The review of the status of the chemical storage and management system for EnRad was contingent on completion of the MSDS review, including the provision of up-to-date MSDS and chemical inventory information. This part of the review process focussed on storage and handling procedures for currently held chemicals. A review of chemical storage considerations including an assessment of any potential reactivity concerns and storage incompatibilities was undertaken. Of particular importance was the assessment of potentially time-sensitive chemicals (see Appendix 4 for testing details) with the intent of creating a register of time sensitive chemicals to manage the additional risks associated with the longer term storage of these chemicals.
Cases where the inappropriate storage of a chemical or mixture of chemical have been identified are detailed individually in this review. This includes the storage and handling of organic waste as distinct from issues associated with pure unused chemicals. Particular reference is made to the potential consequences of inappropriate management of time-sensitive chemicals or unstable chemical mixtures, especially those chemicals involved with the L19 incident.
To assess the suitability of the current storage arrangements regarding the proximity of chemicals to each other, reactivity checks were undertaken using the ChemWatch database and a reactivity database jointly run by Cameo Chemicals and the National Oceanic and Atmospheric Administration (NOAA) branch of the US Department of Commerce (http://cameochemicals.noaa.gov/, accessed 29 January 2009). The results from these reactivity checks are shown in Appendix 5. Electronic spreadsheets have been set up to maintain ongoing details for chemical reactivity and to enable easy updates of information when new chemicals are acquired. These spreadsheets are located on SSDX:
\ENRAD Administration\Darwin_Laboratory\Laboratory Management\EnRad Chemical Inventory\CAMEO-NOAA reactivity templates
Recommendations for improvement in chemical storage procedures arising from the reactivity evaluation process described above are given at the end of this section.
3.1 Time-sensitive chemicals
Organic waste is generated during the process of separation of U and Th isotopes from sample matrices, (Martin & Hancock 2004). This waste consists of 20 mL Xylenes and 5 mL Tri-butyl Phosphate (TBP) per sample. Additionally nitric and hydrochloric acids and water, and traces of metals are likely components of this waste.
The problem with the use of low volatile xylene as the extractant is that it resulted in the build up of vapour (with attendent OH&S risks) in the detergent baths used to clean the funnels between uses. In 2002 a solution was found to this issue. Rinsing the funnels with isopropanol prior to their removal from the fume cupboard was found to effectively strip the xylene from the glass prior to the washing process.
Isopropanol (~10–25 mL per sample) has subsequently been used for rinsing the separating funnels prior to normal washing of the equipment. Typically the isopropanol wash residues have been combined with the extraction wastes described above.
The mixed organic waste was stored in 2.5 L amber glass Winchesters, which had been accumulating in a fume hood in laboratory L.19 in the Enrad Radiochemistry laboratory since the facility was moved to Darwin from Jabiru in August 2002. Some of the waste may have been carried through to Darwin from the Jabiru site, and was thus many years old. A total of 11 x 2.5 L bottles containing 27.5 L in L.19 and L.21, and 1 x 1 L bottle in L.25 containing 150 mL of this type of organic waste was being stored at the time of the L19 incident. At this time there was no documented procedure for the disposal of this type of waste and no formal risk assessment had been made regarding the storage of this waste at the Darwin site.
The fume cupboard in laboratory L19 was primarily used for storing organic reagents in regular usage, and was also occasionally used for sample concentration through precipitation of iron oxyhydroxides as described by Martin and Hancock (2004).
As part of the investigation process following the explosion that occurred in the L19 incident, detailed investigations were conducted into the hazard posed by this waste.
Two possible modes of production of explosive materials by components of this waste were identified:
Organic peroxides are produced in isopropanol if stored for long periods or if not stored correctly (eg when contaminated; Bailey et al 2004b).
Nitric acid can catalyse the hydrolysis of TBP, which can lead to the production of potentially explosive butyl nitrate, or other nitrated organics at higher temperature (Robinson et al 2003).
Details of each of these modes are provided below
18.104.22.168 Peroxidation of Isopropanol
Isopropanol is a time-sensitive chemical, that may form explosive peroxides following prolonged contact with air or sunlight, or if stored for lengthy periods of time (Chemwatch 2008a). The MSDS on file in the laboratory at the time of the L19 incident in mid-2008 was dated 1985 and made no mention of the risk of peroxide formation.
As part of the incident investigation to determine the possible cause of the explosion it was recommended that the winchesters containing the waste should be tested for peroxides. Test strips were purchased (QUANTOFIX® Peroxide 25 and Peroxide 100) for this purpose. However such strips were not suited to testing the organic waste because:
Test strips are difficult to use with water-immiscible, low volatility chemicals – such as Xylenes and TBP – as the strips need to be dried to give a result
There is the potential of explosive material crystallising around the rim of the lid due to long term storage and shaking prior to the attempt to digest the waste. This applies to any test method requiring a sub-sample.
Potential of alternative unstable or explosive material in the waste other than peroxides
Potential for precipitation of peroxides out of solution, or formation of polyperoxides which may lead to readings much lower than what is actually there (Bailey et al 2004a, 2004b). This may lead to the incorrect disposal method being used.
It is recommended that separate waste bottles are used for the Xylenes/TBP waste and the isopropanol washings. The Xylenes/TBP waste shall be disposed of annually, and the isopropanol washings shall be disposed of more regularly: within 12 months of opening the oldest isopropanol source used for washing.
All squirt bottles and open Winchesters of Isopropanol were tested with the Quantofix® peroxide 25 tests strips. The results of this are detailed in Appendix 4. Of note is the low but significant concentration of peroxides (~2 ppm) found in one of the squirt bottles.
It is recommended that squirt bottles of Isopropanol are dated when filled. Excess Isopropanol should disposed into an appropriately labelled storage winchester after 2 months and the squirt bottles cleaned with Decon 90 prior to re-use.
The hydrolysis of TBP by nitric acid is well documented in industrial use of TBP for solvent extraction of uranium and other actinides. This is usually related to the production of Red Oil.
Red oil is as a substance of varying composition containing potentially explosive nitrated organics (Robinson et al 2003). Red oil forms when organic solutions come into contact with concentrated nitric acid at temperatures above 120°C. Although Red oil is relatively stable below 130°C, it can decompose explosively when heated above above 130°C (Robinson et al 2003).
Specific mention was made in the internal investigation report for the L19 incident of the possibility of concentrated nitric acid having being inadvertently added to the digest beaker that exploded. Additionally, photographic evidence indicated the likelihood that the temperature at which the digest was being performed was significantly in excess of 130°C. Although the possibility of Red Oil formation is considered an unlikely cause of the explosion, the mechanism through which TBP is hydrolysed by nitric acid may have led to the formation of explosive material. Given the length of time the waste was stored, these hydrolytic reactions may have been taking place for some time. The formation of butyl nitrate may have increased quantities of explosive material in the waste. Build-up of Butanol and Phosphoric acid in the organic waste may increase the risk of hazardous decomposition products being formed under long term storage.
Considering the unknown characteristics of the mixed waste an experienced external contractor was engaged to provide a quote to remove the waste from site without additional handling by eriss staff. This task was ultimately undertaken by Dangerous Goods Management (DGM), who have removed the waste from the eriss site (to a facility in Darwin). DGM will provide eriss with a certificate of destruction when the material is finally destroyed. Destruction may take up to 18 months due to a low overall volume of this type of waste handled by DGM, and the need to accumulate a sufficiently large volume for economic destruction.
3.1.2 Phosgene formation from Chloroform
Chloroform is used in the EnRad laboratories for 210Po and 210Pb separation.
Although it is not normally mentioned in current MSDS’, chloroform has the potential to form the toxic gas Phosgene which is both stable and soluble in chlororform (Bailey et al 2004b). Additionally, pure chloroform will readily decompose to form free radical degradation products which can lead to the formation of hydrochloric acid (Honeywell International 2008). Therefore prolonged storage of chloroform can lead to a reduction in pH and/or a build-up of phosgene gas in the headspace of the container and the reagent.
Chloroform is generally supplied in 3 forms (Turk 1998):
Unstabilised (ie pure)
Stabilised with Amylene
Stabilised with Ethanol
Bailey et al (2004b) indicate that unstablised chloroform will undergo auto-oxidation to produce phosgene at room temperature in the presence of sunlight, humidity or a metal catalyst. Bailey et al (2004b) also suggest that metals only need to be in trace amounts to catalyse this reaction. This raises the possibility of phosgene formation in unopened bottles of chloroform due to the presence of trace metals and after long-term storage.
Chloroform stabilised with amylene has been shown to be unstable on prolonged storage, even in cool conditions and the absence of air and light (Turk 1998). Ethanol stabilised chloroform is considered to be reasonably well protected from phosgene formation (Honeywell International 2008; ARS 2008), possibly due to ethanol both inhibiting the formation of phosgene and reacting with phosgene to form diethyl carbonate, which is relatively harmless.
Chloroform vapours are dangerous to human health (Chemwatch 2008a) and precautions have been taken to minimise risks posed by these to staff of EnRad. This includes instruction not to open or handle chloroform outside of a fume hood. These precautionary measures are considered sufficient to prevent any significant additional risk being posed to laboratory staff handling chloroform.
Chloroform used at eriss is stablised with 0.6 – 1.0% ethanol and is therefore considered to pose a low risk of having significant levels of phosgene formed. Due to an administrative error in 2004 however, approximately 30 L of ethanol stabilised chloroform was purchased at one time. This excess chloroform has been stored in the flammables store L.36 since delivered. The recommended storage temperature (MSDS, Chemwatch 2008a) is less than 30°C. However, measurement of ambient temperature in the flammables store over a period of two days in October 2008 indicated temperatures in excess of this. This temperature may have reached several degrees higher during hotter days of the year. Although the flammables store should have been ventilated, investigations revealed that the ventilation unit was not functioning and was not being regularly serviced after a change in service contractor for the unit. It was repaired in early 2009 and has now been placed on the service schedule overseen by the facilities manager. At the eriss site there is an additional flammables store (L.35) that is airconditioned, and this should be considered utilisation in future for storage of EnRad flammable stocks.
Based on this information it was considered prudent to test for phosgene formation in older bottles of chloroform. A preliminary pH test was conducted on the only opened bottle of chloroform from L.19 to use as an indicator of potential degradation of the chemical. Even though results from this test indicated no degradation of the chloroform (pH of 7; see Appendix 4 for further details) it was considered prudent to conduct specific testing for phosgene. Phosgene testing was conducted on the opened bottle of chloroform, and one bottle that had not yet been opened. A suitable test method is given by the US Department of Agriculture (ARS 2008) involving the preparation of test strips with 4-(dimethylamino)benzaldehyde and diphenylamine. The test results were negative (see results in Appendix 4).
Despite not finding phosgene in measurable concentrations in this instance, it is recommended that chloroform be integrated into the time-sensitive chemicals management framework as detailed below.
Chloroform has an additional risk in that it may react violently with acetone in the presence of an alkali (Chemwatch 2008a). For this reason the ChemWatch MSDS recommends segregation of these two chemicals. This has also been considered in recommendations for improvements in the storage of both of these organic chemicals.
It is recommended that once current stocks of chloroform have been depleted, sufficient stock should be purchased at any one time to last only an estimated 6 months. Chloroform stock is to be dated upon arrival and when opening each container. In no case should storage of chloroform be for no longer than 18 months after receipt or 12 months after opening, whichever is earliest. Chloroform should be stored in a flammables cabinet inside the laboratory to ensure storage below 30°C.
It is recommended that the L.35 flammables store Manager should be consulted and a risk assessment be undertaken for the transfer of the two EnRad flammables cabinets from the L.36 store to the L.35 store.
It is recommended that acetone be stored separately to other flammable liquids. Co-ordination between users of acetone should be sought to find a common storage cabinet. If this is not possible, flammable cabinets should be sought to store acetone separately to other flammable liquids held at eriss.
3.1.3 Other time-sensitive chemicals
There is one additional time-sensitive chemical (Ultima Gold AB) used by EnRad that has not yet been incorporated into the time sensitive chemicals-management framework. This chemical is used for preparation of samples for liquid scintillation counting, and comprises an organic solvent mixture containing a dissolved scintillant. The manufacturers information indicates that the reagent will degrade through time and hence its perfomance as a reagent will deteriorate. However, this degradation is indicated not to result in the production of unstable byproducts (Chemwatch 2008b).
Current suppliers of this chemical provide expiry date details for each reagent bottle supplied, and these details should be used to incorporate this chemical into an effective time sensitive chemical management framework.