• About this site|
• Site index|
• Contact us|
• Jobs

Investigation of brominated flame retardants present in articles being used, recycled and disposed of in New Zealand

• Executive Summary
• 1. Introduction
• 2. Estimate of the amount of BDE in New Zealand
• 3. Assessment of the fate of BDEs in waste
• 4. Profile of BDE flows in New Zealand
• 5. Risk analysis
• 6. Environmentally sound disposal methods
• 7. Limitations of the study
• 8. Conclusions and Recommendations
• Additional readings
• Appendix A Detailed data tables
• Appendix B Polybrominated Diphenyl Ether (PBDE) structure and physical properties
• Appendix C Analytical equipment and procedures
• Appendix D Certificate of analysis
• Appendix E Methods and analysis of consumer products
• Appendix F Calculating the % BDE present in consumer products
• Appendix G Details of the selected landfills and the sampling of leachate
• Appendix H Current status of the flame retardant industry
• Appendix I Non-dust sources of BDEs more prominent human intake
• Appendix J Completed Questionnaire SC-4/19 for submission of information on new POPs in accordance with SC-4/19
• Appendix L Questionnaire for organisations that manufacture articles containing BFR’s
• Appendix M Excel spreadsheet entitled ‘XRF analyser raw data’
• Appendix N Excel spreadsheets entitled ‘NZ Statistics data giving categories with raw data’

5. Risk analysis

The quantitative estimate of BDE in New Zealand involves a number of assumptions to be made and these have been set out in the relevant Sections above. Because of the uncertainty around the percentage of BDE in used articles in New Zealand we have estimated the amount of BDE currently “In-use” in New Zealand and that contained in landfills by providing a distribution around percentage of BDE in sampled articles as a key risk variable.

The impact of making estimates of uncertain variables is accounted for by using Quantified Risk Analysis (QuRA^TM) to model estimates for key uncertain variables. This involves identifying the likely range (90% confidence interval) that the key uncertain variables may fall within (i.e. low, most likely, and high) assuming a triangular distribution (a simplified/abbreviated version of the bell-shaped distribution). QuRA^TM results in an expected output (average output over 5,000 iterations) and the probability that the output (in this case total mass of BDE in New Zealand landfills) will fall between particular ranges (typically the 5% and 95% levels).

The results of the risk analysis are shown in the table below. Using QuRA^TM to account for uncertainty around key variables gives the following results:

5.1 Sampling and analysis of landfill leachate

To test the hypothesis ‘landfills are a secure depository for PBDE containing polymers’ three landfills were selected for leachate sampling and analysis:

Hampton Downs Landfill Meremere cell 1 operated July 2005 to July 2009

Greenmount Landfill East Tamaki, Auckland operated 1979 to 1 July 2005

Taupo District Council Landfill Broadlands Rd, Taupo operated from 1985

Further details about these landfills and the sampling of them can be found in Appendix G.

5.1.1 Extraction of samples

Brominated flame retardants have a very low solubility, however in the presence of organic acids that accumulate in landfills, the solubility can increase significantly. Co-disposal landfills therefore could potentially discharge significant amounts of brominated flame retardants through their leachate.

Being very non-polar they would also have an affinity to small particulates. Therefore two extracts have been created from each leachate sample:

1. FIL - a particulate sample on Whatman glass microfibre filters GF/F
2. SPE - an extract of the dissolved phase using Waters Oasis HLB LP solid phase extraction (SPE) cartridges which contain a permeable gel with very high affinity for non-polar compounds.

Both required vacuum to be applied to pull the sample through the filter media. The glass microfiber filters blocked quite quickly. The leachate from Taupo DC landfill (TDC) and Greenmount (GM) blocked the filters after 200 ml, so 10 filters were needed to filter 2 litres of leachate. The leachate from Hampton Downs (HD) blocked the filters after just over 100 ml and the SPE gel after 1.7 litres had passed.

Using the Oasis SPE cartridges the vacuum needs to be regulated, to obtain a steady flow of not more than 1 ml/min. Prior to extraction the Oasis cartridges were conditioned with 5 ml Baker Ultra Resi-analysed Methanol (ultra pure for organic residue analysis), as per the Oasis manual.

The filters (as a stack) and the solid phase extraction gel was analysed using a handheld XRF analyser. XRF analysis results of Filters and SPE cartridges showing zinc and total bromine.

Analysing total bromine will over estimate the concentration of brominated flame retardants and certainly the concentration of only the brominated diphenyl ethers.

The filters and SPE cartridges of two landfills, Hampton Downs and Greenmount were analysed for brominated flame retardants including; BDE’s, TBBP-A and HBCD (α, β and gamma). Analysing these compounds is not a routine analysis. The responsible person for this contract has been Prof. Dr. J. de Boer and the research manager drs. S.H. Brandsma. The analysis results have been reviewed by Dr. P.E.G Leonards. After filtration and trapping of the flame retardant on the SPE columns the filters and columns the samples were sent back to IVM for analysis.

Prior to the extraction steps internal standards (BDE58, ¹³C BDE209, ¹³C HBCD and ¹³C TBBP-A) were added to all samples to ensure a proper identification and reliable quantification. One blank and one spike were measured as quality control.

IVM analysed for the PBDEs, TBBP-A and the HBCD isomers in these samples (Table below). Cleanup was performed following IVM protocol using silica columns and gel permeation chromatography (GPC). The purified extracts were analysed by gas chromatography (GC) with electron capture negative ionization technique and mass spectrometry detection (GC/ENCI-MS) for PBDEs, using a 50 meter GC column and measuring the specific bromine m/z 79 and 81. This is a highly sensitive method and was described by De Boer et al. (2001, 2006)^{Footnote 5}. All analyses were carried out under the specific conditions for PBDE analysis as described in De Boer et al. (2001, 2006).

After the PBDEs were measured the purified extracts were redissolved in Methanol and analysed for TBBP-A and HBCD on the LC-MS/MS, using multiple-reaction monitoring (MRM) of the parent and the daughter ion (Boer et al. 2006)^{Footnote 6}. The plastic product was extracted following IEC ACEA protocol “Determination of levels of regulated substances in electro technical products”. However, the detection of the PBDEs, TBBP-A and HBCD was achieved by GC-(NCI)MS and LC-MS/MS following IVM protocol. The recoveries reported by the laboratory of the BDE/TBBP-A and HBCD are > 90%.

From the results we interpret the leachate sample from the Greenmount landfill is very likely diluted as the landfill operated for 25 years, opening in 1979 and being closed in 2005 with the dissolved BDE concentration in the leachate being only 1 % of that found in the leachate from Hampton Downs, cell 1 operating for only 4 years. The site engineer at Greenmount landfill had no full insight in the system and significant difficulties opening a collection well made comparison between different collection wells impossible within the timeframe of this study.

The leachate sample from the closed cell 1 at Hampton downs was taken from a pipe discharging in a collection well. The pipe was pointed out by the site engineer who did have good knowledge of the leachate collection system. This sample is very likely representative of the leachate of this landfill cell which has been in operation from mid 2005 to mid 2009.

Total concentration of the BDE’s analysed in dissolved phase and bound to particulates is 168.9 and 395 ng/ltr respectively. Combined there is 564 ng of BDE leaving the landfill in every liter of leachate. The average daily leachate discharge is 60,000 litres (2 road tankers / day) from cell 1 (closed) and cell 2 (operational). Assuming the BDE concentration in the leachate from Cell 2 is equal to that from cell 1, which is an over estimate, this means daily 0.034 gram BDE is leaving the landfill.

Even adding the small amount of TBBP-A, the annual discharge from the landfill is less than 12.5 gram. This assumes the only other route out of the landfill, evaporation through the 2 meter thick clay cap, of these very low volatility compounds is virtually zero.

At an average BDE concentration of 5% in polymers this equates to 1 piece of plastic of 250 grams entering the environment. This is a very small amount in comparison with the content of the landfill. In the Auckland Regional Waste Survey (2009) the percentage of plastics disposed in landfills is estimated at 8.9 %^{Footnote 7}. During the 4 years of operating cell 1 approximately 2.5 million tons of waste was received. Thus about 222 million kilo’s of plastics are contained in the landfill. The quantity of BDE’s escaping annually represents around 1 billionth of the amount likely held in the landfill cell 1.

This is clearly an infinitesimal volume leading to the conclusion that properly designed and managed landfills are a secure final depository of BDE containing plastics.

5.2 BDEs in the New Zealand recycling system

There are very few articles containing commercial pentaBDE and octaBDE that are recycled in New Zealand. The majority of recycled polymer articles recovered by commercial recycling companies are in the packaging and food-contact categories such as milk and soft drink bottles (recycle classes 1 and 2). These are unlikely to contain BDEs. Recycle category 3, PVC, is self extinguishing. This holds mainly for rigid PVC. Soft PVC contains phthalates to make the PVC pliable, however phthalates are flammable and therefore flame retardants are added to soft PVC in applications where heat or source of ignition is expected.

Category 4 Low density polyethylene is also mainly used in applications where flame retardants are rarely needed. As our research has shown BDEs are limited to a small range of specialist applications and electronics. Some of these fall into the categories 5 and 6, however mainly into 7.

A new form of recycling is that involving expanded polystyrene (EPS). To the author’s knowledge two firms recycle their internal EPS waste, and a third fabricates insulation sheets from recycled EPS.  The latter process involves granulating and steam moulding into desired sheets and blocks that are sold for under floor insulation and as under concrete floor foundation blocks.

When the recycled product is BDE free as most modern EPS is both uses meet current BFR standards. However when older EPS is used, limiting the use to concrete floor foundations would be preferable.

The ban on concentrations of BDE above 0.1% in polymers has had an impact on polymer recycling. As more and more products include recycled polymers, it has become critical to know the BDE concentration in these polymers, either by tracing the origins of the recycled polymers to establish the BDE concentrations, or by measuring the BDE concentrations from samples. Polymers with high BDE concentrations are costly to handle or to discard, whereas polymers with levels below 0.1% have value as recyclable materials.

In our estimates of BDE in the recycling system we have not estimated articles containing less than 0.05% BDE in homogenous polymer parts.

The laboratory analysis of consumer good polymers has found that mixtures of ‘modern’ BFRs like HBCD and TBBP-A can accompany ‘older’ BFRs like PBDEs. We have found an advantage in the manufacturing process using both types of flame retardants and have to conclude that quantities of recycled polymers have been used in the manufacture of these goods. Despite the ‘dilution’ of BDEs in those products the effect is that they remain in our environment and will be influencing our health much longer compared to using BDE free polymers and disposing of the BDE containing polymers to landfills or incinerators.

Read a description of this image

From the bar chart above it is clear that the current New Zealand polymer recycling industry mainly deals with non-BFR containing polymers. BFRs are mostly associated with recycling groups 3 (soft PVC), 6 and 7. Often the specific use for which the articles were intended is very indicative of the presence of BFRs. The age of the article can give some indication of the type of BFR, penta and octa BDEs present in older and largely US derived polymer products, deca-BDE in older as well as more modern polymers (US, EU and NZ made) and non-BDE BFRs in more recent products, increasing from 2006 onwards.

In Figure 5 on the next page the common uses of polymers with recycling numbers 1 – 7 is provided. Brominated flame retardants (BFR’s) are not allowed to be used in polymers which may have contact with food. Clearly the majority of the polymer materials in recycle groups 1 to 6 recovered from household recycling therefore do not contain BFR’s.

Figure 4. Types of plastic; their properties, uses and recycled use

View figure at full size

Read a description of this image

5.3 BDEs in house dust – a case against recycling BDE containing polymers

The aim of the Stockholm convention article 6 is to limit the exposure to persistent organic pollutants, such as the brominated flame retardant group of the poly diphenyl ethers (PBDEs).

Although the relationship between the presence of PBDEs in house / office dust in relation to the concentration of PBDEs in the human body has been poorly researched and certainly beyond the scope of this study, there is nevertheless evidence of the rise of PBDEs in fats of the human body such as in breast milk that reveals a high degree of correlation with the increased use of PBDEs. In the figure below, the increase of PBDEs in human breast milk in Sweden over the period that the use of PBDEs was rapidly rising (1972 – 1997).

Read a description of this image

Figure 5. Often called “the graph that launched a thousand papers” (From Meironyt D, Bergman A. 1999. Analysis of Polybrominated Diphenyl Ethers in Swedish Human Milk, 1972-1997. Journal of Toxicology and Environmental Health. Part A, (58):329-341.)^{Footnote 9}

The US Geological Survey (USGS) reported in 2004 levels of up to 200 ppb in human breast milk in the US.^{Footnote 10}

The same report also indicates Deca-BDE rapidly breaks down in sunlight forming lower brominated PBDEs which are more bio-available and potentially more toxic.

5.4 Natural breakdown and bio-accumulation of PBDE’s

The natural breakdown and bio-accumulation of PBDEs mentioned in the USGS report has also been reported in the study carried out by Renee Sharp and Sonya Lunder of the Environmental Working Group (EWG)^{Footnote 11}. However unlike Europe, Asia and New Zealand the Americas still used a high percentage of commercial Penta PBDE in 2001 as is shown in the table below.

They also found a shift in the ratio of the two main PBDE congeners found in the Penta product: While the Penta mixture contains sixty percent PBDE-99 and forty percent PBDE-47, five out of their ten dust samples contained a significantly higher percentage of PBDE-47. In other words, half of the homes tested had an abnormal ratio of these chemicals as compared to the ratio found in the commercial Penta product.

Because PBDE-99 has five bromines while PBDE-47 only has four, their data suggests that PBDEs are breaking down inside study homes. This is of particular concern because PBDE-47 is more bio-accumulative than PBDE-99 or -209. In EWG’s breast milk study^{Footnote 12}, for example, we found that women had at least twice as much PBDE-47 as PBDE-99 in their bodies^{Footnote 13}. This finding also underscores the concern for Deca which also appears to break down in people’s homes and the environment.

Several other studies also confirm the reductive de-bromination mechanism of the photolytic degradation of PBDEs is rapid in photodecomposition experiments. The data obtained suggests that the photo-degradation of BDE-209 is a sequential de-halogenation mechanism.^{Footnote 14} The concept of de-bromination of deca-BDE in the environment is not accepted by all. Kellyn Betts in the Environmental Science and Technology Journal (September, 2008) reports that Klaus Rothenbacher a toxicologist from British Columbia accepts that while deca-BDE can undergo de-bromination in the lab his conclusions from six recent studies are that there are no indicators of this de-bromination in the environment. Further studies are discussed notably from Jeoff Gearhart whose studies found that deca-BDE present in sealed quartz cuvettes and placed inside cars was found to breakdown to lower molecular weight congeners with as much as 63% of the original bromine not present as a PBDE. The controversy continues and there is the obvious need for further research. The limitations of this study and the time constraints involved mean a full review of available literature is not possible but with the high levels of Deca-BDE found in house dust as was shown above, even if the breakdown occurs slowly or to a small degree, Deca could nevertheless be an important source of exposure to the more toxic and bio-accumulative forms of the PBDEs.

From the BDE analysis results we obtained in this study we can see some articles have not only high Deca-BDE levels but also contain some TBBP-A. The interpretation by researchers at the Institute for Environmental Studies in the Free University of Amsterdam, Holland, is that during the manufacture of these goods or of the polymers to make these goods some recycled polymers have been blended with virgin polymers.

Despite the low levels of commercial Penta BDE imported and used in manufacturing the natural breakdown of commercial Deca BDE could lead to a future rise in presence of the lower brominated compounds which are more soluble, more bio-accumulative and more toxic.

previous | next