Organic micropollutants in a Norwegian water-course

Organic micropollutants in a Norwegian water-course

The Science of the Total Environment, 20 (1981) 277--286 277 Elsevier Scientific Publishing Company, Amsterdam -- Printed in The Netherlands ORGANI...

464KB Sizes 3 Downloads 24 Views

The Science of the Total Environment, 20 (1981) 277--286


Elsevier Scientific Publishing Company, Amsterdam -- Printed in The Netherlands



Department of Chemistry, NLHT, University of Trondheim, N-7055 Dragvoll (Norway) G E O R G E . CARLBERG

Central Institute for Industrial Research, P.O. Box 350, Blindern, Oslo 3 (Norway) (Received May 18th, 1981; accepted in final form July 17th, 1981)


A Norwegian water-course has been investigated for the content of organic mic~opollutants. The investigation was performed using liquid--liquid extractions and analysed by GC and GC--MS using capillary columns. Both qualitative and semi-quantitative results were obtained. The dominating components from the main downstream station were mostly n-alkanes, aromatic hydrocarbons and fatty acids. The main sources for the hydrocarbons are probably runoff from roads and atmospheric fallout, while the fatty acids originate from natural sources. Phenol and cresols together with some unidentified compounds were also present in minor amounts. These compounds probably originate from a large industrial effluent in the region.


In recent years analysis of organic micropollutants in aquatic environments have become more important, due to adverse ecological and health effects from some of these compounds. Different types of compounds are thought to have an entirely different impact on the environment. Investigations have been performed on the Nitelva water course, some 40 km long, in order to relate water quality to specific organic compounds which occur in low concentrations and originate from sources such as domestic sewage, industrial effluents and agricultural and road r u n o f f [ 1--3 ]. Methods to monitor these compounds in aquatic environments have been developed. In this investigation we demonstrate the application of one analytical m e t h o d in determining the input of organic micropollutants from one specific water-course to a lake which has been investigated since 1950 with respect to collective parameters such as phosphorous and nitrogen content, biological and chemical oxygen demand and microbiological activity [4--6]. With growing urbanization the impact on the water-courses in the region 0048-9697/81/0000--0000/$02.50 ©1981 Elsevier Scientific Publishing Company

278 investigated has greatly increased and, in 1975, a monitoring program with 35 sampling stations was established. In 1978 an improvement or stagnation was noticed compared to 1970 for 27 of the stations, due to installation of domestic and industrial :sewage plants [ 5]. Samples were obtained around the main site where the river runs into a lake, 0 y e r e n . The content of specific organic c o m p o u n d s in this sample, analysed by GC and GC--MS, is then related to other sources by selecting six other sampling sites upstream from the main one.


Sampling stations The seven sampling sites are shown on the map in Fig..1. The main site (7) was close to a small town, Lillestr~bm, where the river runs into lake 0yeren. "4%35km Kfel Perhoier~



Sagdalselv a " ~




(4) Sample no.


Sampling stations










Industrial p u r i f i c a t i o n



(7) plant


(5) after d i l u t i o n

with cooling water and river water {7)


Fig. 1. Map showing Nitelva with sampling stations. Two samples from an industrial effluent were investigated, one immediately after treatment in a biologically active sludge plant (5), and the other after dilution with cooling water ( 1 : 2 0 ) and river water (6). The other samples were taken from the river Nitelva and from two sidestreams.

Sampling Samples, 201, taken over a fortnight period in August--September 1979, were collected in acetone-washed glass bottles. A sealed sampling bottle was

279 lowered into the water, then the glass stopper removed in order to avoid sampling of the surface film. Every sample was collected over a 2--4 h time period, and the main sample (7) was collected across the river, using a rowing boat. When the samples could not be extracted immediately after the sampling, they were stored at 4°C in a dark room for a maximum of 2--3 days. This has been shown to result in no degradation of the compounds analysed [7,8].

Extraction and clean-up procedures The extractions were performed in the glass bottles by magnetic stirring for 2 h. The non-polar compounds were extracted with distilled cyclohexane (2 x 250 ml) after pH adjustment to 11 with NaOH, and the extract was then concentrated to 0.5 ml by use of a waterbath and a gentle stream of nitrogen. The chemical recovery with this procedure has been shown to be 95--100% for phthalates and 70--80% for the n-alkanes higher than C ls H32 [ 7]. To investigate the content of sulphuric acid resistant compounds, an aliquot of the cyclohexane extract was treated with equal parts of concentrated H2 SO4 which for the chlorinated compounds [ 7] provided recoveries of 95--100%. The content of compounds resistant to alkaline treatment was investigated according to the method described by Jensen [9]. The polar components were extracted after the removal of non-polar compounds with distilled n-butylacetate (350 ml) after pH adjustment to 2 with H2 SO4. Reduced pressure and a gentle stream of nitrogen was used to concentrate the extracts to 0.5ml. This procedure shows recoveries of more than 80% for the polar compounds [8].

Sample derivatisation The n-butylacetate samples were silylated to give the trimethylsilyl (TMS) derivatives of the polar components. The extracts (50#1) were treated with N,O-bis(trimethylsilyl)-trifluoroacetamide (BSTFA) (5pl) and trimethylchlorosilane (TMCS) (1 pl) by shaking and then left for one hour at 60°C.

Gas chromatography (GC) and gas chromatography--mass spectrometry ( GC--MS) analyses All the extracts were analysed by GC with a flame ionisation detector (FID), while the H2SO4- and NaOH-treated cyclohexane extracts were also analysed by GC with an electron capture detector (ECD). Semi-quantitative estimates were performed by comparison with suitable external standards. Identification of the different compounds was done by GC--MS in electron impact (EI) and chemical ionisation (CI) modes. Methane was used as an ionising gas in CI mode. The analyses were performed using glass capillary columns and 2/~1 of the samples were injected into Grob type injectors [10]. Packed columns were also used in one of the chromatographs with an EC detector. The chromatographs used were Carlo-Erba 2350 (FID), Perkin-Elmer 3920 (ECD) and Hewlett-Packard 5730 (FID + ECD).

280 Two GC--MS systems were used: (i) a Finnigan 9610 gas chromatograph interfaced to a Finnigan 4021 mass spectrometer and an Incos 2000 data system for which mass spectra were recorded with 70 eV electron energy in an EI mode; (ii) Hewlett-Packard 5985A GC--MS with HP 5840A GC and HP 21MX-E data system which was operated in both EI and CI modes with electron energies of 70 eV and 200 eV respectively.

Neutron activation analysis (NAA ) NAA was performed as described by Lunde [11] to determine the total a m o u n t of organochlorinated c o m p o u n d s in the industrial effluent sample (5).


A. Cyclohexane extractable compounds A.1. Analysis by GC and GC--MS. The main sample (7) consists of a complex mixture of non-polar components. The chromatogram is illustrated in Fig. 2 together with that of sample (1) from the "origin" of the river. The components identified are listed in Table 1. The chromatogram of sample (1) shows that the Nitelva has a certain load of organic c o m p o u n d s already at the source, and this sample is used as a background for the other stations. Aromatic hydrocarbons and n-alkanes are the main components in the background sample and are probably derived from atmospheric fallout and rain. Table 1 also gives semi-quantitative results for some components. It can be seen that the difference between samples (1) and (2) is mainly quantitative which is consistent with the fact that, between the t w o stations, there are few point sources in the area that can add new types of compounds. The high a m o u n t of hydrocarbons in sample (2) can probably be related to storm water runoff since this station is located near a road with high traffic density. One of the t w o sidestreams investigated (sampling station (4)) is the possible source for the group of unidentified c o m p o u n d s called C in Table 1. These components together with the remarkable high level of six different phthalates have n o t been related to any known source in the area. It is unlikely that they originate from the agricultural environment of the river. Samples (5) and (6) have been collected in the effluent from the largest industrial plant in the area. The c o m p a n y produces glue, paints and wax resins. The c o m p o u n d s which have been identified in the main sample, but n o t in samples (1) to (4), can all be shown to originate from this particular effluent (Table 1). The c o m p o u n d s called A and B in the table, have n o t been identified and are divided into t w o groups according to their mass spectra; the absolute levels gradually decrease from sample (5) through (6) to (7) due to influences such as sedimentation, evaporation, degradation and dilution. The influence on chemical c o m p o u n d s from these environmental effects varies from species to species. An example of this can be seen from Table 1 by comparing t w o c o m p o n e n t groups in samples (4) and (7). The















j~ ...... 20


Fig. 2. FID gas chromatograms of the cyclohexane extracts of the main sample (7) and of the b a c k ~ o u n d sample (1). Analysed on Carlo-Erba 2350; SE-54 ( 5 0 m × 0.3ram); 40 to 250°C, 3 C/rain. Carrier gas H2, 0.8 kg/cm 2. The numbers in the chromatogram refer to Table 1.

high content of phthalates in sample (4) has decreased to the background level in sample (7), while the content of the unidentified Group C remains at the same concentration. The shorter retention times for the latter components indicate that in this case the evaporation effect is not an important factor, but probably a combination of the other factors is responsible for the result. It follows from this that it may be more important to analyse the individual components in a complex mixture. In order to investigate a possible biogenic source for the n-alkanes, the CPI14-20 (Carbon Preference Index from C14H30 to C20H42)was calculated, and the results are given in Table 2. The CPI gave relatively high values for the sewage effluent, which indicates that the biologically active sludge plant

282 TABLE 1 THE C O N T E N T O F O R G A N I C M I C R O P O L L U T A N T S (pg/l water) IN C Y C L O H E X A N E EXTRACTS

Peak No. in Fig. 2


Sample No. (1)

1 2 3 4

C2 -alkylbenzenes C3 -alkylbenzenes C4-alkylbenzenes n-alkanes



-5 6 7 8 9

hydroxybiphenyl tri-n-butylphosphate phthalates Group A Group B Group C


(0.1 ~0.1 (0.1 0.2 -. (0.1 0.2 . . --

8.3 0.4 0.1 0.6 ~ . . (0.1 0.2 . . . . --

(3) (0.1 0.2 0.1 0.5 -. (0.1 1.1 . . --





0.7 0.4 0.1 0.2 --

17.3 2.0 1.8 2.6 0.6 2.2 0.9 0.3 9.3 7.9 --

4.7 0.2 0.3 0.2 <:0.1 0.2 0.1 0.2 0.3 0.2 --

0.5 0.3 0.3 0.3 -<~0.1 0.1 0.3 0.2 0.1 0.2




Sample No.
















is the biogenic source. All the other samples showed values for CPI14_20 of less than 2.0, which confirms the assumption that the hydrocarbons in these samples originate from non-biogenic sources. The two sidestreams (3) and (4) showed the lowest values, while there was a slight decrease in the values downstream the main river, confirming the increasing pollution in that direction. A.2. Analysis by GC--ECD. The content of sulphuric acid resistant components with ECD-response in the seven samples is shown in Fig. 3 and Table 3. Sample (7) consists mainly of two compounds which have been identified as ~ and ~-benzenehexachloride (BHC) using chemical reactions such as acidic and alkaline treatment [7] together with retention time comparisons. The polychlorinated biphenyls (PCB) content is of the same order of magnitude in all samples, and this is the case for the BHC-isomers as well. The main sample (7) is qualitatively somewhat more complex than the background sample, probably due to the influence of the two side rivers (3) and (4). NAA was used to determine the total amount of chlorinated organics (TOC1) in the industrial effluent. The sewage has not been chlorinated, and




i! (4) I i


iI:i !i





Fig. 3. ECD gas chromatograms of the cyclohexane extracts. Analysed on Perkin-Elmer 3920; 3% SE-30 (2m × 1/8"); 130 to 230°C, 8°C/min. Carrier gas Ar/10% CH4,30 ml/min. The numbers in the chromatograms refer to Table 2, while the numbers within parentheses respresent the sample numbers.



1 2 3 4 5

~-BHC 7-BHC unknown unknown PCB

Sample No. (1)







0.8 1.4 -~0.1 4.6

1.2 1.4 ~0.1 ~0.1 1.8

0.6 1.6 0.1 0.2 2.9

1.0 0.9 ~0.1 0.2 3.1

0.5 -27.0 52.1 4.4

1.3 2.2 0.6 1.1 2.0

2.9 4.0 0.1 0.2 5.0

the gas chromatogram of this effluent contains mainly two unidentified compounds. Semi-quantitative amounts of these compounds are calculated using the response factor of BHC, which is found in the same region in the chromatogram. These compounds together with PCB are responsible for only about 1% of the TOC1 which amounts to 12.5 pg/1. Most of the chlorine not being accounted for is probably bound to high molecular compounds which do not pass the chromatographic system or to more polar compounds.

B. n-Butylacetate extractable compounds analysed by GC--FID and GC--MS Total ion chromatograms of the TMS-derivatives of the polar compounds are compared in Fig. 4, and the identified components are listed in Table 4. The main components in sample (7) are fatty acids. From Table 5 it can be seen that all samples contain fatty acids in the same order of magnitude, and TABLE 4 ORGANIC MICROPOLLUTANTS FOUND IN n-BUTYLACETATE EXTRACTS Peak No. in Fig. 4 1 2 3 4 5 6 7 8 9 10 11


Peak No. in Fig. 4


phenol cresol unidentified (M ÷ 230 from CI mass spectra) benzoic acid unidentified (M + 244 from CI mass spectra) possibly methylated benzoic acid possibly methylated benzoic acid C9 fatty acid C10 fatty acid C12 fatty acid C13 fatty acid

12 13 14 15 16 17 18 19 20 21 22 23

C14 fatty acid Cls fatty acid Cls fatty acid Cls fatty acid C16 unsaturated fatty acid C16 fatty acid Cl7 fatty acid Cls unsaturated fatty acid Cls fatty acid C20 fatty acid phthalate C22 fatty acid


(2) ~

15) 16)

16 ~ 17

t7j I





0 (MIN)

Fig. 4. Chromatograms of the total ion current (EI mode) of the n-butylacetate extracts. Analysed on Hewlett-Packard 5985A GC--MS; SE-54 (20reX 0.35ram); 40 to 250°C, 3°C/min. Carrier gas He, 2 ml/min. The numbers in the chromatograms refer to Table 3, while the numbers within parenthesis represent the sample numbers. TABLE 5 THE CONTENT OF ORGANIC M I C R O P O L L U T A N T S ( p g / I w a ~ r ) IN n-BUTYLACETATE EXTRACTS


fatty acids phenol cresols

Sample No. (1)


9.6 0.01 .

16.4 0.02 .



21.2 0.02 .





14.6 0.03

3.1 7.9 1.0

52.5 5.8 0.6

20.4 0.07 0.1

this suggests that they originate from natural sources or from atmospheric fallout. The industrial effluent was expected to contain components from the river sample (2), since the company uses river water in the production. The reason for the low content of fatty acids in the industrial effluent is probably microbiological degradation in the active sludge plant. Other compounds that can be identified in the main sample are phenol and isomer of cresol. The content of these compounds increased compared


to the background sample (1), and one possible reason is that they originate from the industrial effluent (5) where phenol is one of the major components.


The analytical scheme used in this investigation was found to be well suited for surveying the levels of organic micropollutants in the water-course. The level of organic compounds in the background sample must be due to natural sources or to atmospheric transport of these compounds. There is an increase both qualitatively and quantitatively down the water-course, something which is consistent with the results from previous analyses of collective parameters [4--6]. This increase is probably substantially due to runoff from roads and industrial effluents with one specific industrial effluent being the dominating source.

REFERENCES 1 W. Giger, M. Reinhard, C. Schaffner and F. Ziircher, in L. H. Keith (Editor), Identification & Analysis of Organic Pollutants in Water, Ann Arbor Science, Michigan, 1st edn, 1976, Ch. 26, p. 433. 2 W. L. Budde and J. W. Eichelberger, Anal. Chem., 51 (1979) 567A. 3 R. D. Kloepfer, in L. H. Keith (Editor), Identification & Analysis of Organic Pollutants in Water, Ann Arbor Science, Michigan, 1st edn, 1976, Ch. 24, p. 399. 4 O. Skulberg, Resipientforholdene i Romeriksvassdragene Nitleva, Leira og R~bmua, The Norwegian Institute for Water Research, Oslo, 1972. 5 Vassdragsoverv~aking og forel~pige resultater, AN{~ (Avl~pssambandet Nordre ~yeren) information no. 2, 1979. 6 Sveselva--Harestuvannet--Nitelva Vannkavalitet og forurensingsregnskap, 1976-1977, AN~} -- report. 7 K. Martinsen and G. E. Carlberg, Analyse av organiske mikroforurensninger. Commisioned by NTNF 1550.3610, Report No. 17, Oslo, 1979. 8 A. Bj~brseth, private communication. 9 S. Jensen, A. G. Johnels, M. Olsson and G. Otterlind, in 1st Soviet-Swedish Symposium on the Pollution of the Baltic, Ambio Special Report No. 1, 1972. 10 K. Grob and G. Grob, Chromatographia, 5 (1972) 3. 11 G. Lunde, J. Gether and E. Steinnes, Environ. Sci. Technol., 9 (1975) 155.