntai

nmen

tfro

m tr

eate

dgr

ound

wat

erdi

strib

utio

nsy

stem

Third

par

ty a

ctiv

ity, e

.g.

exca

vato

r

Sab

otag

e

Pip

ing

and

equi

pmen

t fai

lure

s,e.

g. c

orro

sion

As

wat

er is

trea

ted,

wor

st o

utco

me

islo

cal f

lood

ing/

eros

ion.

No

heal

th o

ren

viro

nmen

tal i

mpa

cts.

No

haza

rdou

s co

nseq

uenc

es to

adja

cent

indu

stria

l or r

esid

entia

lar

eas

Rou

tine

mai

nten

ance

of p

ipin

g sy

stem

Pip

ing

on th

e B

IP s

ite w

ithin

sec

urity

fenc

ing

Exc

avat

ion

perm

its p

rior t

o an

yun

derg

roun

d w

ork

on th

e B

IP

No

Page 40: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinn

acle

Ris

k M

anag

emen

t

Oric

a G

TP

PH

A R

epor

t Rev

F.D

oc8

Nov

embe

r 20

0431

Item

No.

Even

tC

ause

sPo

ssib

le C

onse

quen

ces

Prev

entio

n/Pr

otec

tion

Ris

kQ

uant

ified

?

31.

R

elea

se o

fch

lorin

eM

ixin

g of

sto

red

acid

ic li

quid

s(e

g hy

droc

hlor

ic a

cid

or S

MB

S)

with

sto

red

sodi

um h

ypoc

hlor

iteso

lutio

n

Sm

all q

uant

ities

of c

hlor

ine

gas

evol

ved

with

pot

entia

l im

pact

on

loca

lop

erat

ors

No

haza

rdou

s co

nseq

uenc

es to

adja

cent

indu

stria

l or r

esid

entia

lar

eas

Aci

dic

liqui

ds a

re s

tore

d (b

unde

d) a

ndha

ndle

d se

para

tely

from

sod

ium

hypo

chlo

rite

solu

tion,

ther

eby

prev

entin

g an

y m

ixin

g

No

com

mon

dra

ins

exis

t whe

re m

ixin

gof

spi

lls c

an o

ccur

No

32.

R

elea

se o

fco

rrosi

vech

emic

als

Pip

ing,

ves

sels

or e

quip

men

tfa

ilure

s eg

cor

rosi

on

Ope

rato

r err

or

Loca

l im

pact

onl

y

No

haza

rdou

s co

nseq

uenc

es to

adja

cent

indu

stria

l or r

esid

entia

lar

eas

Sto

rage

tank

s in

side

sui

tabl

y si

zed

bund

s

Gen

eral

pla

nt a

rea

spills

are

col

lect

edan

d se

nt to

effl

uent

trea

tmen

t

No

Page 41: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200432

4.3 SAFETY MANAGEMENT SYSTEMS

Safety management systems are intended to minimise the risk from potentiallyhazardous installations by a combination of hardware (i.e. design) and softwarefactors (managements systems such as procedures, policies, plans, trainingetc). To ensure safe operation of the groundwater treatment plant, both thehardware and the software systems must be of high standard.

Orica personnel, having operated chlorine and derivative plants for manydecades at the Botany site, are well aware of the hazardous nature of proposeddangerous goods and hazardous materials. However, it is acknowledged thatthe proposed groundwater treatment plant will necessitate changes to theexisting safety management system.

In general, the Orica procedures, guidelines etc are modified to suit the localsite conditions where required. It is noted that the Orica safety managementsystem is widely used by many companies. The system is typically bought bythese companies and then modified to suit their specific requirements. As such,it is widely regarded as a well prepared, robust system which is suitable for themanagement of safety, health and environmental issues for hazardous industry.

The safety management system is built in layers, as shown below:

Vision and Values

Policy

SH&E Standards

SH&E Model Procedures

Local SH&E Procedures

Operating Procedures / Work Instructions

Senior management define the company's Vision and Values, SH&E Policy andStandards. For Orica and its group companies, there are 19 SH&E Standardsto be followed. SH&E Model Procedures are developed to further detail keyrequirements for control of the company's operations and to provide a model formanagement of SH&E risks and for implementation of the company's SH&EPolicy and Standards. There are currently 117 SH&E Model Procedures.

Local procedures and work instructions are developed to define additionalrequirements to, as far as practicable, control the risks arising from theoperation of each facility and to assure compliance with the company's statutoryOH&S obligations, SH&E Policy and Standards, and the key requirements ofthe SH&E Model Procedures.

The range and detail of local procedures and work instructions are consistentwith the complexity of the activity, the level of risk involved, and the skills andtraining of the people performing the activity.

Page 42: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200433

Sufficient records of activities and events are retained in a secure, retrievablemanner to demonstrate compliance with, and the effectiveness of, localprocedures and to capture adequate information regarding the company'sSH&E impacts.

A Letter of Assurance is prepared annually, detailing the level of compliancewith each of the SH&E Standards and action plans to close any gaps.

The suitability, adequacy and effectiveness of the company SH&E Policy,Standards and SH&E Model Procedures and local SH&E management systemsis reviewed at least every two years.

Given the strength and reputation of the Orica safety management systems, it isexpected that modification of the existing safety management system toaccommodate the groundwater treatment plant should present little difficulty.

For information, some of the more critical procedures associated with the safetymanagement systems at Orica are summarised in Table 5.

Page 43: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200434

Table 5 - Summary of Safety Related Procedures

PROCEDURE PURPOSE

Operations andMaintenance Manuals

To clearly define the method of operations of the plant.

To ensure that accurate information about important aspects of the plantdesign and its operations are available and up to date.

To define for the operators and maintenance team the methods by whichsections of the plant may be safely and efficiently withdrawn from service,repaired and restored to safe efficient operating condition.

To ensure that protective systems are in a good state of repair and functionreliably when required. This includes scheduled testing of trips and alarmsand relief devices.

Operator Training,including safety andemergency training

To enable operators to run the plant to meet objectives safely.

To enable trades personnel to carry out maintenance work so that they arethemselves safe and do not jeopardise the plant safety systems or thesafety of others.

To provide personnel with an understanding of possible hazardoussituations and the ability to respond appropriately.

To provide an understanding of and practice in the use of basic emergencyequipment that might be needed in tackling an emergency (e.g. self-contained breathing apparatus, safety showers).

Permit to Work To safeguard tradesmen (and others) and the plant by ensuring than thatthe plant is safe to work on, that the correct job is done using the rightequipment, that any safety procedures are understood and adhered to, thatoperators know which parts of the plant are being worked on and that theplant is returned to safe condition before being returned to service.

Control of PlantModifications

To ensure that proposed changes to both equipment and operating methodsachieve the desired benefits without any unforeseen and undesirable sideeffects.

Unusual IncidentReporting andInvestigation

To learn from "unusual incidents" that may or may not have had ahazardous outcome, but could have under different circumstances, to beproactive in preventing their occurrence.

EmergencyProcedures

To facilitate effective response to emergencies. To prevent or minimise theeffect of potentially hazardous events by being prepared.

ScheduledManagement Auditingof Procedures

To ensure that operating management is continually aware of how well thedefined procedures and systems affecting safety and loss prevention arebeing followed in practice. To enable corrective action to be taken toimprove adherence to procedures.

Page 44: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200435

Safety management systems in addition to those listed in Table 4 and Table 5are described below.

a. Security

The BIP site is a secure site, with all vehicle entry controlled through thegatehouse at Gate 3, which is manned 24 hours per day.

Security of the site is also achieved by site personnel vigilance, security patrolsby security guards and cameras. In addition, the plant has lighting throughoutthe night to aid observation, the site is fully fenced (adequate construction) andnon-operating gates are locked (e.g. Gate 1 at the south-eastern corner of thesite).

All vehicle access is via Gate 3, either by presenting security passes or bysigning in as a visitor / contractor.

Security personnel are site inducted, have a checklist of areas to inspect andreport (to BIP staff) on unusual incidents.

b. Emergency Response

An audible alarm alerts plant operators if an emergency occurs. Emergencyprocedures exist for the site and are routinely tested via simulated emergencies.

c. Safety Related Procedures

In addition to those procedures listed in Table 5, the following procedures arealso critical to safe operation of the Orica plants:

Replacement materials are tightly specified to ensure the newcomponents are made from compatible materials of construction;

Hazardous areas classification are defined and the site is dematched;and

All critical service hoses are recorded on the hose register and areroutinely inspected and tested.

4.3.1 Safety Software in Risk Assessment

In risk assessments, incidents are assessed in terms of consequences andfrequencies, leading to a measure of risk. Where possible, frequency datacomes from actual experience. However, in many cases, the frequencies usedare generic, based on historical information from a variety of plants andprocesses with different standards and designs.

The quality of the management systems (known as "safety software") in place inthese historical plants will vary. Some will have little or no software, such aswork permits and modification procedures, in place. Others will have exemplarysystems covering all issues of safe operation. Clearly, the generic frequencies

Page 45: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200436

derived from a wide sample represent the failure rates of an "average plant".This hypothetical average plant would have average hardware and softwaresafety systems in place.

If an installation with below average safety software is assessed using genericfrequencies, it is likely that risk will be underestimated. Conversely, if a plant isabove average, the risk will probably be overestimated. However, it isextremely difficult to quantify the effect of software on plant safety.

Therefore, Pinnacle Risk Management adopts a policy which does not attemptto quantitatively account for the presence of and quality of software safetysystems unless specific information is available. It is assumed that the genericfailure frequencies used apply to installations which have safety softwarecorresponding to accepted industry practice. It is believed that this assumptionwill be conservative in that it will overstate the risk from well-managedinstallations such as the Orica sites. Therefore, any quantitative approach isvalid (i.e. conservative) if the safety management within the operation beingassessed is of a high standard.

Page 46: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200437

5 CONSEQUENCE ANALYSIS

GTP

The assessment of risks to both the public as well as to operating personnelaround an industrial development requires the application of the basic stepsoutlined in Section 1.4. Normally, the methodology uses probabilistic risktechniques to determine the likely incident scenarios and leak mechanisms, andthese are outlined in later sections. The use of reliable fluid outflow and jetdispersion models allows the determination of gas discharge rates andconsequent flammable gas cloud sizes.

The typical methodology attempts to take account of all credible, significanthazardous situations that may arise from the operation of processing plants.This is done by first taking a probabilistic approach to vessel and pipe failure forall vessels containing hazardous materials. Specific incidents, identified by avariety of techniques, are then added and the combined data used to generatecomposite risk contours which can be used for both the public and plantpersonnel.

"Specific incidents" are those which are felt or known to be likely in installationsof this kind. Historical information, gathered from similar installationsthroughout the world, is a major source for such incidents. Other potentialincidents (and the way in which they could develop) have been identified via theuse of Hazard Study workshops.

The consequences of an incident are calculated using standard correlations andprobit-type methods which assess the effect of fire radiation, explosionoverpressure and toxicity to an individual, depending on the type of hazard.

Having assembled data on possible incidents, risk analysis requires thefollowing general approach (for individual incidents which are then summatedfor all potential recognised incidents):

Risk = Likelihood x Consequence

In this PHA, however, the approach adopted is to assess each credible,significant hazardous event (i.e. scenario) for its consequential impacts. If it isshown that the GTP does not impose unacceptable consequential impact levelsat the site boundary then, by default, the corresponding risk is broadlyacceptable. If, however, the consequential impact at the nearest site boundaryexceeds values normally associated with irritation, injury and fatality then thescenarios are evaluated for their contribution to off-site risk.

The GTP location is a significant distance from the nearest residential area(approximately 325 m away), so many incidents would not have consequenceswhich would reach as far as this, and hence would have no effects onresidential areas.

Page 47: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200438

In addition to off-site residential risk, there is also risk imposed to theneighbouring industries on the BIP. This risk is also estimated in this PHA andcompared to the HIPAP 4 values (Ref 4). The following information details theneighbouring facilities and the distances involved.

Table 6 – GTP Neighbouring Industries

Direction Company / Area Distance, m

North Qenos (Alkatuff) 25 m (from scrubber)

East Huntsman (drum, IBC storage area) 40 m (from the thermal oxidiser)

South Huntsman (ethylene oxide and surfactants) 70 m (from the thermal oxidiser)

West Qenos (Site Utilities) 50 m (from the thermal oxidiser)

The consequence calculations for this PHA were carried out using commerciallyavailable risk software, TNO’s EFFECTS. The consequence models usedwithin EFFECTS are well known and are fully documented in the TNO YellowBook (Ref 9). A listing of the models is also given in Appendix 4. A briefexplanation and discussion of the consequence modelling follows.

Essentially, for each scenario defined by the analyst, an appropriate releaserate equation is selected based on the release situation and initial state of thematerial. Various outcomes (e.g. jet or flash fires and explosions) are thencalculated based on parameters such as the ease of ignition of the material andamount released. The consequential impact value of interest at a particularlocation can be obtained from the results.

The scenarios modelled in this PHA from Section 4 were chosen based on theability for a release from a pipe and/or vessel to cause adverse effect both on-site and off-site (fatality, injury or irritation).

Outcomes for various types of release cases are modelled within EFFECTS asshown in Table 7. For each release case, probabilities of immediate ignition(leading to jet or pool fires) and delayed ignition (leading to flash fires orexplosions) are calculated for each flammable material depending upon thematerial’s characteristics.

Page 48: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200439

Table 7 – EFFECTS Release Cases

Type of Release Subsequent Event (Note 1)

Gas release Flash fire / explosion

Jet fire

Toxic impact

Instantaneous gas release Flash fire / explosion

Toxic impact

Liquid release Flash fire / explosion

Pool fire

Toxic impact

Instantaneous liquid release Flash fire / explosion

Pool fire

Toxic impact

Pressurised liquefied gas release Flash fire / explosion

Jet fire

Pool fire

Toxic impact

Instantaneous pressurised liquefied gas release Flash fire / explosion

Toxic impact

BLEVE Fireball

Note 1: Safe dispersal is another possible outcome

Assumptions made for the purposes of consequence calculations are alsodescribed in the following sections.

Associated Operations

A different approach was required for analysing risks for the GTP Project’sassociated operations.

Transport risks of the recovered waste EDC liquid are discussed in Section 7.4.

In assessing pipeline and pumping risks it is important to note that the materialshandled (groundwater, salty waste water) are non-hazardous. Therefore risksassociated with any potential releases were confined to potential effects on thebiophysical environment. These are discussed in Section 8. Any risksassociated with these operations will be managed as an integral part of Orica’sSafety Management System which has been described in detail in Section 4.3.

Page 49: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200440

5.1 SCENARIOS MODELLED

The scenarios modelled in this PHA are shown in Table 8. These scenarioswere chosen based on the ability for a release from a pipe and/or vessel tocause significant adverse effect (fatality, injury and/or irritation). As thematerials are flammable and/or toxic, consequence modelling involves thesimulation of various types of fires, explosions and atmospheric dispersion.

Table 8 – Scenarios Modelled

Equipment Item Material Potential Impact

Pipes or vessels from the air stripper tothe plant exhaust vent

EDC or HCl in air Toxicity

Recovered waste EDC liquid Isotainerand transfer system including connectingpiping and hose

EDC Pool fires, toxicity,BLEVE

Natural gas line Natural gas (methane) Fires, UVCE, flash fire

Thermal oxidiser explosion Natural gas (methane) Confined explosion

Caustic scrubber (loss of reflux flow) HCl in air Toxicity

5.2 RELEASE SOURCES

Scenarios involving piping failures have been modelled using three failurecases, corresponding to full pipe fracture, 50mm and 13mm holes. Gasketfailure is likely to result in a gap equivalent to the area between two flange bolts.This has been modelled as a hole with an equivalent diameter of 13mm. Vesselfailures have been modelled as catastrophic rupture and leaks of 50mm, 25mmand 13mm.

These generic failure cases are comparable to those used in a number ofpublished risk assessment studies and described in Lees (Ref 10).

Where the consequential impacts from larger release cases are shown to benegligible, then the smaller release cases are not included in the modelling, asthey will have no contribution to the risk being assessed.

5.3 RELEASE RATES

Release rates were calculated for each release scenario using standardequations based on hole size, pressure, temperature and material state (withinEFFECTS). The maximum release inventory was limited to the contents of theplant equipment plus the amount lost over the duration of the leak (variabledepending on the leak rate).

In case of a liquid loss of containment, part of the material may initially flash offand evaporate, with any remaining liquid evaporating at a lower rate due to

Page 50: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200441

cooling of the liquid spill. Flash and evaporation rate calculations wereperformed by EFFECTS.

5.4 RELEASE DURATION

The assumed time taken to stop and control a release is based on a cautiousbest-estimate of a typical release scenario, rather than always the worst-case(in accordance with quantitative risk assessment principles).

The proposed plant will be highly automated and monitored by instruments. Inthe case of major upset conditions, it will be designed to trip (shut down) quicklyand safely. Many upset conditions would be expected to initiate an automatictrip within 1 minute. Other conditions could occur for approximately 5 or 20minutes depending on the scenario, speed of operator response and type ofoperator response. The selected time for each scenario is defined in therespective consequential analyses.

5.5 METEOROLOGICAL DATA

The meteorological data is comprised of six wind/weather combinations (windspeed/Pasquill stability category) have been used as the basis for all dispersioncalculations. The probability of each combination of wind/weather category andwind direction (data is split into 12 directions) is used in the calculation of clouddrift. The meteorological data used for the risk assessment is contained inAppendix 5.

5.6 TERRAIN EFFECTS AND DEGREE OF CONFINEMENT

Ground roughness affects the turbulent flow properties of wind, hencedispersion of a released material. Terrain effects are taken into account tosome degree in dispersion modelling by use of a parameter known as surfaceroughness length.

A surface roughness factor of 1m was used, corresponding to an area withdensely located low buildings or an industrial area with low structures such asthe BIP site (Ref 11).

As described in Appendix 4, EFFECTS requires the degree of confinement forexplosion calculations. Essentially, this is the proportion of the total mass in thecloud used in the explosion calculation. For example, if 2,000 kg is entered asa total mass and 50% as confinement then 1,000 kg is used in the explosioncalculation.

For this QRA, the following percentage confinement values have been used:

25% for open plant area; and

10% for the natural gas pipeline.

Page 51: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200442

5.7 CONSEQUENCE RESULTS

The following consequence results were obtained via EFFECTS (TNO)modelling.

5.7.1 Pipes or Vessels Failures Leading to Gas Plumes

Thermal Oxidiser Feed Piping

The groundwater containing the EDC and other impurities is pumped to the airstripper. A fan on the inlet to the thermal oxidiser draws air through the airstripper which contacts the groundwater to remove (strip) the impurities. Thesuction side of the fan is under vacuum. If any pipe or vessel on the fan suctionfails whilst the fan is running then air will be drawn in through the holes etc.Hence, the impurities will not be released and a plume will not form.

The discharge pressure of the fan is approximately 7.5 kPag. This low pressurewill limit the flowrate from any holes in piping and vessels. The pressure profileacross the rest of the plant drops to 0 kPag at the exit of the caustic scrubber.

To model potential releases from holes in piping and vessels, the releasedstream is modelled as air and the concentrations of the components of interestare estimated based on the concentration at the point of release.

For the stream entering the thermal oxidiser from the fan, the principlecontaminants are EDC and other chlorinated paraffins and olefins(approximately 660 ppm vol/vol). EDC is the primary contaminant. Itsconcentration in the thermal oxidiser feed is over an order of magnitude morethan the next significant group of contaminants, i.e. carbon tetrachloride andtetrachloroethene. The other components of significance are grouped asaromatics and organic acids, approximately 4 and 580 ppm (vol/vol),respectively. The total flow is 2,124 kmols/hr (50,300 scm/hr).

Neutral gas dispersion modelling of this stream at weather / wind conditions, F2,(i.e. stable weather conditions, F class, and a wind speed of 2 m/s; typicalconditions for modelling maximum plume travel) shows that the ground levelERPG 1 EDC concentration (50 ppm) for a release from a 50 mm hole (at 7.5kPag) travels no further than approximately 10 m from the point of release. Thisis for a release duration of 20 minutes.

The reasons for these relatively short distances are due to the lowconcentration of contaminants in the air stream and the low driving force forleaks from a 50 mm hole. Based on these results, further assessment ofreleases from potential holes (50 mm and smaller) in the thermal oxidiser feedpiping is not warranted, i.e. there will be no fatalities, injuries or irritation effectseither on-site or off site. It is reasonable to expect odour only near the plant.

Note that EDC is the chosen material of interest for the dispersion modelling asits concentration is the highest and its ERPG values are amongst the lowest ofthe stream components.

Page 52: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200443

For catastrophic pipe failure on the discharge of the fan, where the entireflowrate through the fan is released, it is assumed that for such major plantdisturbances, the automatic plant system will detect such conditions (i.e. no flowthrough the plant) and automatically shutdown the process, including the fanbefore the thermal oxidiser. Therefore, a release duration of 1 minute ischosen.

Neutral gas dispersion modelling of this full flow stream at the various weather /wind conditions (Appendix 5) is shown in Table 9.

Table 9 – Thermal Oxidiser Feed Stream, Dispersion Modelling

Wind / Weather Pattern Maximum Distance (m) to:

EDC ERPG 1 EDC ERPG 2

2.3 B 66 31

3.8 D 110 49

5.3 D 89 39

2.3 E 230 92

0.9 F 440 230

2.3 F 390 160

As can be seen from the above data, potential off-site residential impact (i.e. at325 m or further) is limited to a few combinations of wind / weather patterns. Noinjury risk is predicted (i.e. by use of ERPG 2). Irritation (ERPG 1) is onlyexpected for the more stable wind / weather patterns. The risk of EDC irritationimpact on off-site personnel is assessed later in Section 7 of this report.

Note that there are no criteria set in HIPAP 4 (Ref 4) for the risk of toxic irritationand/or injury to personnel on adjacent industrial facilities, i.e. other BIP landusers or neighbouring industries.

With respect to the probability of fatality from releases from the thermal oxidiserfeed pipe, the HSE UK land use planning (LUP) SLOT (specified level oftoxicity) value for EDC is used as there is no widely accepted fatality probitavailable for EDC, the main component of interest.

The HSE has defined the LUP SLOT as:

Severe distress to almost every one in the area;

Substantial fraction of exposed population requiring medical attention;

Some people seriously injured, requiring prolonged treatment; and

Page 53: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200444

Highly susceptible people possibly being killed.

Typically, SLOT values define the toxic load which will result in 1% mortality(Ref 10).

These criteria are fairly broad in scope, reflecting the fact that:

1) There is likely to be considerable variability in the responses of differentindividuals affected by a major accident;

2) There may be pockets of high and low concentrations of a toxic substance inthe toxic cloud release, so that not everyone will get exactly the same degree ofexposure; and

3) The available toxicity data are not usually adequate for predicting precisedose-response effects.

The toxicity expressed by a given substance in the air is influenced by twofactors, the concentration in the air (c) and the duration of exposure (t). Afunctional relationship between c and t can be developed, such that the endproduct of this relationship is a constant:

f(c,t) = constant

This constant is known as the Toxic Load. In the HSE, the Toxic Load relatingto the LUP SLOT is known as the SLOT Dangerous Toxic Load or SLOT DTL.For a number of gases, the relationship between c and t is simple:

Toxic Load = c x t

This relationship is sometimes known as the Haber law. As an example, animaltoxicity data for methyl isocyanate indicates that the LUP SLOT is produced byeach of these c and t pairs:

t (min) 5 10 30 60 120

c (ppm) 150 78 25 12 6

In this example the constant, or SLOT DTL, is 750 ppm.min (that is 150 x 5, 25x 30, etc.).

However, the equation c x t = constant does not apply to all substances, so thefollowing general equation has been developed:

Toxic Load = cn.t

For methyl isocyanate, n in the cn.t relationship is 1. In the case of sulphurdioxide, n = 2 and animal toxicity data suggest that the following pairs of c and twill each produce the LUP SLOT:

Page 54: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200445

t (min) 5 10 30 60 120

c (ppm) 965 682 394 279 197

Here, the constant, or SLOT DTL, is 4.6 x 106 ppm2.min (that is 9652 x 5 or3942 x 30).

For EDC, the HSE LUP SLOT constant is 9 x 104 (where n = 1).

For a 1 minute release, the concentration of EDC in air required to effect aprobability of fatality of 1% is 90,000 ppm. For a 5 minute release, thecorresponding concentration is 18,000 ppm. For a 20 minute release, theconcentration further decreases to 4,500 ppm. As the concentration of EDC inthe thermal oxidiser feed stream is less than these values (up to 660 ppm) thenno fatalities either on-site or off-site from piping failures can be expected, i.e.there is no fatality risk from these dilute emissions.

Thermal Oxidiser Outlet Piping

The stream from the thermal oxidiser has the following flow / composition data:

Mass Rates: Molar Rates:(kmol/hr)

PPM:(vol/vol)

Total Flow 66,070 sm3/hr 2,793 NA

VOCs (as EDC) 1.46 kg/hr 0.014 5

HCl 114.6 kg/hr 4.32 1,550

Chlorine 2.25 kg/hr 0.032 11

Notes: VOC = volatile organic carbon

Neutral gas dispersion modelling of this stream at weather / wind conditions, F2,shows that the ERPG 1 HCl concentration (3 ppm) for a release from a 50 mmhole (at 5 kPag) travels no further than approximately 55 m from the point ofrelease. This is for a release duration of 20 minutes.

Again, the reason for these relatively short distances are due to the lowconcentration of contaminants in the air stream and the low driving force (lowpressure in the pipe) for leaks from a 50 mm hole (estimated leak rate is 0.11kg/s, i.e. relatively low). These calculations show that for the scenariosconsidered, the concentrations to cause concern (ERPG 1) only occur close tothe release, and do not travel far enough to effect neighbouring plants orresidences. Based on these results, further modelling of releases from potentialholes in the thermal oxidiser outlet piping is not warranted, i.e. there will be nofatalities, injuries or irritation effects off site.

Again, note that there are no criteria set in HIPAP 4 (Ref 4) for the risk of toxicirritation and/or injury to personnel on adjacent industrial facilities, i.e. other BIPland users or neighbouring industries.

Page 55: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200446

For catastrophic pipe failures (i.e. full bore pipe ruptures) where the entireflowrate from the thermal oxidiser is released, again, it is assumed that for suchmajor plant disturbances, the automatic plant system will detect such conditions(i.e. no flow through the plant) and automatically shutdown the process,including the fan before the thermal oxidiser. Therefore, a release duration of 1minute is again chosen.

Neutral gas dispersion modelling of this full flow stream at the various weather /wind conditions (Appendix 5) is shown in Table 10. The component of interestis HCl (chlorine is present at a low concentration, i.e. 11 ppm).

Table 10 – Thermal Oxidiser Exit Stream, Dispersion Modelling

Wind / Weather Pattern Maximum Distance (m) to:

HCl ERPG 1 HCl ERPG 2

2.3 B 310 100

3.8 D 560 180

5.3 D 490 150

2.3 E 860 360

0.9 F 1,330 610

2.3 F 1,310 580

As can be seen from the above data, off-site impact (i.e. at 325 m or further) isexpected to occur for most of the wind / weather pattern combinations. The riskof HCl irritation and injury impact on off-site personnel is assessed later inSection 7 of this report.

For HCl, the HSE LUP SLOT constant is 2.37 x 104 (where n = 1).

For a 1 minute release, the concentration of HCl in air required to effect aprobability of fatality of 1% is 23,700 ppm. For a 5 minute release, thecorresponding concentration is 4,740 ppm. For a 20 minute release, theconcentration further decreases to 1,185 ppm. As the concentration of HCl inthe thermal oxidiser exit stream is 1,550 ppm and full flow release durations willbe limited to approximately one minute then given dispersion and consequentdilution of this stream no fatalities either on-site or off-site from the thermaloxidiser exit piping failures can be expected, i.e. there is no fatality risk fromthese HCl emissions.

5.7.2 Recovered Waste EDC Liquid Isotainer Events

EDC has a low vapour pressure at ambient temperature (8 kPa at 20oC). Thenormal boiling point is 83oC. Accidental releases of recovered waste EDC liquid

Page 56: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200447

from the Isotainer (e.g. due to vessel or pipe failure, or operator error) will formpools in the bunded area where the Isotainer is kept. Over time, EDC vapourwill be evolved and hence a toxic plume is generated. If the pools are ignited,however, a pool fire will result.

Prolonged pool fires could propagate to a BLEVE (boiling liquid expandingvapour explosion). For this event, the heat from the pool fire raises thetemperature and hence pressure of the Isotainer contents. The Isotainer safetyrelief valve will lift and material will escape. The Isotainer wall mechanicalintegrity will reduce over time, particularly where flames impact on the vesselwall where EDC vapours are present on the inside. At some point in time(historically greater than 10 minutes), the vessel wall will fail and a catastrophic(i.e. vessel rupture) release will occur. The released material vaporisesinstantaneously, ignites and forms a fireball.

The event generates three significant consequential effects:

Radiant heat from the fireball;

Explosion overpressures from the release of stored energy; and

Missiles can be formed.

As the distance to significant radiant heat levels exceeds the distance tosignificant levels of overpressure, BLEVE models estimate radiant heat onlydue to its dominance in effect distances.

Data used for calculating the effects from an Isotainer BLEVE (fireball) are asfollows:

Quantity of recovered waste EDC liquid 20 te

Vessel burst pressure (PSV setting) 16.8 barg

Recovered waste EDC liquid temperature 20oC

The results from the BLEVE simulation show that the predicted fireball radius isapproximately 81 m. Therefore, the radiant heat levels from a potentialrecovered waste EDC liquid Isotainer BLEVE will not cause any significantimpact on residential occupants but may impact (burn) personnel involved inrunning the GTP and adjacent plants on the BIP (i.e. on-site fatality may occur).The risk of this occurring is analysed in Section 7.

Pool Fires:

Release cases from the recovered waste EDC liquid Isotainer and connectingtransfer system will vary in quantity. Catastrophic vessel failures will releasethe entire Isotainer contents (approximately 20 te) into the bunded area.Releases from transfer system failures will involve smaller quantities andcorrespondingly smaller pool fires. As radiant heat from pool fires is dependenton the size of the pool and hence limited by the bund design, two pool fire

Page 57: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200448

scenarios are modelled, i.e. the entire bund full of recovered waste EDC liquidand a 3 m equivalent diameter pool is chosen for the smaller releases.

Given the location of the GTP and the size of the bunded area, it is notexpected that radiant heat from these scenarios will impact people who are off-site or personnel involved in running the adjacent plants on the BIP. Thesecalculations are included for completeness of the study only.

For the two chosen pool fire calculations, equivalent pool diameters of 9 m and3 m are used. The percentage of combustion energy emitted as radiant heat istaken as 10% and 31% for the 9 m and 3 m cases, respectively (Ref 10).

The values of interest for radiant heat (DIPNR, HIPAP No. 4 and ICI HAZANCourse notes) are shown in Table 11.

Table 11 - Radiant Heat Impact

HEAT FLUX(kW/m2)

EFFECT

1.2 Received from the sun at noon in summer

2.1 Minimum to cause pain after 1 minute

4.7 Will cause pain in 15-30 seconds and second degree burns after 30seconds. Glass breaks

12.6 30% chance of fatality for continuous exposure. High chance of injury

Wood can be ignited by a naked flame after long exposure

23 100% chance of fatality for continuous exposure to people and 10%chance of fatality for instantaneous exposure

Spontaneous ignition of wood after long exposure

Unprotected steel will reach thermal stress temperatures to causefailure

35 25% chance of fatality if people are exposed instantaneously.Storage tanks fail

60 100% chance of fatality for instantaneous exposure

For information, further data on tolerable radiant heat levels is shown in Table12.

Page 58: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200449

Table 12 – Layout Considerations – Tolerable Radiant Heat Levels

Plant Item Tolerable Radiant HeatLevel, kW/m2

Source

Drenched Storage Tanks 38 Ref 10

Special Buildings (Protected) 25 Ref 10

Cable Insulation Degrades 18-20 Ref 10

Normal Buildings 14 Ref 10

Vegetation 12 Ref 10

Plastic Melts 12 Ref 10

Escape Routes 6 Ref 10

Glass Breakage 4 Ref 12

Personnel in Emergencies 3 Ref 10

Plastic Cables 2 Ref 10

Stationary Personnel 1.5 Ref 10

Radiant heat intensity versus distance for each of the pool fire scenarios issummarised in Table 13.

Table 13 – Radiant Heat vs Distance – Pool Fire Scenarios

Pool Fire Scenario: Distance to Specified Radiant Heat Levelfrom the Edge of the Pool Fire, m

23 kW/m2 12.6 kW/m2 4.7 kW/m2 2.1 kW/m2

9 m Diameter - 1 8 11

3 m Diameter - 1 4 6

Note that the predicted surface emissive power (radiant heat flux of the flames)is approximately 17 kW/m2. Therefore, no figure exists for the distance to23 kW/m2.

For assessment of the effects of radiant heat, it is generally assumed that if aperson is subjected to 4.7 kW/m2 of radiant heat and they can take cover withinapproximately 20 seconds then no serious injury, and hence fatality, isexpected. However, exposure to a radiant heat level of 12.6 kW/m2 can resultin fatality for some people for limited exposure durations.

Page 59: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200450

As can be seen from the results given in Table 13, the radiant heat levels frompotential pool fires will not cause any significant impact on people who are off-site or personnel involved in running the adjacent plants on the BIP.

Toxic Releases:

If the same pools of recovered waste EDC liquid are formed in the Isotainerbund as per the above pool fire analysis but not ignited, evaporation of the EDCand subsequent atmospheric dispersion may cause significant concentrations ofEDC off-plot.

A 20 minute evaporation period is assumed. It is assumed that within this time,emergency response action can be taken to mitigate the EDC evaporationand/or dispersal. This may include covering the bund with foam or use of fogsprays from monitors etc.

Neutral gas dispersion modelling of this plume for the full bund case at thevarious weather / wind conditions (Appendix 5) is shown in Table 14. Thepredicted EDC evaporation rate is 0.13 kg/s.

Table 14 – Large Recovered Waste EDC liquid Pool (9 m diameter),Dispersion Modelling

Wind / Weather Pattern Maximum Distance (m) to:

EDC ERPG 1 EDC ERPG 2

2.3 B 46 21

3.8 D 72 31

5.3 D 59 25

2.3 E 146 63

0.9 F 521 202

2.3 F 276 100

As can be seen from the above data, off-site impact (i.e. at 325 m or further) islimited to a one combination of wind / weather (0.9F for irritation effect). Noinjury risk is predicted (i.e. by use of ERPG 2). The risk of EDC irritation impacton off-site personnel is assessed later in Section 7 of this report.

Note that there are no criteria set in HIPAP 4 (Ref 4) for the risk of toxic irritationand/or injury to personnel on adjacent industrial facilities, i.e. other BIP landusers or neighbouring industries.

By application of the HSE land use planning EDC SLOT, again, no off-sitefatalities would be expected from a plume emanating from a pool of recoveredwaste EDC liquid in the Isotainer bund. The dispersion models also show that

Page 60: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200451

the EDC concentration drops below 4,500 ppm (the SLOT value for a 20 minuteexposure duration) within 25 m for the various wind / weather combinations.Hence, neither off-site nor on-site fatalities are expected from this scenario.

For a 3 m diameter pool of recovered waste EDC liquid, the distance to theERPG 1 value for 0.9F is predicted to be 36 m. Therefore, no residential off-siteirritation (or injury) can be expected for any of the wind / weather combinationsfor a plume emanating from a small recovered waste EDC liquid pool. Thepredicted EDC evaporation rate is less than 0.02 kg/s for this case (i.e.relatively low).

5.7.3 Natural Gas Line Failures

Failures associated with the natural gas feed line to the thermal oxidiser willrelease the natural gas to atmosphere and, if ignited, it can form a jet fire, aflash fire and/or an explosion.

The analysis of the potential jet fires from the natural gas feed line to thethermal oxidiser is shown in Table 15. The natural gas pressure is 5 barg(ambient temperature). The line is approximately 80 m long.

Table 15 –Natural Gas Jet Fires

Stream EstimatedLength of Jet,

m

Full bore failure (80 mm) 12

50 mm hole 10

13 mm hole 4

Notes: Jet flames modelled using methane.

As expected for these size jet fires, no adverse radiant heat levels will beimposed off-site or on adjacent operating plants.

Potential vapour cloud explosions and flash fires can occur from the natural gasline failures, i.e. delayed ignition.

The effects from explosion overpressures (Ref 4) are summarised in Table 16.

Page 61: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200452

Table 16 – Effects of Explosion Overpressure

OVERPRESSURE, kPa PHYSICAL EFFECT

3.5 90% glass breakage

No fatality, very low probability of injury

7 Damage to internal partitions & Joinery

10% probability of injury, no fatality

14 Houses uninhabitable and badly cracked

21 Reinforced structures distort, storage tanks fail

20% chance of fatality to person in building

35 Houses uninhabitable, rail wagons & plant items overturned.

Threshold of eardrum damage, 50% chance of fatality for a personin a building, 15% in the open

70 Complete demolition of houses

Threshold of lung damage, 100% chance of fatality for a person in abuilding or in the open

For flash fires, any person inside the flash fire cloud is assumed to be fatallyinjured. As flash fires are of limited duration (typically burning velocity is 1 m/s,Ref 13) then those outside the flash fire cloud have a high probability of survivalwithout serious injury.

The analysis of the potential vapour cloud explosions and flash fires from thenatural gas pipe failures is shown in Table 17. The mass calculated in theflammable range is assumed to be 100% confined, i.e. all this gas is involved inthe explosion calculations. As methane is not a high reactive flammable gasand the quantities involved are relatively small then a weak deflagration isassumed in the explosion calculations (multi-energy method – TNO).

Page 62: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200453

Table 17 – Natural Gas Vapour Cloud Explosions and Flash Fires

Stream Mass ofNatural Gas

in theFlammableRange, kg

Radius ofFlash Fire, m

Distance (m)to 14 kPaExplosion

Overpressure

Distance (m)to 7 kPa

ExplosionOverpressure

Full bore failure (80 mm) 11 < 10 m < 10 m < 10 m

50 mm hole 6 < 10 m < 10 m < 10 m

Notes: 1. Pipeline failures assumed to be isolated within 20 minutes.

2. Radius of flash fires calculated to be the distance to 70 kPa overpressure (assumingthe cloud had exploded, Ref 13)

3. 13 mm holes not modelled as they are too small to generate gas clouds of anysignificant size.

For these releases of natural gas, choked flow exists and rapid jet mixing withair occurs. The result is a relatively small vapour cloud size with limitedconsequential impacts if ignited. The 20 minute release duration also has nosignificant impact on the release. Steady state conditions are reached soonafter the release occurs (i.e. the distance to the LEL does not change at steadystate dispersion conditions). The maximum distance to the LEL for the full borefailure is less than 50 m.

Given these results for the natural gas vapour cloud explosions and flash fires,no adverse consequential impacts will be imposed off-site or at neighbouringplants.

5.7.4 Thermal Oxidiser Explosion

An internal thermal oxidiser explosion is possible due to a failure of the burnermanagement system. For these events to occur, a flammable gasconcentration occurs within the thermal oxidiser whilst it is off-line and a sourceof ignition ignites it. This is a known hazard with burners, boilers etc and thecontrol and start-up systems include prevention measures such as purging theinternal space prior to ignition.

Internal explosions have the potential to cause harm through overpressures andpossibly missiles. TNO (Ref 9) has developed a methodology for estimatingoverpressures from internal explosions and it is used as follows.

For an internal thermal oxidiser explosion to occur, the gas inside the unit mustbe in the flammable range, i.e. between the lower and upper explosion limits.The entire volume of the thermal oxidiser is assumed to contain flammable gas.For some internal explosions, this will be a conservative assumption.

Therefore, the scenario modelled is an internal explosion of 200 m3 (i.e. theinternal volume of the unit).

Page 63: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200454

From the explosion modelling, the distances to specified overpressure levels forthe thermal oxidiser internal explosion scenario are shown in Table 18.

Table 18 – Distance to Specified Levels of Explosion Overpressure forPotential Internal Thermal Oxidiser Explosion Scenario

Distance to Specified Overpressure, m

70 kPa 35 kPa 21 kPa 14 kPa 7 kPa 3.5 kPa

ThermalOxidiser < 10 14 19 25 42 80

It was also estimated that the explosion overpressure for this event at thenearest residential areas to the GTP (about 325 m away) would not exceed0.7 kPa. At this low overpressure, there would be expected to be no injuriesand only very minor building damage, if any.

Given the site layout and the location of the thermal oxidiser (approximately50 m from neighbouring industrial BIP facilities), no fatalities or significanteffects from this explosion scenario are expected to people who are off-site oron adjacent plants.

As with the Isotainer BLEVE case, it is possible that missiles could begenerated from an internal thermal oxidiser explosion. The risk of theseexplosion events is analysed in Section 7.

5.7.5 Caustic Scrubber Failure

Should the caustic scrubber fail to absorb the HCl carryover from the HClabsorber, e.g. loss of reflux flow and the plant does not trip, then the thermaloxidiser outlet stream will be vented with HCl mist. The exhaust vent height is20 m.

From Section 5.7.1, the stream from the thermal oxidiser (70oC) has thefollowing flow / composition data:

Mass Rates: Molar Rates:(kmol/hr)

ppm:(vol/vol)

Total Flow 66,070 scm/hr 2,793

VOCs (as EDC) 1.46 kg/hr 0.014 5

HCl 114.6 kg/hr 4.32 1,550

Chlorine 2.25 kg/hr 0.032 11

Notes: VOC = volatile organic carbon

Page 64: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200455

The HCl absorber will be designed for a 75 wt% recovery, i.e. 28.7 kg/hr of HClwill flow to the caustic scrubber. This is approximately 281 ppm HCl in the gasstream.

Neutral gas dispersion modelling of this stream being vented from the causticscrubber at weather / wind conditions, F2 and E2.3, shows that the ERPG 1 HClconcentration (3 ppm) does not form at ground level due to the elevated point ofdispersal. The ERPG 1 concentration will exist for approximately 400 mdownwind of the plant vent only at heights greater than 15 m above the ground.This is for a release duration of 5 minutes (manual shutdown from response toplant alarms expected). Consequently, there would not be any off-site or on-site impacts.

Note that if the HCl absorber has failed, a subsequent failure of the causticscrubber will vent gas with essentially the same composition as the thermaloxidiser outlet gas as per the analysis in Section 5.7.1. The difference for thiscase, however, will be the elevation of the vent pipe as compared to acatastrophic pipe failure exit the thermal oxidiser at potentially a lower elevation.This means that the ground level HCl concentration in this case would be evenlower and would therefore be below the ERPG 1 criterion.

5.7.6 Thermal Oxidiser Feed / Product Exchanger Failure

This scenario involves a failure of the thermal oxidiser feed / product exchanger,i.e. a hole allows feed gas to pass directly to the product gas which is vented viathe HCl and caustic scrubbers. Given the results from the elevated vent case inSection 5.7.5, the low driving force for feed gas to pass into the product gas, theability of the scrubbers to knock-out some feed gas impurities and that the holesize is, historically, most likely to be small then this scenario is not included inthe consequence modelling.

5.8 CALCULATION OF FATALITY DUE TO FIRES, EXPLOSIONS AND TOXICRELEASES

The following methodologies are used by TNO for the calculation of fatality:

Flash Fires: 100% fatality within the diameter of the fireball and no fatalityeffect outside of the fireball (due to the short duration).

Jet Fires: 100% fatality within the dimensions of the flame. The flame ismodelled as a rectangle. Outside the flame, heat radiation levelsare calculated at particular points using the view factor method.Fatality is calculated using the probit equation below with anexposure duration of 20 seconds:

Probit = -36.38 + 2.56 ln(tQ1.33)

t exposure duration (sec)

Page 65: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200456

Q heat flux (W/m2)

Note that this probit is only valid for very short exposure durations(less than 1 minute). For the purposes of this risk assessment it isassumed a person has 20 seconds to escape from heat radiation(i.e. an exposure duration of 20 seconds).

Pool Fires: 100% fatality within the diameter of the pool fire. Effects past theedge of the pool fire are calculated via the probit above for jetfires.

Explosions: Riskcurves explosion fatality by first calculating a flash fire andthen assuming 100% fatality within the fire diameter. Theexplosion fatality effects are taken into account by assuming anaverage 1.25% fatality up to the 10 kPa radius. The 1.25%method is based on an analysis of fatalities that occurred in theLPG Mexico City disaster in 1984 where the majority of peoplewere in fact killed by falling rubble etc.

BLEVE: 100% fatality within the diameter of the fireball. As for explosions,overpressure effects can cause an additional 1.25% fatality up tothe 10 kPa overpressure radius. However, for BLEVEs the peakoverpressure is normally within the fireball radius so overpressureeffects do not actually contribute to the fatality calculations.

Toxic Impact:As per the description given in Appendix 4 using probit equations.

Page 66: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200457

6 FREQUENCY / LIKELIHOOD ANALYSIS

The frequency of an event is defined as the number of historical occurrences ofthe event over a specified time period; with the period in risk analysis generallybeing taken as one year. Likelihood is the expected number of events for afuture time basis.

Two approaches have been used to estimate the frequencies of hazardousevents. The first method is to use statistical data relating to failure of wholesystems or equipment items. Secondly, complex events can be broken downinto contributing factors and the overall event likelihood estimated from theknown frequencies of the smaller events using techniques such as fault treeanalysis. In this PHA, previous fault trees exist for specific events and henceare not reconstructed here.

Assumptions made for the purposes of likelihood calculations are describedbelow.

6.1 GENERIC EQUIPMENT FAILURE FREQUENCIES

For piping and equipment failures, frequencies have been estimated either fromdata compiled and published by ICI (Ref 14) or from frequency estimatespublished by the Institution of Chemical Engineers (Ref 15).

Table 19 - Generic Equipment Failure Frequencies

Type of Failure Failure Rate (x 106) per year

Pipelines

13 mm hole

50 mm hole

3 mm gasket (13 mm hole equivalent)

Guillotine fracture (full bore):

< 50 mm

> 50 mm but < 100 mm

> 100 mm

3 / m

0.3 / m

5 / joint

0.6 / m

0.3 / m

0.1 / m

Vessels

13 mm hole

25 mm hole

50 mm hole

Catastrophic failure - Pressure Vessel

6

3

3

1

Hoses

Hose failure – moderate to large leak 4 x 10-6 per hour (Ref: TNO, Riskcurves UserManual)

Page 67: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200458

6.2 BLEVE LIKELIHOOD

Detailed calculations by ICI Engineering for ethylene oxide storages and OricaEngineering (Ref 16) into the causes and likelihoods of a BLEVE show that thelikelihood of a BLEVE for an unprotected vessel is in the order of 10-7/yr. Thatis, a fire occurs (e.g. from a leak in connecting piping which is ignited andimpinges on the vessel to cause the BLEVE) and no mitigation is available. Forprotected vessels (e.g. fire insulation and/or a water spray system are installed)the likelihood of a BLEVE is in the order of 10-8/yr.

For the recovered waste EDC liquid Isotainer, a BLEVE likelihood of 10-7/yr isincluded in the modelling work (i.e. no specific allowances for protectivesystems are included).

6.3 THERMAL OXIDISER INTERNAL EXPLOSION LIKELIHOOD

Explosions involving combustion systems do occur and can result in multiplefatalities. For example, the gas feed systems to boilers fail and an explosivemixture within the boiler is subsequently ignited. During the 1970’s and early1980’s, the likelihood of explosions involving these types of systems was in theorder of 1 every 1,000 years (ICI Mond Reliability Data, Ref 14). However, withthe advent of more robust burner management systems, conformance to gassupply codes and standards, and the use of nitrogen for purging lines, thefrequency of incidents has reduced in recent years.

Analysis of burner explosions was performed by Orica Engineering for the HCBPHA (Ref 17). It was concluded that for a similar type of burner managementsystem, the likelihood of internal explosions was approximately 5 x 10-6 peryear. This value is consistent with the performance of modern designs and isused in this study.

6.4 CAUSTIC SCRUBBER FAILURE LIKELIHOOD

The likelihood of the caustic scrubber failure is approximated as no detaileddesign is available. A simple layer of protection analysis (Ref 18) is performedas follows:

Expected demand likelihood, e.g. loss of reflux flow, say, once per year

Estimated probability of failure for a redundant, diversifieddesigned, high integrity, trip system in this application 0.0001 (based on

Ref 19)

Probability of vent analyser failure 0.1 (Ref 19)

Probability of unsuccessful fast operator response 0.5 (Ref 19)

Therefore, the caustic scrubber failure case is approximated as five times every100,000 years (or once every 20,000 years).

Page 68: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200459

6.5 DOMINO INCIDENTS

A potentially hazardous event within a plant can cause further incidents in thesame plant, or in some cases in other plants. The secondary event is called adomino event. With any large site there is potential for a severe incident in onearea to cause a knock-on or domino incident in another area.

From the analysis conducted in Section 5, radiant heat and overpressures frompotential hazardous events are not sufficient to cause propagation to nearbyadjacent operating plants. For the BLEVE case, the duration is approximately11 seconds, hence propagation to other piping and vessel failures within thistimeframe is unlikely.

Missiles from the BLEVE could cause propagation, however, the low likelihoodof 1 x 10-7 per year for the BLEVE event is further reduced when the probabilityof a missile hitting a selected target is considered. Therefore, the risk ofpropagation is broadly acceptable. Emissions of gas streams will not cause anysignificant propagation events.

It is also possible that an event in the neighbouring plants may cause a knockon event in the GTP operations. However, as shown in Section 5, theconsequences of many of these potential hazardous events associated with theGTP operations will not have any off-site impact. The risk of these events is,however, estimated in Section 7 of this report.

The proposed facility is close to Sydney's Kingsford Smith Airport and there ispotential for aircraft impact on the BIP site. In 1990, the Australian Centre ofAdvanced Risk and Reliability Engineering Ltd (ACARRE) considered the risksassociated with increased operations at Kingsford Smith Airport due to the thirdrunway. The ACARRE study examined the likely frequency of aircraft crashingonto various sites within the Port Botany region, including the Botany site (thenICI Australia Pty Ltd). The result from the ACARRE study was used todetermine the risk from aircraft crashes with a potential for knock-on effects atthe Botany site (Ref 20). The conclusion was that a low level of risk exists forthis event.

Earthquake events may also cause plant damage sufficient to cause a release.Frequencies and consequences of these events have also been estimated inRef 20 and again, the conclusion being a low level of risk exists for this event.

Generally, other external events that may lead to propagation of incidentsinclude:

Subsidence Landslide

Burst Dam Vermin/insect infestation

Storm and high winds Forest fire

Storm surge Rising water courses

Page 69: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200460

Flood Storm water runoff

Breach of security Lightning

Tidal waves Forest fire

None of these contributory events pose any significant risk to the GTP or itsassociated operations (note: security issues as previously discussed in Section4.3).

Page 70: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200461

7 RISK ANALYSIS

7.1 RISK CRITERIA

The risk criteria applying to new developments in NSW are summarised inTable 20 below (from Ref 4).

Table 20 - Risk Criteria, New Plants

Description Risk Criteria

Fatality risk to sensitive uses, including hospitals, schools, aged care 0.5 x 10-6 per year

Fatality risk to residential and hotels 1 x 10-6 per year

Fatality risk to commercial areas, including offices, retail centres,warehouses

5 x 10-6 per year

Fatality risk to sporting complexes and active open spaces 10 x 10-6 per year

Fatality risk to be contained within the boundary of an industrial site 50 x 10-6 per year

Injury risk - incident heat flux radiation at residential areas should notexceed 4.7 kW/m2 at frequencies of more than 50 chances in amillion per year or incident explosion overpressure at residentialareas should not exceed 7 kPa at frequencies of more than 50chances in a million per year

50 x 10-6 per year

Injury risk - frequency at which toxic concentrations in residentialareas should not exceed a level which would be seriously injurious tosensitive members of the community following a relatively shortperiod of exposure

10 x 10-6 per year

Irritation risk - frequency at which toxic concentrations in residentialareas should not cause irritation to eyes or throat, coughing or otheracute physiological responses in sensitive members of thecommunity

50 x 10-6 per year

Property damage risk – incident heat flux radiation at neighbouringpotentially hazardous installations or at land zoned to accommodatesuch installations should not exceed 23 kW/m2 at frequencies ofmore than 50 chances in a million per year

50 x 10-6 per year

Property damage risk – incident explosion overpressure atneighbouring potentially hazardous installations, at land zoned toaccommodate such installations or at nearest public buildings shouldnot exceed 14 kPa at frequencies of more than 50 chances in amillion per year

50 x 10-6 per year

The risk associated with the proposed GTP and associated operations will beestimated in the following sections and compared to the above criteria.

Page 71: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200462

7.2 MITIGATING FEATURES FOR OFF-SITE INDIVIDUALS

By convention in Australia, mitigation factors are not taken into account in theestimation of individual fatality risk. An individual is considered to be locatedpermanently at a particular location, and no scope for shelter or escape isfactored into the calculations. The risk results are essentially the risk at alocation, not necessarily to a particular individual and are therefore consideredhighly conservative.

7.3 RISK RESULTS

7.3.1 Individual Fatality Risk

Off-site:

From the consequence analysis in Section 5, there are no identified events thathave the potential to cause off-site fatality. Therefore, the risk to sensitive landusers, residential areas, commercial areas, sporting areas and open spaces islower than the DIPNR HIPAP 4 criteria. Correspondingly, societal risk is alsonegligible.

On-site:

From the consequence analysis in Section 5, the one event which may possiblylead to fatalities on the BIP is a BLEVE of the recovered waste EDC liquidIsotainer. As the likelihood for this event is taken as 1 x 10-7 per year, given a100% chance of fatality (conservative), then the corresponding industrial risk toother BIP users is below 50 x 10-6 per year. Therefore, the fatality risk to bothBIP and other neighbouring industries is lower than the DIPNR HIPAP 4criterion.

Missiles:

There are two possible events that could generate missiles, i.e. a recoveredwaste EDC liquid Isotainer BLEVE and an internal thermal oxidiser explosion.The likelihood of each event is 1 x 10-7 per year and 5 x 10-6 per year,respectively. The combined likelihood is thus 5.1 x 10-6 per year. As theprobability of missile generation for an explosion event is generally taken as 0.1or lower (Ref 10), then the risk of fatality from any potential missiles will be lessthan 1x 10-6 per year and hence is considered broadly acceptable. Note thatthis does not include the probability of a person being struck by a missile if itwas generated (i.e. conservative).

7.3.2 Injury Risk

Off-site:

From the consequence analysis in Section 5 and using the ERPG 2 value forinjury risk due to exposure to HCl (conservative given the duration of exposure

Page 72: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200463

expected), the catastrophic failures associated with the piping and equipmentdownstream of the thermal oxidiser have the potential for off-site injury for wind/ weather classes 2.3E, 0.9F and 2.3F.

The likelihood of these failures occurring is estimated as follows:

There are four vessels (thermal oxidiser recuperator, quench, acid absorber andcaustic scrubber) and approximately 55 metres of piping. Therefore, thelikelihood of catastrophic failures occurring is estimated to be (from Table 19):

= (4 x 1 x 10-6) + 55 x 0.1 x 10-6 per year

= 10 x 10-6 per year

The probability of the three wind / weather conditions existing and blowing froman arc from northwest to southwest (i.e. a 90o arc such that the wind is blowingin the direction of the residential area along Denison Street) is estimated to be10% (see Appendix 5 for the wind / weather probability data).

Therefore, the likelihood of injury from HCl emissions is 1 x 10-6 per year. Thisis lower than the DIPNR HIPAP 4 criterion.

On-site:

There is no on-site criterion for injury risk in HIPAP 4.

7.3.3 Irritation Risk

Off-site:

From the consequence analysis in Section 5 and using the ERPG 1 value forirritation risk due to exposure to HCl (conservative given the duration ofexposure expected), the catastrophic failures associated with the piping andequipment downstream of the thermal oxidiser have the potential for off-siteirritation for wind / weather classes 3.8D, 5.3D, 2.3E, 0.9F and 2.3F.

Also, off-site irritation is possible from catastrophic failures associated with thepiping entering the thermal oxidiser (i.e. EDC releases) for wind / weatherclasses 0.9F and 2.3F and from a plume emanating from a bund spill ofrecovered waste EDC liquid (release from the recovered waste EDC liquidIsotainer and transfer system) for 0.9F only.

The likelihood of these failures occurring is estimated as follows:

Firstly, as above, there are four vessels (thermal oxidiser recuperator, quench,acid absorber and caustic scrubber) and approximately 55 metres of piping afterthe thermal oxidiser.

Secondly, there is approximately 30 metres of piping from the air blower to thethermal oxidiser.

Page 73: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200464

Lastly, it is assumed that only large releases from the Isotainer have thepotential to fill the bund. If a fault in the transfer system occurs, it is assumedthat the loss of flow will be detected and operator response will prevent a largerelease occurring (e.g. by isolating the valve at the Isotainer outlet).

Therefore, the likelihood of catastrophic failures occurring is estimated to be(from Table 19):

= (4 x 1 x 10-6) + 55 x 0.1 x 10-6 per year plus

30 x 0.1 x 10-6 per year plus

(1 x 10-6 + 3 x 10-6 + 3 x 10-6), i.e. Isotainer catastrophic failure, 50 mmand 25 mm holes

= 20 x 10-6 per year

This value is below the HIPAP criterion for residential irritation likelihood(without even taking into consideration the probability of the relevant wind /weather conditions existing such that the wind blows in the direction of DenisonStreet).

On-site:

There is no on-site criterion for injury risk in HIPAP 4.

7.3.4 Property Damage

Given the explosion overpressure and radiant heat analyses in Section 5 of thisreport, the risk of property damage due to propagation is negligible, i.e. thepredicted explosion overpressures at the neighbouring facilities are less than14 kPa and the predicted radiant heat at the neighbouring facilities is less than23 kW/m2 (the latter takes into consideration the likelihood of a recovered wasteEDC liquid Isotainer BLEVE being approximately 1 x 10-7 per year).

7.3.5 Cumulative Risk

Cumulative risk within the Botany/Randwick Industrial Complex was consideredby the Department of Infrastructure, Planning and Natural Resources (then theDepartment of Planning) in 1985, and was updated in 1999. Subsequent tothis, an overview report on the Botany / Randwick Industrial Area was issued byDIPNR in 2001. The calculated risk levels presented in this PHA will have onlya minor impact on the cumulative risk results formerly calculated for theBotany/Randwick area. This can be expected given the relatively fewpotentially hazardous events that can significantly affect off-site risk.

Page 74: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200465

7.4 TRANSPORT RISK

It is proposed to transport an Isotainer containing recovered waste EDC liquidfrom storage at Port Botany (at Terminals Pty Ltd) to the GTP, on average,once every two weeks. The proposed route is:

From the Terminals facility, turn right onto Friendship Road, travel toBumborah Road and then left onto Botany Road

Turn right into Beauchamp Road

Turn left into Denison Street

Turn left into the BIP at Gate 3

In 1995, ICI Engineering (Ref 21) performed a transport risk assessment forliquid EDC movements in the reverse direction, i.e. from the BIP site toTerminals Pty Ltd. This transport risk assessment reviewed the risk associatedwith 2,500 road tankers (i.e. not Isotainers which are more robustly designed)per year. This figure considerably exceeds the proposed GTP recovered wasteEDC liquid Isotainer movements of approximately 25 per year.

The 1995 transport risk assessment included potential hazardous eventsassociated with loading and unloading activities as well as incidents along thetransport route. Both fires and toxicity impacts were considered.

The results of the 1995 transport risk assessment found that the risk associatedwith the movement of EDC via 2,500 road tankers per year was broadlyconsidered acceptable. Given that the GTP transport of recovered waste EDCliquid involves significantly less movements via the identical route then it isconcluded that the risk from recovered waste EDC liquid transport for thisProject is also broadly acceptable.

7.5 RISK FROM COMBUSTION PRODUCTS

There is a potential risk to those attending a fire emergency (and possibly off-site) of effects from toxic products of combustion from a recovered waste EDCliquid fire, e.g. carbon oxides, HCl and smoke, as well as vaporised EDC (i.e.not combusted).

Impact from toxic products of combustion will only be significant, generally, localto the fire. As stated in Lees (Ref 10):

“The hot products of combustion rising from a fire typically have a temperaturein the range 800-1200oC and a density a quarter that of air.”

Hence, a buoyant plume is formed (as seen when smoke is emitted from achimney) and the combustion products rise and are dispersed as per theprevailing wind / weather conditions. Several runs of the Brigg’s Plume Model(Ref 9) for various combinations of weather / wind conditions and fire

Page 75: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200466

temperatures show that the plume rises from a recovered waste EDC liquidbund fire to at least one hundred metres and then disperses via passivedispersion in the down wind direction. Momentum effects continue to cause theplume to rise whilst it is dispersing. The results are shown in Table 21. Theresults also show that plume rise is insensitive to fire temperature variations of800oC +/- 100oC (not shown).

Table 21 – Fire Plume Rise Modelling

Wind / Weather Initial Height ofPlume, m

5.3 D 102

2.3 E 235

0.9 F 600

2.3 F 235

Unless a temperature inversion exists where reverse atmospheric currents canoccur (i.e. air slumps to the ground as opposed to air eddies that rise), no effectat ground level is expected. Note that dispersion models best account fortemperature inversions by using F class stability (i.e. typically when theadiabatic lapse rate is positive). The models, however, do not include theprovision for air slumping to ground.

It is noted that if a temperature inversion exists where the combustion productscan return to the ground, emergency response may be required to clear /control the area. Given the low likelihood of recovered waste EDC liquidreleases calculated in this PHA (approximately 7 x 10-6 per year), the probabilityof ignition of flammable liquids being less than 10% (Ref 15) and the probabilityof a temperature inversion occurring in the Sydney area (less than 19%, Bureauof Meteorology information provided to Pinnacle Risk Management), the risk ofirritation from products of combustion is broadly considered acceptable, as thetotal risk is less than the residential irritation criterion, even after adding theirritation effects from Section 7.3.3.

Page 76: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200467

8 RISK TO THE BIOPHYSICAL ENVIRONMENT

The main concern for risk to the biophysical environment is generally witheffects on whole systems or populations. Whereas any adverse effect on theenvironment is obviously undesirable, it is difficult to envisage an incidentscenario at the GTP that would threaten a whole system or population. For thisproposal, the risk to people in the vicinity is the main concern.

For completeness, risks to the biophysical environment due to loss ofcontainment events are summarised below.

8.1 ESCAPE OF MATERIALS TO ATMOSPHERE

The likelihood of releases of gas and products of combustion were estimated aspart of the risk assessment and shown to be low. For any release of theprocess air stream, the contaminant’s concentrations are low and hencesignificant environmental impact is not expected. Certainly, from the analysis inthis report, no incident scenarios were identified where whole systems orpopulations could be affected by a release to the atmosphere.

8.1.1 HCl and Chlorine

Failures of the HCl column would include loss of circulation and loss ofinventory (make-up). Alarms and trips would be installed to protect againstthese failures. A sudden increase in the HCl load on the system from thethermal oxidiser beyond the design case is not feasible given the constantgroundwater feed rate.

Failure of the correct operation of the caustic scrubbing column could lead to arelease of HCl and chlorine, although as shown in Section 5 of this PHA, thechlorine rate is very low. Performance failure could be due to loss of circulation,failure of good contacting due to a problem with the packing or loss of availablecaustic in the scrubber.

Failure detection would initiate a shutdown of the GTP. Detection would be bylow circulation flow, pressure drop across the column and by ORP meter tocheck caustic strength. In addition, an HCl meter would be installed in the stackto provide a further protection against system failure.

8.1.2 Dioxins

There has been significant review by the project team members re thepossibility of dioxins forming immediately downstream of the thermal oxidiser.Pinnacle Risk Management understands that the approach taken by the projectteam members is to select a process where the possibility of dioxins forming isminimised. The chosen technology, as Pinnacle Risk Managementunderstands, is accepted by authorities such as the USA and GermanGovernments as being acceptable with respect to dioxin targets being met. Thefollowing discussion details the possibility of dioxin formation.

Page 77: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200468

Incorrect Temperature in the Thermal Oxidiser:

The temperature in the thermal oxidiser will be controlled by measuring thetemperature in an appropriate spot and adjusting the fuel gas to maintain thesetpoint. If the temperature in the thermal oxidiser gets too high or low thethermal oxidiser will shut down. This will be measured by multiple (2)independent devices and performed by the independent Safety InstrumentedSystem. This system will be designed to meet the requirements of IEC 61508and IEC 61511. With a thermal oxidiser operating temperature of 1000°C, it isenvisaged that there would be alarms at 975oC and 1,025oC, with trip points at950oC and 1050oC, respectively. There will also be a carbon monoxide monitoron the exit gas which is used to indicate good reaction conditions.

The consequences of too high a temperature will be equipment damage withthe potential for loss of containment.

If the temperature is too low, then there is the possibility of reduced destructionefficiency. This in turn could lead to emission of contaminants, such as EDC,VC and benzene. It is also possible, that there could be products of incompletecombustion (PIC). In turn, these are associated with the potential for dioxinformation. The likelihood of dioxin formation will increase depending on otherconditions, such as correct quench operation. If the quench is operatingproperly (i.e. rapid quench through the temperature range 450oC to 250oC), theneven if there are PICs, there is a reduced chance of dioxin formation.

De Novo Dioxin Generation:

There are three main routes for the release of dioxins from a thermal oxidiser:

1. If dioxins enter the thermal oxidiser (some thermal oxidisers are used fordestroying material containing dioxins) and if they are not destroyed, thensome may leave. Note that there are no dioxins in the feed to the GTP.

2. If there are solids involved (e.g. municipal waste destruction), then the ashparticles present in the reaction zone can act as catalytic sites for theformation of dioxins. For the GTP, there are no solids in the feed to thethermal oxidiser.

3. The third mechanism involves the formation of dioxins de novo due to therecombination of molecules after the reaction zone. This is thought to ofteninvolve PICs but may not require them. However, if the gases spend toolong in the temperature range 450oC to 250oC, particularly in the presence ofsome metals such as copper which act as catalysts, then dioxins may beformed.

To protect against this, temperatures will be measured through the gas coolingtrain, to ensure that the gas enters the quench system at greater than 600oC,and that the exit from the quench is less than 100oC. Events that may lead toexcessive cooling pre-quench may include the steam system running at too lowa temperature (pressure) due to a control failure (alarms and trips in place) or afailure of the off-gas preheater (a steam preheater on the off-gas stream before

Page 78: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200469

the recuperator to ensure that cold spots are not formed in the recuperatorwhich could lead to corrosion, again protected with alarms and trips). Anotherevent is a steam tube failure in the waste heat boiler (WHB). This could lowerthe temperature of the off-gas. A serious tube failure would cause pressureloss in the steam and be detected as above, a smaller steam leak would dropthe temperature and be detected by low exit temperature from the WHB.

Burner Issues:

If the burner does not operate properly, then there could be poor combustion.The burner is designed to operate for extended periods without maintenance(self-cleaning) and would be checked on a routine basis. The fuel and airsupply are both clean and should not cause any fouling problems. Further, thecarbon monoxide measurement previously mentioned would warn of decreasesin burner performance.

The recovered waste EDC liquid would be injected into the burner as a liquidfuel for atomisation and reaction. The dual fuel (gas plus liquid) is a standardapplication. If the liquid injection performance decreases, then this would beobserved through changes in the backpressure and by routine visualobservation of the flame. The nozzle would be subject to routine maintenance.

A poorly performing liquid nozzle may lead to droplets in the reaction zone andpotentially PICs. This may be associated with carbon monoxide and bealarmed, or may be observed visually. As per the above discussion, even ifPICs were present, then as long as the quench system is performing well, thenthere is a reduced chance of dioxin formation.

To approximate the likelihood of possible dioxin formation, a simple layer ofprotection analysis (Ref 18) is performed as follows. It is noted that only limitedthermal oxidiser design and safety systems information is available at thisstage.

Expected demand likelihood, e.g. temperature controlfailure, burner upset or steam system malfunction, say, 1 in every 5 years

Estimated probability of failure for a redundant designed,high integrity, temperature trip system in this application 0.001 (based on

Ref 19)

Probability of inadequate quench operation 0.05 (Ref 19)

Probability of unsuccessful operator responseto abnormal offgas composition alarm 0.1 (Ref 19)

Therefore, the possibility of dioxin formation is approximated as once every1,000,000 years if due to quench maloperation and once every 50,000 years ifoccurring before the quench. Whilst the figures are low, they should bereviewed once further thermal oxidiser design and protection system details areknown.

Page 79: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200470

8.2 ESCAPE OF MATERIALS TO SOIL, WATERWAYS OR SEWERAGESYSTEM

Spills of liquids at the GTP or associated operations, e.g. ground water, HCl,caustic and recovered waste EDC liquid from tanks and the recovered wasteEDC liquid Isotainer, will be contained in bunded areas. Should these spillsoccur then they can be neutralised / controlled and pumped away from therelevant bunded area as appropriate. Ground water spills can be rerun throughthe plant.

The likelihood of releases from the extraction wells and connecting pipelines isminimised by covered pits, underground pipes, signage and some materials areto be corrosion resistant whilst others will include allowances for corrosion. Ifany spills occur, the impact will be local to the spill site, as the contaminantconcentrations are low (less than 1% in total).

GTP Water Discharge:

Water will discharge to the environment once it has been cleaned. To ensurethat no environmental damage occurs, the water will need to have adequateremoval of volatile material in the air stripping system, iron removal, non-volatileorganic removal, and nutrient removal.

Failures of these systems might be due to incorrect air or water flows, lowgroundwater temperature, loss of caustic or flocculent addition, exhaustion ofthe carbon beds or incorrect sodium hypochlorite addition.

If a failure is detected, the water system will automatically go into recycle mode,so that product water is fed into the groundwater feed tank and then no freshgroundwater is admitted into the system. Protection against failure of thesesystems will be by measuring correct air and water flows and temperatures inthe air stripping system; checking for correct pH adjustment and flows of causticand flocculent in the iron removal stage, ORP measurement and sodiumhypochlorite flow in the nutrient removal stage. These automatic systems willbe supplemented by regular laboratory analysis of key performance indicators.

Further, there will be online pH measurement (with alarm and tripping) as a finalcheck of discharge water.

As with releases to the atmosphere, no incident scenarios were identified wherewhole systems or populations (or even part thereof) could be affected by arelease to the soil, waterways or sewerage system at the GTP. A summary ofthe types of potential hazardous events, their causes and safeguards,associated with the water discharge systems is shown in Table 22.

Page 80: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinn

acle

Ris

k M

anag

emen

t

Oric

a G

TP

PH

A R

epor

t Rev

F.D

oc8

Nov

embe

r 20

0471

Tabl

e 22

– W

ater

Dis

char

ge H

azar

d Id

entif

icat

ion

Wor

d D

iagr

am

Item

No.

Even

tC

ause

sPo

ssib

le C

onse

quen

ces

Prev

entio

n/Pr

otec

tion

Com

men

ts

1.

Loss

of

cont

ainm

ent f

rom

the

disc

harg

e lin

e

Cor

rosi

on, t

hird

par

tyda

mag

e, m

echa

nica

lfa

ilure

Leak

s in

to g

roun

d ar

ound

the

pipe

(pip

e is

mai

nly

unde

rgro

und)

Pip

e tra

vers

es in

dust

rial a

reas

and

road

s

Effl

uent

wat

er s

peci

ficat

ion

conf

orm

s to

AN

ZEC

C m

arin

edi

scha

rge

guid

elin

es s

obi

olog

ical

impa

ct w

ill be

low

,m

ain

effe

ct is

from

slig

htly

salty

wat

er

Pip

e is

cas

t iro

n w

ith in

tern

al p

last

icsl

eeve

, exi

stin

g be

nds

will

bere

plac

ed b

y ne

w fi

breg

lass

ben

dsw

hich

will

hav

e ex

tern

al s

teel

slee

ves

Pip

e w

ill b

e hy

drot

este

d pr

ior t

o us

e

Pip

e w

ill b

e in

clud

ed in

'Dia

l Bef

ore

You

Dig

' sys

tem

Pip

e w

ill h

ave

stan

dard

cat

hodi

cpr

otec

tion

agai

nst c

orro

sion

Fibr

egla

ss b

ends

nee

d to

be fi

tted

to a

llow

inse

rtion

of

plas

tic s

leev

e

No

know

n th

reat

ened

or

enda

nger

ed p

lant

or a

nim

alsp

ecie

s lik

ely

to b

e af

fect

edby

und

ergr

ound

leak

s in

indu

stria

l are

as

2.

Out

of s

peci

ficat

ion

disc

harg

e - t

oosa

lty

Mal

func

tion

of w

ater

treat

men

t sys

tem

Mor

e sa

lty th

an n

orm

al w

ater

ente

rs th

e do

ck

Pos

sibl

e ef

fect

s on

mar

ine

spec

ies

from

ove

rly s

alty

wat

er- l

ocal

& a

cute

effe

cts

only

Vio

latio

n of

per

mitt

eddi

scha

rge

leve

ls

Wat

er tr

eatm

ent c

ontro

l, al

arm

and

trip

syst

ems

Unl

ikel

y to

be

mor

e sa

lty th

an s

eaw

ater

due

to n

atur

e of

pro

cess

No

know

n th

reat

ened

or

enda

nger

ed s

peci

es li

kely

to b

e af

fect

ed b

y th

ese

flow

s in

to d

ock

wat

ers

3.

Out

of s

peci

ficat

ion

disc

harg

e -

too

dilu

te

Mal

func

tion

of w

ater

trea

tmen

t sys

tem

Mor

e di

lute

than

nor

mal

wat

eren

ters

the

dock

No

effe

ct -

sim

ilar

effe

cts

tost

orm

wat

er fl

ows

in c

hann

el

Wat

er tr

eatm

ent c

ontr

ol, a

larm

and

trip

sys

tem

s

Page 81: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinn

acle

Ris

k M

anag

emen

t

Oric

a G

TP

PH

A R

epor

t Rev

F.D

oc8

Nov

embe

r 20

0472

Item

No.

Even

tC

ause

sPo

ssib

le C

onse

quen

ces

Prev

entio

n/Pr

otec

tion

Com

men

ts

4.

Out

of s

peci

ficat

ion

disc

harg

e -

exce

ssam

mon

ia

Mal

func

tion

of w

ater

trea

tmen

t sys

tem

Am

mon

ia n

ot r

emov

ed in

wat

er tr

eatm

ent

Man

y m

arin

e or

gani

sms

sens

itive

to a

mm

onia

Cou

ld h

arm

or

kill

mar

ine

orga

nism

s lo

cal t

o th

edi

scha

rge

poin

t if

conc

entr

atio

n at

dis

char

gehi

gh e

noug

h (a

cute

effe

cts

only

)

Vio

latio

n of

per

mitt

eddi

scha

rge

leve

ls

Wat

er tr

eatm

ent c

ontr

ol, a

larm

and

trip

sys

tem

sN

o kn

own

thre

aten

ed o

ren

dang

ered

spe

cies

like

lyto

be

affe

cted

by

thes

eflo

ws

into

doc

k w

ater

s

5.

Out

of s

peci

ficat

ion

disc

harg

e, i.

e.

Exc

ess

sodi

umhy

poch

lorit

e or

othe

r ch

emic

als

harm

ful t

o m

arin

eor

gani

sms

Wro

ng p

H

Mal

func

tion

of w

ater

trea

tmen

t sys

tem

Mal

func

tion

of m

ain

plan

t cau

sing

brea

kthr

ough

of

chem

ical

s no

rmal

lyre

mov

ed u

pstr

eam

of

wat

er tr

eatm

ent

Exc

essi

ve w

ater

trea

tmen

tch

emic

als

disc

harg

ed

Wat

er d

isch

arge

too

acid

ic o

rto

o al

kalin

e

Cou

ld h

arm

or

kill

mar

ine

orga

nism

s lo

cal t

o th

edi

scha

rge

poin

t if

conc

entr

atio

n at

dis

char

gehi

gh e

noug

h (a

cute

effe

cts

only

)

Vio

latio

n of

per

mitt

eddi

scha

rge

leve

ls

Mai

n pl

ant a

nd w

ater

trea

tmen

tco

ntro

l, al

arm

and

trip

sys

tem

s

Car

bon

filte

rs in

wat

er tr

eatm

ent

plan

t will

red

uce

conc

entr

atio

ns o

fm

ost c

hlor

inat

ed s

peci

es

No

know

n th

reat

ened

or

enda

nger

ed s

peci

es li

kely

to b

e af

fect

ed b

y th

ese

flow

s in

to d

ock

wat

ers

Page 82: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200473

9 CONCLUSION AND RECOMMENDATIONS

The risk associated with the proposed GTP and associated operations at theBIP and environs has been assessed and compared against the DIPNR riskcriteria.

The results of this PHA show that the risk associated with the proposed GTPand associated operations complies with DIPNR guidelines for tolerable fatality,injury, irritation and societal risk. Also, transport risk, risks to biophysicalenvironment, the risk of propagation and the impact on cumulative risk in thePort Botany / Randwick area from releases are broadly acceptable. Theseconclusions apply to both off-site (e.g. residential areas) and on-site (i.e.neighbouring industrial facilities) risk.

The primary reason for the low risk levels from operation of the GTP is thatsignificant consequential impacts from potential hazardous events do not reachthe nearest site boundary or, for the neighbouring industrial facilities, theirlikelihood is acceptably low.

For the pumping wells and pipeline operations associated with the GTPoperation, the primary reason for the negligible risk levels is that the materialshandled are non-hazardous. Nevertheless, as described in detail in the PHA,considerable effort will be made to minimise the chances of leaks from thesesystems.

As with most PHA’s, limited detailed design information is currently available.Correspondingly, some of the analysis in this report is based on assumedconditions. The assumptions made in this analysis are to be reviewedthroughout the project design stage and updated in the Final Hazard Analysis.Therefore, no specific recommendations are made.

It is assumed that the GTP and associated operations will be reviewed via theHAZOP methodology.

Page 83: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 2004A1.1

Appendix 1

GTP Process Flow Diagrams

Preliminary Hazard Analysis, GroundwaterTreatment Plant, Orica Australia Pty Ltd,

Botany Industrial Park

Page 84: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …
Page 85: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …
Page 86: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …
Page 87: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 2004A2.1

Appendix 2

Typical GroundwaterComposition

Preliminary Hazard Analysis, GroundwaterTreatment Plant, Orica Australia Pty Ltd,

Botany Industrial Park

Page 88: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 2004A2.2

Appendix 2 – Typical Groundwater Composition.

Note that the following groundwater composition is indicative data used fordesign purposes only.

NameFeedwater

Compositionmg/L

NameFeedwater

Compositionmg/L

pH 4.5 - 6 Mid = 5 Magnesium 16.59

1.1.1.2-Tetrachloroethane 0.001 Chloride 617.0

1.1.2.2-Tetrachloroethane 1.46 Sulphate 184.2

1.1.2-Trichloroethane 0.82 Alkalinity as CaCO3 57.3

1.1-Dichloroethane 0.53 Reactive Silica 10.2

1.1-Dichloroethene 0.47Total Hardness asCaCO3 194.0

1.2-Dichloroethane 169.3 Arsenic 0.018

Carbon disulphide 4.12 Cadmium 0.00015

Carbon tetrachloride 15.8 Chromium 0.0017

Chloroethane 0.001 Copper 0.0013

Chloroform 5.01 Iron 13.5

Chloromethane 0.001 Iron (FSRIB) 43.9

cis-1.2-Dichloroethene 1.29 Lead 0.00010

Methylene chloride 0.08 Mercury 0.0000087

Tetrachloroethene 13.6 Nickel 0.0018

trans-1.2-Dichloroethene 0.24 Zinc 0.017

Trichloroethene 7.22 Ammonia as N 10.6

Vinyl chloride 5.98 Nitrate as N 0.05

Hexachloroethane 0.053 Nitrite as N 0.03

Hexachlorobutadiene 0.023Total Phosphorus asP 0.95

Benzene 1.17 BOD 72.5

Toluene 0.0035 Sulphide 5.54

Page 89: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 2004A2.3

NameFeedwater

Compositionmg/L

NameFeedwater

Compositionmg/L

2-Methylphenol 0.0005Suspended Solids(SS) 0.50

3- & 4-Methylphenol 0.0032 Selenium - Filtered 0.02

Chlorobenzene 0.0016Manganese -Filtered 0.14

1.2-Dichlorobenzene 0.043 Fluoride 0.080

1.3-Dichlorobenzene 0.0010 Bromide 0.35

1.4-Dichlorobenzene 0.031 Formic acid 2.00

2.4-Dichlorophenol 0.10 Acetic Acid 45.6

2.6-Dichlorophenol 0.030 Propionic Acid 0.36

1.2.4-Trichlorobenzene* 0.001 Butyric Acid 10.4

2.4.5-Trichlorophenol 0.001 Valeric Acid 0.00

2.4.6-Trichlorophenol 0.02 Hexanoic Acid 5.44

2-Chlorophenol 0.01 2,4-D 0.083

Phenol 0.002Aluminium -Filtered 1.21

Sodium 395.9 Barium - Filtered 0.10

Potassium 9.37 Cyanide (Free) 0.000023

Calcium 50.33 Silver - Filtered 0.00029

Page 90: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.doc8 November 2004A3.1

Appendix 3

Selected MSDS’s

Preliminary Hazard Analysis, GroundwaterTreatment Plant, Orica Australia Pty Ltd,

Botany Industrial Park

Page 91: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Product name: EDC Substance Key: 000030703003

Issued: 29.09.1999 Version: 1.2 Page: 1 of 7

0DWHULDO�6DIHW\�'DWD�6KHHW

1. IDENTIFICATION OF THE SUBSTANCE/PREPARATION AND THECOMPANY/UNDERTAKING

Product name: EDC

Synonyms: EDC1,2-Dichloroethane, Ethylene dichlorideCAS-No.: 107-06-2

Molecular Formula: C2-H4-Cl2

Supplier: Orica Australia Pty LtdACN: 004 117 828Street Address: 1 Nicholson Street

Melbourne 3000Australia

Telephone: + 61 3 9665 7111Facsimile: + 61 3 9665 7937

Emergency telephone number: 1 800 033 111 (ALL HOURS)

2. COMPOSITION/INFORMATION ON INGREDIENTS

Recommended use: Production of vinyl chloride, solvent, metal degreasing, paint and varnish remover.

Appearance: Clear, colourless liquid. Chloroform-like odour.

3. HAZARDS IDENTIFICATION

Hazardous according to criteria of Worksafe Australia.

Hazard CategoryT ToxicXi Irritant

R-phrase(s)R11 Highly flammable.R22 Harmful if swallowed.R36/37/38 Irritating to eyes, respiratory system and skin.R45(2) May cause cancer.

Classified as Dangerous Goods by the criteria of the Australian Dangerous Goods Code (ADG Code) fortransport by road or rail.Class: 3 Flammable LiquidSubsidiary Risk 1: 6.1 Toxic

Poisons Schedule (Aust)/Toxic Substance (NZ): S6

Page 92: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Product name: EDC Substance Key: 000030703003

Issued: 29.09.1999 Version: 1.2 Page: 2 of 7

0DWHULDO�6DIHW\�'DWD�6KHHW

This material is a Scheduled Poison S6 and must be stored, maintained and used in accordance with therelevant regulations.

4. FIRST AID MEASURES

Poison Information Centres in each State capital city can provide additional assistance for scheduled poisons.

Ingestion: Immediately rinse mouth with water. Give water to drink. Do NOT induce vomiting. Seekimmediate medical assistance.

Eye contact: Immediately irrigate with copious quantities of water for at least 15 minutes. Eyelids to be heldopen. Remove clothing if contaminated and wash skin. Seek immediate medical assistance.

Skin contact: Immediately wash contaminated skin with plenty of soap and water. Remove contaminatedclothing and wash before re-use. If swelling, redness, blistering or irritation occurs seek medical advice.

Inhalation: Remove victim from exposure - avoid becoming a casualty. Remove contaminated clothing andloosen remaining clothing. Allow patient to assume most comfortable position and keep warm. Keep at rest untilfully recovered. If breathing laboured and patient cyanotic (blue), ensure airways are clear and have qualifiedperson give oxygen through a face mask. If breathing has stopped apply artificial respiration at once. In eventof cardiac arrest, apply external cardiac massage. Seek immediate medical advice.

Notes to physician: Treat symptomatically. DO NOT give stimulants. Symptoms may include cyanosis, fall of blood pressure, vomiting, diarrhoea, cardiovascular collapse, and coma. If exposure is severe, these rapidly progress to pulmonary oedema and respiratory difficulty. Renal and hepatic failure are possible complications.

5. FIRE-FIGHTING MEASURES

Specific hazards: Highly flammable liquid. May form flammable vapour mixtures with air. Avoid all ignitionsources. Flameproof equipment necessary in area where this chemical is being used. Nearby equipment mustbe earthed. Vapour may travel a considerable distance to source of ignition and flash back.

Fire fighting further advice: Highly flammable liquid. On burning will emit toxic fumes including those ofhydrogen chloride and phosgene. Heating can cause expansion or decomposition leading to violent rupture ofcontainers. If safe to do so, remove containers from path of fire. Keep containers cool with water. Fire fightersto wear self contained breathing apparatus if risk of exposure to vapour or products of combustion.

Suitable extinguishing media: Water fog (or if unavailable fine water spray), foam, dry agent (carbon dioxide,dry chemical powder).

6. ACCIDENTAL RELEASE MEASURES

Avoid inhalation of vapours. Work up wind or increase ventilation. Shut off all possible sources of ignition. Cleararea of all unprotected personnel. Wear protective equipment to prevent skin and eye contamination andinhalation of vapours. Contain - prevent run off into drains and waterways. Use absorbent (soil, sand or other

Page 93: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Product name: EDC Substance Key: 000030703003

Issued: 29.09.1999 Version: 1.2 Page: 3 of 7

0DWHULDO�6DIHW\�'DWD�6KHHW

inert material). Collect and seal in properly labelled drums for disposal. Wash area down with excess water.Material MUST BE thoroughly dispersed and diluted with excess water before allowing any entry to drains. Usewater spray to disperse vapour. If contamination of sewers or waterways has occurred advise local emergencyservices.

7. HANDLING AND STORAGE

Storage: Store in a cool place and out of direct sunlight. Store in a well ventilated area. Store away fromoxidising agents, acids, alkalis and foodstuffs. Store away from sources of heat or ignition. Keep containersclosed at all times - check regularly for leaks.

This material is a Scheduled Poison S6 and must be stored, maintained and used in accordance with therelevant regulations.

8. EXPOSURE CONTROLS / PERSONAL PROTECTION

National occupational exposure limits

TWA STEL Carcin-ogenCategory

Notices

10 ppm 40 mg/m3

As published by National Occupational Health and Safety Commission (Worksafe Australia).

Exposure Standard (TWA) is the time-weighted average airborne concentration over an eight-hour working day,for a five-day working week over an entire working life. According to current knowledge this concentration shouldneither impair the health or, not cause undue discomfort to, nearly all workers.

These Exposure Standards are guides to be used in the control of occupational health hazards. All atmosphericcontamination should be kept to as low a level as is workable. These Exposure Standards should not be usedas fine dividing lines between safe and dangerous concentrations of chemicals. They are not a measure ofrelative toxicity.

Engineering measures: Ensure ventilation is adequate to maintain air concentrations below ExposureStandards. If inhalation risk exists, use with local exhaust ventilation or while wearing organic vapour respiratoror air supplied mask. Vapour heavier than air - prevent concentration in hollows or sumps. DO NOT enterconfined spaces where vapour may have collected. Keep containers closed when not in use.

Personal protection equipment: Orica Personal Protection Guide No.1, 1998: H - OVERALLS, SAFETYSHOES, CHEMICAL GOGGLES, GLOVES (S), RESPIRATOR.

Avoid skin and eye contact and inhalation of vapour. Wear overalls, chemical goggles and impervious gloves.Use with adequate ventilation. If inhalation risk exists wear organic vapour respirator meeting the requirementsof AS/NZS 1715 and AS/NZS 1716.Available information suggests that gloves made from Teflon (TM) (not neoprene or rubber) should be suitablefor intermittent contact. (6) However, due to variations in glove construction and local conditions, a finalassessment should be made by the user. Always wash hands before smoking, eating, drinking or using thetoilet. Wash contaminated clothing and other protective equipment before storage or re-use.

Page 94: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Product name: EDC Substance Key: 000030703003

Issued: 29.09.1999 Version: 1.2 Page: 4 of 7

0DWHULDO�6DIHW\�'DWD�6KHHW

9. PHYSICAL AND CHEMICAL PROPERTIES

Form / Colour / Odour: Clear, colourless liquid. Chloroform-like odour.

Solubility: Slightly soluble in water.

Specific Gravity : 1.26 @ 20 °C Relative Vapour Density : 3.4 @ 20 °C Flash point : 13 °C CC Boiling Point : 83.5 °C Vapour Pressure (20 C) : 8.1 kPa Decomp. Point (C) : N Av Flammability Limits (%) : 6.2-16 pH : N App Autoignition Temp (C) : 440 Viscosity : N Av % Volatile by volume : 100 Evaporation Rate : N Av Solubility in water (g/L) : 8.7 @ 20C (n-Butyl acetate=1) (Typical values only - consult specification sheet) N Av = Not available N App = Not applicable

10. STABILITY AND REACTIVITY

Stability: Avoid contact with oxidising materials, acids and alkalis. Attacks many metals in the presence ofwater. Attacks many plastics. Can produce toxic fumes (hydrogen chloride and phosgene) on contact with hotsurfaces or under the influence of electrostatic charges.

11. TOXICOLOGICAL INFORMATION

Main symptoms: No adverse health effects expected if the product is handled in accordance with this SafetyData Sheet and the product label. Symptoms that may arise if the product is mishandled are:

Ingestion: Swallowing can result in nausea, vomiting and central nervous system depression. If the victim isuncoordinated there is a greater likelihood of vomit entering the lungs and causing subsequent complications.

Eye contact: An eye irritant.

Skin contact: Contact with skin will result in irritation. Will have a degreasing action on the skin. Repeated orprolonged skin contact may lead to irritant contact dermatitis.

Inhalation: Vapour is irritant to mucous membranes and respiratory tract. Inhalation of vapour can result inheadaches, dizziness and possible nausea. Inhalation of high concentrations can produce central nervoussystem depression, which can lead to loss of co-ordination, impaired judgement and, if exposure is prolonged,unconsciousness.

Long Term Effects: Evidence from animal tests indicates the repeated or prolonged exposure to this chemicalcould result in liver, kidney and central nervous system disorders. Animal studies (rat and mouse) haveindicated this chemical to be carcinogenic by oral administration, while inhalation studies have beeninconclusive. There is no adequate evidence for carcinogenicity in humans, however in view of the animal datathe material should be considered a possible human carcinogen.

Page 95: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Product name: EDC Substance Key: 000030703003

Issued: 29.09.1999 Version: 1.2 Page: 5 of 7

0DWHULDO�6DIHW\�'DWD�6KHHW

Acute toxicity / Chronic toxicity Oral LD50 (mouse): 413 mg/kg; Oral LD50 (rat): 670 mg/kg. (1) Inhalation LC50 (rat): 1000 ppm/7 hr. (1) 1,2-Dichloroethane has been shown to cause nervous system disorders, liver and kidney disease, lung and heart effects. (2) There is sufficient evidence that 1,2-dichloroethane causes cancer in animals (rats and mice) by oral and skin routes. Long term inhalation studies on rats and mice at 150 ppm for 78 weeks have been inconclusive with regards to carcinogenic effects. (2) In view of these cancer findings the possibility of cancer in humans cannot be excluded, however specific evidence associated with inhalation exposure and the occurrence of cancer in humans has not been found in the literature. (2) This material has been classified by the International Agency for Research on Cancer (IARC) as a Group 2B agent. Group 2B - The agent is possibly carcinogenic to humans. (3) Positive in IN VITRO mutagenicity assays. (4) Evidence from animal studies suggests that 1,2-dichloroethane probably does not produce birth defects or affect reproduction. (2,5)

12. ECOLOGICAL INFORMATION

Avoid contaminating waterways. LC50 values for fish exposed for 1-4 days ranged between 85 and 550 mg/litre water, with bioaccumulation unlikely. A no observed adverse effect level of 11 mg/litre was found for DAPHNIA MAGNA, following long term exposure. (5)

n-Octanol/Water Partition Coefficient, P: log P = 1.48 (5)

13. DISPOSAL CONSIDERATIONS

Refer to State Land Waste Management Authority. Advise flammable nature. Empty containers must bedecontaminated and destroyed. Normally suitable for incineration by approved agent.

14. TRANSPORT INFORMATION

Classified as Dangerous Goods by the criteria of the Australian Dangerous Goods Code (ADG Code) fortransport by road or rail.

UN-No: 1184Class: 3 Flammable LiquidSubsidiary Risk 1: 6.1 ToxicHazchem code: 2YEEPG: 3A2Packing group: Packing Group 2

Proper shipping name: ETHYLENE DICHLORIDE

Page 96: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Product name: EDC Substance Key: 000030703003

Issued: 29.09.1999 Version: 1.2 Page: 6 of 7

0DWHULDO�6DIHW\�'DWD�6KHHW

Segregation Dangerous Goods: Not to be loaded with explosives (class 1), flammable gases (class 2.1) ifboth are in bulk, toxic gases (class 2.3), nitromethane (class 3), spontaneously combustible substances (class4.2), oxidising agents (class 5.1), organic peroxides (class 5.2), radioactive substances (class 7), food and foodpackaging in any quantity, however, exemptions may apply.

15. REGULATORY INFORMATION

Hazardous according to criteria of Worksafe Australia.

Hazard CategoryT ToxicXi Irritant

R-phrase(s)R11 Highly flammable.R22 Harmful if swallowed.R36/37/38 Irritating to eyes, respiratory system and skin.R45(2) May cause cancer.

S-phrase(s)S16 Keep away from sources of ignition - No smoking.S23 Do not breathe vapour.S24/25 Avoid contact with skin and eyes.S26 In case of contact with eyes, rinse immediately with plenty of water and seek medical

advice.S29 Do not empty into drains.S45 In case of accident or if you feel unwell, seek medical advice immediately (show label

where possible).S53 Avoid exposure - obtain special instructions before use.

Poisons Schedule (Aust)/Toxic Substance (NZ): S6

16. OTHER INFORMATION

Literary reference

(1), In ’Registry of Toxic Effects of Chemical Substances 1999’ (Ed. D. Sweet),KI05250000, (US Dept. of Health & Human Services: Cincinatti 1999). (2), In ’Toxicological Profile for 1,2-Dichloroethane Update’, US Department of Health and Human Services, May, 1994. (3), In ’IARC Monographs on the Evaluation of Carcinogenic Risk to Humans Suppl. 7’ (Eds International Agency for Research on Cancer), (International Agency for Research on Cancer: U.K. 1987). (4), In ’IARC Monographs on the Evaluation of Carcinogenic Risk to Humans Vol 20’ (Eds International Agency for Research on Cancer), (International Agency for Research on Cancer: U.K. 1979). (5), In ’Environmental Health Criteria 62 - 1,2-Dichloroethane’, World HealthOrganization, Geneva, 1987.

Page 97: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Product name: EDC Substance Key: 000030703003

Issued: 29.09.1999 Version: 1.2 Page: 7 of 7

0DWHULDO�6DIHW\�'DWD�6KHHW

(6), In ’Quick Selection Guide to Chemical Protective Clothing Third Edition’,(Eds. Forsberg, K. and Mansdon, S.Z.), (Van Nostrand Reinhold, New York, 1997). This chemical is listed on the Australian Inventory of Chemical Substances (AICS). This Material Safety Data Sheet has been prepared by SHE Pacific Pty Ltd on behalf of Orica Ltd and its subsidiary companies. Contact Point: SHE Pacific Pty Ltd, MSDS Services Within Australia: Telephone 1 800 624 132 Facsimile (03) 9665 7929 Outside Australia: Telephone +61 3 9665 7500 Facsimile +61 3 9665 7929

Reason(s) For Issue: Five yearly update; Change in First Aid Measures; Change inToxicological Information; Change in Fire-Fighting Measures; Change in Storage andTransport requirements. Safety Data Sheets are updated frequently. Please ensure that you have a current copy.

This MSDS summarises at the date of issue our best knowledge of the health and safety hazard information ofthe product, and in particular how to safely handle and use the product in the workplace. Since Orica Limited andits subsidiaries cannot anticipate or control the conditions under which the product may be used, each usermust, prior to usage, review this MSDS in the context of how the user intends to handle and use the product inthe workplace.

If clarification or further information is needed to ensure that an appropriate assessment can be made, the usershould contact this company.

Our responsibility for product as sold is subject to our standard terms and conditions, a copy of which is sent toour customers and is also available upon request.

Page 98: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Material Safety Data Sheet

This material is hazardous according to criteria of NOHSC.Classified as Dangerous Goods by the criteria of the Australian Dangerous Goods Code (ADG Code) for Transport by Road and Rail.

1. Identification of the substance/preparation and of the company/undertaking

Product Name: WASTE EDC (FROM CONTAMINATED GROUNDWATER)

Supplier: Orica Australia Pty LtdABN: 004 117 828Street Address: 1 Nicholson Street,

Melbourne 3000Australia

Telephone Number: +61 3 9665 7111Facsimile: +61 3 9665 7937

Emergency Telephone: 1 800 033 111 (ALL HOURS)

2. Composition/information on ingredients

Product Description: Waste.

Components / CAS Number Proportion Risk PhrasesEthylene dichloride107-06-2

>60% R11, R22, R36/37/38, Carc.Cat.2 R45

Vinyl chloride75-01-4

<10% R12, Carc. Cat 1 R45

1,1,2-Trichloroethane79-00-5

<10% R20/21/22

Benzene71-43-2

<1% R11, Carc. Cat.1 R45, R48/23/24/25

Chloroform67-66-3

<1% R22, R38, Carc. Cat. 3 R40, R48/20/22

Water7732-18-5

<1% -

1,2-Dichloroethene (E)-156-60-5

<10% R11, R20, R52/53

1,1,2,2-Tetrachloroethane79-34-5

<1% R26/27, R51/53

3. Hazards identification

Risk Phrases: Highly Flammable. Harmful by inhalation, in contact with skin and if swallowed. Irritating to eyes, respiratory system and skin. May cause cancer.

Poisons Schedule: S7 Dangerous Poison.

Product Name: WASTE EDC (FROM CONTAMINATED GROUNDWATER)Substance No: 000000017093 Issued: 13/07/2004 Version: 1

Page 1 of 7

Page 99: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Material Safety Data Sheet

4. First-aid measures

For advice, contact a Poisons Information Centre (Phone eg. Australia 131 126; New Zealand 0 800 764766) or a doctor.

Inhalation: Remove victim from area of exposure - avoid becoming a casualty. Remove contaminated clothing and loosen remaining clothing. Allow patient to assume most comfortable position and keep warm. Keep at rest until fully recovered. If patient finds breathing difficult and develops a bluish discolouration of the skin (which suggests a lack of oxygen in the blood - cyanosis), ensure airways are clear of any obstruction and have a qualified person give oxygen through a face mask. Apply artificial respiration if patient is not breathing. Seek immediate medical advice.

Skin Contact: If skin or hair contact occurs, immediately remove any contaminated clothing and wash skin and hair thoroughly with running water and soap. If swelling, redness, blistering or irritation occurs seek medical assistance.

Eye Contact: If in eyes, hold eyelids apart and flush the eye continuously with running water. Continue flushing until advised to stop by the Poisons Information Centre or a doctor, or for at least 15 minutes.

Ingestion: Rinse mouth with water. If swallowed, do NOT induce vomiting. Give a glass of water. Seek immediate medical assistance.

Notes to physician: Treat symptomatically.

5. Fire-fighting measures

Specific Hazards: Highly flammable liquid. Avoid all ignition sources. All potential sources of ignition (open flames, pilot lights, furnaces, spark producing switches and electrical equipment etc) must be eliminated both in and near the work area. Do NOT smoke. May form flammable vapour mixtures with air. Vapour may travel a considerable distance to source of ignition and flash back.

Fire-fighting advice: On burning will emit toxic fumes, including those of oxides of carbon . Heating can cause expansion or decomposition of the material, which can lead to the containers exploding. If safe to do so, remove containers from the path of fire. Keep containers cool with water spray. If safe to do so, remove containers from path of fire. Fire fighters to wear self-contained breathing apparatus and suitable protective clothing if risk of exposure to vapour or products of combustion.

Suitable Extinguishing Media: Foam, dry agent (carbon dioxide, dry chemical powder).

6. Accidental release measures

Shut off all possible sources of ignition. Clear area of all unprotected personnel. Slippery when spilt. Avoid accidents, clean up immediately. Wear protective equipment to prevent skin and eye contact and breathing in vapours. Work up wind or increase ventilation. Contain - prevent run off into drains and waterways. Use absorbent (soil, sand or other inert material). Use a spark-free shovel. Collect and seal in properly labelled containers or drums for disposal. If contamination of sewers or waterways has occurred advise local emergency services.

7. Handling and storage

Handling advice: Avoid skin and eye contact and breathing in vapour. Take precautionary measures against static discharges.

Product Name: WASTE EDC (FROM CONTAMINATED GROUNDWATER)Substance No: 000000017093 Issued: 13/07/2004 Version: 1

Page 2 of 7

Page 100: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Material Safety Data Sheet

Storage advice: Store in a cool, dry, well ventilated place and out of direct sunlight. Store away from sources of heat or ignition. Store away from foodstuffs. Store away from incompatible materials described in Section 10. Keep containers closed when not in use - check regularly for leaks.

This material is a Scheduled Poison S7 and must be stored, maintained and used in accordance with the relevant regulations.

8. Exposure controls/personal protection

Occupational Exposure Limits:No value assigned for this specific material by the National Occupational Health and Safety Commission. However, Exposure Standard(s) for constituent(s):

Ethylene dichloride: 8hr TWA = 40 mg/m3 (10 ppm) Vinyl chloride, monomer: 8hr TWA = 13 mg/m3 (5 ppm), Carcinogen Category 1 1,1,2-Trichloroethane: 8hr TWA = 55 mg/m3 (10 ppm), Sk 1,1,2,2-Tetrachloroethane: 8hr TWA = 6.9 mg/m3 (1 ppm), Sk Chloroform: 8hr TWA = 10 mg/m3 (2 ppm), Carcinogen Category 3 Benzene: 8hr TWA = 3.2 mg/m3 (1 ppm), Carcinogen Category 1

As published by the National Occupational Health and Safety Commission.

TWA - The time-weighted average airborne concentration over an eight-hour working day, for a five-day working week over an entire working life.

`Sk' Notice – absorption through the skin may be a significant source of exposure. The exposure standard is invalidated if such contact should occur.

Carcinogen Category 1 – established human carcinogen. There is sufficient evidence to establish a causal association between human exposure and the development of cancer.

Carcinogen Category 3 – substances suspected of having carcinogenic potential. The available information is not adequate for making a satisfactory assessment.

These Exposure Standards are guides to be used in the control of occupational health hazards. All atmospheric contamination should be kept to as low a level as is workable. These exposure standards should not be used as fine dividing lines between safe and dangerous concentrations of chemicals. They are not a measure of relative toxicity.

Engineering Control Measures:Ensure ventilation is adequate and that air concentrations of components are controlled below quoted Exposure Standards. Vapour heavier than air - prevent concentration in hollows or sumps. DO NOT enter confined spaces where vapour may have collected. If inhalation risk exists: Use with local exhaust ventilation or while wearing organic vapour/particulate respirator. Keep containers closed when not in use.

Personal Protective Equipment:Orica Personal Protection Guide No. 1, 1998: H - OVERALLS, SAFETY SHOES, CHEMICAL GOGGLES, GLOVES,

Product Name: WASTE EDC (FROM CONTAMINATED GROUNDWATER)Substance No: 000000017093 Issued: 13/07/2004 Version: 1

Page 3 of 7

Page 101: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Material Safety Data Sheet

RESPIRATOR.

Wear overalls, chemical goggles and impervious gloves. Use with adequate ventilation. If inhalation risk exists wear organic vapour/particulate respirator or air supplied mask meeting the requirements of AS/NZS 1715 and AS/NZS 1716. Always wash hands before smoking, eating, drinking or using the toilet. Wash contaminated clothing and other protective equipment before storage or re-use.

9. Physical and chemical properties

Physical state: LiquidColour: ColourlessOdour: PleasantSolubility: Slightly soluble in water.Specific Gravity: ca. 1.2Relative Vapour Density (air=1): >1Vapour Pressure (20 °C): ca. 8 kPaFlash Point (°C): 13 (for Ethylene dichloride)Flammability Limits (%): 6.2-16 (for Ethylene dichloride)Autoignition Temperature (°C): 440 (for Ethylene dichloride)Boiling Point/Range (°C): 83.5 (for Ethylene dichloride)pH: Not applicableViscosity: 0.82 cP @ 20°C

10. Stability and reactivity

Stability: Incompatible with oxidising agents , acids , and alkalis . Attacks metals in the presence of moisture.

11. Toxicological information

No adverse health effects expected if the product is handled in accordance with this Safety Data Sheet and the product label. Symptoms or effects that may arise if the product is mishandled and overexposure occurs are:

Ingestion: Swallowing can result in nausea, vomiting and central nervous system depression. If the victim is showing signs of central system depression (like those of drunkeness) there is greater likelihood of the patient breathing in vomit and causing damage to the lungs.

Eye contact: An eye irritant.Skin contact: Contact with skin will result in irritation. Will have a degreasing action on the skin. Repeated or prolonged

skin contact may lead to irritant contact dermatitis. Component/s of this material can be absorbed through the skin with resultant toxic effects.

Inhalation: Breathing in vapour will produce respiratory irritation. Breathing in vapour can result in headaches, dizziness, drowsiness, and possible nausea. Breathing in high concentrations can produce central nervous system depression, which can lead to loss of co-ordination, impaired judgement and if exposure is prolonged, unconsciousness.

Long Term Effects:Available evidence from animal studies indicate that repeated or prolonged exposure to this material could result in effects on the liver , kidneys , and central nervous system. May cause cancer.

Product Name: WASTE EDC (FROM CONTAMINATED GROUNDWATER)Substance No: 000000017093 Issued: 13/07/2004 Version: 1

Page 4 of 7

Page 102: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Material Safety Data Sheet

Toxicological Data:No LD50 data available for the product. For the constituent Ethylene dichloride :

Oral LD50 (rat): 500 mg/kg.Oral LD50 (mice): 413 mg/kg.Oral LD50 (rabbit): 860 mg/kg.

12. Ecotoxicological information

Avoid contaminating waterways.

13. Disposal considerations

Refer to Waste Management Authority. Dispose of material through a licensed waste contractor. Advise flammable nature. Normally suitable for incineration by an approved agent.

14. Transport information

Road and Rail TransportClassified as Dangerous Goods by the criteria of the Australian Dangerous Goods Code (ADG Code) for Transport by Road and Rail.

UN No: 1992Class-primary 3 Flammable LiquidSubrisk 1: 6.1 ToxicPacking Group: IIProper Shipping Name: FLAMMABLE LIQUID, TOXIC, N.O.S. (CONTAINS ETHYLENE DICHLORIDE AND VINYL

CHLORIDE)

Hazchem Code: 3WE

Marine TransportClassified as Dangerous Goods by the criteria of the International Maritime Dangerous Goods Code (IMDG Code) for transport by sea.

UN No: 1992Class-primary: 3 Flammable LiquidSubrisk 1: 6.1 ToxicPacking Group: IIProper Shipping Name: FLAMMABLE LIQUID, TOXIC, N.O.S. (CONTAINS ETHYLENE DICHLORIDE AND VINYL

CHLORIDE)

Air TransportClassified as Dangerous Goods by the criteria of the International Air Transport Association (IATA) Dangerous Goods

Product Name: WASTE EDC (FROM CONTAMINATED GROUNDWATER)Substance No: 000000017093 Issued: 13/07/2004 Version: 1

Page 5 of 7

Page 103: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Material Safety Data Sheet

Regulations for transport by air.

UN No: 1992Class-primary: 3 Flammable LiquidSubrisk 1: 6.1 ToxicPacking Group: IIProper Shipping Name: FLAMMABLE LIQUID, TOXIC, N.O.S. (CONTAINS ETHYLENE DICHLORIDE AND VINYL

CHLORIDE)

15. Regulatory information

Classification: This material is hazardous according to criteria of NOHSC.T : Toxic Xi: Irritant

Risk Phrase(s): R11: Highly Flammable.R20/21/22: Harmful by inhalation, in contact with skin and if swallowed.R36/37/38: Irritating to eyes, respiratory system and skin.Carc. Cat. 1. R45: May cause cancer.

Safety Phrase(s): S16: Keep away from sources of ignition - No smoking.S23: Do not breathe vapour/mist/aerosol.S24/25: Avoid contact with skin and eyes.S26: In case of contact with eyes, rinse immediately with plenty of water and seek medical advice.S36/37/39: Wear suitable protective clothing, gloves and eye/face protection.

Poisons Schedule: S7 Dangerous Poison.

All the constituents of this material are listed on the Australian Inventory of Chemical Substances (AICS).

16. Other information

This material safety data sheet has been prepared by SH&E Shared Services, Orica.

Reason(s) for Issue:First Issue Primary MSDS

This MSDS summarises to our best knowledge at the date of issue, the chemical health and safety hazards of the material and general guidance on how to safely handle the material in the workplace. Since Orica Limited cannot anticipate or control the conditions under which the product may be used, each user must, prior to usage, assess and control the risks arising from its use of the material.

If clarification or further information is needed, the user should contact their Orica representative or Orica Limited at the

Product Name: WASTE EDC (FROM CONTAMINATED GROUNDWATER)Substance No: 000000017093 Issued: 13/07/2004 Version: 1

Page 6 of 7

Page 104: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Material Safety Data Sheet

contact details on page 1.

Orica Limited's responsibility for the material as sold is subject to the terms and conditions of sale, a copy of which is available upon request.

Product Name: WASTE EDC (FROM CONTAMINATED GROUNDWATER)Substance No: 000000017093 Issued: 13/07/2004 Version: 1

Page 7 of 7

Page 105: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Material Safety Data Sheet

This material is hazardous according to criteria of NOHSC.Not classified as Dangerous Goods by the criteria of the Australian Dangerous Goods Code (ADG Code) for transport by Road and Rail.

1. Identification of the substance/preparation and of the company/undertaking

Product Name: SOUTHLANDS GROUNDWATER - CENTRAL PLUME

Supplier: Orica Australia Pty LtdABN: 004 117 828Street Address: 1 Nicholson Street

Melbourne 3000 Australia

Telephone Number: +61 3 9665 7111Facsimile: +61 3 9665 7937

Emergency Telephone: 1 800 033 111 (ALL HOURS)

2. Composition/information on ingredients

Product Description: Contaminated ground water for storage, treatment, and/or experimentation.

Components / CAS Number Proportion Risk PhrasesNon hazardous component(s)-

>99% -

Trichloroethylene79-01-6

<0.1% R36/38, Muta. Cat.3 R40, Carc. Cat.2 R45, R52/53, R67

Vinyl chloride75-01-4

<0.01% R12, Carc. Cat 1 R45

1,1,2-Trichloroethane79-00-5

<0.01% R20/21/22

Ethylene dichloride107-06-2

>0.1-<0.5% R11, R22, R36/37/38, Carc.Cat.2 R45

3. Hazards identification

Risk Phrases: May cause cancer.

Poisons Schedule: S7 Dangerous Poison.

4. First-aid measures

For advice, contact a Poisons Information Centre (Phone eg. Australia 131 126; New Zealand 0 800 764766) or a doctor. Not to be available except to authorised or licensed persons.

Product Name: SOUTHLANDS GROUNDWATER - CENTRAL PLUMESubstance No: 000000008892 Issued: 07/10/2003 Version: 1

Page 1 of 5

Page 106: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Material Safety Data Sheet

Inhalation: Remove victim from area of exposure - avoid becoming a casualty. Seek medical advice if effects persist.

Skin Contact: If skin contact occurs, remove contaminated clothing and wash skin with running water. If irritation occurs seek medical advice.

Eye Contact: If in eyes, wash out immediately with water. In all cases of eye contamination it is a sensible precaution to seek medical advice.

Ingestion: Rinse mouth with water. If swallowed, do NOT induce vomiting. Give a glass of water. Seek medical advice.

Notes to physician: Treat symptomatically.

5. Fire-fighting measures

Specific Hazards: Non-combustible material.Fire-fighting advice: Decomposes on heating emitting toxic fumes. Fire fighters to wear self-contained

breathing apparatus and suitable protective clothing if risk of exposure to products of decomposition.

Suitable Extinguishing Media: Not combustible, however, if material is involved in a fire use: Extinguishing media appropriate to surrounding fire conditions.

6. Accidental release measures

Slippery when spilt. Avoid accidents, clean up immediately. Contain - prevent run off into drains and waterways. Use absorbent (soil, sand or other inert material). Collect and seal in properly labelled containers or drums for disposal.

7. Handling and storage

Handling advice: Avoid skin and eye contact and breathing in vapour, mists and aerosols.

Storage advice: Store in a cool, dry, well ventilated place and out of direct sunlight. Store away from foodstuffs. Keep containers closed when not in use - check regularly for leaks.

This material is a Scheduled Poison S7 and must be stored, maintained and used in accordance with the relevant regulations.

8. Exposure controls/personal protection

Occupational Exposure Limits:No value assigned for this specific material by the National Occupational Health and Safety Commission. However, Exposure Standard(s) for constituent(s):

Ethylene dichloride: 8hr TWA = 40 mg/m3 (10 ppm) 1,1,2-Trichloroethane: 8hr TWA = 55 mg/m3 (10 ppm), Sk Trichloroethylene: 8hr TWA = 270 mg/m3 (50 ppm), 15 min STEL = 1080 mg/m3 (200 ppm) Vinyl chloride, monomer: 8hr TWA = 13 mg/m3 (5 ppm), Carcinogen Category 1

Product Name: SOUTHLANDS GROUNDWATER - CENTRAL PLUMESubstance No: 000000008892 Issued: 07/10/2003 Version: 1

Page 2 of 5

Page 107: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Material Safety Data Sheet

As published by the National Occupational Health and Safety Commission.

TWA - The time-weighted average airborne concentration over an eight-hour working day, for a five-day working week over an entire working life.

STEL (Short Term Exposure Limit) – the average airborne concentration over a 15 minute period which should not be exceeded at any time during a normal eight hour work day. According to current knowledge this concentration should neither impair the health of, nor cause undue discomfort to, nearly all workers.

Carcinogen Category 1 – established human carcinogen. There is sufficient evidence to establish a causal association between human exposure and the development of cancer.

`Sk' Notice – absorption through the skin may be a significant source of exposure. The exposure standard is invalidated if such contact should occur.

These Exposure Standards are guides to be used in the control of occupational health hazards. All atmospheric contamination should be kept to as low a level as is workable. These exposure standards should not be used as fine dividing lines between safe and dangerous concentrations of chemicals. They are not a measure of relative toxicity.

Engineering Control Measures:Ensure ventilation is adequate and that air concentrations of components are controlled below quoted Exposure Standards. Keep containers closed when not in use.

Personal Protective Equipment:Orica Personal Protection Guide No. 1, 1998: B - OVERALLS, SAFETY SHOES, SAFETY GLASSES, GLOVES.

Wear overalls, safety glasses and impervious gloves. Always wash hands before smoking, eating, drinking or using the toilet. Wash contaminated clothing and other protective equipment before storage or re-use.

9. Physical and chemical properties

Physical state: LiquidOdour: SlightSolubility: Miscible in water.Specific Gravity: ca. 1 @20°CFlash Point (°C): Not applicable.Boiling Point/Range (°C): ca. 100pH: ca. 7 (1% aqueous solution)

10. Stability and reactivity

Stability: The material is stable under normal ambient and anticipated storage and handling conditions of temperature and pressure.

11. Toxicological information

Product Name: SOUTHLANDS GROUNDWATER - CENTRAL PLUMESubstance No: 000000008892 Issued: 07/10/2003 Version: 1

Page 3 of 5

Page 108: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Material Safety Data Sheet

No adverse health effects expected if the product is handled in accordance with this Safety Data Sheet and the product label. Symptoms or effects that may arise if the product is mishandled and overexposure occurs are:

Ingestion: No adverse effects expected, however large amounts may cause nausea and vomiting.Eye contact: May be an eye irritant.Skin contact: Contact with skin may result in irritation.Inhalation: Breathing in mists or aerosols may produce respiratory irritation.

Long Term Effects:May cause cancer.

Toxicological Data:No LD50 data available for the product. However, for constituent(s) Ethylene dichloride :

Oral LD50 (rat): 670 mg/kg.

12. Ecotoxicological information

Avoid contaminating waterways.

13. Disposal considerations

Refer to Waste Management Authority. Dispose of material through a licensed waste contractor.

14. Transport information

Road and Rail TransportNot classified as Dangerous Goods by the criteria of the Australian Dangerous Goods Code (ADG Code) for transport by Road and Rail.

Marine TransportNot classified as Dangerous Goods by the criteria of the International Maritime Dangerous Goods Code (IMDG Code) for transport by sea.

Air TransportNot classified as Dangerous Goods by the criteria of the International Air Transport Association (IATA) Dangerous Goods Regulations for transport by air.

15. Regulatory information

Classification: This material is hazardous according to criteria of NOHSC.T : Toxic

Product Name: SOUTHLANDS GROUNDWATER - CENTRAL PLUMESubstance No: 000000008892 Issued: 07/10/2003 Version: 1

Page 4 of 5

Page 109: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Material Safety Data Sheet

Risk Phrase(s): Carc. Cat. 2. R45: May cause cancer.

Safety Phrase(s): S45: In case of accident or if you feel unwell, seek medical advice immediately (show the label whenever possible).S53: Avoid exposure – obtain special instructions before use.

Poisons Schedule: S7 Dangerous Poison.

16. Other information

`Registry of Toxic Effects of Chemical Substances'. Ed. D. Sweet, US Dept. of Health & Human Services: Cincinatti, 2003.

This material safety data sheet has been prepared by SH&E Shared Services, Orica.

Reason(s) for Issue:Revised Primary MSDSChange in Formulation

This MSDS summarises to our best knowledge at the date of issue, the chemical health and safety hazards of the material and general guidance on how to safely handle the material in the workplace. Since Orica Limited cannot anticipate or control the conditions under which the product may be used, each user must, prior to usage, assess and control the risks arising from its use of the material.

If clarification or further information is needed, the user should contact their Orica representative or Orica Limited at the contact details on page 1.

Orica Limited's responsibility for the material as sold is subject to the terms and conditions of sale, a copy of which is available upon request.

Product Name: SOUTHLANDS GROUNDWATER - CENTRAL PLUMESubstance No: 000000008892 Issued: 07/10/2003 Version: 1

Page 5 of 5

Page 110: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Material Safety Data Sheet

Based on available information, not classified as hazardous according to criteria of NOHSC.Not classified as Dangerous Goods by the criteria of the Australian Dangerous Goods Code (ADG Code) for transport by Road and Rail.

1. Identification of the substance/preparation and of the company/undertaking

Product Name: SOUTHLANDS GROUND WATER

Supplier: Orica Australia Pty LtdABN: 004 117 828Street Address: 1 Nicholson Street

Melbourne 3000 Australia

Telephone Number: +61 3 9665 7111Facsimile: +61 3 9665 7937

Emergency Telephone: 1 800 033 111 (ALL HOURS)

2. Composition/information on ingredients

Product Description: Contaminated ground water for storage, treatment or experimentation.The major proportion of this material comprises of water. However, there are organic and inorganic contaminants at trace levels. The major contaminants of concern have been listed below:

Components / CAS Number Proportion Risk PhrasesEthylene dichloride107-06-2

<0.1% R11, R22, R36/37/38, Carc.Cat.2 R45

Carbon tetrachloride56-23-5

<0.05% R23/24/25, Carc. Cat.3 R40, R48/23, R52/53, R59

Tetrachloroethylene127-18-4

<0.025% Carc. Cat. 3 R40, R51/53

Vinyl chloride75-01-4

<0.02% R12, Carc. Cat 1 R45

Trichloroethylene79-01-6

<0.005% R36/38, Muta. Cat.3 R40, Carc. Cat.2 R45, R52/53, R67

Chloroform67-66-3

<0.005% R22, R38, Carc. Cat. 3 R40, R48/20/22

3. Hazards identification

Poisons Schedule: S7 Dangerous Poison.

4. First-aid measures

For advice, contact a Poisons Information Centre (Phone eg. Australia 131 126; New Zealand 0 800 764766) or a doctor.

Product Name: SOUTHLANDS GROUND WATERSubstance No: 000000001802 Issued: 22/03/2004 Version: 2

Page 1 of 5

Page 111: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Material Safety Data Sheet

Not to be available except to authorised or licensed persons.

Inhalation: Remove victim from area of exposure - avoid becoming a casualty. Remove contaminated clothing and loosen remaining clothing. Allow patient to assume most comfortable position and keep warm. Keep at rest until fully recovered. If patient finds breathing difficult and develops a bluish discolouration of the skin (which suggests a lack of oxygen in the blood - cyanosis), ensure airways are clear of any obstruction and have a qualified person give oxygen through a face mask. Apply artificial respiration if patient is not breathing. Seek immediate medical advice.

Skin Contact: If skin or hair contact occurs, remove contaminated clothing and flush skin and hair with running water. If irritation occurs seek medical advice.

Eye Contact: If in eyes, wash out immediately with water. In all cases of eye contamination it is a sensible precaution to seek medical advice.

Ingestion: Rinse mouth with water. If swallowed, do NOT induce vomiting. Give a glass of water. Seek medical advice.

Notes to physician: Treat symptomatically.

5. Fire-fighting measures

Specific Hazards: Non-combustible material.Fire-fighting advice: Non-combustible material.

Suitable Extinguishing Media: Not combustible, however, if material is involved in a fire use: Extinguishing media appropriate to surrounding fire conditions.

6. Accidental release measures

Slippery when spilt. Avoid accidents, clean up immediately. Contain - prevent run off into drains and waterways. Use absorbent (soil, sand or other inert material). Collect and seal in properly labelled containers or drums for disposal.

7. Handling and storage

Handling advice: Avoid skin and eye contact and breathing in vapour, mists and aerosols.

Storage advice: Store in a cool, dry, well ventilated place and out of direct sunlight. Store away from foodstuffs. Keep containers closed when not in use - check regularly for leaks.

This material is a Scheduled Poison S7 and must be stored, maintained and used in accordance with the relevant regulations.

8. Exposure controls/personal protection

Occupational Exposure Limits:No value assigned for this specific material by the National Occupational Health and Safety Commission. However, Exposure Standard(s) for constituent(s):

Carbon tetrachloride: 8hr TWA = 0.63 mg/m3 (0.1 ppm), Carcinogen Category 2, Sk

Product Name: SOUTHLANDS GROUND WATERSubstance No: 000000001802 Issued: 22/03/2004 Version: 2

Page 2 of 5

Page 112: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Material Safety Data Sheet

Chloroform: 8hr TWA = 10 mg/m3 (2 ppm), Carcinogen Category 3 Ethylene dichloride: 8hr TWA = 40 mg/m3 (10 ppm) Perchloroethylene: 8hr TWA = 340 mg/m3 (50 ppm), 15 min STEL = 1020 mg/m3 (150 ppm), Carcinogen Category 3 Trichloroethylene: 8hr TWA = 270 mg/m3 (50 ppm), 15 min STEL = 1080 mg/m3 (200 ppm) Vinyl chloride, monomer: 8hr TWA = 13 mg/m3 (5 ppm), Carcinogen Category 1

As published by the National Occupational Health and Safety Commission.

TWA - The time-weighted average airborne concentration over an eight-hour working day, for a five-day working week over an entire working life.

STEL (Short Term Exposure Limit) – the average airborne concentration over a 15 minute period which should not be exceeded at any time during a normal eight hour work day. According to current knowledge this concentration should neither impair the health of, nor cause undue discomfort to, nearly all workers.

`Sk' Notice – absorption through the skin may be a significant source of exposure. The exposure standard is invalidated if such contact should occur.

Carcinogen Category 1 – established human carcinogen. There is sufficient evidence to establish a causal association between human exposure and the development of cancer.

Carcinogen Category 2 - probable human carcinogen. There is sufficient evidence to provide a strong presumption that human exposure may result in the development of cancer. This evidence is generally based on appropriate long term animal studies, limited epidemiological evidance or other relevant information.

Carcinogen Category 3 – substances suspected of having carcinogenic potential. The available information is not adequate for making a satisfactory assessment.

These Exposure Standards are guides to be used in the control of occupational health hazards. All atmospheric contamination should be kept to as low a level as is workable. These exposure standards should not be used as fine dividing lines between safe and dangerous concentrations of chemicals. They are not a measure of relative toxicity.

Engineering Control Measures:Ensure ventilation is adequate and that air concentrations of components are controlled below quoted Exposure Standards. Keep containers closed when not in use.

Personal Protective Equipment:Orica Personal Protection Guide No. 1, 1998: B - OVERALLS, SAFETY SHOES, SAFETY GLASSES, GLOVES.

Wear overalls, safety glasses and impervious gloves. Always wash hands before smoking, eating, drinking or using the toilet. Wash contaminated clothing and other protective equipment before storage or re-use.

9. Physical and chemical properties

Physical state: LiquidOdour: SlightSolubility: Partially miscible with water.

Product Name: SOUTHLANDS GROUND WATERSubstance No: 000000001802 Issued: 22/03/2004 Version: 2

Page 3 of 5

Page 113: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Material Safety Data Sheet

Specific Gravity: 1 @20°CRelative Vapour Density (air=1): Not applicableVapour Pressure (20 °C): Not applicableFlash Point (°C): Not applicableFlammability Limits (%): Not applicableAutoignition Temperature (°C): Not applicableBoiling Point/Range (°C): 100pH: 7 (1% aqueous solution)

Freezing Point/Range (°C): 0

10. Stability and reactivity

Stability: The material is stable under normal ambient and anticipated storage and handling conditions of temperature and pressure.

11. Toxicological information

No adverse health effects expected if the product is handled in accordance with this Safety Data Sheet and the product label. Symptoms or effects that may arise if the product is mishandled and overexposure occurs are:

Ingestion: No adverse effects expected, however large amounts may cause nausea and vomiting.Eye contact: May be an eye irritant.Skin contact: Repeated or prolonged skin contact may lead to irritation.Inhalation: Breathing in mists or aerosols may produce respiratory irritation.

Long Term Effects:No information available for the product.

Toxicological Data:No LD50 data available for the product.

12. Ecotoxicological information

Avoid contaminating waterways.

13. Disposal considerations

Refer to Waste Management Authority.

14. Transport information

Road and Rail TransportNot classified as Dangerous Goods by the criteria of the Australian Dangerous Goods Code (ADG Code) for transport by Road and Rail.

Product Name: SOUTHLANDS GROUND WATERSubstance No: 000000001802 Issued: 22/03/2004 Version: 2

Page 4 of 5

Page 114: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Material Safety Data Sheet

Marine TransportNot classified as Dangerous Goods by the criteria of the International Maritime Dangerous Goods Code (IMDG Code) for transport by sea.

Air TransportNot classified as Dangerous Goods by the criteria of the International Air Transport Association (IATA) Dangerous Goods Regulations for transport by air.

15. Regulatory information

Classification: Based on available information, not classified as hazardous according to criteria of NOHSC.Poisons Schedule: S7 Dangerous Poison.

16. Other information

This material safety data sheet has been prepared by SH&E Shared Services, Orica.

Reason(s) for Issue:5 Yearly Revised Primary MSDS

This MSDS summarises to our best knowledge at the date of issue, the chemical health and safety hazards of the material and general guidance on how to safely handle the material in the workplace. Since Orica Limited cannot anticipate or control the conditions under which the product may be used, each user must, prior to usage, assess and control the risks arising from its use of the material.

If clarification or further information is needed, the user should contact their Orica representative or Orica Limited at the contact details on page 1.

Orica Limited's responsibility for the material as sold is subject to the terms and conditions of sale, a copy of which is available upon request.

Product Name: SOUTHLANDS GROUND WATERSubstance No: 000000001802 Issued: 22/03/2004 Version: 2

Page 5 of 5

Page 115: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Material Safety Data Sheet

Based on available information, not classified as hazardous according to criteria of NOHSC.Not classified as Dangerous Goods by the criteria of the Australian Dangerous Goods Code (ADG Code) for transport by Road and Rail.

1. Identification of the substance/preparation and of the company/undertaking

Product Name: GTP GROUNDWATER

Supplier: Orica Australia Pty LtdABN: 004 117 828Street Address: 1 Nicholson Street,

Melbourne 3000Australia

Telephone Number: +61 3 9665 7111Facsimile: +61 3 9665 7937

Emergency Telephone: 1 800 033 111 (ALL HOURS)

2. Composition/information on ingredients

Product Description: Waste material. Opaque liquid, with a slight chlorinated hydrocarbon odour.

Components / CAS Number Proportion Risk PhrasesOther ingredient(s)-

to 100% -

Ethylene dichloride107-06-2

<0.1% R11, R22, R36/37/38, Carc.Cat.2 R45

3. Hazards identification

Poisons Schedule: None allocated.

4. First-aid measures

Inhalation: Remove victim from area of exposure - avoid becoming a casualty. Seek medical advice if effects persist.

Skin Contact: If skin contact occurs, remove contaminated clothing and wash skin with soap and water. If irritation occurs, seek medical advice.

Eye Contact: If in eyes, wash out immediately with water. In all cases of eye contamination it is a sensible precaution to seek medical advice.

Ingestion: Immediately rinse mouth with water. If swallowed, do NOT induce vomiting. Give a glass of water. Seek immediate medical assistance.

Notes to physician: Treat symptomatically.

5. Fire-fighting measures

Product Name: GTP GROUNDWATERSubstance No: 000000019161 Issued: 05/10/2004 Version: 1

Page 1 of 5

Page 116: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Material Safety Data Sheet

Specific Hazards: Non-combustible material.Fire-fighting advice: Decomposes on heating emitting toxic fumes. Fire fighters to wear self-contained

breathing apparatus and suitable protective clothing if risk of exposure to products of decomposition.

Suitable Extinguishing Media: Not combustible, however, if material is involved in a fire use: Water fog (or if unavailable fine water spray), foam, dry agent (carbon dioxide, dry chemical powder).

6. Accidental release measures

Wear protective equipment to prevent skin and eye contact and breathing in vapours/dust. Cover with damp absorbent (inert material, sand or soil). Collect and seal in properly labelled containers, bags or drums for disposal or re-use. Wash area down with excess water.

7. Handling and storage

Handling advice: Avoid skin and eye contact and breathing in vapour.

Storage advice: Store in a cool, dry, well ventilated place and out of direct sunlight. Store away from incompatible materials described in Section 10. Keep containers closed when not in use - check regularly for spills.

8. Exposure controls/personal protection

Occupational Exposure Limits:No value assigned for this specific material by the National Occupational Health and Safety Commission. However, Exposure Standard(s) for constituent(s):

Ethylene dichloride: 8hr TWA = 40 mg/m3 (10 ppm)

As published by the National Occupational Health and Safety Commission.

TWA - The time-weighted average airborne concentration over an eight-hour working day, for a five-day working week over an entire working life.

STEL (Short Term Exposure Limit) – the average airborne concentration over a 15 minute period which should not be exceeded at any time during a normal eight hour work day. According to current knowledge this concentration should neither impair the health of, nor cause undue discomfort to, nearly all workers.

These Exposure Standards are guides to be used in the control of occupational health hazards. All atmospheric contamination should be kept to as low a level as is workable. These exposure standards should not be used as fine dividing lines between safe and dangerous concentrations of chemicals. They are not a measure of relative toxicity.

Engineering Control Measures:

Product Name: GTP GROUNDWATERSubstance No: 000000019161 Issued: 05/10/2004 Version: 1

Page 2 of 5

Page 117: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Material Safety Data Sheet

Ensure ventilation is adequate and that air concentrations of components are controlled below quoted Exposure Standards. Use with local exhaust ventilation or while wearing organic vapour respirator. Keep containers closed when not in use.

Personal Protective Equipment:Orica Personal Protection Guide No. 1, 1998: G - OVERALLS, SAFETY SHOES, SAFETY GLASSES, GLOVES, RESPIRATOR.

Wear overalls, safety glasses and impervious gloves. If risk of inhalation exists, wear organic vapour respirator meeting the requirements of AS/NZS 1715 and AS/NZS 1716. Wash contaminated clothing and other protective equipment before storage or re-use. Always wash hands before smoking, eating, drinking or using the toilet.

9. Physical and chemical properties

Physical state: LiquidColour: OpaqueOdour: Slight Chlorinated HydrocarbonSolubility: Miscible with water.Specific Gravity: ca. 1.0 (water=1)Relative Vapour Density (air=1): Not availableVapour Pressure (20 °C): Not availableFlash Point (°C): Not applicableFlammability Limits (%): Not applicableAutoignition Temperature (°C): Not applicableMelting Point/Range (°C): ca. 0Boiling Point/Range (°C): ca. 100Decomposition Point (°C): Not availablepH: Not available

10. Stability and reactivity

Stability: Stable under normal conditions of use.

11. Toxicological information

No adverse health effects expected if the product is handled in accordance with this Safety Data Sheet and the product label. Symptoms or effects that may arise if the product is mishandled and overexposure occurs are:

Ingestion: Swallowing can result in nausea, vomiting, diarrhoea, and abdominal pain.Eye contact: May be an eye irritant.Skin contact: Contact with skin may result in irritation. Repeated or prolonged skin contact may lead to irritation.Inhalation: Breathing in vapour may produce respiratory irritation.

Long Term Effects:No information available for the product.

Toxicological Data:No LD50 data available for the product.

Product Name: GTP GROUNDWATERSubstance No: 000000019161 Issued: 05/10/2004 Version: 1

Page 3 of 5

Page 118: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Material Safety Data Sheet

12. Ecotoxicological information

Avoid contaminating waterways.

13. Disposal considerations

Refer to Waste Management Authority.

14. Transport information

Road and Rail TransportNot classified as Dangerous Goods by the criteria of the Australian Dangerous Goods Code (ADG Code) for transport by Road and Rail.

Marine TransportNot classified as Dangerous Goods by the criteria of the International Maritime Dangerous Goods Code (IMDG Code) for transport by sea.

Air TransportNot classified as Dangerous Goods by the criteria of the International Air Transport Association (IATA) Dangerous Goods Regulations for transport by air.

15. Regulatory information

Classification: Based on available information, not classified as hazardous according to criteria of NOHSC.Poisons Schedule: None allocated.

All the constituents of this material are listed on the Australian Inventory of Chemical Substances (AICS).

16. Other information

This material safety data sheet has been prepared by SH&E Shared Services, Orica.

Reason(s) for Issue:First Issue Primary MSDS

This MSDS summarises to our best knowledge at the date of issue, the chemical health and safety hazards of the

Product Name: GTP GROUNDWATERSubstance No: 000000019161 Issued: 05/10/2004 Version: 1

Page 4 of 5

Page 119: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Material Safety Data Sheet

material and general guidance on how to safely handle the material in the workplace. Since Orica Limited cannot anticipate or control the conditions under which the product may be used, each user must, prior to usage, assess and control the risks arising from its use of the material.

If clarification or further information is needed, the user should contact their Orica representative or Orica Limited at the contact details on page 1.

Orica Limited's responsibility for the material as sold is subject to the terms and conditions of sale, a copy of which is available upon request.

Product Name: GTP GROUNDWATERSubstance No: 000000019161 Issued: 05/10/2004 Version: 1

Page 5 of 5

Page 120: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.doc8 November 2004A4.1

Appendix 4

Description of EFFECTS

Preliminary Hazard Analysis, GroundwaterTreatment Plant, Orica Australia Pty Ltd,

Botany Industrial Park

Page 121: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.doc8 November 2004A4.2

Appendix 4 – Description of EFFECTS

EFFECTS is a software package developed by TNO to perform consequentialimpact calculations. The programme performs consequence calculations foruser defined hazardous incident / release scenarios.

Consequence models in EFFECTS are based on the models in the well-known"Yellow Book" (Ref A4.1). The models used by EFFECTS are fully described inthe Yellow Book and are also described briefly below.

A4.1 Release Rates

EFFECTS can model release behaviour for compressed gas, liquid or 2-phasereleases from vessels, pipelines or total vessel rupture. Data supplied by theanalyst includes the type of release, location of release with respect to vesselgeometry, pipe lengths etc. and initial conditions of the fluid (i.e. before release).The maximum spreading area of a pool is also supplied by the analyst, as is thesubstrate material (i.e. concrete, sand, water etc).

A4.2 Atmospheric Conditions

Meteorological data used in the dispersion calculations is input by the user.The weather data is divided into 6 Pasquill stability classes and averagewindspeeds, with 12 directional probabilities for each windspeed / stability classcombination.

A4.3 Dispersion

The dispersion model used is selected by the analyst, depending on releaseconditions and behaviour, e.g. velocity, density of fluid.

For passive dispersions, a Gaussian plume dispersion model is used.

The SLAB model is used for dense gas dispersion calculations. Dispersionfrom a ground level evaporating pool, a horizontal or vertical jet or aninstantaneous release can be treated. The model predicts dispersion behaviourby solving the conservation equations for mass, momentum and energy. Thecloud is treated as a steady state plume, a transient "puff" or a combination ofthe two, depending on the release duration.

In the case of a finite duration release, cloud dispersion is initially describedusing the steady state plume model as long as the source is active. Once thesource has been shut off, subsequent dispersion is calculated by the transientpuff model. For instantaneous releases, the transient puff model is used for theentire calculation.

The input to the dispersion model is the release rate calculated by the releaserate and / or pool evaporation models. In the case of a time varying release, therelease rate is averaged over the release duration. Alternatively, the analystmay choose to use a time varying release rate as an input to the calculation(this lengthens the calculation time). For this study the release rates have been

Page 122: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.doc8 November 2004A4.3

averaged over the release duration for the most part, however if a significantreduction in release rate occurs over the release duration, the time step optionhas been used.

A4.4 Toxic Impact

The effect of exposure to toxic materials is predicted using probit equations.The calculated probit can be mathematically transformed (using the errorfunction), allowing a probability of fatality to be predicted for a particular dose.EFFECTS contains a set of default probits for common toxic materials, howevera user defined probit may also be used. The equations are of the form:

Probit = A + B ln(cnt)

t exposure time (min)

c concentration (ppm)

Probability of fatality = 12 (1 + erf

Pr −520.5 )

The dispersion results are presented in terms of dose rather than concentration.Dose is calculated within EFFECTS by integrating the concentration at aparticular location over the exposure duration. The concentration may vary overtime. The duration is the lesser of the time taken for the cloud to pass or themaximum exposure duration (a parameter set by the analyst). Typically, amaximum exposure duration of 1 hour is set.

A4.5 Fires and Heat Radiation

Heat radiation effects are calculated based on flame surface emissive power(which is dependent on the quantity of material, its heat of combustion, flamedimensions and the fraction of heat radiated). The heat flux at a particulardistance from a fire is calculated using the view factor method. The view factortakes into account the distance from the flame to the target, the flamedimensions and the orientation angle between the flame and the target.

The effect of heat radiation on a person is calculated from the probit equationand the probability of fatality predicted by transforming the probit as for toxicimpact calculations. However in this case the dose is a thermal dose ratherthan toxic dose.

Probit = -36.38 + 2.56 ln(tQ1.33)

t exposure time (sec)

Q heat flux (W/m2)

Note that this probit is only valid for very short exposure durations (less than 1minute). For the purposes of this risk assessment it is assumed a person has20 seconds to escape from heat radiation (i.e. an exposure duration of 20seconds).

Page 123: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.doc8 November 2004A4.4

A4.6 Explosion and Overpressures

The Multi Energy method is used to predict the overpressures from flammablegas explosions. The key feature of the Multi-Energy method is that theexplosion is not primarily defined by the fuel air mixture but by the environmentin which the vapour disperses.

Partial confinement is regarded as a major cause of blast in vapour clouddeflagrations. Blast of substantial strength is not expected to occur in openareas. Strong blast is generated only in places characterised by partialconfinement while other large parts of the cloud burn out without contributing tothe blast effects. The vapour cloud explosion is not regarded as an entity but isdefined as a number of sub-explosions corresponding to various sources ofblast in the vapour cloud, i.e. each confined part of the cloud is calculated as aseparate vapour cloud explosion.

The initial strength of the blast is variable, depending on the degree ofconfinement and on the reactivity of the gas. The initial strength is representedas a scale of 1 to 10 where 1 means slow deflagration and 10 meansdetonation. For explosions in process plant environments, the initial strength isthought to lie between 4 to 7 on the scale.

The TNT equivalence model has traditionally been used to predict explosioneffects. For comparison, applying the Multi-Energy method with a blast strengthof 10 will give similar overpressures as application of the TNT method to theobstructed part of the cloud with an efficiency factor of 20%.

References

A4.1 "The Yellow Book", Methods for the Calculation of the Physical Effects ofthe Escape of Dangerous Material, CPR 14E, Parts 1& 2, Committee forthe Prevention of Disasters, 3rd edition 1997

Page 124: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.doc8 November 2004A5.1

Appendix 5

Meteorological Data

Preliminary Hazard Analysis, GroundwaterTreatment Plant, Orica Australia Pty Ltd,

Botany Industrial Park

Page 125: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.doc8 November 2004A5.2

Appendix 5 – Meteorological Data

Six typical Botany wind/weather combinations (wind speed / Pasquill stabilitycategory) have been used as the basis for all dispersion calculations. Theprobability of each combination of wind/weather category and wind direction(any of 12 directions) is used in the risk modelling.

The wind/weather data used is a consolidated (reduced) version of a detailedstudy of prevailing wind/weather conditions at Botany carried out by P Zib andassociates. The data was reduced in conjunction with DIPNR some years agoto the typical wind/weather categories (wind speed / Pasquill stabilityconditions).

Direction WindFrom

Windspeed (m/s) / Stability Class

2.3 B 3.8 D 5.3 D 2.3 E 0.9 F 2.3 F

346-015 3.91 4.44 1.72 7.63 10.38 8.15

016-045 6.85 7.77 11.07 9.69 5.09 8.89

046-075 9.74 8.00 10.98 6.76 3.60 4.34

076-105 6.98 7.70 3.84 4.09 4.23 4.76

106-135 5.08 7.50 3.73 6.24 4.27 5.42

136-165 8.26 10.40 8.38 7.51 4.46 6.78

166-195 10.02 12.21 23.72 7.31 4.93 3.49

196-225 6.31 4.52 8.51 3.32 4.96 2.16

226-255 7.86 6.48 8.00 6.33 9.50 5.00

256-285 20.86 14.04 17.18 15.12 8.56 14.81

286-315 9.56 12.26 2.66 19.17 12.74 24.87

316-345 4.62 4.76 0.60 6.71 26.85 11.04

Sum (%) 100 100 100 100 100 100

Proportion (%) 5.3 39.2 29.8 14.3 3.8 7.6

Page 126: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200474

10 REFERENCES

1 Department of Urban Affairs & Planning (NSW) Hazardous IndustryPlanning Advisory Paper No 6 – Guidelines for Hazard Analysis, 1992

2 Department of Urban Affairs & Planning (NSW), Applying SEPP (StateEnvironmental Planning Policy) 33

3 DIPNR, Director-General’s Requirements for Groundwater TreatmentPlant EIS, 2004

4 Department of Urban Affairs & Planning (NSW) Hazardous IndustryPlanning Advisory Paper No 4 – Risk Criteria for Land Use SafetyPlanning, 1992

5 Aker Kvaerner, Orica Australia Pty Ltd, Project Definition Study forBotany Groundwater Treatment Plant, Rev P1, February 2004

6 URS, Background Paper, Botany Groundwater Project, ProposedGroundwater Treatment Plant at Botany Site, 25 May 2004

7 KBR, Preliminary Hazard Analysis, Steam Stripping Unit, Orica Australia,2004

8 http://www.aiha.org/Committees/documents/erpglevels.pdf, AIHA ERPGwebsite, 2004

9. Yellow Book, Methods for the Calculation of the Physical Effects of theEscape of Dangerous Material, CPR 14E, Parts 1& 2, Committee for thePrevention of Disasters, TNO, 3rd edition 1997.

10 Lees F.P., Loss Prevention in the Process Industries, 2nd Edition 1996

11 Gas Dispersion, Process SHE Guide No 20, 10th November 1998, ICIEngineering UK

12 TNO, Methods for the Determination of Possible Damage (The GreenBook), 1992

13 ICI HAZAN Course Manual, 1997

14 ICI Engineering Department, Process Safety Guide 14 - Reliability Data,ICI PLC (UK)

15 A W Cox, F P Lees and M L Ang, Classification of Hazardous Locations,Institution of Chemical Engineers, 1990

16 Orica Engineering, Elgas, Victoria, Quantitative Risk Analysis, 2002

17 Orica Engineering, Preliminary Hazard Analysis for the HCB DestructionPlant, 2003

Page 127: PRELIMINARY HAZARD ANALYSIS GROUNDWATER TREATMENT …

Pinnacle Risk Management

Orica GTP PHA Report Rev F.Doc8 November 200475

18 AIChE CCPS, Layer of Protection Analysis, 2001

19 ICI, Process SHE Guide Number 4

20 SHE Pacific, Final Hazard Analysis, Replacement Chloralkali Plant,Orica Australia Pty Ltd, Botany Industrial Park, NSW, 9 March 2000

21 ICI Engineering, Risk Assessment, Transport of Ethylene Dichloride fromICI Botany Site to Terminals Pty Ltd, September 1995