- Email: [email protected]
Improved extraction procedures for polychlorinated biphenyls in solid samples with aqueous sodium hydroxide under autoclave conditions
Chemosphere 40 (2000) 587±591
Improved extraction procedures for polychlorinated biphenyls in solid samples with aqueous sodium hydroxide under autoclave conditions Yoshio Akimoto*, Yoshio Inouye School of Pharmaceutical Sciences, Toho University, Miyama 2-2-1, Funabashi-shi, Chiba 274-8510, Japan Received 29 April 1999; accepted 30 June 1999
Abstract The ecacy of the extraction of polychlorinated biphenyls (PCBs) from varnish-in®ltrated insulating papers as a model of solid materials with an aqueous sodium hydroxide (NaOH) by autoclaving at 121°C for 30 min was compared with those for the conventional procedures, i.e., saponi®cation with 1 N ethanolic NaOH in a boiling water bath for 60 min and extraction with benzene in a Soxhlet apparatus. The newly invented autoclaving method was found to be superior to the conventional ones, yielding approximately 5- to 6-fold cumulative PCB congeners without being accompanied by extended decomposition. Therefore, the autoclave-based sample treatment is recommended for more accurate determination of PCBs in the case of PCB-impregnated solid materials such as hardened oils and resin-coated or -in®ltrated papers instead of being treated conventionally. Ó 2000 Elsevier Science Ltd. All rights reserved. Keywords: Polychlorinated biphenyls; Sample treatment; Autoclave; Solid samples; Insulating paper
1. Introduction Polychlorinated biphenyls (PCBs) were industrialized for the ®rst time in 1929 in USA and have been widely used for their insulating and ®re retardant properties as dielectric oil, machine oil, pressure sensitive copy paper, plasticizer, coating materials, etc. On the other hand, due to their possible chronical toxicities represented by the interference with bodyÕs endocrine systems, slow degradation in the environment (Jensen, 1966) and bioaccumulation in the food chain (EPA, 1987), they pose a threat to our ecosystem. Nowadays they are under strict control in terms of production, usage, storage and so on to avoid an environmental scattering.
*
Corresponding author. Tel.: +81-92-541-0161; fax: +81-92553-5698. E-mail address: [email protected] (Y. Akimoto).
Because the long-term storage has a great risk of accidental environmental diusion, several methods have been developed for the destruction of PCBs primarily in oils of transformers and capacitors (Mille, 1982; Horing and Masters, 1984; Barton and Mordy, 1984; GRA&I, 1985; Boyd et al., 1986; Staszak and Malinowski, 1987; Tessitore et al., 1987; Mincher et al., 1992; De Filippis, et al., 1997). Unlike oils, the vessels and solid contents contaminated with PCBs are either re-used or dumped after appropriate measures are taken, for which the analysis of PCBs in these items should be as accurate as possible to avoid an underestimation of PCB content resulting in unsuitable handlings. In the present study, we invented a new method for the extraction of PCBs on an autoclave dissolution of solid samples in an aqueous sodium hydroxide (NaOH). The newly introduced method proved to be more effective for the extraction of PCBs from varnish-in®ltrated insulating paper of a transformer in which PCBs
0045-6535/00/$ - see front matter Ó 2000 Elsevier Science Ltd. All rights reserved. PII: S 0 0 4 5 - 6 5 3 5 ( 9 9 ) 0 0 2 9 0 - 8
588
Y. Akimoto, Y. Inouye / Chemosphere 40 (2000) 587±591
were used as dielectric oil, when compared with the conventional procedures.
equipped with a 63 Ni electron capture detector (GC± ECD) and an SPB-5 capillary column (Supelco, USA, 0.32 mm i.d. ´ 30 m, 0.25 lm ®lm) were used.
2. Experimental
2.3. Analytical conditions
2.1. Reagents and materials
For the analysis of PCBs by GC±ECD, 2 ll of sample solution was loaded by splitless injection on the capillary column. The injection port temperature was 250°C and the detector temperature was 280°C. Column temperature conditions were as follows: 100°C for 2 min isothermal, 15°C/min to 200°C, 3°C/min to 280°C and isothermal at 280°C for the rest of run time. Helium carrier gas ¯ow rate was 2 ml/min, and nitrogen makeup gas was ¯owed at 50 ml/min. The qualitative analysis of the standard PCB contents in a sample solution was achieved by comparing the retention times of GC peaks with those obtained with the PCB standard solution, and the cumulative peak areas normalized to that of an internal standard P5CBz were used for the quantitative proposes.
Hexane, benzene, ethanol and anhydrous sodium sulfate (Na2 SO4 ) of pesticide analysis grade were purchased from WAKO Pure Chemical, Japan. Kanechlor (KC) 300, 400, 500 and 600 and pentachlorobenzene (P5CBz) were purchased from GL Sciences, Japan. Silica gel (Silica gel 60 extra pure, 70±230 mesh) was obtained from E. Merck and activated at 140°C for 14 h before use. Nitrogen and helium gases with respective purities of over 99.99999% and over 99.998% were obtained from Nihon Sanso, Japan. The other chemicals of reagent grade were obtained from WAKO. Distilled water was used after being deionized and then extracted twice with hexane. The mixture of KC 300, 400, 500 and 600 at each ®nal concentration of 2 lg/ml (PCB standard solution) and the 100 ng/ml P5CBz solution were prepared in ethanol. Varnish-in®ltrated insulating paper of a transformer in which PCBs were used as dielectric oil was obtained from Central Research Institute of Electric Power Industry, Japan. The insulating paper was cut into 5 ´ 5 mm pieces. 2.2. Apparatus Wide-mouth Pyrex bottles (50 ml) with heat-stable Te¯on-lined caps were used. Kuderna±Danish (KD) concentrator was used for the concentration of organic solvents. Autoclave (HA24, Hirayama Manufacturing Co.) and gas chromatograph (GC-14B, Shimadzu)
2.4. Conditions for the autoclave treatment Half ml of the PCB standard solution containing each 1 lg of KC 300, 400, 500 and 600 was diluted with 50 ml of 0, 0.5, 1 or 2 N aqueous NaOH in a wide-mouth bottle tightly capped and autoclaved at 121°C for 30, 60 or 90 min. After being allowed to stand at room temperature and spiked with 200 ll of P5CBz solution, the autoclaved solution was transferred into a 200 mlseparatory funnel. The bottle was rinsed twice with 5 ml of ethanol and then 10 ml of hexane, and the rinsing solutions and 50 ml of 10% aqueous sodium chloride were combined with the original extract in the separatory funel. After being shaken, the hexane layer was collected, and the water layer was re-extracted with 10 ml of hexane. The hexane extracts were combined, dried
Fig. 1. GC±ECD chromatogram of PCBs standard solution spiked with P5CBz.
Y. Akimoto, Y. Inouye / Chemosphere 40 (2000) 587±591
over Na2 SO4 and concentrated to approximately 0.5 ml using KD concentrator and nitrogen gas ¯ow. The concentrate was loaded onto a silica gel column (10 ´ 1 cm i.d. equilibrated with hexane) and the column was eluted with 100 ml of hexane. After the addition of 100 ll of octane to prevent exsiccation of the solvent, the elute was concentrated to about 200 ll using KD concentrator and nitrogen gas ¯ow.
589
poured. The bottle was capped tightly and autoclaved at 121°C for 30 min. Hexane extract was obtained by the procedures described in Section 2.4. 2.5.2. Alkaline ethanol treatment (The Pharmaceutical Society of Japan, 1996.) The insulating paper was placed in a 100 ml-Erlenmeyer ¯ask equipped with a condenser and immersed in 30 ml of 1 N ethanolic NaOH solution. After being placed in a boiling water bath for 2 h, the whole was allowed to stand at room temperature and spiked with 200 ll of P5CBz solution. The mixture was transferred into a 300 ml-separatory funnel. The ¯ask was rinsed twice with 5 ml of ethanol and then 10 ml of hexane, and the rinsing solutions and 200 ml of water were added into the separatory funnel. After being shaken, the hexane layer was collected and the water layer was reextracted with 10 ml of hexane. The hexane extracts were combined.
2.5. Extraction of PCBS from solid materials Approximately 250 mg of varnish-in®ltrated insulating paper was treated with the following three extraction procedures. The individual extracts were analyzed by GC±ECD as in the case of the PCB standard solution. 2.5.1. Autoclave treatment The insulating paper was placed in the wide-mouth bottle, into which 50 ml of 1 N aqueous NaOH was
Table 1 Eects of alkaline concentrations and heating period on total recovery of PCBsa Alkali concentration
Heating period 30 min
Water 0.5 N NaOH 1 N NaOH 2 N NaOH a
60 min
90 min
Recovery (%)
SD (%)
Recovery (%)
SD (%)
Recovery (%)
SD (%)
98.4 103.0 101.7 95.0
4.3 3.6 2.3 7.6
97.3 97.2 96.8 93.3
1.8 0.6 1.5 3.0
98.2 96.8 94.5 94.8
1.1 2.6 1.4 4.5
Each value represents mean s and standard deviation (SD) in three experiments.
Table 2 In¯uences of alkali concentration and heating period on recovery of individual PCB congenersa Peak no.b
RTc (min)
Heating peroid 30 min
3 10 14 20 23 25 29 31 35 39 40 a
10.92 12.58 13.96 15.39 16.30 17.19 18.30 19.03 20.07 21.30 22.68
60 min
90 min
30 min
60 min
With 0.5 N NaOH (%)
With 1 N NaOH (%)
108.9 93.5 111.4 102.5 98.5 99.9 96.0 107.8 114.4 103.5 107.8
97.1 101.2 110.3 104.3 103.6 98.0 95.7 105.0 112.6 99.6 102.0
93.8 103.2 100.3 95.7 98.2 91.5 99.7 88.6 91.0 95.7 98.3
103.2 102.7 98.9 105.2 94.8 89.0 93.9 90.1 98.6 88.0 95.5
98.5 108.4 105.3 100.5 92.6 95.6 104.7 87.8 95.6 92.5 95.7
Each value represents mean in three experiments. Selected peaks are numerically shown on the GC±ECD chromatogram (Fig. 1). c Retention times on the GC±ECD chromatogram. b
90 min
30 min
60 min
90 min
With 2 N NaOH (%) 107.3 103.4 99.4 101.3 94.2 88.9 83.9 88.1 93.0 89.6 86.4
97.7 105.9 105.2 94.9 96.6 93.2 107.5 97.7 104.4 99.2 98.9
93.2 90.7 95.8 103.5 93.4 96.0 96.4 105.3 97.0 87.8 96.6
86.2 92.2 93.2 95.1 90.2 92.6 85.4 90.7 96.5 81.0 84.7
590
Y. Akimoto, Y. Inouye / Chemosphere 40 (2000) 587±591
Table 3 Determination of PCB congeners in insulating paper after autoclave treatment, alkaline ethanol treatment and Soxhlet extraction Peak no.a
RTb (min)
Autoclave treatment 3
IS
d
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 Total Relative ratio (%)e a
Alkaline ethanol treatment
RV
Area (´10 )
RV
Area (´103 )
RVc (´103 )
9.33
36
1.00
53
1.00
69
1.00
9.98 10.36 10.92 11.15 11.33 11.72 11.83 11.93 12.34 12.58 12.82 13.32 13.43 13.96 14.15 14.33 14.88 15.04 15.18 15.39 15.83 16.00 16.30 16.95 17.19 17.33 17.70 18.09 18.30 18.68 19.03 19.25 19.48 19.73 20.07 20.29 20.50 21.00 21.30 22.68 23.13 23.38 24.83 25.82 27.73
25 18 81 43 122 62 59 54 56 116 13 75 93 67 70 53 24 57 30 102 36 19 144 18 104 79 35 15 134 18 53 32 38 8 42 21 42 10 154 68 28 38 14 38 18
17 4 17 12 8 0 0 10 11 26 5 12 36 29 25 9 7 22 8 38 11 7 84 3 30 28 14 16 55 6 11 7 15 0 23 8 20 0 34 17 21 9 9 14 6
0.25 0.06 0.25 0.17 0.12 0.00 0.00 0.14 0.16 0.38 0.07 0.17 0.52 0.42 0.36 0.13 0.10 0.32 0.12 0.55 0.16 0.10 1.22 0.04 0.43 0.41 0.20 0.23 0.80 0.09 0.16 0.10 0.22 0.00 0.33 0.12 0.29 0.00 0.49 0.25 0.30 0.13 0.13 0.20 0.09
68.39 100
0.69 0.50 2.25 1.19 3.39 1.72 1.64 1.50 1.56 3.22 0.36 2.08 2.58 1.86 1.94 1.47 0.67 1.58 0.83 2.83 1.00 0.53 4.00 0.50 2.89 2.19 0.97 0.42 3.72 0.50 1.47 0.89 1.06 0.22 1.17 0.58 1.17 0.28 4.28 1.89 0.78 1.06 0.39 1.06 0.50 784
3
18 3 25 12 38 0 0 14 14 35 5 15 44 32 31 21 8 23 9 40 12 7 58 5 30 30 12 14 45 8 14 7 14 0 21 5 12 0 29 25 17 9 3 8 22 14.98 21.9
Peak number shown on the chromatogram. Retention time on the chromatogram. c Peak area ratio of each PCB congener to P5CBz. d Internal standard P5CBz. e Percentage of cumulative peak area rations to that for the autoclave treatment. b
c
Soxhlet extraction
Area (´10 )
2462
c
0.34 0.06 0.47 0.23 0.72 0.00 0.00 0.26 0.26 0.66 0.09 0.28 0.83 0.60 0.58 0.40 0.15 0.43 0.17 0.75 0.23 0.13 1.09 0.09 0.57 0.57 0.23 0.26 0.85 0.15 0.26 0.13 0.26 0.00 0.40 0.09 0.23 0.00 0.55 0.47 0.32 0.17 0.06 0.15 0.42 744
10.78 15.8
Y. Akimoto, Y. Inouye / Chemosphere 40 (2000) 587±591
2.5.3. Soxhlet extraction The insulating paper was extracted using a Soxhlet extractor with 80 ml of benzene for 24 h. The 200 ll of P5CBz solution was spiked and the benzene extract was washed twice with 50 ml of water. 3. Results and discussion 3.1. Optimization of conditions for autoclave treatment Fig. 1 shows a chromatogram of the PCB standard solution spiked with P5CBz by GC±ECD and numerical peaks were those selected for estimation of stabilities during the preceding sample treatment. The total recovery of PCBs for the autoclave treatment in water ranged from 97.3% to 98.4% depending on the heating period as shown in Table 1. This indicates that PCBs remained almost completely intact during the autoclave treatment. Under the alkaline conditions, however, the total recovery of PCBs slightly decreased as the NaOH concentration increased or the heating period elongated. To clarify the reason for the decrease in the total PCB recovery, some of the PCB congeners corresponding to the numerical peaks in Fig. 1 were analyzed individually and the results are shown in Table 2. The recoveries of PCB congeners with rather short retention times such as peak 4, 10, 20 and 23 were satisfactory under any conditions. In contrast, the decrease in the recovery was notable for the peaks with much longer retention times as the NaOH concentration increased or the heating period became longer. Since the congeners giving longer retention times were regarded to be highly chlorinated, it was considered that the decrease in the total PCB recovery would result from the instability of a part of highly chlorinated congeners under the conditions of high NaOH concentration and lasted heating at 121°C. Consequently, it was proposed that samples for PCBs analysis should be treated with 1 N and lower NaOH for 60 min and shorter in an autoclave. 3.2. Ecacy of autoclave treatment for extraction of PCBS from solid material The solvent colored brown after autoclave treatment. A large portion of the varnish was broken away from the foundation paper and suspended as soft swollen matter. The solvent after the alkaline ethanol treatment colored pale brown. The varnish was sparingly broken away and a few particles was suspended. In the case of Soxhlet extraction, the extract was almost colorless and the insulating paper was slightly whitened without being broken away. Based on these, the autoclave treatment would have the highest potency to dissolve coating and in®ltrating resins.
591
Table 3 shows the relative peak areas of PCB congeners normalized to that of P5CBz and the cumulative peak areas after the conventional treatments were shown as the percentages of that of the autoclave treatment. Four PCB congeners corresponding to peaks 6, 7, 34 and 38 in Fig. 1 were detected only after autoclaving. The sum of peak area ratios of PCB congeners to that of an internal standard P5CBz was largest in the case of autoclaving; It was 4.6 and 6.3 times of those for the alkaline ethanol saponi®cation and Soxhlet extraction, respectively. It was proven in this paper that the extent of recovery of PCB congeners from the solid materials was highly dependent on the extraction conditions, showing the conventional methods were not sucient for this purpose. We recommend that the newly invented method employing an autoclave is more suitable for the estimation of PCB contents in the solidi®ed or hardened materials than those conventionally used.
References Barton, T.G., Mordy, J.A., 1984. The destruction of halogenated organic chemicals by plasma pyrolysis. Can. J. Pharmacol. 62, 978±978. Boyd, J.W., Schoe®eld, W.R., Schramm, D.E., 1986. Mobile, electric reactor detoxi®es PCBS. Hazar. Mater. Waste Manag. Magaz. 4, 32±36. De Filippis, P., Chianese, A., Pochetti, F., 1997. Removal of PCBs from mineral oils. Chemosphere 35, 1659±1667. EPA, 1987. Development of advisory levels for polychlorinated biphenyl (PCBs) cleanup project summary. US Environmental Protection Agency, EPA/600/S6-86/002. GRA&I, 1985. Experimental process for destruction of hazardous waste using solar energy. Final Technical Report, Phase 1. Govt. Reports Announcements & Index (GRA&I), Issue 12. Horing, A.W., Masters, H., 1984. Destruction of PCB-contaminated soils with a high-temperature ¯uid-wall (HTFW) reactor. Govt. Reports Announcements & Index (GRA&I), Issue 11. Jensen, S., 1966. Report of a new chemical hazard. New Scientist. 32, 612. Mille, G.J., 1982. Chemical decomposition of PCBS: the acurex process, 1981 PCB Seminar; Electric Power Research Institute, Dallas, TX. Mincher, B.J., Meikrantz, D.H., Arbon, R.E., Murphy, R.J., 1992. High energy decomposition of halogenated hydrocarbons. Final Report, FY 91, Govt Reports Announcements & Index (GRA&I), Issue 11. Staszak, C.N., Malinowski, K.C., 1987. The destruction of hazardous organic waste materials using the modar oxidation process. Second International Conference on New Frontiers for Hazardous Waste Management. pp. 97±106. Tessitore, J.L., Cross, F., Munoz, J.H., 1987. PCB destruction by incineration. Pollution Engineering 19, 70±75. The Pharmaceutical Society of Japan, 1996. Standard Methods of Analysis for Hygiene Chemists 1990. Kanehara, Tokyo (Chapter 2.4.2.2).
Improved extraction procedures for polychlorinated biphenyls in solid samples with aqueous sodium hydroxide under autoclave conditions Yoshio Akimoto*, Yoshio Inouye School of Pharmaceutical Sciences, Toho University, Miyama 2-2-1, Funabashi-shi, Chiba 274-8510, Japan Received 29 April 1999; accepted 30 June 1999
Abstract The ecacy of the extraction of polychlorinated biphenyls (PCBs) from varnish-in®ltrated insulating papers as a model of solid materials with an aqueous sodium hydroxide (NaOH) by autoclaving at 121°C for 30 min was compared with those for the conventional procedures, i.e., saponi®cation with 1 N ethanolic NaOH in a boiling water bath for 60 min and extraction with benzene in a Soxhlet apparatus. The newly invented autoclaving method was found to be superior to the conventional ones, yielding approximately 5- to 6-fold cumulative PCB congeners without being accompanied by extended decomposition. Therefore, the autoclave-based sample treatment is recommended for more accurate determination of PCBs in the case of PCB-impregnated solid materials such as hardened oils and resin-coated or -in®ltrated papers instead of being treated conventionally. Ó 2000 Elsevier Science Ltd. All rights reserved. Keywords: Polychlorinated biphenyls; Sample treatment; Autoclave; Solid samples; Insulating paper
1. Introduction Polychlorinated biphenyls (PCBs) were industrialized for the ®rst time in 1929 in USA and have been widely used for their insulating and ®re retardant properties as dielectric oil, machine oil, pressure sensitive copy paper, plasticizer, coating materials, etc. On the other hand, due to their possible chronical toxicities represented by the interference with bodyÕs endocrine systems, slow degradation in the environment (Jensen, 1966) and bioaccumulation in the food chain (EPA, 1987), they pose a threat to our ecosystem. Nowadays they are under strict control in terms of production, usage, storage and so on to avoid an environmental scattering.
*
Corresponding author. Tel.: +81-92-541-0161; fax: +81-92553-5698. E-mail address: [email protected] (Y. Akimoto).
Because the long-term storage has a great risk of accidental environmental diusion, several methods have been developed for the destruction of PCBs primarily in oils of transformers and capacitors (Mille, 1982; Horing and Masters, 1984; Barton and Mordy, 1984; GRA&I, 1985; Boyd et al., 1986; Staszak and Malinowski, 1987; Tessitore et al., 1987; Mincher et al., 1992; De Filippis, et al., 1997). Unlike oils, the vessels and solid contents contaminated with PCBs are either re-used or dumped after appropriate measures are taken, for which the analysis of PCBs in these items should be as accurate as possible to avoid an underestimation of PCB content resulting in unsuitable handlings. In the present study, we invented a new method for the extraction of PCBs on an autoclave dissolution of solid samples in an aqueous sodium hydroxide (NaOH). The newly introduced method proved to be more effective for the extraction of PCBs from varnish-in®ltrated insulating paper of a transformer in which PCBs
0045-6535/00/$ - see front matter Ó 2000 Elsevier Science Ltd. All rights reserved. PII: S 0 0 4 5 - 6 5 3 5 ( 9 9 ) 0 0 2 9 0 - 8
588
Y. Akimoto, Y. Inouye / Chemosphere 40 (2000) 587±591
were used as dielectric oil, when compared with the conventional procedures.
equipped with a 63 Ni electron capture detector (GC± ECD) and an SPB-5 capillary column (Supelco, USA, 0.32 mm i.d. ´ 30 m, 0.25 lm ®lm) were used.
2. Experimental
2.3. Analytical conditions
2.1. Reagents and materials
For the analysis of PCBs by GC±ECD, 2 ll of sample solution was loaded by splitless injection on the capillary column. The injection port temperature was 250°C and the detector temperature was 280°C. Column temperature conditions were as follows: 100°C for 2 min isothermal, 15°C/min to 200°C, 3°C/min to 280°C and isothermal at 280°C for the rest of run time. Helium carrier gas ¯ow rate was 2 ml/min, and nitrogen makeup gas was ¯owed at 50 ml/min. The qualitative analysis of the standard PCB contents in a sample solution was achieved by comparing the retention times of GC peaks with those obtained with the PCB standard solution, and the cumulative peak areas normalized to that of an internal standard P5CBz were used for the quantitative proposes.
Hexane, benzene, ethanol and anhydrous sodium sulfate (Na2 SO4 ) of pesticide analysis grade were purchased from WAKO Pure Chemical, Japan. Kanechlor (KC) 300, 400, 500 and 600 and pentachlorobenzene (P5CBz) were purchased from GL Sciences, Japan. Silica gel (Silica gel 60 extra pure, 70±230 mesh) was obtained from E. Merck and activated at 140°C for 14 h before use. Nitrogen and helium gases with respective purities of over 99.99999% and over 99.998% were obtained from Nihon Sanso, Japan. The other chemicals of reagent grade were obtained from WAKO. Distilled water was used after being deionized and then extracted twice with hexane. The mixture of KC 300, 400, 500 and 600 at each ®nal concentration of 2 lg/ml (PCB standard solution) and the 100 ng/ml P5CBz solution were prepared in ethanol. Varnish-in®ltrated insulating paper of a transformer in which PCBs were used as dielectric oil was obtained from Central Research Institute of Electric Power Industry, Japan. The insulating paper was cut into 5 ´ 5 mm pieces. 2.2. Apparatus Wide-mouth Pyrex bottles (50 ml) with heat-stable Te¯on-lined caps were used. Kuderna±Danish (KD) concentrator was used for the concentration of organic solvents. Autoclave (HA24, Hirayama Manufacturing Co.) and gas chromatograph (GC-14B, Shimadzu)
2.4. Conditions for the autoclave treatment Half ml of the PCB standard solution containing each 1 lg of KC 300, 400, 500 and 600 was diluted with 50 ml of 0, 0.5, 1 or 2 N aqueous NaOH in a wide-mouth bottle tightly capped and autoclaved at 121°C for 30, 60 or 90 min. After being allowed to stand at room temperature and spiked with 200 ll of P5CBz solution, the autoclaved solution was transferred into a 200 mlseparatory funnel. The bottle was rinsed twice with 5 ml of ethanol and then 10 ml of hexane, and the rinsing solutions and 50 ml of 10% aqueous sodium chloride were combined with the original extract in the separatory funel. After being shaken, the hexane layer was collected, and the water layer was re-extracted with 10 ml of hexane. The hexane extracts were combined, dried
Fig. 1. GC±ECD chromatogram of PCBs standard solution spiked with P5CBz.
Y. Akimoto, Y. Inouye / Chemosphere 40 (2000) 587±591
over Na2 SO4 and concentrated to approximately 0.5 ml using KD concentrator and nitrogen gas ¯ow. The concentrate was loaded onto a silica gel column (10 ´ 1 cm i.d. equilibrated with hexane) and the column was eluted with 100 ml of hexane. After the addition of 100 ll of octane to prevent exsiccation of the solvent, the elute was concentrated to about 200 ll using KD concentrator and nitrogen gas ¯ow.
589
poured. The bottle was capped tightly and autoclaved at 121°C for 30 min. Hexane extract was obtained by the procedures described in Section 2.4. 2.5.2. Alkaline ethanol treatment (The Pharmaceutical Society of Japan, 1996.) The insulating paper was placed in a 100 ml-Erlenmeyer ¯ask equipped with a condenser and immersed in 30 ml of 1 N ethanolic NaOH solution. After being placed in a boiling water bath for 2 h, the whole was allowed to stand at room temperature and spiked with 200 ll of P5CBz solution. The mixture was transferred into a 300 ml-separatory funnel. The ¯ask was rinsed twice with 5 ml of ethanol and then 10 ml of hexane, and the rinsing solutions and 200 ml of water were added into the separatory funnel. After being shaken, the hexane layer was collected and the water layer was reextracted with 10 ml of hexane. The hexane extracts were combined.
2.5. Extraction of PCBS from solid materials Approximately 250 mg of varnish-in®ltrated insulating paper was treated with the following three extraction procedures. The individual extracts were analyzed by GC±ECD as in the case of the PCB standard solution. 2.5.1. Autoclave treatment The insulating paper was placed in the wide-mouth bottle, into which 50 ml of 1 N aqueous NaOH was
Table 1 Eects of alkaline concentrations and heating period on total recovery of PCBsa Alkali concentration
Heating period 30 min
Water 0.5 N NaOH 1 N NaOH 2 N NaOH a
60 min
90 min
Recovery (%)
SD (%)
Recovery (%)
SD (%)
Recovery (%)
SD (%)
98.4 103.0 101.7 95.0
4.3 3.6 2.3 7.6
97.3 97.2 96.8 93.3
1.8 0.6 1.5 3.0
98.2 96.8 94.5 94.8
1.1 2.6 1.4 4.5
Each value represents mean s and standard deviation (SD) in three experiments.
Table 2 In¯uences of alkali concentration and heating period on recovery of individual PCB congenersa Peak no.b
RTc (min)
Heating peroid 30 min
3 10 14 20 23 25 29 31 35 39 40 a
10.92 12.58 13.96 15.39 16.30 17.19 18.30 19.03 20.07 21.30 22.68
60 min
90 min
30 min
60 min
With 0.5 N NaOH (%)
With 1 N NaOH (%)
108.9 93.5 111.4 102.5 98.5 99.9 96.0 107.8 114.4 103.5 107.8
97.1 101.2 110.3 104.3 103.6 98.0 95.7 105.0 112.6 99.6 102.0
93.8 103.2 100.3 95.7 98.2 91.5 99.7 88.6 91.0 95.7 98.3
103.2 102.7 98.9 105.2 94.8 89.0 93.9 90.1 98.6 88.0 95.5
98.5 108.4 105.3 100.5 92.6 95.6 104.7 87.8 95.6 92.5 95.7
Each value represents mean in three experiments. Selected peaks are numerically shown on the GC±ECD chromatogram (Fig. 1). c Retention times on the GC±ECD chromatogram. b
90 min
30 min
60 min
90 min
With 2 N NaOH (%) 107.3 103.4 99.4 101.3 94.2 88.9 83.9 88.1 93.0 89.6 86.4
97.7 105.9 105.2 94.9 96.6 93.2 107.5 97.7 104.4 99.2 98.9
93.2 90.7 95.8 103.5 93.4 96.0 96.4 105.3 97.0 87.8 96.6
86.2 92.2 93.2 95.1 90.2 92.6 85.4 90.7 96.5 81.0 84.7
590
Y. Akimoto, Y. Inouye / Chemosphere 40 (2000) 587±591
Table 3 Determination of PCB congeners in insulating paper after autoclave treatment, alkaline ethanol treatment and Soxhlet extraction Peak no.a
RTb (min)
Autoclave treatment 3
IS
d
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 Total Relative ratio (%)e a
Alkaline ethanol treatment
RV
Area (´10 )
RV
Area (´103 )
RVc (´103 )
9.33
36
1.00
53
1.00
69
1.00
9.98 10.36 10.92 11.15 11.33 11.72 11.83 11.93 12.34 12.58 12.82 13.32 13.43 13.96 14.15 14.33 14.88 15.04 15.18 15.39 15.83 16.00 16.30 16.95 17.19 17.33 17.70 18.09 18.30 18.68 19.03 19.25 19.48 19.73 20.07 20.29 20.50 21.00 21.30 22.68 23.13 23.38 24.83 25.82 27.73
25 18 81 43 122 62 59 54 56 116 13 75 93 67 70 53 24 57 30 102 36 19 144 18 104 79 35 15 134 18 53 32 38 8 42 21 42 10 154 68 28 38 14 38 18
17 4 17 12 8 0 0 10 11 26 5 12 36 29 25 9 7 22 8 38 11 7 84 3 30 28 14 16 55 6 11 7 15 0 23 8 20 0 34 17 21 9 9 14 6
0.25 0.06 0.25 0.17 0.12 0.00 0.00 0.14 0.16 0.38 0.07 0.17 0.52 0.42 0.36 0.13 0.10 0.32 0.12 0.55 0.16 0.10 1.22 0.04 0.43 0.41 0.20 0.23 0.80 0.09 0.16 0.10 0.22 0.00 0.33 0.12 0.29 0.00 0.49 0.25 0.30 0.13 0.13 0.20 0.09
68.39 100
0.69 0.50 2.25 1.19 3.39 1.72 1.64 1.50 1.56 3.22 0.36 2.08 2.58 1.86 1.94 1.47 0.67 1.58 0.83 2.83 1.00 0.53 4.00 0.50 2.89 2.19 0.97 0.42 3.72 0.50 1.47 0.89 1.06 0.22 1.17 0.58 1.17 0.28 4.28 1.89 0.78 1.06 0.39 1.06 0.50 784
3
18 3 25 12 38 0 0 14 14 35 5 15 44 32 31 21 8 23 9 40 12 7 58 5 30 30 12 14 45 8 14 7 14 0 21 5 12 0 29 25 17 9 3 8 22 14.98 21.9
Peak number shown on the chromatogram. Retention time on the chromatogram. c Peak area ratio of each PCB congener to P5CBz. d Internal standard P5CBz. e Percentage of cumulative peak area rations to that for the autoclave treatment. b
c
Soxhlet extraction
Area (´10 )
2462
c
0.34 0.06 0.47 0.23 0.72 0.00 0.00 0.26 0.26 0.66 0.09 0.28 0.83 0.60 0.58 0.40 0.15 0.43 0.17 0.75 0.23 0.13 1.09 0.09 0.57 0.57 0.23 0.26 0.85 0.15 0.26 0.13 0.26 0.00 0.40 0.09 0.23 0.00 0.55 0.47 0.32 0.17 0.06 0.15 0.42 744
10.78 15.8
Y. Akimoto, Y. Inouye / Chemosphere 40 (2000) 587±591
2.5.3. Soxhlet extraction The insulating paper was extracted using a Soxhlet extractor with 80 ml of benzene for 24 h. The 200 ll of P5CBz solution was spiked and the benzene extract was washed twice with 50 ml of water. 3. Results and discussion 3.1. Optimization of conditions for autoclave treatment Fig. 1 shows a chromatogram of the PCB standard solution spiked with P5CBz by GC±ECD and numerical peaks were those selected for estimation of stabilities during the preceding sample treatment. The total recovery of PCBs for the autoclave treatment in water ranged from 97.3% to 98.4% depending on the heating period as shown in Table 1. This indicates that PCBs remained almost completely intact during the autoclave treatment. Under the alkaline conditions, however, the total recovery of PCBs slightly decreased as the NaOH concentration increased or the heating period elongated. To clarify the reason for the decrease in the total PCB recovery, some of the PCB congeners corresponding to the numerical peaks in Fig. 1 were analyzed individually and the results are shown in Table 2. The recoveries of PCB congeners with rather short retention times such as peak 4, 10, 20 and 23 were satisfactory under any conditions. In contrast, the decrease in the recovery was notable for the peaks with much longer retention times as the NaOH concentration increased or the heating period became longer. Since the congeners giving longer retention times were regarded to be highly chlorinated, it was considered that the decrease in the total PCB recovery would result from the instability of a part of highly chlorinated congeners under the conditions of high NaOH concentration and lasted heating at 121°C. Consequently, it was proposed that samples for PCBs analysis should be treated with 1 N and lower NaOH for 60 min and shorter in an autoclave. 3.2. Ecacy of autoclave treatment for extraction of PCBS from solid material The solvent colored brown after autoclave treatment. A large portion of the varnish was broken away from the foundation paper and suspended as soft swollen matter. The solvent after the alkaline ethanol treatment colored pale brown. The varnish was sparingly broken away and a few particles was suspended. In the case of Soxhlet extraction, the extract was almost colorless and the insulating paper was slightly whitened without being broken away. Based on these, the autoclave treatment would have the highest potency to dissolve coating and in®ltrating resins.
591
Table 3 shows the relative peak areas of PCB congeners normalized to that of P5CBz and the cumulative peak areas after the conventional treatments were shown as the percentages of that of the autoclave treatment. Four PCB congeners corresponding to peaks 6, 7, 34 and 38 in Fig. 1 were detected only after autoclaving. The sum of peak area ratios of PCB congeners to that of an internal standard P5CBz was largest in the case of autoclaving; It was 4.6 and 6.3 times of those for the alkaline ethanol saponi®cation and Soxhlet extraction, respectively. It was proven in this paper that the extent of recovery of PCB congeners from the solid materials was highly dependent on the extraction conditions, showing the conventional methods were not sucient for this purpose. We recommend that the newly invented method employing an autoclave is more suitable for the estimation of PCB contents in the solidi®ed or hardened materials than those conventionally used.
References Barton, T.G., Mordy, J.A., 1984. The destruction of halogenated organic chemicals by plasma pyrolysis. Can. J. Pharmacol. 62, 978±978. Boyd, J.W., Schoe®eld, W.R., Schramm, D.E., 1986. Mobile, electric reactor detoxi®es PCBS. Hazar. Mater. Waste Manag. Magaz. 4, 32±36. De Filippis, P., Chianese, A., Pochetti, F., 1997. Removal of PCBs from mineral oils. Chemosphere 35, 1659±1667. EPA, 1987. Development of advisory levels for polychlorinated biphenyl (PCBs) cleanup project summary. US Environmental Protection Agency, EPA/600/S6-86/002. GRA&I, 1985. Experimental process for destruction of hazardous waste using solar energy. Final Technical Report, Phase 1. Govt. Reports Announcements & Index (GRA&I), Issue 12. Horing, A.W., Masters, H., 1984. Destruction of PCB-contaminated soils with a high-temperature ¯uid-wall (HTFW) reactor. Govt. Reports Announcements & Index (GRA&I), Issue 11. Jensen, S., 1966. Report of a new chemical hazard. New Scientist. 32, 612. Mille, G.J., 1982. Chemical decomposition of PCBS: the acurex process, 1981 PCB Seminar; Electric Power Research Institute, Dallas, TX. Mincher, B.J., Meikrantz, D.H., Arbon, R.E., Murphy, R.J., 1992. High energy decomposition of halogenated hydrocarbons. Final Report, FY 91, Govt Reports Announcements & Index (GRA&I), Issue 11. Staszak, C.N., Malinowski, K.C., 1987. The destruction of hazardous organic waste materials using the modar oxidation process. Second International Conference on New Frontiers for Hazardous Waste Management. pp. 97±106. Tessitore, J.L., Cross, F., Munoz, J.H., 1987. PCB destruction by incineration. Pollution Engineering 19, 70±75. The Pharmaceutical Society of Japan, 1996. Standard Methods of Analysis for Hygiene Chemists 1990. Kanehara, Tokyo (Chapter 2.4.2.2).