Örebro University
School of Science and Technology Josefin Persson
2009
Development and evaluation of methods for analysis of
TBECH and HBCD using HRGC/HRMS and
2
Abstract
The two additive brominated flame retardants, tetrabromoethylcyclohexane (TBECH) and hexabromocyclododecane (HBCD) are used to prevent fire to start and spread. They are simply mixed with material and are most likely to leach out in the environment, because of non-covalently binding to the material. TBECH can exist as four pairs of enantiomers, α-, β-, and δ-TBECH. The technical HBCD can exist as three pairs of enantiomers, α-, β- and γ-HBCD and two meso forms δ- and ε-γ-HBCD. None of these compounds are produced in Sweden, but they are imported to industries. TBECH has been found in Beluga blubber and can accumulate in zebrafish. HBCD has been found in water environments and can be toxic to and bioaccumulate in water-living animals.
In this study, a method was developed for separation and detection of α-, β-, γ- and δ-TBECH on HRGC/HRMS. All TBECH-isomers could be separated with the developed method. How much of the TBECH isomers that were recovered after applying existing extraction and clean-up procedures, normally applied for clean-clean-up and extraction of PCBs and PCDD/Fs, was evaluated. Low recovered amounts (6.8-35.5 %) of TBECH-isomers added in known amounts to three different whale samples indicate severe evaporation losses and possibly photolytic degradation. None of the four enantiomers were detected in the three whale samples.
For HBCD analysis, both the chromatography and MS/MS parameters were optimised for δ- and ε- HBCD yielding good chromatography and sensitivity. However, due to technical difficulties during the time-period of this project, no whale samples could be analysed for HBCD on UPLC/MS/MS.
3
Index
1. Introduction ... 4
1.1. Aims ... 4
1.2. Brominated flame retardants ... 4
1.3. Tetrabromoethylcyclohexane ... 5
1.4. Hexabromocyclododecane ... 5
2. Materials and methods ... 7
2.1. Extraction and cleanup for TBECH analysis ... 7
2.2. Extraction and cleanup for HBCD analysis ... 8
2.3. HRGC/HRMS ... 9
2.4. UPLC/MS/MS ... 10
2.5. Detection of TBECH ... 10
3. Result and discussion ... 10
3.1. TBECH method development ... 10
3.2. Evaluation of clean-up procedure... 12
3.3. Application of the method ... 14
3.4. HBCD method development ... 16
3.5. Application of the method ... 18
4. Conclusions and future perspectives... 19
5. Acknowledgments ... 19
4
1. Introduction
1.1. Aims
The aims for this work were to evaluate an existing sample clean-up procedure and to develop methods for detection of two brominated flame retardants. α-, β-, γ- and δ-tetrabromoethylcyclohexane (TBECH) with high resolution gas chromatography coupled to high resolution mass spectrometry (HRGC/HRMS) and α-, β-, γ-, δ- and ε- hexabromocyclododecane (HBCD) with ultra performance liquid chromatography coupled to triple quadrupole mass spectrometry (UPLC/MS/MS). To evaluate the develop methods α-, β-, γ- and δ-TBECH and α-, β-, γ-, δ- and ε-HBCD were analysed in fat samples from Fin whales or Winke whales from Iceland. The whale samples were extracted and fractionated using open column chromatography and then analysed on HRGC/HRMS and UPLC/MS/MS, respectively.
1.2. Brominated flame retardants
Brominated flame retardants (BFR) are cyclic carbon compounds which contains bromine. These compounds have been used for over 30 years to prevent fire to start and spread in textiles, electronics, building materials and toys [1, 2] and approximately 200 000 tons/year are being produced globally [3]. BFRs are not produced in Sweden, instead 100 tons of BFRs were imported to the industries in 2007 [4]. BFRs are found both in the air and biota, because of leakage from industries. Concentrations of BFRs have been found in birds and fish, mostly in water-living animals. BFRs have also been found in human tissues and breast milk [1, 2] and in arctic mammalians [9]. Around 70 different BFRs are used today [1, 2]. BFRs are grouped into three different groups depending on how they are used in polymers; brominated monomers, reactive and additive. Two additive BFRs are 1, 2-Dibromo-4-(1, 2-dibromoethyl)cyclohexane (TBECH) and 1, 2, 5, 6, 9, 10-hexabromocyclododecane (HBCD) [5]. Additive BFRs are mixed with the material (polymer) and are more likely to leach out in nature during use and disposal, because of their non-covalently binding with the material. Therefore, research concerning these compounds is of importance [5, 6].
5
1.3. Tetrabromoethylcyclohexane
1, 2-Dibromo-4-(1, 2-Dibromoethyl)cyclohexane or tetrabromoethylcyclohexane (TBECH) is a BFR with the molecular formula C8H12Br4. TBECH is used as an additive in polystyrene and polyurethane products [7]. The world production and use of TBECH are unknown. TBECH can exist as four pairs of enantiomers, α-, β-, γ-, δ-TBECH (see Figure 1) [8] because of four chiral carbons that can be in R or S configuration [7]. The nomenclature is based on the elution order from a DB-5
capillary column [9]. The enantiomers are thermal sensitive and can transform to each other (thermal interconversion) at temperature above 120°C and to detect all four enantiomers the initial temperature must be 120 °C or below. Despite that, GC/MS is considered to be the most suitable technique for analysing TBECH enantiomers [7].
Figure 1. The four pairs of enantiomers of TBECH. From left; α- TBECH, β-TBECH, γ-TBECH and δ-TBECH.
Studies have shown that TBECH can bind to the human androgen receptor and activate it in vitro, which can cause health problem[10]. Today there are few reports on TBECH in environment but in one study TBECH was found in Beluga blubber from the Canadian Arctic [9] and has found to accumulate in zebrafish [7].
1.4. Hexabromocyclododecane
1, 2, 5, 6, 9, 10-hexabromocyclododecane or hexabromocyclododecane (HBCD) is a BFR that is used as an additive in plastic materials and textiles [11, 12]. HBCD is produced from cyclododeca-1, 5, 9 -triene (CDT) by bromination [5, 6]. The usage of HBCD in the world is unknown, but the use has decreased in Sweden from 80 ton in 1998 to 6 tons in 2007 [4]. The three pairs of enantiomers, (±) α-, β- and γ-HBCD are the dominantly forms of HBCD. HBCD has a molecular formula of C12H18Br6 and a
6 structure containing six chiral carbons that can be in R or S configuration [12]. Since the spatial arrangement is different for the three enantiomers their physical and
chemical properties can vary, such as hydrophobicity and water solubility, which leads to different ability to accumulate and spread in the environment [13]. In a recent study 16 different possible HBCD has been found in theory, but the technical mixture contains the three pairs of enantiomers α-, β- and γ-HBCD and two meso forms δ- and ε-HBCD (see Figure 2) [6].
Figure 2. The three pairs of enantiomers of HBCD and the two meso forms that has been found in technical
mixture. From the top; (±) α-HBCD, (±) β-HBCD and (±) γ-HBCD. At the bottom left is δ-HBCD and at bottom right is ε-HBCD.
7 HBCD is thermally sensitive and breaks down at temperatures above 160 °C. Even though GC has been used to separate the enantiomers of HBCD, the most suitable technique is probably LC [5, 6]. HBCD can cause long-time effects in water
environments and can be toxic to water-living organisms [1]. In food chains studies, it has been found that HBCD bioaccumulate in water environments [6] and can
bioaccumulate both in terrestrial and aquatic organisms [13]. HBCD has been found in abiotic samples, such as ambient air and river sediment [6]. HBCD has also been found in polar bears from Greenland and Svalbard [6]. HBCD can cause human allergic reactions after physical contact with textiles treated with HBCD [1] and low levels of HBCD has been found in human breast milk and human blood [6]. The dominating isomer of HBCD is γ-HBCD in the technical mixture, but if the
temperature has been above 160 °C during production isomerisation of γ-HBCD to α-HBCD occurs and α-α-HBCD is the dominating isomers. This is reflected in samples and measured concentrations [6, 13].
2. Materials and methods
2.1. Extraction and cleanup for TBECH analysis
Fat from Fin whale or Winke whale from Iceland was first extracted and then going through a sequential clean-up procedure that normally is applied for PCB and PCDD/F analysis. To evaluate the extraction and clean-up procedure, three whale samples were extracted and fractionated with and without the addition of known amounts of all four TBECH-isomers. Sample 1 (un-spiked) correspond to sample 4 with the only
difference that sample 4 were spiked with all TBECH-sisomers. In the same way, sample 2 corresponds to sample 5 and sample 3 to sample 6. 5 g of homogenate (fat from whale grinded with Na2SO4 to remove water) was placed in a glass column. Before elution 25 µl internal standard (13C PCB mix with the following congeners #28, #52, #70, #101, #105, #118, #138, #153, #156, #170, #180, #194, #202 and #206 with concentrations around 120 pg/µl for each congener) (Wellington Laboratories, Guelp, Canada) was added to all samples and to the blank sample consisting of Na2SO4. Samples 4 to 6 were spiked with 50 µl α/β TBECH standard (50 pg/µl)
8 (Wellington Laboratories) and 50 µl γ/δ TBECH standard (50 pg/µl) (Wellington Laboratories). The homogenate was eluted with n-hexane: dichloromethane (1:1) and the eluate was collected in glass flasks with known weights. The organic solvents were evaporated using a rotary evaporator and the flasks were left in the fume hood until constant weights of the flasks containing whale fat were reached. The fat weights were registered before the fat was dissolved in a small volume of n-hexane and then
fractioned on a multilayer column which eliminated lipids and other polar molecules in the sample. The analytes were eluted with n-hexane into glass flasks and the solvent was evaporated to 1-3 ml on a rotary evaporator. The samples were then fractionated on an aluminium oxide column into two fractions, a non-planar PCB faction that was eluted with n-hexane: dichloromethane (49:1) and a planar dioxin fraction that was eluted with n-hexane: dichloromethane (1:1). The dioxin faction was further
fractionated on a carbon column, yielding a PCB faction (by elution with n-hexane) and a dioxin faction (by elution with toluene). The PCB fraction from the aluminium oxide column and the carbon column was collected in the same glass flasks. Before evaporation to 1 ml on a rotary evaporator, 25 µl tetradecane was added to all flasks. Then the samples were further treated on a minisilica column to remove remaining polar compounds in the extract. The PCB fraction and the dioxin fraction were transferred with n-hexane to 8 ml flasks and evaporated with nitrogen. The samples were then transferred to GC vials containing 25 µl recovery standard (13C PCB mix with #81, #114 and #178 congener with concentrations around 120 pg/µl for each) ( Wellington Laboratories). Also two quantifications standards containing internal standard (13C PCB, 120 pg/µl), recovery standard (13C PCB, 120 pg/µl), α/β TBECH standard (50 pg/µl), γ/δ TBECH standard (50 pg/µl) and tetradecane were prepared. For calculations, only the labelled PCB congeners with similar retention times as the TBECH enantiomers were used. The samples were protected from UV-light by covering them by aluminium foil throughout the whole analytical procedure.
2.2. Extraction and cleanup for HBCD analysis
The samples for HBCD analysis were fat from Fin whale or Winke whale from Iceland and were prepared identically to the TBECH extracts (see section 2.1) but by adding 25 µl internal standard δ-HBCD (5 µg/ml) (Wellington Laboratories) prior to
9 the fat extraction. No tetradecane was added to the final extracts and after transferring the extract to the 8 ml vial the extract were evaporated to dryness and then dissolved in 500 µl methanol. The samples were then transferred to LC vials containing 25 µl recovery standard (ε-HBCD, 5 µg/ml) obtained from Wellington Laboratories. Also, three quantification standards with different concentrations (25 ng/ml, 75 ng/ml and 150 ng/ml) of δ-HBCD and ε-HBCD (each dissolved in 30 % methanol and 70 % water) and a standard containing 25 µl internal standard and 25 µl recovery standard dissolved in methanol were prepared. Standards available for HBCD analysis were only δ-and ε-HBCD. Since no labelled standards were available δ-HBCD was used as internal standard and ε-HBCD was used as recovery standard. The probability to find these enantiomers in biota should be low as α-HBCD is the dominant enantiomer reported from the literature. If any of the α-, β-, γ-isomers of HBCD were found in the whale samples their areas would had been summarised and calculated with the relative response factor 1 against the internal standard (δ-HBCD).
2.3. HRGC/HRMS
A gas chromatograph (6990N Network GC, Agilent Technologies, Waldbron,
Germany) coupled to a Micromass Auto Spec-Ultima (Waters Corporation, Midford, USA) high resolution mass spectrometer was used for analysing TBECH. 1 µl of sample was injected by on-column injection on a SilGuard BPX-5 column (30m x 250 µm x 0,1 µm; SGE) equipped with a 3 m long guard column (i.d: 320 µm). As carrier gas helium was used and a temperature program was developed (see section 3.1). TBECH was ionised using electron impact in positive mode ((+) EI). The isomers of TBECH were detected by single ion monitoring (SIM) with the m/z 264.9226 [M] and 266.9207 [M+2]. The samples were quantified using isotope dilution.
2.4. UPLC/MS/MS
For HBCD analysis an Acquity TM Ultra performance LC coupled to a Quattro Premier XE triple quadrupole mass spectrometer (Waters Corporation) was used. 10 µl of sample was injected to be separated on an Acquity BEH C18 column (2.1 mm x
10 100 mm x 1.7 µm) with a flowrate of 0.125 ml/min. As mobile phase two solutions were used, methanol (B) and 30:70 % methanol: water (A). These solutions were used for a gradient. Initial composition was 40 % A and 60 % B. The composition was changed linearly in 7 minutes to 10 % A and 90 % B. After 7.1 minutes the composition was reverted to the initial setting and the system was allowed to
equilibrate for 6 minutes. The complete time for analysis was 15 minutes. The samples were ionised with negative electrospray ((-) ESI) and the cone voltage and collision energy were set to 15 V and 20 V, respectively. The capillary voltage was set to 2.80 kV. The source temperature and desolvation temperature were set to 120 °C and 300 °C. The cone gas flow was set to 50 L/Hr and the deslovation gas flow was set to 700 L/Hr. HBCD was detected with multiple reaction monitoring (MRM) measuring the transitions 640.53→78.7 and 640.53→80.8.
2.5. Detection of TBECH
For detect of the TBECH-isomers in the samples the retention times in the
quantification standard and the internal standard were compared. Also, the isotope ration between the fragment ions, 264 and 266, were compared between the
quantification standard and the spiked and unspiked samples. The limit of detection (LOD) of the method was estimated by following formula;
3. Result and discussion
3.1. TBECH method development
Method development for detection of α-, β-, γ- and δ-TBECH was performed on high resolution gas chromatography coupled to high resolution mass spectrometry
(HRGC/HRMS). First the mass spectrometer was set to measure the molecules with the m/z 264.9226 [M] and 266.9207 [M+2] by single ion monitoring (SIM). These
11 masses are fragments from the precursor molecule with the loss of two bromine
molecules. So, 264.9926 correspond to the fragment [M-HBr2] and 266.9207 correspond to the fragment [M-HBr2+2], which has the highest intensity in the mass spectra of TBECH.
In a previous study [14] different GC-columns were evaluated for best separation efficiency of the four TBECH enantiomers. The results indicated that a thin phase (0.1 μm) BPX-5 column would be best suited for the separation of TBECH enantiomers. In this study a 30 m, thin phase BPX-5 column was evaluated in combination with on-column injection. Since these compounds are thermally instable and are known to suffer from thermal interconversion between the TBECH isomers the on-column technique seems to be optimal for sample introduction of TBECH.
Different temperature programs were tested both to see which settings resulted in best resolution of the α-, β-, γ- and δ-enantiomers and also to evaluate how the different settings affected the degradation of the isomers. Also, the influence of the initial temperature on the thermal interconversion between the TBECH isomers and degradation was evaluated for three different temperatures, i.e. 100, 110 and 120°C. No improvements were seen when varying the initial temperature and the optimal temperature program giving best separation was the following; 120 °C with a hold for 2 minutes followed by a temperature ramp of 2 °C/minute up to 181 °C. Then the temperature was increased to 300 °C with the rate of 35 °C/minute and final hold for 6 minutes. In figure 3, the separation of the four TBECH enantiomers obtained with the described temperature program is shown.
12 DL09-009: std TBECH Time 26.00 27.00 28.00 29.00 30.00 31.00 32.00 33.00 34.00 35.00 % 0 100 26.00 27.00 28.00 29.00 30.00 31.00 32.00 33.00 34.00 35.00 % 0 100 26.00 27.00 28.00 29.00 30.00 31.00 32.00 33.00 34.00 35.00 % 0 100
09051806 Voltage SIR 8 Channels EI+
303.9597 1.24e8 Area 34.32 14239150 27.37 11955843 32.40 14071450 27.91 45060 28.23 26915 31.14 34953 28.71 5596 29.33 4617 31.51 5735 33.61 108821 33.87 5845 35.05 477598
09051806 Voltage SIR 8 Channels EI+
266.9207 2.03e7 Area 30.14 3101862 29.62 2428763 29.45 1460 33.32 1501142 33.15 1756660 32.87 878809 32.37 125929 31.90 83407 31.45 21414 30.54 21670 33.69 1535134.041580 34.36 5668 34.76 3314
09051806 Voltage SIR 8 Channels EI+
264.9228 1.06e7 Area 30.14 1565531 29.63 1316841 29.48 3094 27.18 1369 33.32 829467 33.15 852815 32.90 440432 32.40 87352 30.47 35482 31.122848131.595645231.98 7626 33.689770 34.05 7139 34.69 3687 34.96 2020
Figure 3. Chromatogram showing the separation of the α-, β-, γ- and δ-isomers of TBECH. Labelled PCB
congeners were used as internal and recovery standard.
3.2. Evaluation of clean-up procedure
To evaluate the existing clean-up procedure the recoveries were calculated for each sample (see Table 1). Also, to further evaluate the clean-up procedure the recovered amounts of added TBECH isomers in the spiked whale samples (sample 4 to 6) were calculated (see Table 2).The recovery in the blank sample was around 30 %. In the un-spiked whale samples the recoveries varied between 34 to 75% and in the spiked whale samples the recoveries varied between 97-138%. Acceptable boundaries for recoveries are normally between 50-120% showing problems with the clean-up procedure. Because the laboratory ran out of aluminium oxide and of long delivery-times the blank sample and the un-spiked whale samples were covered with
aluminium foil and kept in the fume-hood for almost two weeks before the clean-up could be resumed. The low recoveries in these samples could therefore be explained
α-TBECH β-TBECH γ-TBECH δ-TBECH IS (PCB # 52) RS (PCB # 81) IS (PCB # 70) Isotope ratio 2:1
13 by evaporation losses of the labelled PCB congeners. However, the high recoveries in sample 6 are still unexplained.
The recovered amounts of added amounts of TBECH isomers varied between 6.8 to 35.5%. The low recovered amounts are probably due to large evaporation losses during the clean-up procedure and possibly, to some extent, due to photolytic degradation of the TBECH isomers. The results imply that the adapted method was insufficient for the TBECH isomers and that the use of labelled TBECH isomers would be very useful to have better control of the clean-up procedure.
Table 1. Recoveries of PCB #52 and PCB #70 for all whale samples (spiked and un-spiked) and blank (Na2SO4).
Sample IS (PCB # 52) IS (PCB # 70) DL09-009:1 56.4 60.9 DL09-009:2 68.5 75.2 DL09-009:3 34.5 56.5 DL09-009:4 (Spiked) 97.4 106.2 DL09-009:5 (Spiked) 114.5 116.4 DL09-009:6 (Spiked) 137.2 138.9 DL09-009:7 (Blank) 32.4 34.1
14
Table 2. Recovered amounts of the four enantiomers of TBECH in the spiked whale samples after extraction
and clean-up. Results are presented both in concentrations (pg/g lipid) and in percentage (%).
Sample α-TBECH (pg/g lipid) β-TBECH (pg/g lipid) γ-TBECH (pg/g lipid) δ-TBECH (pg/g lipid) Added α/β TBECH standard (pg/g lipid) Added γ/δ TBECH standard (pg/g lipid) Recovered amount of added TBECH isomers (%) α β γ δ DL09-009:4 142.2 177.3 73.9 68.0 500 500 28.4 35.5 14.8 13.6 DL09-009:5 76.1 145.6 59.2 55.3 500 500 15.2 29.1 11.8 11.1 DL09-009:6 48.9 135.3 33.8 47.2 500 500 9.8 27.1 6.8 9.4
3.3. Application of the method
Three different whale samples were run to apply the evaluated clean-up procedure and the developed HRGC/HRMS method on real samples. In all of the whale samples one peak showed the same retention time as the α-TBECH isomer based on retention time comparisons with the quantification standard and the internal standards, see example in Figure 4. However, when comparing isotope ratios between the fragment ions, i.e. mass 264 and 266, the ratios differed between the signals in the quantification standard and the signals for the suspected peak in the whale samples (see Figures 4 and 5). Since the co-eluting peak did not have the correct isotope ratio the unidentified peak can not be positively identified as the α-TBECH isomer.
All four TBECH enantiomers could be identified and quantified in the spiked whale samples, see Figure 5. However, the chromatograms show large contributions of other
15 compounds having the same mass as were monitored in these samples (see Figure 4 and 5).
The LOD (limit of detection) for the method was estimated to 47.3 pg/g. In the Beluga blubber study the method detction limit (MDL) was determined to 0.8 pg/g with the signal noise ratio 3:1 [9]. The MDL in the Beluga blubber study is around 500 times lower compared to the LOD in this study. This indicates that no concentrations of the four enantiomers of TBECH were detected in the whale samples.
DL09-009: 1 Time 26.00 27.00 28.00 29.00 30.00 31.00 32.00 33.00 34.00 35.00 % 0 100 26.00 27.00 28.00 29.00 30.00 31.00 32.00 33.00 34.00 35.00 % 0 100 26.00 27.00 28.00 29.00 30.00 31.00 32.00 33.00 34.00 35.00 % 0 100
09051809 Voltage SIR 8 Channels EI+
303.9597 1.31e8 Area 34.33 11647301 27.59 5895963 27.21 684 25.53 603 32.51 7571649 28.12 12800 31.21 6184 28.70 1362 29.05 1380 29.54 1230 30.49 1064 29.87 1193 30.06 882 31.77 2600 32.17 1400 33.65 51370 33.84 5448
09051809 Voltage SIR 8 Channels EI+
266.9207 3.70e7 Area 34.68 1348850 33.69 505783 33.62;119966 32.86 144094 30.27 182105 29.17 173894 26.94 91247 25.62 21922 25.17 1519 26.16 12579 27.63 15021 27.16 3332 28.357834 27.82 3444 29.93 49419 31.39 15015131.98 70009 34.41 70483 34.18 31769 34.98 271482
09051809 Voltage SIR 8 Channels EI+
264.9228 1.24e7 Area 34.69 530078 33.69 405986 33.62 81213 32.84 135910 29.98 54562 26.94 60688 25.64 14673 26.15 11388 29.00 45614 28.37 6623 27.47 2111 27.83 1523 29.18 5995129.53 9490 30.19 60592 31.40 25137 2248732.06 32.22 5345 33.23 32689 34.41 69483 33.89 22890 34.98 77938 35.08 1667
Figure 4. Chromatogram showing a whale sample without added amounts of TBECH-isomers. Based on
retention time comparisons one peak was tentatively identified as the α enantiomer of TBECH.
α-TBECH? IS (PCB # 70) IS (PCB # 52) RS (PCB # 81) Isotope ratio 1:1
16 DL09-009: 4 Time 26.00 27.00 28.00 29.00 30.00 31.00 32.00 33.00 34.00 35.00 36.00 % 0 100 26.00 27.00 28.00 29.00 30.00 31.00 32.00 33.00 34.00 35.00 36.00 % 0 100 26.00 27.00 28.00 29.00 30.00 31.00 32.00 33.00 34.00 35.00 36.00 % 0 100
09051812 Voltage SIR 8 Channels EI+
303.9597 1.23e8 Area 34.32 9865011 32.37 10455870 27.25 7780832 31.06 32811 27.83 9943 31.60 4790 33.80 1333968 33.60 31419 32.88 17282 7568535.25 35.07 52225 35.54 3760436.09 42316 36.37 5977
09051812 Voltage SIR 8 Channels EI+
266.9207 9.75e6 Area 35.78 686562 34.51 477683 33.63 645326 30.08 865164 29.56 509190 26.82 209214 28.89 61467 26.99;33487 27.77 5885 1501928.67 33.44 84055 33.14 216159 32.48 46620 31.96 68307 31.14 48882 30.38 5276 34.17 46418 35.47 238857 34.69 200179 34.95 105278 35.95 31013
09051812 Voltage SIR 8 Channels EI+
264.9228 6.50e6 Area 33.63 407815 33.56 125250 30.09 459741 29.57 291674 27.03 132904 27.44 27445 29.24 12995 27.66 7671 28.77 13526 28.26 6480 33.44 67692 32.72 84178 32.47 32684 31.88 14040 31.23 25502 30.51 7721 35.78 391729 34.95 69453 34.68 37509 33.77 15298 35.16 25111
Figure 5. Chromatogram of sample (spiked with α/β TBECH standard and γ/δ TBECH standard) where four
peaks was possible identified as α-, β-, γ- and δ-TBECH (see Figures 7 and 8).
3.4. HBCD method development
A method for detection of α-, β-, γ-, δ- and ε-HBCD was developed using ultra performance liquid chromatography coupled to triple quadrupole mass spectrometry (UPLC/MS/MS). Detection on the mass spectrometer was optimised by tuning for the precursor ion and product ions. The bromine trace with the highest intensity [M-H+6] -(m/z 640.53) (see Figure 6) was chosen as the precursor ion, and the product ions found were 79Br (m/z 78.7) and 81Br (m/z 80.8). Also, the cone voltage was optimised for the precursor ion and the collision energy was optimised for the product ions. From these data a multiple reaction monitoring (MRM) method was developed using
negative electrospray ((-) ESI) ionisation.
α-TBECH β-TBECH γ-TBECH
δ-TBECH
IS (PCB # 52) IS (PCB # 70) RS (PCB # 81)
Isotope ratio 2:1
17
Figure 6. Full scan spectra on UPLC/(-)ESI/MS/MS showing the bromine pattern for the precursor ion [M+4]- ( m/z 640.53) with six bromine molecules.
When direct infusion of a standard solution of δ-HBCD (500 ng/ml) was performed, HBCD formed three adduct clusters, corresponding to m/z +60, +45 and +36. Tentative structures could be [M+HAc]-, [M+HCOO]- and [M+HCl]-. Since the mobile phase at first contained NH4Ac it was removed to avoid some of the adduct formation [15].
Next a method for the liquid chromatograph was created. A 50 mm column was used at first, but the retention time of δ-HBCD was around 1-2 minutes which was not sufficient since α, β and γ elutes before δ-HBCD [16]. To obtain a slower system a longer column (100 mm) was used and the retention time was increased to around 5 minutes (see Figure 7). The instruments limit of detection (LOD) for ε-HBCD was calculated to 6.8 ng/ml and for δ-HBCD 8.0 ng/ml with the signal to noise ratio 3.
18 Std. 75 dHBCD Time 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50 5.00 5.50 6.00 6.50 7.00 7.50 8.00 8.50 % 0 100 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50 5.00 5.50 6.00 6.50 7.00 7.50 8.00 8.50 % 0 100 09051111 Sm (Mn, 2x3) MRM of 2 Channels ES- TIC (dHBCD) 904 Area 5.23 65 09051107 Sm (Mn, 2x3) MRM of 2 Channels ES- TIC (dHBCD) 740 Area 4.76 62
Figure 7. Chromatogram for standard solution δ-HBCD 75 ng/ml (bottom) and ε-HBCD 75 ng/ml (top).
The mobile phase composition and the flowrate were also changed to optimise the separation. Further development showed that the chromatography was improved when adding 30 % water to the vials. The largest problem was that the LC back pressure increased to max (14 000 psi) during the run. The reason for the increased back pressure could be instrumental problem and the solution would be to change the column.
3.5. Application of the method
To evaluate the method whale samples were extracted and fractioned and analysed (see section 2.3). However, during the run the intensity suddenly decreased. One reason for the lower intensity can be that the HBCD has been debrominated during storage, i.e. it breaks down relative fast in methanol. Another reason can be that the HBCD seemed to have formed adducts with water or methanol which reduce the signals of the product ions in the mass spectrometry. Due to time limitations further studies to elucidate this problem were not possible.
ε-HBCD
19
4. Conclusions and future perspectives
A HRGC/HRMS method was developed for the separation of α-, β-, γ- and δ-TBECH. Unfortunately, no TBECH-isomers could be detected in the whale samples. The results from the evaluation of the applied extraction and clean-up procedure showed that only very low amounts of added TBECH-isomers could be recovered after the whole procedure. Tentatively, this could be a result from evaporation losses and photolytic degradation. In a future perspective, a new clean-up procedure ought to be developed to increase the recovered amounts of TBECH during extraction and clean-up.
A method was developed for separation and detection of δ- and ε-HBCD on
UPLC/MS/MS. However, no samples could be analysed due to instrumental problems. More effort needs to be directed towards the detection of HBCD, reducing adduct formation and evaluate storage stability. Moreover, standards for all enantiomers as well as labelled standards are needed for future studies.
5. Acknowledgments
Professor Bert van Bavel, PhD, Örebro University (MTM), for letting me be a part of your laboratory group.
Jessika Hagberg, PhD, Örebro University (MTM), for tutoring me during this work and with all help with the gas chromatography. Also, for the feedback on the report.
Anna Kärrman, PhD, Örebro University (MTM), for tutoring me during this work and with all help with the liquid chromatography. Also, for the feedback on the report.
20 6. References 1. Naturskyddsföreningen, http://www.naturskyddsforeningen.se/natur-och-miljo/miljogifter/liten-kemikalieordlista/organiska-miljogifter/bromerade-flamskyddsmedel/, 2009-04-04 2. Livsmedelsverket, http://www.slv.se/templates/SLV_Page.aspx?id=11490&epslanguage=SV , 2009-04-04
3. Kemikalieinspektionen, http://www.kemi.se/templates/Material____3959.aspx, KemI Rapport 4/03, 2009-04-28
4. Kemikalieinspektionen, http://www.kemi.se/templates/Page____3697.aspx, 2009-03-28
5. Alaee M, Arias P, Sjödin A and Bergman Å, An overview of commercially used brominated flame
retardants, their applications, their use patterns in different countries/regions and possible modes of release, Environment International 29 (2003), 683-689
6. Law J.R, Kohler M, Heeb V.N, Gerecke C.A, Schmid P, Voorspoels S, Covaci A, Becher G, Janák K and Thomsen C, Hexabromocyclododecane challenges scientists and regulators, Environmental science and technology, 2005, 281-287
7. Arsenault G, Lough A, Marvin C, McAlees A, McCridle R, MacInnis G, Pleskach K, Potter D, Riddell N, Sverko E, Tittlemier S and Tomy G, Structure characterization and thermal stabilities of the isomers
of the brominated flame retardant 1,2-dibromo-4-(1,2-dibromoethyl)cyclohexane, Chemosphere 72
(2008), 1163-1170
8. Wellington Laboratories, http://www.well-labs.com/pdfs/tbech.pdf, 2009-04-04
9. Tomy T.G, Pleskach K, Arsenault G, Potter D, McCrindle R, Marvin H.C, Sverko E and Tittlemier S,
Identification of the Novel Cycloaliphatic Brominated Flame Retardant
1,2-Dibromo-4-(1,2-dibromoethyl)cyclohexane in Canadian Arctic Beluga (Delphinapterus leucas), Environ. Sci. Technol.
42, 2008, 543-549
10. Larsson A, Eriksson A.L, Andersson L.P, Ivarson P and Olsson PE, Identification of the Brominated
Flame Retardant 1,2-Dibromo-4-(1,2-dibromoethyl)cyclohexane as an Androgen Agonist, J. Med.
Chem. 49, 2006, 7366-7372
11. Köppen R, Becker R, Jung C and Nehls I, On the thermally induced isomerisation of
21
12. Heeb V.N, Schweizer B.W, Mattrel P, Haag R, Grecke C. A, Kohler M, Schmid P, Zennegg M and Wolfensberger M, Solid-state conformation and absolute configuration of (+) and (-) α-, β- and
γ-hexabromocyclododecane (HBCDs), Chemosphere 68 (2007), 940-950
13. Koeppen R, Becker R, Emmerling F, Jung C and Nehls I, Enantioselective Preparative HPLC
Separation of the HBCD-Stereoisomers from the Technical Product and Their Absolute Structure Elucidation Using X-Ray Crystallography, Chirality 19 (2007), 214-222
14. Le Goff D, Analytical development of a GC-MS method for brominated flame retardant: TBECH, Diploma thesis, 2008, Örebro University
15. Riddell N, Arsenualt G, Chittim B, MacInnis G, Marvin C, McAlees A, McCrindle R and Tomy G,
Identification of the Dominant Molecular Ion Adducts Present in the LC-Ms spectra of the HBCD Diastereomers, Wellington Laboratories, 2006,
http://www.well-labs.com/pdfs/HBCD%20poster%20Setac%202006.pdf
16. Arsenault G, Konstantinov A, McAlees A, McCrindle R, Riddell N and Yeo B, Delta(δ)- and
Epsilon(ε)-1, 2, 5, 6, 9, 10-Hexabromocyclododecane, Wellington Laboratories, 2007,
