See discussions, stats, and author profiles for this publication at: [Link]

net/publication/11166340

Distribution and levels of brominated flame retardants in sewage sludge

Article in Chemosphere · October 2002

DOI: 10.1016/S0045-6535(02)00113-3 · Source: PubMed

CITATIONS READS

180 65

3 authors, including:

Tomas Öberg

94 PUBLICATIONS 2,569 CITATIONS

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Environmental ChemOinformatics View project

All content following this page was uploaded by Tomas Öberg on 11 March 2021.

The user has requested enhancement of the downloaded file.

This PDF-file is an author generated postprint of an article published in
Chemosphere 48, 805-809 (2002). The definitive published version is available at
[Link]

DISTRIBUTION AND LEVELS OF BROMINATED FLAME

RETARDANTS IN SEWAGE SLUDGE

Karin Öberg1*, Kristofer Warman2 and Tomas Öberg1

1
T. Öberg Konsult AB, G. Brov. 13, 371 60 Lyckeby, Sweden. Email: info@[Link]
2
ALcontrol AB, Box 154, 611 24 Nyköping, Sweden. Email: [Link]@[Link]

* Current address - California Institute of Technology. Email: karin@[Link]

ABSTRACT

One hundred and sixteen sewage sludge samples from 22 municipal wastewater

treatment plants in Sweden were analysed for brominated flame retardants.

Polybrominated diphenyl ethers (PBDEs) were in the range n.d.-450 ng/g wet weight,

tetrabromobisphenol A (TBBPA) varied between n.d.-220 ng/g wet weight, 2,4,6-

tribromophenol was in the range n.d.-0.9 ng/g wet weight and polybrominated biphenyls

were not detected (except for a possible analytical interference). There was a significant

variation in the samples among plants. Influence from industries and other local sources

can therefore be assumed. The correlation pattern indicated contribution from three

different types of technical products; composed of either low-brominated PBDEs, deca-

BDE or TBBPA

Keywords: BFRs, brominated diphenyl ethers, PBDE, TBBPA, waste water treatment,

principal component analysis, ANOVA

1
INTRODUCTION

Brominated flame retardants (BFRs) are widely distributed in the environment, and their

persistence and/or potential for bioaccumulation have caused concern (WHO, 1997).

The content of BFRs in sewage sludge from wastewater treatment plants gives an

indication of the general exposure and use of these compounds. Several investigators

have reported findings of the most widely used BFRs in this matrix, and other parts of

the wastewater disposal system (Antusch, 1999, Hagenmaier et al, 1992, Hellström,

2000, Kuch et al, 2001, Nylund et al, 1992). In this study we will give a general

overview of distribution and levels of BFRs in sewage sludge from Swedish plants.

METHODS

Samples

Sewage sludge samples from Swedish wastewater treatment plants are frequently

analysed for various pollutants. Results are here reported for 116 sewage sludge

samples, from 22 municipal wastewater treatment plants, which arrived to the

ALcontrol laboratory in Nyköping, Sweden, between October 1999 and September

2000. The sludge samples were collected by the plant owner’s staff and sent directly to

the laboratory.

Chemical analysis

Sample preparation, extraction and clean-up procedures were adapted with minor

modifications from Nylund et al (1992) and Sellström et al (1995 and 1998). Chemicals

used are listed in Table 1.

2
The samples were analysed for brominated flame retardants (BFRs) by high-resolution

gas chromatography/mass spectrometry (HRGC/MS) with negative chemical ionisation

(NCI). A HP 5973 mass spectrometer was used for all analyses and the ion source

temperature was kept at 150 °C and the electron energy at 133 eV. The gas

chromatographic separations were carried out using J&W DB5MS columns, with (5%-

Phenyl)-methylpolysiloxane as stationary phase. 30 m x 0.25 mm i.d. columns were

used for all analyses except for decabromodiphenyl ether and decabromobiphenyl where

a 15 m x 0.25 mm i.d. column was used instead. The temperature programmes were run

from 80 °C to 325 °C and helium (AGA, Stockholm, Sweden) was used as carrier gas.

Quantification was done by the internal standard method relative to a multilevel

calibration for all components. 3,3’,4,4’-Tetrabromodiphenyl ether, 13C-2,3,3',4,4',5'-

hexabromobiphenyl and pentabromophenol were used as internal standards.

Analytical data were originally reported in either ng/g wet weight (n=94) or ng/g dry

weight (n=22), but for comparison all values have here been recalculated to wet weight.

Data analysis

The analytical data were described and evaluated using exploratory data analysis

(EDA), analysis of variance (ANOVA), linear regression and principal component

analysis (PCA). Tukey (1977) review the EDA-methods. Ståhle and Wold (1989)

describe ANOVA, and Draper and Smith (1981) describe regression analysis. PCA is

reviewed by Jackson (1991) and Martens and Næs (1989), together with other bilinear

methods for multivariate data analysis.

3
The software Statistica (StatSoft, Inc., USA) was used for EDA, ANOVA, linear

regression and descriptive statistics. The principal component analysis was carried out

with the software Unscrambler (CAMO ASA, Norway).

RESULTS

The analytical results are summarised in Table 2. Medians, quartiles and ranges are

better suited as measures of location and spread than the arithmetic means and standard

deviations, since the distributions for all analytical variables were highly skewed.

In Figure 1 the distributions of the sum of PBDEs for the different wastewater treatment

plants are shown with Box-Whisker plots. Here the data is also summarised as medians,

quartiles and ranges.

Clear-cut spatial trends were not evident, but samples from the northern part of Sweden

generally showed lower levels of the analysed compounds. The samples with the

highest concentrations of decabromodiphenyl ether and tetrabromobisphenol A were

collected in wastewater treatment plants with possible contribution from textile

respectively electronics industry. The short time-span (12 months) covered by these

samples does not permit us to evaluate possible temporal trends.

The variations of the sum of PBDEs among and within the wastewater treatment plants

were evaluated with analysis of variance (ANOVA). The measurement results were log-

transformed in order to fulfil the normality assumption (see Dean, 1981). The variation

among plants was statistically significant, Table 3.

4
The correlation pattern between the different PBDEs and TBBPA was evaluated using

principal component analysis. The measurement results for deca-BDE and TBBPA were

log-transformed prior to the data analysis, and all variables were auto-scaled (55

samples analysed for all PBDEs and TBBPA). The low-brominated PBDEs did not

require any transformations, since their distributions were only slightly skewed within

this subset. A model with three significant components (cross validated) could explain

93 % of the variance. The 3D-loadings plot in Figure 2 show how the different

analytical variables correlate. Tetra-, penta- and hexabrominated diphenyl ethers

correlate well, while decabromodiphenyl ether and tetrabromobisphenol A show other

distribution patterns distinctly separated from each other. The correlation coefficients

between the different lower-brominated diphenyl ethers vary between 0.92-0.98. The

linear relationship between PBDE #47 and #99 is shown in Figure 3.

The congener pattern for tetra-, penta- and hexabrominated diphenyl ethers is stable as

shown in Table 4. This corresponds well with the composition reported for the

commercial product Bromkal 70-5DE (Huber and Ballschmiter, 2001, Sjödin et al,

1998, Sundström and Hutzinger, 1976).

DISCUSSION

The results presented show that there are considerable variations in the concentrations

of brominated flame retardants in sewage sludge, and that there is a statistically

significant variation among municipal waste water treatment plants. Influence from

local sources can therefore be assumed. Two examples of industries that can give

specific contributions are the electronics (TBBPA) and textile industries (deca-BDE),

and the highest concentrations of each pollutant were indeed found in samples with such

5
a connection. Similar observations regarding TBBPA were also made by Kuch et al

(2001). The correlation pattern of BFRs in the analysed samples indicates that there are

three major constituents, i.e. contribution from three distinctly different types of

technical products, composed of either low-brominated PBDEs, deca-BDE or TBBPA.

Norris et al (1973), later confirmed by Watanabe and Tatsukawa (1987), showed that

decabromodiphenyl ether can undergo photolytic debromination. There are however no

indications in this study that the bromination pattern for PBDEs is changed during

wastewater treatment, and only weak or non-significant correlations were found

between the low-brominated PBDEs and deca-BDE. This is not to be expected either

since significant UV-exposure can only occur after the sewage sludge has left the

treatment plant.

BFRs are only one example of the many hundreds of organic pollutants that occur in

sewage sludge. Analyses of this matrix are therefore attractive as a monitoring tool for

the current usage and emissions of persistent organic pollutants prior to environmental

transformations and degradation. Spatial and temporal variations may give an indication

of local and general sources as well as the effectiveness of product control measures.

REFERENCES

Antusch, E., 1999. Localisation of organic pollutant emissions into local sewers by

sluice-skin examinations (In German, Lokalisierung organischer Schadstoffemissionen

in kommunale Abwasserkanäle durch Sielhautuntersuchungen). Doctoral Dissertation.

University of Karlsruhe, Germany.

6
Dean, R.B., 1981. Use of log-normal statistics in environmental monitoring. In

Chemistry in Water Reuse, Vol. 1, (ed.) Cooper, W.J., Ann Arbor Science Publishers,

Ann Arbor, MI.

Draper, N.R. and Smith, H., 1981. Applied regression analysis. John Wiley & Sons,

New York.

Hagenmaier, H., She, J., Benz, T., Dawidowsky, N., Dusterhöft, L., and Lindig, C.,

1992. Analysis of sewage sludge for polyhalogenated dibenzo-p-dioxins, dibensofurans,

and diphenyl ethers. Chemosphere 25, 1457-1462.

Hellström, T., 2000. Brominated flame retardants (PBDE and PBB) in sewage sludge –

a problem? (In Swedish, Bromerade flamskyddsmedel (PBDE och PBB) i slam – ett

problem?). Report M 113, VAV, Stockholm, Sweden.

Huber, S., and Ballschmiter, K., 2001. Characterisation of five technical mixtures of

brominated flame retardants. Fersenius J. Anal. Chem. 371, 882-890.

Jackson, J.E., 1991. A user’s guide to principal components. John Wiley & Sons, New

York.

Kuch, B., Hagenmaier, H., and Körner, W., 2001. Determination of brominated flame

retardants in sewage sludges and sediments in South- West Germany. Poster presentation at

the 11th Annual Meeting of SETAC Europe, May 6-10 2001, Madrid, Spain.

7
Martens, H., and Næs, T., 1989. Multivariate calibration. John Wiley & Sons, New

York.

Norris, J.M., Ehrmantraut, J.W., Gibbons, C.L., Kociba, R.J., Schwetz, B.A., Rose, J.O.,

Humiston, C.G., Jewett, G.L., Crummett, W.B., Gehring, P.J., Tirsell, J.B., and Brosier,

J.S., 1974. Toxicological and environmental factors involved in the selection of

decabromodiphenyl oxide as a fire retardant chemical. JFF/Combustion Toxicology 1,

52-77.

Nylund, K., Asplund, L., Jansson, B., and Jonsson, P. , 1992. Analysis of some

polyhalogenated organic pollutants in sediment and sewage sludge. Chemosphere 24,

1721-1730.

Sjödin, A., Jakobsson, E., Kirkegaard, A., Marsh, G., and Sellström, U., 1998. Gas

chromatographic indentification and quantitation of polybrominated diphenyl ethers in a

commercial product, Bromkal 70-5DE. J. Chromatogr. A 822, 83-89.

Ståhle, L. And Wold, S, 1989. Analysis of variance (ANOVA). Chemom. Intell. Lab.

Syst. 6, 259-272.

Sundström, G., and Hutzinger, O., 1976. Environmental chemistry of flame retardants

V. The composition of Bromkal R 70-5 DE – a pentabromodiphenyl ether preparation.

Chemosphere 3, 187-190.

Tukey, J. W., 1977. Exploratory Data Analysis. Addison-Wesley, Reading, MA.

8
Watanabe, I., and Tatsukawa, R., 1987. Formation of brominated dibenzofurans from

the photolysis of flame retardant decabromobiphenyl ether in hexane solution by UV

and sun light. Bull. Environ. Contam. Toxicol. 39, 953-959.

WHO, 1997. Flame Retardants: A General Introduction. IPCS, Environmental Health

Criteria 192. WHO, Geneva.

9
Table 1: Chemicals used

Chemicals Supplier / quality

Acetic anhydride Merck p.a
Dichloromethane Rathburn, HPLC Grade
n-Hexane Merck LiChrosolv
Methyl tertiary butyl ether (MTBE) Merck p.a
Potassium hydroxide KOH Merck p.a
Pyridine Fluka GC quality
Sodium chloride NaCl Merck p.a
Sulfuric acid H2SO4 conc. Merck p.a.
Toluene Distilled in the laboratory

PBDE:
2,2’,4,4’-Tetrabromodiphenyl ether CIL 50 µg/ml in nonane
2,2’,3,4,4’-Pentabromodiphenyl ether CIL 50 µg/ml in nonane
2,2’,4,4’,5-Pentabromodiphenyl ether CIL 50 µg/ml in nonane
2,2’,4,4’,6-Pentabromodiphenyl ether CIL 50 µg/ml in nonane
2,2’,3,4,4’,5’-Hexabromodiphenyl ether CIL 50 µg/ml in nonane
2,2’,4,4’,5,5’-Hexabromodiphenyl ether CIL 50 µg/ml in nonane
2,2’,4,4’,5,6’-Hexabromodiphenyl ether CIL 50 µg/ml in nonane
Decabromodiphenyl ether Fluka purum
3,3’,4,4’-Tetrabromodiphenyl ether (IS) CIL 50 µg/ml in nonane

PBB:
Firemaster BP-6 Accu Standard tech.
Decabromobiphenyl Chem Service
13
C-2,3,3',4,4',5'-Hexabromobiphenyl (IS) CIL 40 µg/ml in nonane

TBBPA:
Tetrabromobisphenol A Aldrich 97%
2,4,6-Tribromophenol Aldrich 98%
Pentabromophenol (IS) Aldrich 99%

10
Table 2: Brominated flame retardants in sewage sludge, ng/g wet weight

Analyte n Min Lower Median Upper Max

quartile quartile
2,2’,4,4’-Tetrabromodiphenyl ether (PBDE #47) 105 <0.3 4.9 7.0 10.5 48
2,2’,3,4,4’-Pentabromodiphenyl ether (PBDE #85) 105 <0.3 <0.3 0.42 0.61 2.7
2,2’,4,4’,5-Pentabromodiphenyl ether (PBDE #99) 105 <0.3 6.7 10 14 320
2,2’,4,4’,6-Pentabromodiphenyl ether (PBDE #100) 105 <0.3 1.0 1.7 2.5 11
2,2’,3,4,4’,5’-Hexabromodiphenyl ether (PBDE #138) 105 <0.3 <0.3 <0.3 <0.3 <1
2,2’,4,4’,5,5’-Hexabromodiphenyl ether (PBDE #153) 105 <0.3 0.54 0.86 1.2 5.5
2,2’,4,4’,5,6’-Hexabromodiphenyl ether (PBDE #154) 105 <0.3 0.42 0.72 0.99 5
Decabromodiphenyl ether (PBDE #209) 114 <0.6 5.2 11 21 390
Tetrabromobisphenol A (TBBPA) 57 <0.3 0.51 2.0 4.0 220
2,4,6-Tribromophenol 57 <0.3 <0.3 <0.3 0.36 0.9
2,3',4,4',5-Pentabromobiphenyl (PBB #118) 21 <0.3 <0.3 <0.3 <0.3 <0.3
2,2’,3,4,4’,5’-Hexabromobiphenyl (PBB #138) 21 <0.3 <0.3 <0.3 <0.3 <0.3
2,2’,4,4’,5,5’-Hexabromobiphenyl (PBB #153) 21 <0.3 <0.3 <0.3 0.78* 1.1*
2,3',4,4',5,5'-Hexabromobiphenyl (PBB #167) 21 <0.3 <0.3 <0.3 <0.3 <0.3
2,2',3,4,4',5,5'-Heptabromobiphenyl (PBB #180) 21 <0.3 <0.3 <0.3 <0.3 <0.3
Decabromobiphenyl (PBB #209) 21 <0.3 <0.3 <0.3 <0.3 <0.3
* Incomplete separation from PBDE#154.

11
Table 3: ANOVA table, sum of PBDEs in 103 samples with complete analyses

Source of df Sums of squares Mean squares F p

variation (SS) (MS)
Among plants 21 8.85 0.421 1.90 0.021
Within plants 81 17.94 0.222
Total 102 26.79 0.263

12
Table 4: Congener pattern for Br4-Br6DEs (%) in 71 samples with concentrations above
the detection limit

BDE #47 BDE #85 BDE #99 BDE #100 BDE #153 BDE #154
Mean 34 2.1 48 8.2 4.3 3.4
Sdev 3.3 0.37 3.0 0.65 0.70 0.44

13
Figure 1: Box-Whisker plots for each wastewater treatment plant (22 plants and 1-13

samples from each)

Figure 2: Loadings plot for PBDEs and TBBPA (55 samples analysed for all PBDEs

and TBBPA)

Figure 3: PBDE #99 vs. PBDE #47, ng/g wet weight in 104 sewage sludge samples (r2=0.94)

14
 500

400
Sum PBDEs ng/g wet weight

300

200

100

Min-Max
25%-75%
0 Median value

15
X-loadings

-Y
0.8 X
PBDE#209
0.6

0.4

0.2

-0.2
PBDE#47
-0.4 PBDE#154 PBDE#99
PBDE#100
PBDE#153 TBBP A
-0.6
PBDE#85
0
0.05
0.10
0.15
0.20
0.25
0.30
0.35 0.6 0.8
0.40 0.2 0.4
0.45 -0.2 0

BFR-model, X-expl: 68%,14%,11%

16
 80
V
i
e
w
p
u
b
l

60
i
c
a
t
i
o
n
s

PBDE99
t

40
a
t
s

20

0
0 10 20 30 40 50
PBDE47

17