Preliminary Assessment of Exposure to Asbestos from a Simulation of Removal of Caulking Compounds at Wardelocking Reservoir

P a g e 1 Lou Roberts Manager Safety, Environment and Aboriginal Affairs Water Corporation 629 Newcastle Street LEEDERVILLE WA 6007 Dear Lou Preliminary Assessment of Exposure to Asbestos from a Simulation of Removal of Caulking Compounds at Wardelocking Reservoir 1. Introduction The Water Corporation completed a project which involved the renovation of a 1958 reservoir in Minnivale in the north east wheatbelt near Dowerin. On completion of the renovation it became known that the reservoir contained chrysotile asbestos within the caulking material located between the concrete slabs and the fascia panels. The removal of the old caulking compound during the renovation was performed using several different tools and included using an angle grinder as well as a rotary floor grinder to level the surface of the concrete. It is the author s understanding that some of the workers undertaking the work were not wearing respirators.. It is also the Author s understanding that the reservoir was essentially sealed with little air exchange during the renovation process. Once it was known that the caulking compound contained chrysotile asbestos, the workers involved in the work expressed a concern about the possible exposure to the asbestos. The Water Corporation was requested by one of the contracting companies involved to undertake a simulation activity. The Commissioner for Worksafe WA granted permission for this simulation to be performed on condition that strict personal exposure control was used. The author was engaged to facilitate this simulation activity including overseeing all sampling and analysis. The Wardelocking Reservoir near Wagin was selected for the simulation because it was currently off-line and contained chrysotile asbestos in the caulking compound as per the Minnivale reservoir. The company contracted to do the simulation was a licensed asbestos removalist which had extensive experience removing all types of asbestos including the most hazardous friable limpet asbestos. 1 Cumnock Place, Duncraig WA 6023 Telephone: (08) 9447 9911 or (08) 9448 5462 Facsimile: (08) 9447 9911 Mobile: 0438 001 955 Email: lglossop@bigpond.net.au

P a g e 2 The simulation took place on the 10th of December 2015 and representatives from the contracting companies involved in work at Minnivale attended the site to observe the work practices. 2. Simulation Using Worst-Case Scenario It was recommended by the author that a worst-case scenario be used for the simulation even though this would most likely overestimate the actual exposures. The worst-case scenario had the following conditions: a) A small totally enclosed air tight enclosure. b) Three people close to each other continuously using angle grinders to remove the caulking compound. c) Personal monitoring. d) No dust suppression. e) No air movement inside the enclosure. f) High volume sampling rate to have low detection limit. g) Analysis using a technique that can detect the thinnest fibres. At the Minnivale reservoir worker exposures would have been less due to; 1. The volume of the air in the reservoir was greater than the Wardelocking enclosure so would have had a diluting effect on exposures. 2. There was an opening in the reservoir wall that would have allowed some air exchange with outside air which would also have had a diluting effect on exposures. 3. People doing the work at Minnivale would not have worked as intensely or as closely together as the simulation resulting in lower exposures at Minnivale. 3. Analytical Method The filters used for the monitoring were analysed by high magnification Scanning Electron Microscopy in conjunction with Energy Dispersive Spectrometry to determine the mineralogy of each particle. This method detects some fibres that are so small that the normal technique of analysis using optical phase contrast microscopy will not detect them. The results showed some chrysotile fibres that were in this situation. The air samples collected had to be diluted 20 times to enable the filters to be counted accurately because of the dust loading. The diluted samples were able to be measured on the Scanning Electron Microscope. The Energy Dispersive Spectrometer was able to identify/speciate all fibres on the filters. All chrysotile fibres meeting the respirable fibre definition were counted. 4. Particle Speciation The analysis showed many particles that met the respirable fibre definition were non-asbestos minerals such as quartz, chlorite which originated from the grinding of the concrete. 1 Cumnock Place, Duncraig WA 6023 Telephone: (08) 9447 9911 or (08) 9448 5462 Facsimile: (08) 9447 9911 Mobile: 0438 001 955 Email: lglossop@bigpond.net.au

P a g e 3 5. Results Chrysotile asbestos was detected in nearly all the samples with the highest results for the people doing the grinding. The concentrations ranged from below the limit of detection (<0.01 f/ml) up to 0.1 f/ml. Most exposures were around 0.03 f/ml. The Workplace Exposure Standard for asbestos is currently 0.1 f/ml for all forms of asbestos even though the hazard from chrysotile is significantly lower than the other regulated asbestos types. In 1995 the National Occupational Health and Safety Commission (NOHSC) had a Workplace Exposure Standard of 0.1 f/ml for crocidolite and amosite (considered much more hazardous than chrysotile), but 1.0 f/ml for chrysotile in recognition of the different hazard and risk. Chrysotile asbestos is not as hazardous because of the fibres being curly and not straight so being trapped higher in the lung as well as them being far more biosoluble in the lung. Chrysotile is believed to dissolve within 1 to 2 years whereas crocidolite and amosite are essentially in the lungs for life. Also chrysotile is several orders of magnitude less likely to cause mesothelioma. In 2003, the Workplace Exposure Standard for all forms of asbestos was changed to 0.1 f/ml which was not based on scientific evidence, but to simplify the management and control of asbestos. Often there were different types of asbestos types used in the same asbestos products and this was especially true for asbestos re-enforced high pressure water pipe. 6. Discussion The simulation was designed with a worst-case scenario of exposure as mentioned above. The simulation monitoring showed that grinding caulking material was capable of producing exposures to chrysotile fibres. Before the simulation was done, it was thought by the author that few chrysotile fibres would be produced, but the monitoring and analysis showed some free chrysotile fibres in the respirable range are produced by this aggressive grinding technique. Based on these results of the simulation, the author believes that it is highly likely that people performing the caulking removal work at Minnivale using grinding and sanding would have had significantly lower exposures to chrysotile compared to the simulation because of the reasons mentioned above in Section two. This exposure would have been below the Workplace Exposure Standard for asbestos because the worst case simulation results were at or below the Workplace Exposure Standard. 7. Information that Should be Provided to the Contracting Companies It is important that the results of the simulation are provided to all workers involved in the work, so that they can understand that any exposure they may have had would not have exceeded the Workplace Exposure Standard. 1 Cumnock Place, Duncraig WA 6023 Telephone: (08) 9447 9911 or (08) 9448 5462 Facsimile: (08) 9447 9911 Mobile: 0438 001 955 Email: lglossop@bigpond.net.au

P a g e 4 Some employees may be anxious about the results and it would be necessary to provide an information session to discuss the results. The author recommends that employees should be given the opportunity to ask questions about the simulation and their possible past exposures. Yours sincerely Laurie Glossop B.Sc Ph.D COH MAIOH FAIOH Principal Consultant 22 nd December 2015 1 Cumnock Place, Duncraig WA 6023 Telephone: (08) 9447 9911 or (08) 9448 5462 Facsimile: (08) 9447 9911 Mobile: 0438 001 955 Email: lglossop@bigpond.net.au

Assessment of Exposure to Asbestos from Simulation of Removal of Caulking Compounds at Wardelocking Reservoir December 2015 Prepared for: Water Corporation

Assessment of Exposure to Asbestos from Simulation of Caulking Compounds at Wardelocking Reservoir December 2015 Prepared for: Water Corporation DOCUMENT CONTROL Report Title Project Name Assessment of Exposure to Asbestos from Simulation of Removal of Caulking Compounds at Wardelocking Reservoir Asbestos in Caulking Compounds Job Number Client Water Corporation Report Number DOCUMENT DISTRIBUTION Document File Name Document Status Distributed to Date distributed Limitations and disclaimer: This report documents the work undertaken by Glossop Consultancy. This document may contain confidential information. As such, the document is intended only for the use by Water Corporation. It is not for public circulation or publication or to be used by any third party without the express written permission of either the Water Corporation or Glossop Consultancy. This report should be read in full. While the findings presented in this report are based on information that Glossop Consultancy considers reliable unless stated otherwise, the accuracy and completeness of source information cannot be guaranteed, although Glossop Consultancy has taken reasonable steps to verify the accuracy of such source data. Glossop Consultancy has made no independent verification of this information beyond the agreed scope of works and Glossop Consultancy assumes no responsibility for any inaccuracies or omissions outside of Glossop Consultancy s direct control. Furthermore, the information compiled in this report addresses the specific needs of the Water Corporation, so may not address the needs of third parties using this report for their own purposes. Thus, Glossop Consultancy and their employees accept no liability for any losses or damage for any action taken or not taken on the basis of any part of the contents of this report. Those acting on information provided in this report do so entirely at their own risk. This report does not purport to give legal advice. Legal advice can only be given by qualified legal practitioners. December 2015 Page 2 of 23

Executive Summary A bituminous based caulking compound used at Minnivale Reservoir to seal concrete slab joints to hold water, as well as facia joints, contained chrysotile (white asbestos) as a re-enforcing material within the caulking compound. This reservoir had recently been renovated requiring the caulking compound to be removed and replaced. The Water Corporation, and the contractors undertaking the renovation, were not aware that the caulking compound contained chrysotile (known as white asbestos) and therefore this led to work on asbestos containing material using prohibited methods such as removal by way of grinding. It is believed those grinding, as well as other personnel who accessed the enclosed work area, were not adequately protected from airborne contaminants by use of personal protective equipment including respirators or disposable overalls. Once it became known that the caulking compound contained chrysotile contractors undertaking removal works expressed their concerns about potential exposure to airborne chrysotile fibres. Given the hazard was unknown at the time, works had taken place without air monitoring and therefore data representative of actual conditions throughout the Minnivale renovation were not available. The only way to estimate likely exposure was to perform a simulation of the work in a controlled environment. After obtaining approval from the Worksafe WA Commissioner, a simulation of the removal of the caulking compound at the Wardelocking Reservoir was performed by a licensed asbestos contractor on the 10 December 2015, and was overseen by a specialist asbestos consultancy, EOSH. The simulation was performed as a worst case scenario to determine the highest potential exposure scenario. Personal exposure and static air monitoring was performed during the removal of the caulking compound using grinders that were used throughout the Minnivale renovation. These samples were analysed using a dilution/elutriation process due to the high loading of dust on the filters. After the dilution the re-filtered dust was analysed using Scanning Electron Microscopy (SEM) and Energy Dispersive Spectrometry (EDS). Angle grinding of the caulking compound produced free chrysotile fibres which were not hypothesised as a result of grinding of the caulking compound. The results of the analysis of the personal and positional samples analysed showed similar concentrations of free fibres. Furthermore, these results obtained for both personal and positional samples analysed indicate that chrysotile fibres produced as a result of grinding were extremely small and remain suspended in the air for extended periods of time. All of the measured concentrations were at or below the Workplace Exposure Standard (WES) of 0.1 f/ml during worst case scenario simulation, which indicates that fibre concentrations produced during work at Minnivale Reservoir are likely to be lower than results produced by the simulation. However, as chrysotile is a known and regulated Class One carcinogen, exposures should be kept to levels as low as reasonably practicable and always below the WES. December 2015 Page 3 of 23

Assessment of Exposure to Asbestos from Simulation of Caulking Compounds at Wardelocking Reservoir WATER CORPORATION December 2015 December 2015 Page 4 of 23

TABLE of CONTENTS Executive Summary... iii 1.1 Objective... 9 2 Analysis of Caulking Compound... 9 2.1 Analysis by SEM and EDS... 10 3 Simulation Methodology... 12 3.1 Enclosure at Wardelocking... 13 3.2 Equipment Used for Removal of Caulking Compound... 15 3.3 Removal Work Methodology... 15 3.4 Monitoring Exposure during Simulation... 16 3.5 Sample Analysis Methodology... 17 4 Results... 19 5 Discussion... 20 5.1 Understanding Results... 20 6 Information to be Provided to Employees Involved in Work at Minnivale... 21 7 Appendix A UWA SEM Analysis Report... 22 8 Appendix B - Scientific Assessment of Exposure to Asbestos from Caulking Compounds... 23 December 2015 Page 5 of 23

List of Tables Table 1: Sample Results Table 2: UWA SEM Analysis Report List of Figures Figure 1: Location of Wardelocking Reservoir 8 Figure 2: Photograph of Caulking Compound Wardelocking 9 Figure 3: Photograph of Caulking Compound Wardelocking Enlarged 400% 10 Figure 4: SEM Images of Wardelocking Caulking Compound 11 Figure 5: SEM Images of Wardelocking Caulking Compound 11 Figure 6: Enclosure 13 Figure 7: Smoke testing commencing internally 14 Figure 8: Smoke testing at full extent 14 Figure 9: Smoke testing external image 15 Figure 10: Personal Protective Equipment Used During Simulation 16 Figure 11: SEM Dust Loading 17 Figure 12: SEM dust loading post elutriation 18 December 2015 Page 6 of 23

Introduction The Water Corporation completed a project which involved the renovation of Minnivale Reservoir near Wyalkatchem. On completion of the renovation it became known that the reservoir contained chrysotile (white asbestos) within the caulking material located between the concrete slabs and the fascia panels. The removal of the old caulking compound during the renovation was performed using several different tools including an angle grinder and a rotary floor grinder to level the surface of the concrete. The author understands that some of the workers undertaking the work were not wearing adequate respiratory protective equipment. Furthermore, it was identified that the reservoir was essentially sealed with minimal air exchange during the renovation process. Once it was known that the caulking compound contained chrysotile (white asbestos), the workers involved in the work expressed a concern about the possible exposure to asbestos fibres. The Water Corporation was requested by one of the contracting companies involved to undertake a simulation activity to determine if any exposure to chrysotile may have occurred. The Commissioner for Worksafe WA granted permission for this simulation to be performed on condition that strict personal exposure control was adopted. The author was engaged to facilitate this simulation activity including overseeing all sampling and analysis. Wardelocking Reservoir was selected for the simulation because it was currently off-line and contained chrysotile asbestos in the caulking compound as per the Minnivale Reservoir. The company contracted to do the simulation (Delta Group) holds an unrestricted asbestos license and additionally has extensive experience removing all types of asbestos including friable limpet asbestos which is considered the most hazardous form. The simulation took place on the 10 December 2015 and representatives from the contracting companies involved in work at Minnivale attended the site to observe the work practices of the simulation. The Location of the Wardelocking Reservoir can be seen in the Figure 1 below: December 2015 Page 7 of 23

Figure 1: Location of Wardelocking Reservoir December 2015 Page 8 of 23

1.1 Objective The objective of the simulation of the removal of the caulking compound was to determine whether any chrysotile is able to become free fibre during the removal of the caulking compound, and if so, what, was the likely personal exposure concentration to airborne asbestos. The simulation was undertaken to attempt to reproduce a worst-case scenario to determine the highest level of risk from this activity. 2 Analysis of Caulking Compound The caulking compound at the Minnivale Reservoir and Wardelocking Reservoir were both analysed by a NATA accredited laboratory by pulling fibres from the bulk samples taken of the caulking compound and analysing them under Polarised Light Microscopy (PLM). Results identified that the caulking compounds at both reservoirs contained chrysotile. The concentration of chrysotile within the caulking compounds was not determined. A photo of a sample of caulking compound retrieved from Wardelocking Reservoir is shown in Figure 2 below. Figure 2: Photograph of Caulking Compound Wardelocking White flecks of chrysotile are clearly visible in the photograph. Figure 3 illustrates the chrysotile fibres in more detail (enlarged 400%). The caulking compound at this stage was not very flexible and was easy to break by hand. It would not be described as friable/brittle, but was able to be broken by hand and was oxidised to some degree. December 2015 Page 9 of 23

Figure 3: Photograph of Caulking Compound Wardelocking Enlarged 400% 2.1 Analysis by SEM and EDS A bulk sample of the caulking compound collected at Wardelocking was dissolved in a solvent and the solid residue analysed by Scanning Electron Microscopy (SEM) and Energy Dispersive Spectrometry (EDS). SEM and EDS were used to analyse the sample as it provided highly magnified images of particles and the composition of any such particles. The results of the SEM and EDS analysis identified that the caulking compound was made of predominantly bitumen with the chrysotile as a reinforcing fibre. The SEM images are below in Figure 4 and Figure 5 below: December 2015 Page 10 of 23

Figure 4: SEM Images of Wardelocking Caulking Compound Figure 5: SEM Images of Wardelocking Caulking Compound The two SEM images above show the fibre bundles of chrysotile (white asbestos). Only chrysotile was found in the caulking compounds. Nearly all the chrysotile fibres seen were December 2015 Page 11 of 23

large fibre bundles and not individual chrysotile fibrils. The straight looking fibre bundle on the right in Figure 5 was determined by EDS to be chrysotile and not a form of amphibole asbestos. If the above chrysotile fibre bundles are aggressively cut or ground they can produce smaller fibre bundles and possibly individual fibrils of about 0.03µm in diameter. 3 Simulation Methodology The author recommended that the simulation be conducted using a worst-case methodology even though this would most likely overestimate the actual exposures at Minnivale. The worst-case scenario had the following conditions: a) A small totally enclosed air tight enclosure. b) Two people close to each other continuously using angle grinders to remove the caulking compound. c) One person conducting sweeping tasks of residual dust on the floor of the reservoir post grinding. d) Personal monitoring on both grinders and sweeping personnel. e) No dust suppression. f) No air movement inside the enclosure. g) High volume sampling to achieve the lowest analytical limit of detection for the method. h) Analysis using a technique that can detect the thinnest fibres. At the Minnivale reservoir actual worker exposures would likely have been lower than simulated worst case results due to; 1. The volume of the air in the Minnivale enclosure was greater than the volume in the Wardelocking enclosure. If the quantity of airborne fibre was considered equal in both enclosures then the overall concentration in the Minnivale enclosure would have been lower given a larger volume of air would have a diluting effect on the concentration. It is believed however, the concentration of airborne fibre in the Minnivale enclosure was likely to be less than the concentration measured in the simulation due to simulated activities being representative of worst case scenarios. 2. There was an opening in the reservoir wall at the Minnivale site that would have allowed some air exchange with fresh outside air which would also have had a diluting effect on exposures. 3. People doing the work at Minnivale would not have worked as intensively or as closely together as the simulation resulting in lower exposures at Minnivale. December 2015 Page 12 of 23

3.1 Enclosure at Wardelocking The enclosure at Wardelocking had to be big enough to allow sufficient caulking to be removed, but not so big as to make it impractical to construct. As there was no roof at Wardelocking a roof had to be constructed over the enclosure to simulate the conditions at Minnivale. Additionally, a completely enclosed structure was required to ensure no asbestos fibres would be released into the surrounding environment. A photograph of the enclosure/structure is shown in the Figure 6 below: Figure 6: Enclosure The enclosure was 12 metres long, 8 metres wide and 3 metres high. The enclosure was air tight which was confirmed by smoke testing. Photographs of the smoke test can be seen in Figure 7 and Figure 8 below. December 2015 Page 13 of 23

Figure 7: Smoke testing commencing internally Figure 8: Smoke testing at full extent The enclosure was checked for air tightness/sealing by use of a smoke generator which filled the enclosure with a dense cloud of smoke. Following identification of minor leaks around two scaffold bars protruding through the lining no smoke was observed leaking from any point around the structure after those leaks were sealed. The enclosure was considered airtight for the simulation. Figure 9 below shows the outside of the enclosure during the smoke testing process: December 2015 Page 14 of 23

Figure 9: Smoke testing external image Figure 9 above also shows the two negative pressure HEPA filter units (black boxes in the middle of the photograph). These HEPA filters were used to clear the smoke before starting the removal of the caulking material. When the smoke was removed from the enclosure and filtered through the HEPA system, no smoke could be seen exiting the HEPA exhaust. The HEPA filters were considered to be effectively filtering particulates with aerodynamic equivalent diameters similar to asbestos fibres. The original removal work at Minnivale would not have been undertaken in an enclosure as small in volume as the unit constructed for the simulation. 3.2 Equipment Used for Removal of Caulking Compound The simulation used the most aggressive form to remove the caulking compound. It is understood that at the Minnivale Reservoir, angle grinders were used to remove the caulking compound located between concrete joints. Therefore, during the simulation, angle grinders were used to remove the caulking material. During the simulation the angle grinders removed large particles of caulking compound, but there was also some respirable particulate produced. As the simulation continued increased dust levels were evident within the enclosure. 3.3 Removal Work Methodology Two people used angle grinders in close proximity to each other. They essentially ground out the caulking compound from adjacent parallel joints. The grinding was done nearly continuously for four hours throughout the air monitoring process. Toward the end of the four December 2015 Page 15 of 23

hour period, sweeping was performed by another operative to simulate the work conducted at Minnivale during the renovation. No dust suppression was used within the enclosure during simulated removal work and the HEPA units were inactive. During the simulation, workers performing the caulking removal utilised Powered Air Purifying Respirators (PAPR) with P3 particulate filters and disposable overalls as shown in Figure 10 below. Figure 10: Personal Protective Equipment Used During Simulation 3.4 Monitoring Exposure during Simulation The duration of air monitoring had to be long enough to collect an adequate volume of air to achieve a low detection limit for the analytical method. This meant the flowrate had to be set at two litres per minute which increased the loading of dust on the filters. Two different types of samples were collected: Personal Monitoring (exposure) Positional/Static Monitoring. December 2015 Page 16 of 23

Three workers performing grinding and sweeping tasks participated in personal monitoring. Each sample required at least 200 litres of air to pass through the filter. 3.5 Sample Analysis Methodology Ideally, dust overloads on filters should be avoided to facilitate adequate sample analysis however, due to the unexpected production of high levels of dust as a result of grinding very dry, oxidised, aged caulking material all samples were heavily loaded. Due to this, standard analysis was not possible with the samples collected on the day of the simulation. Figure 11 below shows the typical loading on the filters. Figure 11: SEM Dust Loading Dust loading on the filters was high and there were typically particles stacked on top of each other. This result precluded standard analysis because of the loading of dust. In Figure 11 above you can see several chrysotile fibres in the middle of the SEM image. As the dust loading was too high a dilution/elutriation process had to be utilised. The filters were sonicated into water containing a dispersant with the sonication time being two minutes. The filter became totally clean after this period of sonication. The total volume of water for the sonication was 400mL and a 20mL aliquot was taken from the 400mL. This December 2015 Page 17 of 23

aliquot was then filtered onto a 0.2µm nucleopore filter which allows for high quality SEM analysis. The process is an approved German Method of elutriation for bulk samples: BGIA 7485 Method for the analytical determination low mass concentrations of asbestos fibres in sprays, powders and dusts with SEM / EDS. After the dilution/elutriation the filters looked like the image in Figure 12 below: Figure 12: SEM dust loading post elutriation A chrysotile bundle can be seen in the middle of this SEM image above. The light grey materials are pieces of bitumen that account for the vast majority of the particles on the filter. The typical fibre count is provided in the University of Western Australia s Centre for Microscopy, Characterisation and Analysis SEM Analysis Report located in Table 2, Appendix B. The analysis results identify that chrysotile was present as small fibre bundles with high aspect ratios. Chrysotile fibrils are typically about 0.02 to 0.05µm in diameter so the fibres in these images are considered fibre bundles. The results also identified the presence of other respirable fibres, such as quartz. These are most likely to have come from incidental grinding of the concrete slabs. December 2015 Page 18 of 23

4 Results All the filters were diluted/elutriated to determine the concentration of chrysotile fibres on each sample. The concentrations of chrysotile measured by SEM are in the Table1. Table 1: Sample Results Sample No Date of Sample Sample Type Chrysotile Total Total Chrysotile fibre count x 20 * Sample Volume (L) Fibre Concentration (F/mL) B0027 10/12/2015 ASB Stat 2 40 630 0.02 B0028 10/12/2015 ASB Stat 4 80 798 0.03 B0029 10/12/2015 ASB Stat 4 80 812 0.03 B0030 10/12/2015 ASB Stat 4 80 814 0.03 B0031 10/12/2015 Personal Op 1 - grinding 2 40 276 0.05 B0032 10/12/2015 Personal Op 1 - grinding 1 20 240 0.03 B0033 10/12/2015 B0034 10/12/2015 Personal Op 2 - grinding Personal Op 2 - grinding 0 0 250 <0.01 1 20 241.8 0.03 B0035 10/12/2015 Personal Op 3 general area 2 40 236 0.06 B0036 10/12/2015 Personal Op 2 - grinding 0 0 262 <0.01 B0037 10/12/2015 Personal Op 2 - grinding 0 0 262 <0.01 B0038 10/12/2015 Personal Op 1 - grinding 1 20 264 0.03 B0039 10/12/2015 Personal Op 1 - grinding 4 80 264 0.1 B0040 10/12/2015 Personal Op 3 crack chaser blade 3 60 248 0.09 *Due to sample dilution 20:1 December 2015 Page 19 of 23

As can be seen by Table 1, the number of chrysotile fibres counted is small because of the dilution/elutriation method adopted to create a countable filter. The multiplying of the number of the chrysotile fibres by the dilution factor to estimate the actual concentration is high so the concentrations (results) are considered in analytical terms, low precision, but accurate. There is a large uncertainty with the concentrations however results are considered relevant for estimating exposures within an acceptable range. Another important result are the positional samples. All positional samples were well away from the grinding activity and all contained chrysotile in similar concentrations. This highlights that the chrysotile fibres remain airborne and are dispersed like a gas - this was expected. Emphasis on the different concentrations should be avoided because the precision is low. 5 Discussion In summary, the exposures within the enclosure are essentially the same. The angle grinding of the caulking compound has produced non-encapsulated chrysotile fibres ( free fibres ). Chrysotile (white asbestos) was detected in nearly all the samples. The concentrations ranged from below the limit of detection (<0.01 f/ml) up to 0.1 f/ml. Most exposures were around 0.03 f/ml. The Workplace Exposure Standard (WES) for asbestos is currently 0.1 f/ml for all forms of asbestos even though the hazard from chrysotile is significantly lower than the other regulated asbestos types. In 1995 the National Occupational Health and Safety Commission (NOHSC) had a Workplace Exposure Standard of 0.1 f/ml for crocidolite and amosite (considered much more hazardous than chrysotile), but 1.0 f/ml for chrysotile in recognition of the different hazard and lower associated risk. Chrysotile is not as hazardous because of the fibres being curly and not straight allowing the fibres to be trapped higher in the lung. Chrysotile fibres are also far more biosoluble in the lung. Chrysotile is believed to dissolve within one to two years whereas the amphiboles crocidolite and amosite are essentially in the lungs for life. Also chrysotile is several orders of magnitude less likely to cause mesothelioma. In 2003, the Workplace Exposure Standard for all forms of asbestos was changed to 0.1 f/ml, which was not based on scientific evidence, but to simplify the management and control of asbestos. Often there were different types of asbestos types used in the same asbestos products and this was especially true for asbestos re-enforced high pressure water pipe. Applying different standards for mixed asbestos scenarios was considered impractical. 5.1 Understanding Results The simulation was designed with a worst-case scenario of exposure as mentioned above. The simulation monitoring showed that grinding the caulking material was capable of producing exposures to chrysotile fibres. Before the simulation was conducted, it was thought by the author that few chrysotile fibres would be produced, but the monitoring and analysis showed some free chrysotile fibres in the respirable range are produced by this aggressive grinding technique. December 2015 Page 20 of 23

Based on the results of the simulation, the author believes that it is highly likely that people performing the caulking removal work at Minnivale using grinding and sanding would have had significantly lower exposures to chrysotile compared to the simulation because of the reasons mentioned throughout the report. This exposure would very likely have been below the Workplace Exposure Standard for asbestos because the worst case simulation results were at or below the Workplace Exposure Standard. 6 Information to be Provided to Employees Involved in Work at Minnivale It is important that the results of the simulation are provided to all workers involved in the Minnivale renovation so that they can understand that any exposure they may have had would not have exceeded the Workplace Exposure Standard. Some employees may be anxious about the results and it would be necessary to provide an information session to discuss the results. The author recommends that employees should be given the opportunity to ask questions about the simulation and their possible past exposures. Laurie Glossop B.Sc Ph.D COH MAIOH FAIOH Principal Consultant Glossop Consultancy December 2015 Page 21 of 23

7 Appendix A UWA SEM Analysis Report Table 2: UWA SEM Analysis Report December 2015 Page 22 of 23

8 Appendix B - Scientific Assessment of Exposure to Asbestos from Caulking Compounds There are very few scientific papers on the assessment of exposures from removing caulking compounds. This is possibly due to the level of risk considered to be low. One paper that has some measurement data is: Journal of Exposure Analysis and Environmental Epidemiology (2004) 14, 234 244. Occupational exposure to airborne asbestos from coatings, mastics, and adhesives. Dennis J Paustenbach, Amy Sage, Michael Bono and Fionna Mowat This paper had similar/comparable concentrations of chrysotile when removing mastics/adhesives. Over the past few years, a question has arisen about the degree of exposure to airborne asbestos associated with the application, cleanup, and tear-out of glues and mastics used between 1940 and the present. These liquid products were used either to adhere insulation to pipes and boilers or to cover the insulation so as to protect it. In this study, four asbestoscontaining products, a coating, two mastics, and an adhesive, which were representative of the various classes of products that have been used historically, were tested to determine the airborne concentration of asbestos fibers released during five different activities (application, spill cleanup, sanding, removal, and sweep cleaning). Each activity was performed for 30 min (often in triplicate). Personal (n=172) and area (n=280) air samples were collected during the tests, and each was analyzed for total fiber concentrations using phase contrast microscopy (PCM), and for asbestos fiber count using transmission electron microscopy (TEM). A measurable concentration of asbestos fibers was detected in six of the 452 samples collected (0.0017 0.0184 fibers/ml). The observed asbestos fibers counts for each product were similar to background. Only one asbestos fiber was detected in an indoor background sample; no asbestos fibers were identified in any of the outdoor background samples. The (raw) PCM-total fiber concentrations were adjusted based on TEM analyses that reported fraction of asbestos fibers (to derive a PCM-asbestos concentration) and by the fraction of the 8-h workday that a worker spends performing the activity (to derive a calculated TWA). For the coatings, mastics, and adhesives evaluated in the present study, the calculated TWAs using hypothetical work scenarios were well below the current Occupational Safety and Health Administration (OSHA) Permissible Exposure Limit (PEL) of 0.1 fibers/ml. The calculated TWAs ranged from 0.03 to 0.009 fibers/ml. The actual concentration of airborne asbestos due to these products is almost certainly much less than the TWAs, and may be so low as to not be measurable. These results support the historical view that these products, over the past 50 years, did not pose an occupational health hazard under foreseeable uses. December 2015 Page 23 of 23