JEQ Grow Your Career With ASA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
 QUICK SEARCH:   [advanced]


     


Published online 20 April 2005
Published in J Environ Qual 34:735-760 (2005)
DOI: 10.2134/jeq2004.0333
© 2005 American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Related articles in JEQ
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Web of Science (12)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Brändli, R. C.
Right arrow Articles by Tarradellas, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Brändli, R. C.
Right arrow Articles by Tarradellas, J.
Agricola
Right arrow Articles by Brändli, R. C.
Right arrow Articles by Tarradellas, J.
Related Collections
Right arrow Municipal Wastes
Right arrow Ecological Risk Assessment
Right arrow Organic Compounds
Right arrow Soil Pollution
Right arrow Municipal Waste

REVIEWS AND ANALYSES

Persistent Organic Pollutants in Source-Separated Compost and Its Feedstock Materials—A Review of Field Studies

Rahel C. Brändlia,b, Thomas D. Buchelib,*, Thomas Kuppera, Reinhard Furrerc, Franz X. Stadelmannb and Joseph Tarradellasa

a Ecole Polytechnique Fédérale de Lausanne (EPFL), Laboratory of Environmental Chemistry and Ecotoxicology (CECOTOX), Faculty of Architecture, Civil and Environmental Engineering, CH-1015 Lausanne, Switzerland
b Agroscope FAL Reckenholz, Swiss Federal Institute for Agroecology and Agriculture, Reckenholzstrasse 191, CH-8046 Zürich, Switzerland
c National Center for Atmospheric Research, Boulder, CO 80307-3000

* Corresponding author (Thomas.Bucheli{at}fal.admin.ch)

Received for publication August 27, 2004.

    ABSTRACT
 TOP
 ABSTRACT
 INTRODUCTION
 DATA SELECTION
 DATA PROCESSING
 STATISTICS
 CRITICAL ASSESSMENT OF SAMPLE...
 DATA EVALUATION
 PERSISTENT ORGANIC POLLUTANTS...
 OTHER ORGANOCHLORINES
 OTHER COMPOUNDS
 CONCLUSIONS
 REFERENCES
 
Composting and the application of compost to the soil follow the principle of recycling and sustainability. Compost can also have a positive effect on physical, chemical, and biological soil parameters. However, little is known about the origin, concentration, and transformation of persistent organic pollutants (POPs) in compost. We therefore compiled literature data on some priority POPs in compost and its main feedstock materials from more than 60 reports. Our data evaluation suggests the following findings. First, median concentrations of {Sigma} 16 polycyclic aromatic hydrocarbons (PAHs), {Sigma} 6 polychlorinated biphenyls (PCBs), and {Sigma} 17 polychlorinated dibenzo-p-dioxins and -furans (PCDD/Fs) were higher in green waste (1803, 15.6 µg/kg dry wt., and 2.5 ng international toxicity equivalent [I-TEQ]/kg dry wt.) than in organic household waste (635, 14.6 µg/kg dry wt., and 2.2 ng I-TEQ/kg dry wt.) and kitchen waste (not available [NA], 14.9 µg/kg dry wt., 0.4 ng I-TEQ/kg dry wt.). The POP concentrations in foliage were up to 12 times higher than in other feedstock materials. Second, in contrast, compost from organic household waste and green waste contained similar amounts of {Sigma} 16 PAHs, {Sigma} 6 PCBs, and {Sigma} 17 PCDD/Fs (1915, 39.8 µg/kg dry wt., and 9.5 ng I-TEQ/kg dry wt., and 1715, 30.6 µg/kg dry wt., and 8.5 ng I-TEQ/kg dry wt., respectively). Third, concentrations of three-ring PAHs were reduced during the composting process, whereas five- to six-ring PAHs and {Sigma} 6 PCBs increased by roughly a factor of two due to mass reduction during composting. {Sigma} 17 PCDD/Fs had accumulated by up to a factor of 14. Fourth, urban feedstock and compost had higher POP concentrations than rural material. Fifth, the highest concentrations of POPs were usually observed in summer samples. Finally, median compost concentrations of POPs were greater by up to one order of magnitude than in arable soils, as the primary recipients of compost, but were well within the range of many urban soils. In conclusion, this work provides a basis for the further improvement of composting and for future risk assessments of compost application.

Abbreviations: ACY, acenapththylene • ANT, anthracene • ASE, accelerated solvent extraction • BaA, benzo[a]anthracene • BaP, benzo[a]pyrene • BbF, benzo[b]fluoranthene • BBP, butylbenzylphthalate • BkF, benzo[k]fluoranthene • BPE, benzo[ghj]perylene • CBz, chlorobenzene • CHR, chrysene • CP, chlorophenol • CPA, chlorinated paraffin • DAD, diode array detector • DBA, dibenzo[a,h]anthracene • DBP, dibutylphthalate • DCP, dichlorophenol • DEHP, di(2-ethylhexyl)phthalate • ECD, electron captor detector • FD, fluorescence detector • FLT, fluoranthene • GC, gas chromatography • GPC, gel permeation chromatography • HCB, hexachlorobenzene • HCH, hexachlorohexane • HPLC, high performance liquid chromatography • IPY, indeno[1,2,3-cd]pyrene • I-TEQ, international toxicity equivalent • LAS, linear alkylbenzene sulfonate • MS, mass spectrometry • MSW, municipal solid waste • NA, not available • NER, non-extractable residue • PAH, polycyclic aromatic hydrocarbon • PBDE, polybrominated diphenylether • PCB, polychlorinated biphenyl • PCBz, pentachlorobenzene • PCDD/F, polychlorinated dibenzo-p-dioxin and -furan • PCP, pentachlorophenol • PHE, phenanthrene • POP, persistent organic pollutant • PYR, pyrene • TCBz, tetrachlorobenzene • TCP, tetrachlorophenol • T-OCDD/F, tetra- to octachloro dibenzo-p-dioxin and -furan • TrCP, trichlorophenol


    INTRODUCTION
 TOP
 ABSTRACT
 INTRODUCTION
 DATA SELECTION
 DATA PROCESSING
 STATISTICS
 CRITICAL ASSESSMENT OF SAMPLE...
 DATA EVALUATION
 PERSISTENT ORGANIC POLLUTANTS...
 OTHER ORGANOCHLORINES
 OTHER COMPOUNDS
 CONCLUSIONS
 REFERENCES
 
MODERN SOCIETIES PRODUCE considerable amounts of waste. Some 40 million Mg of municipal solid waste (MSW) are collected annually for recycling in the 15 member states of the European Union (before 2004). This corresponds to approximately 18% of the total municipal waste produced (European Commission, 2003). In Switzerland, the annual production of MSW is about 660 kg per capita, of which 45% is recycled and 55% is incinerated or disposed of (Kettler, 2002). To save incineration costs and landfill capacities, recycling of MSW is essential. Policies aiming at reducing volumes of MSW have resulted in the collection and composting (i.e., the controlled conversion of organic material into humus by composting plants) of nonseparated or partly separated (both non-organic and organic) refuse from households. The problems arising in connection with such mixed waste compost have been recognized since the beginning of the 1980s: it contains higher percentages of impurities and contaminants such as heavy metals and organic pollutants than compost from source-separated organic waste (Lisk et al., 1992a; Grossi et al., 1998; Hogg et al., 2002). In consequence, the production of mixed-waste composts is currently being phased out in most European countries. This is also due to the implementation of statutory quality standards or voluntary quality assurance programs by compost plant operators, which demand composting of source-separated biodegradable wastes from gardens or kitchens.

A plethora of different terms are in use within the disciplines of waste management and compost science and utilization to describe different types of wastes and compost feedstock materials. Even more confusingly, the vocabulary varies considerably between national and international directives and guidelines (German Ordinance on Biowastes, 1998; European Union, 1999; Hogg et al., 2002; United States Composting Council, 2002; Pain and Menzi, 2003; Amlinger et al., 2004; European Compost Network, 2004; USEPA, 2004). In this review, we largely use the following terms: "kitchen waste" (crude organic waste originating from private kitchens), "organic household waste" (a mixture of kitchen waste, garden waste, and small amounts of paper), "green waste" (organic waste from private gardens and public green areas), "green waste compost," and "organic household waste compost." Note that organic household waste always contains kitchen waste and needs to be mixed with green waste to achieve sufficiently aerated composting. The resulting compost, known as organic household waste compost, cannot therefore be directly compared with either organic household waste or with kitchen waste.

About 17 x 106 Mg of organic household and green waste are collected per annum in the 15 European Union member states (before 2004), which is about 35% of the total estimated recoverable potential. This results in about 9 x 106 Mg of compost (Hogg et al., 2002). The amount of separately collected organic household and green waste differs considerably between the countries. Thus more than 60% of the organic waste is separately collected in Germany, Austria, and Denmark, whereas this fraction hardly reaches 1% in Greece, Portugal, Spain, and Ireland (Hogg et al., 2002).

The risk of compost pollution by organic pollutants remains even if separately collected organic waste is composted, because of aerial deposition or accidental [e.g., via the increasingly ubiquitous plastic debris (Thompson et al., 2004)] and deliberate input (e.g., pesticide application) of such chemicals to organic materials. Organic pollutants may also accumulate during composting because of a mass reduction of about 40 to 60% due to mineralization (Gronauer et al., 1997; Schleiss, 2003). Consequently, a large number of compounds such as polychlorinated biphenyls (PCBs), polycyclic aromatic hydrocarbons (PAHs), polychlorinated dibenzo-p-dioxins and -furans (PCDD/Fs), brominated flame retardants, phthalates, linear alkylbenzene sulfonates (LAS), pesticides, and nonylphenols have been detected in compost derived from source-separated organic residues (Berset and Holzer, 1995; Ulén, 1997; Tørsløv et al., 1997; Hund et al., 1999; Büyüksönmez et al., 2000; Vanni et al., 2000; Zethner et al., 2000; Vergé-Leviel, 2001; Marb et al., 2003). Note that heavy metals are frequently found in compost (Lisk et al., 1992a; Tørsløv et al., 1997; Grossi et al., 1998; Krogmann, 1999; Zethner et al., 2000; Marb et al., 2001, 2003; Hogg et al., 2002; Herter et al., 2003; Amlinger et al., 2004) but are not covered by this report.

Although the concentrations of organic pollutants measured in compost may be regarded as rather low, the total input to its main recipient, namely the soil, cannot be neglected. Since the nutrient content of compost is relatively low, considerable quantities (8–10 Mg dry wt./ha/yr) are required to obtain a sufficient fertilizing effect (Herter et al., 2003; Timmermann et al., 2003). Even higher amounts are applied if the compost is used as a soil conditioner. Consequently, the input of PAHs, PCBs, and PCDD/Fs to the soil via compost application may be equal to or even higher than that introduced by aerial deposition or the application of other organic fertilizers such as sewage sludge or manure (Herter et al., 2003). To assure a sustainable soil quality, the input of organic pollutants to the soil has to be minimized. Conversely, the risk of soil pollution due to compost application has to be balanced with the soil-improving qualities of the compost, such as the stability of the soil particles, its pore volume, water capacity, organic matter content, and the related carbon and nitrogen content (Timmermann et al., 2003).

In view of this ambivalent role of compost in soil protection and improvement, it seems essential to increase our knowledge on the actual contamination level of compost and to evaluate whether the concentrations of pollutants in compost derived from source-separated organic residues can be reduced. This requires a knowledge of the sources of the pollutants and of potentially contaminated feedstock materials. Although there is an extensive review of the literature on pesticide concentrations in compost, mostly from the United States (Büyüksönmez et al., 1999, 2000), persistent organic pollutants (POPs) relevant to the environment such as PAHs, PCBs, PCDD/Fs, and other organochlorines have not yet been thoroughly reviewed. Moreover, most of the data and publications relevant to this subject have been published in the form of government reports, theses, or dissertations. This prevents the efficient dissemination of this knowledge among scientists and decision-makers. We have reviewed and analyzed more than 60 such studies to make these results more accessible. Only some 25 of them proved to be of sufficiently high scientific and information quality to be entered into a quantitative database which served as a basis for statistical evaluation (see below for details). Most of the data included originate from European studies. However, it should be noted that the present compilation by no means claims to be complete. There may well be more (non-European) reports of interest to our study, but since not officially published, this "gray literature" is difficult to access.

In summary, this review aims to contribute to the following aspects of compost research:


    DATA SELECTION
 TOP
 ABSTRACT
 INTRODUCTION
 DATA SELECTION
 DATA PROCESSING
 STATISTICS
 CRITICAL ASSESSMENT OF SAMPLE...
 DATA EVALUATION
 PERSISTENT ORGANIC POLLUTANTS...
 OTHER ORGANOCHLORINES
 OTHER COMPOUNDS
 CONCLUSIONS
 REFERENCES
 
This review focuses on reports on POP data in composts and their feedstock materials gathered from field studies. These data were collected in a database and statistically analyzed as specified below. The database did not include the extensive data originating from studies dealing with the composting of soil with other organic wastes for remediation purposes, laboratory composting experiments, experiments with artificially spiked (isotope-labeled) compounds, and compost containing sewage sludge. However, such results were occasionally used for comparison with the results obtained from the database analysis.

The criteria for including data in the database were a clear definition of the feedstock to exclude mixed-waste composts and MSW composts, composting or anaerobic digestion as a treatment process, field data, and reporting of single-compound concentration on a dry-weight basis. These rigorous demands reduced the more than 60 studies originally reviewed to about 25. Table 1 shows the 25 studies integrated into the database and subjected to detailed data analysis and Table 2 lists the remaining reports. Although they are not suitable for further evaluation (for the reasons given in Table 2), they are added here to (i) illustrate the vast amount of information available on this topic, (ii) allow this often gray literature to be tracked in future, and (iii) save the time and effort of other researchers seeking similar information.


View this table:
[in this window]
[in a new window]
 
Table 1. Field studies on the occurrence of organic pollutants in compost and its feedstock. These data were compiled in the database for statistical analysis.

 

View this table:
[in this window]
[in a new window]
 
Table 2. Field studies on the occurrence of organic pollutants in compost and its feedstock. For the reasons indicated, these data were not included in the database.

 
Very little data were found on the fate of POPs during anaerobic digestion and its concentrations in related products (Nilsson et al., 2000; Vorkamp et al., 2002). Our investigations were consequently focused on composting and the compost itself.


    DATA PROCESSING
 TOP
 ABSTRACT
 INTRODUCTION
 DATA SELECTION
 DATA PROCESSING
 STATISTICS
 CRITICAL ASSESSMENT OF SAMPLE...
 DATA EVALUATION
 PERSISTENT ORGANIC POLLUTANTS...
 OTHER ORGANOCHLORINES
 OTHER COMPOUNDS
 CONCLUSIONS
 REFERENCES
 
Information about the type of feedstock, catchment areas, collection time (season), sample preparation, and analysis as well as the original pollutant concentrations was extracted from the original references. The grouping of the data into subclasses of organic wastes and respective composts for subsequent comparison was based on the following criteria/restrictions: (i) the presence of sufficient data per category, (ii) consideration of common practices and peculiarities in the composting process, and (iii) initial hypotheses relating to processes potentially relevant for the occurrence of POPs in compost.

The feedstocks kitchen waste, organic household waste, and green waste were separated on the basis of the hypothesis of different POP exposures: accidentally discharged nonbiodegradable waste (e.g., plastics, electrical equipment, insulation material, hydraulic liquids, etc., known as "impurities") is thought to have a considerable influence on kitchen and organic household waste, whereas aerial deposition is suspected to be the main input pathway for green waste. The direct application of pesticides could play a role in both types of feedstocks. After the initial screening of the literature, the number of individual samples (>10) allowed four more feedstocks to be identified, namely grass, shrub clippings, bark, and foliage. Compost containing organic household waste and kitchen waste (organic household waste compost) and compost derived from green waste were compared to assess the influence of kitchen waste on POP concentrations in the compost. The evolution of organic pollutants during composting was evaluated by comparing their concentrations in green waste with those in the corresponding compost. Note that the comparison of organic household waste and its compost is difficult, since the former feedstock is often mixed with considerable amounts of green waste or other structuring material to achieve a balanced mix and thus a sufficiently aerated composting process. The composting process was further compared with relevant laboratory studies.

Persistent organic pollutant concentrations are expected to be higher in urban compost and feedstock materials due to higher emissions (Buehler et al., 2001; Meijer et al., 2003a; Krauss and Wilcke, 2003; Jaward et al., 2004). To account for these different emission patterns, the collecting areas were grouped as urban, semiurban, and rural according to information given in the original reports. Moreover, POP emissions vary according to the seasons (Halsall et al., 1995; Rahman et al., 1998; Schauer et al., 2003). The volatilization and recondensation processes of these semivolatile compounds depend on the temperature (Lee and Jones, 1999; Meijer et al., 2003a). Seasonally varying concentrations of POPs in both feedstock and compost can therefore be expected and were evaluated by comparing the data gathered during the different seasons (spring, summer, autumn, winter). Unfortunately, seasonal information was included in only few datasets and in those cases it was rarely specified whether the reporting time referred to the time of sampling or of feedstock collection. The catchment area and seasonal data were cross-checked to exclude possible process interference.

Both the concentrations of individual compounds and their sums were evaluated for all samples in the database. The summation was performed as follows: (i) sum of 16 USEPA priority PAHs ({Sigma} 16 PAHs); (ii) sums of three-ring PAHs (acenaphthylene [ACY], acenaphthene, fluorene, phenanthrene [PHE], anthracene [ANT]), four-ring PAHs (fluoranthene [FLT], pyrene [PYR], benzo[a]anthracene [BaA], chrysene [CHR]), five-ring PAHs (benzo[b]fluoranthene [BbF], benzo[k]fluoranthene [BkF], benzo[a]pyrene [BaP], dibenzo[a,h]anthracene) [DBA]), and six-ring PAHs (indeno[1,2,3-cd]pyrene [IPY], benzo[ghj]perylene [BPE]); (iii) sum of the PCBs #28, #52, #101, #138, #153, and #180 ({Sigma} 6 PCBs); and (iv) sum of 17 polychlorinated dibenzo-p-dioxins and -furans ({Sigma} 17 PCDD/Fs). Since data on {Sigma} 17 PCDD/F concentrations in the feedstock were scarce and some publications reported the sum of all tetra- to octa-chlorinated dibenzo-p-dioxins and -furans (T-OCDD/Fs), these data were also taken into account.

Certain PAH ratios are representative for specific PAH emission sources and could contain information on the origin of PAHs (Yunker et al., 2002; Bucheli et al., 2004). The following ratios were calculated from the individual PAH of the original references: ANT/(ANT and PHE), BaA/(BaA and CHR), FLT/(FLT and PYR), and IPY/(IPY and BPE). The first two ratios differentiate between petrogenic and pyrogenic sources, with numbers below 0.1 and 0.2 respectively being indicative of the former, and numbers above 0.1 and 0.35 respectively pointing to the latter source. The last two ratios separate petrogenic (<0.4 and 0.2, respectively) from liquid fossil-fuel combustion (0.4 to 0.5 and 0.2 to 0.5 respectively), and biomass and coal burning ( > 0.5 in both cases). For further details on PAH ratios and other molecular markers as well as typical numbers in specific emission sources, see Yunker et al. (2002) and Bucheli et al. (2004). Generally, PAH ratios have to be judged carefully and evaluated in concert. They may be biased in the present case, especially for lower-weight PAHs and particularly in composted samples due to possible preferential biodegradation or volatilization. Moreover, the borderline ratio between liquid fossil-fuel combustion and biomass or coal-burning derived from pure emission-source analysis is suspected to somewhat overestimate the latter source when applied to soils (Bucheli et al., 2004) and possibly to composts.


    STATISTICS
 TOP
 ABSTRACT
 INTRODUCTION
 DATA SELECTION
 DATA PROCESSING
 STATISTICS
 CRITICAL ASSESSMENT OF SAMPLE...
 DATA EVALUATION
 PERSISTENT ORGANIC POLLUTANTS...
 OTHER ORGANOCHLORINES
 OTHER COMPOUNDS
 CONCLUSIONS
 REFERENCES
 
The datasets were depicted as box-plots, and mean concentrations were included to better visualize the data distribution. Some datasets contained extremely high pollutant levels. They were considered as outliers (Table 1) and were not taken into account in the data analysis. All subsequently given concentrations represent median numbers unless otherwise stated. The Wilcoxon–Mann–Whitney test (Wilcoxon, 1945) was chosen to compare the means of two datasets (level of significance 0.025). This test does not require the samples to be normally distributed. The test is also robust to outliers.


    CRITICAL ASSESSMENT OF SAMPLE PREPARATION AND ANALYTICAL METHODS
 TOP
 ABSTRACT
 INTRODUCTION
 DATA SELECTION
 DATA PROCESSING
 STATISTICS
 CRITICAL ASSESSMENT OF SAMPLE...
 DATA EVALUATION
 PERSISTENT ORGANIC POLLUTANTS...
 OTHER ORGANOCHLORINES
 OTHER COMPOUNDS
 CONCLUSIONS
 REFERENCES
 
Reliable sample preparation and analytical methods are preconditions for the reliable interpretation of pollutant concentrations. So we will start this review by looking at the analytical methodologies used. To the best of our knowledge, no standardized analytical method, guideline, or best laboratory practice has acquired international acceptance for the analysis of POPs in compost and/or its feedstocks. The treatment of compost samples was often similar to or adopted from soil or sewage-sludge analysis. A few national attempts to harmonize and standardize methods have been made by the research and education foundation of the United States Composting Council (2002), the Austrian legislator (Bundesminister für Land- und Forstwirtschaft, Umwelt und Wasserwirtschaft, 2001; Österreichisches Normungsinstitut, 2002), and the German association of agricultural research and analysis stations (Verband Deutscher Landwirtschaftlicher Untersuchungs- und Forschungsanstalten, 1995). These rather recent guidelines have clearly not yet made any significant impact on the majority of the scientific or analytical community. As apparent from Table 1, compost samples were prepared and analyzed very differently by various investigators. Sample preparation mostly included drying, homogenization and/or grinding, and milling as well as storage. However, detailed method validations, mostly for soil analysis, have shown that most sample preparation steps are prone to artifacts. Thus the analysis of semivolatile compounds often involves drying at elevated temperatures, which may lead to some volatilization of the analyte (Auer and Malissa, 1990; Chiarenzelli et al., 1996; Berset et al., 1999; Desaules and Dahinden, 2000). Water removal at room temperature may also contaminate samples with pollutants ubiquitously present in ambient air (Alcock et al., 1994; Cousins et al., 1997). Drying and freezing may alter the interaction of the analyte with the sample matrix (Northcott and Jones, 2001). Even though freeze-drying is faster and minimizes contamination by ambient air, the risk of losses due to volatilization (Berset et al., 1999), cross-contamination (own results, unpublished), and the alteration of analyte–solid interaction remains. In most cases, the samples were homogenized by mixing, milling, or grinding. Very few studies took countermeasures against sample heating, which often takes place in this step. The size fraction selected for analysis varied considerably (0.1 to 2 mm). Many studies did not even bother to provide information on the homogenization step and their final size fraction. The same was true for storage conditions such as the storage containers used, the temperature, light conditions, and storage time. In conclusion, it seems that no artifact-free sample preparation technique exists, although some measures may be more suitable than others. In our own current work (Brändli et al., unpublished data, 2004), we attempt to reduce analyte losses, contamination, and matrix alterations by performing chemical drying instead of air-drying and homogenizing the samples while at the same time cooling them with ice.

The most commonly used technique for extracting hydrophobic organic contaminants from compost was Soxhlet extraction (Table 1). However, use was also made of alternative techniques such as sonication, saponification with liquid–liquid extraction, solid–liquid extraction, and accelerated solvent extraction (ASE). A wide range of solvents was applied: hexane, toluene, dichloromethane, methanol, acetone, and some mixtures of these. Numerous studies have been performed to evaluate and compare the different extraction techniques (Popp et al., 1997; Schantz et al., 1997; Berset et al., 1999; Dupeyron et al., 1999; Hubert et al., 2000; Martens et al., 2002; Song et al., 2002; Hollender et al., 2003). Unfortunately, none of these concerned compost samples. Few general conclusions can be drawn from the various comparative studies. Some of them found significant differences in extraction potencies of individual techniques (at least for certain sample types and compound classes) (Popp et al., 1997; Schantz et al., 1997; Dupeyron et al., 1999; Hubert et al., 2000; Martens et al., 2002; Hollender et al., 2003). But others did not (Heemken et al., 1997; Berset et al., 1999), or did so only at high levels of contamination (Song et al., 2002). Overall, the studies are difficult to compare, as they used different samples and chose different apparatus conditions and solvents.

Given the high content of organic matter in compost samples of up to 60%, a rigorous cleanup is crucial. Various silica gels, Al2O3, and/or Florisil columns were often used. Gas chromatography (GC)–mass spectrometry (MS) was the most frequently selected technique for detecting semivolatiles such as PAHs, PCBs, PCDD/Fs, and organochlorines, but other techniques were also used, such as GC–electron captor detector (ECD) and high performance liquid chromatography (HPLC)–fluorescence detection. In view of the different sample-preparation methods and the wide range of analytical methods used, it is likely that some of the data variability and statistical uncertainty observed in the compost literature compiled here originates from analytical differences.


    DATA EVALUATION
 TOP
 ABSTRACT
 INTRODUCTION
 DATA SELECTION
 DATA PROCESSING
 STATISTICS
 CRITICAL ASSESSMENT OF SAMPLE...
 DATA EVALUATION
 PERSISTENT ORGANIC POLLUTANTS...
 OTHER ORGANOCHLORINES
 OTHER COMPOUNDS
 CONCLUSIONS
 REFERENCES
 
Polycyclic Aromatic Hydrocarbons
Feedstock
Median concentrations of {Sigma} 16 PAHs were 421 µg/kg dry wt. (n = 18) in bark, 635 µg/kg dry wt. (n = 69) in organic household waste, 1803 µg/kg dry wt. (n = 31) in green waste, and 4710 µg/kg dry wt. (n = 13) in foliage (Fig. 1) . The significantly elevated PAH concentrations for all individual compounds as well as for {Sigma} 16 PAHs (Table 3) in foliage and green waste may be explained by an increased exposure time and the filtering effects of these feedstock materials for semivolatile organic compounds (Wania and McLachlan, 2001; Horstmann and McLachlan, 1998). Alternatively, green waste could have contained some fractions of highway clippings exposed to vehicle exhaust emissions.



View larger version (11K):
[in this window]
[in a new window]
 
Fig. 1. Sum of 16 USEPA polycyclic aromatic hydrocarbons ({Sigma} 16 PAHs; µg/kg dry wt.) in kitchen waste (not available [NA]), organic household waste (n = 69, one study only), green waste (n = 31), foliage (n = 13, one study only), shrub clippings (NA), bark (n = 18, one study only), grass (NA), compost containing organic household waste (n = 78), and compost originating from green waste (n = 23). Line: median; dotted line: mean; box: 25th and 75th percentile; lines with whiskers: 10th and 90th percentile; dots: outside values.

 

View this table:
[in this window]
[in a new window]
 
Table 3. P values of the comparison of {Sigma} 16 polycyclic aromatic hydrocarbon (PAH) concentrations in different feedstocks and composts.

 
Median ranges of the diagnostic PAH ratios for emission sources ANT/(ANT and PHE) (0.11–0.19), FLT/(FLT and PYR) (0.55–0.59), BaA/(BaA and CHR) (0.34–0.57), and IPY/(IPY and BPE) (0.44–0.52) reveal that the PAHs in all feedstock types originated overwhelmingly from pyrogenic sources (i.e., liquid fossil fuel combustion or biomass burning). Potential contamination of feedstocks with crude oil or unburned fossil fuels can largely be excluded. Based on typical numbers from PAH emission sources (Yunker et al., 2002; Bucheli et al., 2004), about 90% of all FLT/(FLT and PYR) and roughly 50% of all IPY/(IPY and BPE) figures indicate that, besides liquid fossil fuel combustion, biomass burning and coal combustion might have contributed to the presence of PAHs in the feedstocks. The PAHs might not have originated solely from atmospheric deposition, but could also stem from other minor contributions such as charcoal ash. However, no such indications were given in the original literature.

Compost
The PAH concentration in compost containing organic household waste was slightly higher than in green waste compost for all 16 compounds except for naphthalene. However, on a 95% level the difference is significant only for ANT, PYR, and CHR. The median {Sigma} 16 PAH concentration for green waste compost was 1715 µg/kg dry wt. (n = 23), whereas compost containing organic household waste had a median of 1915 µg/kg dry wt. (n = 78) (Fig. 1). The difference for the {Sigma} 16 PAHs is not statistically significant (Table 3). This is not surprising since organic household waste is often blended with considerable amounts of green waste as a structuring material to achieve a sufficiently aerated composting process. Median source diagnostic PAH ratios in organic household waste [ANT/(ANT and PHE): 0.16, FLT/(FLT and PYR): 0.57, BaA/(BaA and CHR): 0.41, and IPY/(IPY and BPE): 0.48] and green waste compost [ANT/(ANT and PHE): 0.14, FLT/(FLT and PYR): 0.59, BaA/(BaA and CHR): 0.40, and IPY/(IPY and BPE): 0.53)] are not significantly different from the respective numbers in the feedstocks, except for BaA/(BaA and CHR), which decreased from about 0.55 to about 0.40. These numbers lead to a qualitative apportionment of PAH sources similar to that stated above for the feedstocks (i.e., predominantly pyrogenic sources with some contribution from biomass burning).

The median concentration of {Sigma} 16 PAHs is not significantly different in green waste and its respective compost (1803 and 1715 µg/kg dry wt.; Fig. 1). However, a more detailed inspection of the PAH data revealed that the median concentrations of three- and four-ring PAHs were higher in green waste (379 and 869 µg/kg dry wt. respectively) than in its corresponding compost (127 µg/kg dry wt., p = 0.001, and 773 µg/kg dry wt., p = 0.257). In contrast, higher concentrations of five- and six-ring PAHs were measured in compost (380 and 232 µg/kg dry wt.) than in feedstock (328 and 119 µg/kg dry wt.), the difference being significant for the six-ring PAHs on a 90% level (p = 0.0449). We therefore hypothesize the degradation/and or volatilization of the lower-fused PAHs, whereas the accumulation of the higher-fused rings may be due to mass reduction (40–60%) during composting. This hypothesis is supported by laboratory composting studies using 14C-labeled PAHs (van Raaij et al., 1996; Vergé-Leviel, 2001; Carlstrom and Tuovinen, 2003). Phenanthrene (a three-ring PAH) was mineralized to CO2 more efficiently than fluoranthene (four rings). Benzo[a]pyrene (five rings) was not mineralized at all. However, the formation of non-extractable residues (NER) was observed in all studies. Hund et al. (1999) observed a degradation of PYR (four rings) to about 50% during composting and 36% NER formation. Similar results were reported by Hartlieb and Klein (2001). Decreasing biodegradation with an increasing number of fused rings in a mixture of soil and mixed-waste compost and a lower degradation of higher-fused rings during mixed waste composting or soil remediation was reported by several authors (Martens, 1982; Wischmann and Steinhart, 1997; Dahosch, 1998; Loser et al., 1999). Delayed mineralization of PAHs was reported to occur after the thermophilic phase (Vergé-Leviel, 2001; Ertunc et al., 2002; Hartlieb et al., 2003). However, losses may be considerable in this phase due to volatilization (Joyce et al., 1998; Hund et al., 1999). It is generally impossible to distinguish between the formation of NER and degradation when working with nonlabeled substances, so a possible degradation is difficult to quantify.

Catchment Areas
Higher median PAH concentrations ({Sigma} 16 PAHs = 1390 µg/kg dry wt., n = 12, and 4149 µg/kg dry wt., n = 7) were measured in urban organic household waste and green waste than in rural material ({Sigma} 16 PAHs = 584 µg/kg dry wt., n = 45, and 1693 µg/kg dry wt., n = 20). Semirural concentrations (1060 µg/kg dry wt., n = 12) of organic household waste were recorded between urban and rural concentrations, whereas semirural levels (994 µg/kg dry wt., n = 4) of green waste were even lower than rural ones. Urban compost containing organic household waste had higher median PAH concentrations ({Sigma} 16 PAHs = 2698 µg/kg dry wt., n = 17) than rural compost ({Sigma} 16 PAHs = 827 µg/kg dry wt., n = 15). Semirural levels laid between the two extremes. The dataset of green waste compost was too small to be evaluated for different catchment areas. In summary, these findings correspond well with the higher emissions of PAHs found in urban areas than rural ones (Buehler et al., 2001; Schauer et al., 2003).

Inspection of PAH ratios for source diagnosis revealed no significant differences in the major PAH contributors to composts or feedstocks separated in urban or rural areas. In particular, the data did not point to elevated contributions of liquid fossil-fuel combustion compared to biomass burning in the urban areas with their heavier traffic concentrations, as might have been suspected.

Seasonality
The seasonal PAH concentration pattern in organic household waste conflicted with the usual temporal succession in the environment: higher concentrations were measured in summer (952 µg/kg dry wt., n = 17) and spring (938 µg/kg dry wt., n = 15) than in autumn (557 µg/kg dry wt., n = 18) and winter (494 µg/kg dry wt., n = 18). In general, PAH concentrations in the environment are higher in winter than in summer as observed in air and plant material (Smith et al., 2001), rainwater (Gans et al., 1999), roof runoffs (Shu and Hirner, 1997; Gans et al., 1999), and sewage plant effluents (Pham et al., 1999). These higher concentrations are attributed to higher emissions from heating systems (Schauer et al., 2003) and combustion (Lee and Jones, 1999). In contrast, concentrations of green waste were higher in spring (3169 µg/kg dry wt., n = 9) and winter (2846 µg/kg dry wt., n = 4) than in summer (2435 µg/kg dry wt., n = 8) and autumn (1037 µg/kg dry wt., n = 10). Even though the number of measurements is rather low (4 < n < 10), these findings correspond with the seasonal pattern in the environment. The difference between organic household waste and green waste may be explained by their varying origins (e.g., imported kitchen waste such as tropical or citrus fruits vs. more local green waste), the shorter exposure time of kitchen waste to air, its possible lower affinity to semivolatile and hydrophobic compounds (e.g., root and tuber vegetable residues as compared to grass and leaves), and the small number of measurements.

The picture of inversed seasonal PAH concentration was even more pronounced in compost: the highest concentrations in compost containing organic household waste and green waste compost were reported in summer (3186 µg/kg dry wt., n = 7, NA) and autumn (2576 µg/kg dry wt., n = 13, and 3498 µg/kg dry wt., n = 3). In contrast, the levels in spring (1869 µg/kg dry wt., n = 15, and 774 µg/kg dry wt., n = 7) and winter (1165 µg/kg dry wt., n = 26, and 1715 µg/kg dry wt., n = 7) were lower. These findings correspond to those of another study on seasonal patterns of PAHs in compost (Breuer et al., 1997).

It still remains largely unclear why the PAH concentrations in compost were highest in seasons with lower emissions: in view of the varying residence time of green and organic household waste in the composting process (a few weeks to more than a year), a defined lag of about half a year is implausible. A possible higher mass reduction during composting in summer compared to winter due to a change of prevailing feedstocks [more easily degradable material such as grass in summer compared to more ligneous material in winter (Krogmann, 1994; Krauss et al., 1996; Krogmann, 1999; Taube, 2001)] cannot fully account for this discrepancy. The effects of elevated ambient temperatures in summer that may influence the composting, the PAH-degradation kinetics, and thus the total mass reduction are considered minimal.

Soil Concentration
A thorough overview of the literature on the PAH content in temperate topsoils was provided by Wilcke (2000). Median levels of {Sigma} 16 PAHs were as follows: arable soil: 216 µg/kg, grassland: 194 µg/kg, forest soils: 410 µg/kg, and urban soils: 1103 µg/kg. The median concentration ({Sigma} 16 PAHs) in the compost literature compiled here was 1870 µg/kg (n = 101). This suggests that the PAH concentrations in compost are up to one order of magnitude higher than in the recipient soils. As a complement to the concentrations, however, the actual input loads need to be assessed and compared to inputs from other sources such as fertilizers (e.g., sewage sludge, manure) or atmospheric deposition (Herter et al., 2003).

Polychlorinated Biphenyls
Feedstock
Median concentrations of {Sigma} 6 PCBs were 5.4 µg/kg dry wt. (n = 20) in bark, 9.3 µg/kg dry wt. (n = 39) in grass, 9.8 µg/kg dry wt. (n = 12) in shrub clippings, 14.6 µg/kg dry wt. (n = 82) in organic household waste, 14.9 µg/kg dry wt. (n = 8) in kitchen waste, 15.6 µg/kg dry wt. (n = 41) in green waste and 36.6 µg/kg dry wt. (n = 29) in foliage (Fig. 2) . The differences were significant for {Sigma} 6 PCBs between bark and foliage and the other feedstock materials, grass and green waste, and organic household waste (Table 4). This picture was identical for most individual PCBs. The fact that the highest concentrations were found in foliage may, as for PAHs, be explained by its filtering effect on semivolatile organic compounds (Horstmann and McLachlan, 1998; Wania and McLachlan, 2001). In contrast to PAHs, whose concentrations in green waste outweighed those in organic household waste, PCB concentrations were similar in organic household and green waste. This fact, as well as the higher 90th percentile and outside concentrations (Fig. 2), might be explained by fractions of accidentally discharged impurities (e.g., plastics) in organic household waste.



View larger version (11K):
[in this window]
[in a new window]
 
Fig. 2. Sum of six polychlorinated biphenyls ({Sigma} 6 PCBs; µg/kg dry wt.) in kitchen waste (n = 8, one study only), organic household waste (n = 82), green waste (n = 41), foliage (n = 29), shrub clippings (n = 12, one study only), bark (n = 20), grass (n = 39), compost containing organic household waste (n = 124), and compost originating from green waste (n = 55). Line: median; dotted line: mean; box: 25th and 75th percentile; lines with whiskers: 10th and 90th percentile; dots: outside values.

 

View this table:
[in this window]
[in a new window]
 
Table 4. P values of the comparison of {Sigma} 6 polychlorinated biphenyl (PCB) concentrations in different feedstocks and composts.

 
Compost
Higher PCB concentrations were measured in compost containing organic household waste ({Sigma} 6 PCBs = 39.8 µg/kg dry wt., n = 124) than that containing green waste ({Sigma} 6 PCBs = 30.6 µg/kg dry wt., n = 55) (Fig. 2). Except for PCB #52, and as for the PAHs (see above), this difference is not significant (Table 4). This similarity in median concentration may be at least partially due to the large portions of green waste often mixed into organic household waste (compost). However, some studies showed that organic household waste contained significantly more PCBs than green waste (e.g., Krauss, 1994). This is also indicated by the elevated 90th percentile and outside values in Fig. 2 for organic household waste compost and may again be explained on the basis of improperly separated organic waste and the presence of impurities such as plastics in organic household waste.

A possible accumulation of PCBs is hypothesized during green waste composting. The median concentration in green waste was 15.6 µg/kg dry wt. ({Sigma} 6 PCBs, n = 41) whereas the corresponding compost contained 30.6 µg/kg dry wt. ({Sigma} 6 PCBs, n = 55) (Fig. 2). The difference of roughly a factor of two is significant on a 95% level for PCBs #101, #138, #153, and #180, but not for {Sigma} 6 PCBs (p = 0.093; Table 4), and corresponds to the respective mass reduction during composting. The increase of PCB #28 by a factor of 1.3 and of PCB #52 by a factor of 1.6 is less pronounced and not significant. It may be due to some volatilization or degradation of these less stable congeners. This hypothesis is supported by laboratory composting studies: Vergé-Leviel (2001) found only small mineralization rates of 14C-labeled PCB #52, although 38% of the compound was volatilized during composting. Van Raaij et al. (1996) found that 14C-labeled PCB #77 remained stable during composting and reported 10% of NER after composting. Dahosch (1998) found that PCB levels did not decrease during the composting of mixed waste and Lazzari et al. (1999) even observed an increase in PCBs during the composting of sewage sludge and crushed ligneous waste. This was explained by the degradation of organic matter. Block (1998) reported PCB degradation of up to 40% in heavily contaminated soil composted with yard trimmings. However, remediation of heavily contaminated soil may be successful whereas degradation is probably much slower or does not occur at all at lower concentrations. Furthermore, since nonlabeled substances were used, the reduction could also be due to the formation of NER. In conclusion, it seems that PCBs are unlikely to degrade during composting and accumulate instead (due to mass reduction). Some volatilization of the lower chlorinated compounds such as PCBs #28 and #52 appears possible.

Catchment Areas
In organic household waste and green waste, no significant concentration differences were found between rural ({Sigma} 6 PCBs = 12.0 µg/kg dry wt., n = 46, and 15.3 µg/kg dry wt., n = 19) and urban ({Sigma} 6 PCBs = 29.0 µg/kg dry wt., n = 12, and 21.9 µg/kg dry wt., n = 9) catchment areas. But the difference of over a factor of two between urban and rural organic household waste is still considerable. Semirural concentrations of organic household waste (18.1 µg/kg dry wt., n = 12) laid between the rural and urban values, whereas concentrations of green waste were in the same range as rural concentrations (13.2 µg/kg dry wt., n = 4). In compost containing organic household waste, the differences between rural and urban regions were significant on a 90% confidence level for all congeners except for PCBs #28 and #52 and for {Sigma} 6 PCBs (p = 0.0359). Semirural concentrations of organic household waste compost laid between rural and urban levels. Due to a lack of data, green waste compost could not be evaluated according to catchment area. These findings may be attributed to the generally higher PCB burden in urban areas, such as was reported in air (Harner et al., 2004; Jaward et al., 2004) and soil (Langenkamp and Part, 2001; Krauss and Wilcke, 2003; Schmid et al., 2005).

Seasonality
Higher concentrations of organic household waste, although not significantly so, were measured in spring and summer ({Sigma} 6 PCBs = 19.0 µg/kg dry wt., n = 18, and 14.9 µg/kg dry wt., n = 18) than in autumn and winter samples ({Sigma} 6 PCBs = 11.8 µg/kg dry wt. n = 27, and 8.8 µg/kg dry wt., n = 18). The PCB concentrations in green waste differed little and ranged from 13.2 µg/kg dry wt. ({Sigma} 6 PCBs, n = 15) in spring to 20.9 µg/kg dry wt. ({Sigma} 6 PCBs, n = 4) in winter. However, a distinct seasonal change in PCB concentrations of compost containing organic household waste and green waste compost was observed. In both composts the highest concentrations were measured in summer ({Sigma} 6 PCBs = 60 µg/kg dry wt., n = 45, and 49.8 µg/kg dry wt., n = 22) and autumn ({Sigma} 6 PCBs = 39.3 µg/kg dry wt., n = 12, and 60 µg/kg dry wt., n = 6). Lower levels were reported in spring ({Sigma} 6 PCBs = 27.5 µg/kg dry wt., n = 15, and 10.0 µg/kg dry wt., n = 10) and winter ({Sigma} 6 PCBs = 22.9 µg/kg dry wt., n = 30, and 23.7 µg/kg dry wt., n = 9). In contrast to the PAHs, the concentration gradient over time in compost follows that found in other environmental compartments (Halsall et al., 1995; Lohmann et al., 2000). It is therefore suspected to be governed by the annual temperature regime, thus leading to increased release and volatilization of PCBs at higher temperatures in summer and autumn (Lee and Jones, 1999; Lohmann et al., 2000) and may, as for the PAHs, also reflect a possible variation in the degradation efficiency of organic waste.

Soil Concentration
In background soil samples from 191 global surface (0–5 cm) soils, the {Sigma} 6 PCBs of the mean concentrations of individual congeners was 2.5 µg/kg (Meijer et al., 2003b). German soil samples from private gardens, urban parks, and industrial/traffic sites exhibited higher {Sigma} 6 PCBs concentrations (30, 80, and 48 µg/kg, respectively) than samples from arable land, grassland, and forest (10 to 18 µg/kg) (Langenkamp and Part, 2001). The median compost concentration found in the present literature review is 38 µg/kg (n = 179). As with PAHs, this concentration is roughly higher by a factor of 15 than the background soil concentrations, but well within the range of many soils in urban areas. The PCB soil input loads from other recycling fertilizers lay within the same range as the compost. The input by atmospheric deposition varies greatly and can exceed that due to recycling fertilizers (Herter et al., 2003).

Polychlorinated Dibenzo-p-Dioxins and -Furans
Feedstock
Few measurements were reported for PCDD/Fs in organic waste (Table 1). Median concentrations of {Sigma} 17 PCDD/Fs were 0.44 ng I-TEQ/kg dry wt. (n = 9) in kitchen waste, 0.48 ng I-TEQ/kg dry wt. (n = 12) in shrub clippings, 0.76 ng I-TEQ/kg dry wt. (n = 4) in bark, 0.81 ng I-TEQ/kg dry wt. (n = 36) in grass, 2.22 ng I-TEQ/kg dry wt. (n = 9) in organic household waste, 2.54 ng I-TEQ/kg dry wt. (n = 9) in green waste, and 3.64 ng I-TEQ/kg dry wt. (n = 28) in foliage (Fig. 3) . The elevated concentrations in foliage correspond to similar findings for PAHs and PCBs (see above). However, on a concentration basis (ng/kg dry wt.), which might be the more relevant unit for evaluating environmental distribution processes, the median concentration in foliage {Sigma} 17 PCDD/Fs was lower (82.1 ng/kg dry wt.) than, for instance, in green waste (216.6 ng/kg dry wt.). Due to a lack of data, {Sigma} T-OCDD/Fs (in ng/kg dry wt.) was also taken into account. The {Sigma} T-OCDD/F concentrations were 58 ng/kg dry wt. (n = 22) in bark, 64 ng/kg dry wt. (n = 12) in shrub clippings, 70 ng/kg dry wt. (n = 9) in kitchen waste, 98 ng/kg dry wt. (n = 38) in grass, 197 ng/kg dry wt. (n = 79) in organic household waste, 235 ng/kg dry wt. (n = 28) in foliage, and 292 ng/kg dry wt. (n = 42) in green waste. Significant differences were observed between bark and organic household waste, green waste, and foliage; shrub clippings and the former three feedstocks; kitchen waste and organic household waste, green waste, and foliage; as well as between grass and green waste (Table 5).



View larger version (11K):
[in this window]
[in a new window]
 
Fig. 3. Sum of 17 polychlorinated dibenzo-p-dioxins and -furans ({Sigma} 17 PCDD/Fs; ng international toxicity equivalent [I-TEQ]/kg dry wt.) in kitchen waste (n = 9, one study only), organic household waste (n = 9), green waste (n = 9), foliage (n = 28), shrub clippings (n = 12, one study only), bark (n = 4), grass (n = 36), compost containing biowaste (n = 124), and compost originating from green waste (n = 61). Line: median; dotted line: mean; box: 25th and 75th percentile; lines with whiskers: 10th and 90th percentile; dots: outside values.

 

View this table:
[in this window]
[in a new window]
 
Table 5. P values of the comparison of polychlorinated dibenzo-p-dioxin and -furan (PCDD/F) and tetra- to octachloro dibenzo-p-dioxin and -furan (T-OCDD/F) concentrations in different feedstocks and composts.

 
Compost
Except for octachloro dibenzo-p-dioxin and octachloro dibenzofuran, all single congeners followed the same pattern: concentrations of organic household waste compost were in general higher than those of green waste compost. However, {Sigma} 17 PCDD/F concentrations varied little between compost containing organic household waste (9.6 ng I-TEQ/kg dry wt., n = 124; Fig. 3) and that containing green waste (8.5 ng I-TEQ/kg dry wt., n = 61) and the differences are far from significant (Table 5). This finding contrasts with the significant differences found between the respective feedstocks and may, as with PAHs and PCBs, be caused by the blending of organic household waste and kitchen waste with green waste for facilitated composting.

During the composting process, an increase of PCDD/Fs ({Sigma} 17 PCDD/Fs and {Sigma} T-OCDD/Fs) concentrations was observed (Fig. 3). This increase is much more pronounced than with PAHs and PCBs and higher than the usual mass reduction of about 40 to 60%. To explain this, the formation of PCDD/Fs during composting has been hypothesized. Dahosch (1998) found PCDD/Fs persistent during MSW composting and even observed the formation of octachloro dibenzo-p-dioxins. Similar results were found by Krauss et al. (1994), Öberg et al. (1994), and Fiedler et al. (1994). The possibility of PCDD/F-formation during composting is still controversially discussed and has also been stated as being unlikely to occur (Öberg et al., 1994).

Catchment Areas
For both organic household waste and green waste, higher concentrations of {Sigma} T-OCDD/Fs were reported in urban (229 ng/kg dry wt., n = 12, and 1343 ng/kg dry wt., n = 11) than in rural material (175 ng/kg dry wt., n = 47, p = 0.2950, and 236 ng/kg dry wt., n = 19, p = 0.0569). In both feedstocks, semirural concentrations (148 ng/kg dry wt., n = 12, and 119 ng/kg dry wt., n = 4) were even lower than rural ones. This may be due to the fact that all PCDD/F data in organic household waste and green waste data originated from the same source (Sihler et al., 1999). The elevated concentrations of PCDD/Fs in urban feedstocks correspond well with the findings for PCBs and PAHs (see above). No difference was found between rural and urban composts containing organic household waste ({Sigma} 17 PCDD/Fs = 8.4 ng I-TEQ/kg dry wt., n = 14, and {Sigma} 17 PCDD/Fs = 7.9 ng I-TEQ/kg dry wt., n = 24). Semirural levels ({Sigma} 17 PCDD/Fs = 9.3 ng I-TEQ/kg dry wt., n = 13) were even slightly higher than rural concentrations in this type of compost. In green waste compost, the difference between rural ({Sigma} 17 PCDD/Fs = 5.23 ng I-TEQ/kg dry wt., n = 8) and urban ({Sigma} 17 PCDD/Fs = 10.58 ng I-TEQ/kg dry wt., n = 16) samples is significant for the {Sigma} 17 PCDD/Fs (p = 0.011), but not for most single compounds.

Seasonality
In organic household waste, higher PCDD/F concentrations ({Sigma} T-OCDD/Fs) were measured in spring (293 ng/kg dry wt., n = 17) and summer (249 ng/kg dry wt., n = 18) than in winter (148 ng/kg dry wt., n = 18) and autumn (180 ng/kg dry wt., n = 26). The difference between the spring concentration and that of autumn and winter is significant. In green waste, the highest concentrations were reported in spring (694 ng/kg dry wt., n = 17) and winter (611 ng/kg dry wt., n = 4). Summer (191 ng/kg dry wt., n = 9) and autumn (170 ng/kg dry wt., n = 9) concentrations were lower. Even though median concentrations varied considerably, the differences were not significant. The high levels in spring might be explained by higher emissions of PCDD/Fs during the heating period due to domestic burning of wood/coal (Lohmann et al., 2000). In compost containing organic household waste, however, the highest concentrations were observed in summer (17 PCDD/Fs = 11.3 ng I-TEQ/kg dry wt., n = 46) and autumn (8.1 ng I-TEQ/kg dry wt., n = 13), which conflicts with the expected emission pattern dominated by waste incineration which is largely independent of the season (Dettwiler et al., 1997). In green waste compost, PCDD/F concentrations were similar in summer ({Sigma} 17 PCDD/Fs 8.4 ng I-TEQ/kg dry wt., n = 22), autumn (7.7 ng I-TEQ/kg dry wt., n = 6), and winter (9.0 ng I-TEQ/kg dry wt., n = 9), but lower in spring (4.2 ng I-TEQ/kg dry wt., n = 8). In view of the small numbers of measurements, we will not interpret these results further.

Soil Concentration
The {Sigma} 17 PCDD/F concentrations in soil were observed in the range of 2 to 5 ng I-TEQ/kg (Dettwiler et al., 1997; Langenkamp and Part, 2001). Schmid et al. (2005) reported {Sigma} 17 PCDD/F numbers between 1.1 and 11.4 ng I-TEQ/kg dry wt. with a median of 2.4 ng I-TEQ/kg dry wt. (n = 23). The median concentration for {Sigma} 17 PCDD/Fs in compost was 9.25 ng I-TEQ/kg dry wt. (n = 185). As with PAHs and PCBs, this number is in the upper range of the measured concentrations in soils.


    PERSISTENT ORGANIC POLLUTANTS COMPARISON
 TOP
 ABSTRACT
 INTRODUCTION
 DATA SELECTION
 DATA PROCESSING
 STATISTICS
 CRITICAL ASSESSMENT OF SAMPLE...
 DATA EVALUATION
 PERSISTENT ORGANIC POLLUTANTS...
 OTHER ORGANOCHLORINES
 OTHER COMPOUNDS
 CONCLUSIONS
 REFERENCES
 
The comparison of PCB and PAH concentrations in feedstock and compost (Fig. 4) shows that these two compound classes corresponded in most samples: the higher the PCB the higher the PAH concentration. This points to similar environmental behavior and the same input pathway, probably aerial deposition. However, there is a group of mostly green waste samples which had considerably high PAHs ({Sigma} 16 PAHs = 6000–14000 µg/kg dry wt.) but relatively low PCB concentrations (Fig. 4: data points clearly above the regression line, which was derived from the total of all data). These samples may originate from road clippings or contain ashes. High PCB concentrations with concomitantly lower PAH levels, as mostly observed in some organic household waste and organic household waste compost samples, may be explained by higher fractions of impurities in organic household waste.



View larger version (22K):
[in this window]
[in a new window]
 
Fig. 4. Sum of 16 polycyclic aromatic hydrocarbons ({Sigma} 16 PAHs; µg/kg dry wt.) versus sum of six polychlorinated biphenyls ({Sigma} 6 PCBs; µg/kg dry wt.) measured in the same samples of (a) organic household waste (n = 68), green waste (n = 29), and bark (n = 18), and (b) compost containing organic household waste (n = 67) and compost originating from green waste (n = 20). No concomitant PCB and PAH analysis is available for kitchen waste, foliage, shrub clippings, and grass. For illustration, the regression line (forced through zero) from all available data is added.

 
Higher PCDD/F levels were reported for most samples with increasing PCB concentrations (Fig. 5) , again pointing to similar behavior and input pathways. However, certain samples of the foliage and the organic household waste compost groups exhibited elevated PCB concentrations ({Sigma} 6 PCBs = 40–70 and 120–250 µg/kg dry wt.) but relatively low PCDD/F levels (Fig. 5). As for the PCB–PAH concentration, these higher PCB concentrations can probably be attributed to impurities.



View larger version (23K):
[in this window]
[in a new window]
 
Fig. 5. Sum of 17 polychlorinated dibenzo-p-dioxins and -furans ({Sigma} 17 PCDD/Fs; ng international toxicity equivalent [I-TEQ]/kg dry wt.) versus sum of six polychlorinated biphenyls ({Sigma} 6 PCBs; µg/kg dry wt.) measured in the same samples of (a) kitchen waste (n = 7), organic household waste (n = 9), green waste (n = 9), foliage (n = 28), shrub clippings (n = 12), bark (n = 2), and grass (n = 36), and (b) compost containing organic household waste (n = 111) and compost originating from green waste (n = 50). For illustration, the regression line (forced through zero) from all available data is added.

 

    OTHER ORGANOCHLORINES
 TOP
 ABSTRACT
 INTRODUCTION
 DATA SELECTION
 DATA PROCESSING
 STATISTICS
 CRITICAL ASSESSMENT OF SAMPLE...
 DATA EVALUATION
 PERSISTENT ORGANIC POLLUTANTS...
 OTHER ORGANOCHLORINES
 OTHER COMPOUNDS
 CONCLUSIONS
 REFERENCES
 
No organochlorine data could be found in the literature on compost feedstock. However, a considerable number of these compounds were analyzed in compost (Table 6). Most of the organochlorine pesticides have rarely been detected and median concentrations were very low or zero. Therefore no evaluation according to feedstock, catchment areas, and seasonality was performed.


View this table:
[in this window]
[in a new window]
 
Table 6. Further organic compounds analyzed in compost.

 
The ubiquitous pesticide {Sigma} DDT (sum of o,p'-DDD, o,p'-DDE, o,p'-DDT, p,p'-DDD, p,p'-DDE, and p,p'-DDT) was also detected in compost (8.1 µg/kg dry wt., n = 45). For comparison, Meijer et al. (2001) measured 1.36 µg/kg dry wt. (sum of p,p'-DDE, o,p'-DDT, and p,p'-DDT) in a UK soil sample and Webber and Wang (1995) detected DDT (sum of p,p'-DDE, o,p'-DDT, and p,p'-DDT) in 10 out of 30 Canadian soil samples, with concentrations ranging from 1 to 70260 µg/kg dry wt. Hexachlorobenzene (HCB), a precursor in the production of various organochlorine compounds, was used as a fungicide for the treatment of seeds and is linked to combustion processes and to the metallurgical industry. Applications of HCB are believed to have been phased out in North America and Western Europe (Bailey, 2001). It is listed in Annex C of the Stockholm convention on POPs that requires measures to reduce or eliminate releases from unintentional production. The median HCB concentration in compost was 1.2 µg/kg dry wt. (n = 142). Hexachlorobenzene was detected in agricultural soils at a concentration of 1 to 6 µg/kg dry wt. (Marb et al., 2003). Pentachlorobenzene (PCBz) and tetrachlorobenzene (TCBz) are precursors of pesticides. They were measured at 0.74 µg/kg dry wt. (n = 74) and 0.44 µg/kg dry wt. (n = 43) in compost. Target values for PCBz in Dutch soil are 2.5 and 10 µg/kg dry wt., respectively (Netherlands Ministry of Housing, Spatial Planning and Environment, 1994). Pentachlorophenol (PCP) is used as a wood-protection product with fungicide and bactericide properties. Its usage at a concentration of 1% or higher is prohibited within the EU (European Union, 1991). The median PCP concentration in compost was 14 µg/kg dry wt. (n = 100).


    OTHER COMPOUNDS
 TOP
 ABSTRACT
 INTRODUCTION
 DATA SELECTION
 DATA PROCESSING
 STATISTICS
 CRITICAL ASSESSMENT OF SAMPLE...
 DATA EVALUATION
 PERSISTENT ORGANIC POLLUTANTS...
 OTHER ORGANOCHLORINES
 OTHER COMPOUNDS
 CONCLUSIONS
 REFERENCES
 
Biphenyl, a fungicide predominately used in citrus fruits, was detected in all the samples analyzed (Table 6). Marb et al. (2003) argued that this compound is used by the chemical industry and suspected that its major input pathway to compost was via organic household waste. However, the dataset was too small to check this hypothesis. Other citrus fungicides, namely o-phenylphenol and thiabendazole, were detected in compost at 20 µg/kg dry wt. (n = 28) and 7.1 µg/kg dry wt. (n = 23), respectively. Other pesticides, such as cyfluthrin, deltamethrin, and fenpropathrin, were rarely measured and reported.

The median concentration of di(2-ethylhexyl)phthalate (DEHP) in compost was 300 µg/kg dry wt. The DEHP concentration was significantly higher in compost containing organic household waste (1300 µg/kg dry wt., n = 51) than in green waste compost (84 µg/kg dry wt., n = 28) (p = 0.0036). This can be explained by the higher plastic content of organic household waste compared to green waste. It supports the hypothesis derived from the above findings for PCBs that organic household waste might contain more such impurities than green waste. In a German study, the DEHP concentration in agricultural soils ranged from 300 to 700 µg/kg (Langenkamp and Part, 2001). In Denmark, DEHP concentrations of 12 to 40 µg/kg dry wt. were measured in uncultivated and arable soils (Vikelsøe et al., 2002).

A recent study showed that the median concentration of polybrominated diphenylethers (PBDE) (expressed as the sum of IUPAC congeners #17, #28, #47, #66, #71, #85, #99, #100, #138, #153, #154, #183, and #190) in compost was 12.2 µg/kg dry wt. (n = 10) (Marb et al., 2003). Polybrominated diphenylethers, used as a flame retardant, are detected at increasing concentrations in the environment (Rahman et al., 1998; De Boer et al., 2000; de Wit, 2002; Hites, 2004). Its soil concentrations ranged from 0.065 to 12 µg/kg dry wt. (sum of all congeners) in remote areas (Hassanin et al., 2004).

Chlorinated paraffins (sum of short- and medium-chained compounds) are used as secondary plasticizers as well as in paints and adhesives: they were detected in organic household waste at a concentration of 16041 µg/kg dry wt. (n = 17) (Nilsson, 2000). No quantitative compost and soil data could be found, but concentrations of between 5 and 200 µg/kg dry wt. were reported in sediments (Bolliger and Randegger-Vollrath, 2003).


    CONCLUSIONS
 TOP
 ABSTRACT
 INTRODUCTION
 DATA SELECTION
 DATA PROCESSING
 STATISTICS
 CRITICAL ASSESSMENT OF SAMPLE...
 DATA EVALUATION
 PERSISTENT ORGANIC POLLUTANTS...
 OTHER ORGANOCHLORINES
 OTHER COMPOUNDS
 CONCLUSIONS
 REFERENCES
 
This review compiles environmentally relevant POP data from some 25 individual studies in compost and its feedstocks. The sample preparation and analytical procedures reported varied widely. Of the seven types of feedstocks differentiated here, foliage contained the highest concentrations of PAHs, PCBs, and PCDD/Fs. Bark, shrub clippings, and grass exhibited the lowest levels of POPs, followed by kitchen waste, organic household waste, and green waste. The elevated concentrations of these semivolatile compounds in green waste and foliage may be explained by the efficient filter characteristics of these materials. Organic household waste and green waste composts contained similar amounts of POPs, because the former is often blended with significant amounts of green waste to achieve an aerobic composting process. Whereas the PAH concentration in compost was between the levels observed for the different feedstock materials, PCB concentrations in compost were at the higher end of feedstock concentrations. This suggests degradation/volatilization of the lower-fused PAHs, but no such indications were observed for the heavier PAHs and PCBs. Most strikingly, PCDD/F compost concentrations were elevated by a factor of 2 to 14 compared with feedstock levels. This is more than can be expected from the organic mass reduction occurring during composting. Comparison of the main POP compound classes investigated suggests that atmospheric deposition may be a relevant input source for most of these semivolatile pollutants. The PAH or PCB concentrations can sometimes be relatively elevated, probably due to inputs from traffic emissions, ashes, or impurities. Specific contamination might also occur due to pesticide application. Urban feedstock and compost was generally elevated in PAH, PCB, and PCDD/F levels in comparison with rural samples. This corresponds to the pattern observed in other environmental compartments. Seasonal differences were observed for PAHs, PCBs, and PCDD/Fs, the concentrations generally being highest during summer. This is in accordance with the seasonal variation observed in the environment for PCBs, but not for PAHs and PCDD/Fs. The POP level in compost was compared with representative data from soils (i.e., its primary recipient). Concentrations of PAHs, PCBs, and PCDD/Fs were up to one order of magnitude higher than those reported for arable soils, but of the same order of magnitude as for many urban soils. Overall, these findings may provide a best-possible current-state basis for future measures to reduce the POP content in compost as well as to assess the risks of compost application.


    ACKNOWLEDGMENTS
 
Our thanks are due to Hans Jörg Bachmann (FAL) and Martin Kohler (EMPA) for their useful comments on the manuscript. We would also like to acknowledge the Swiss Agency for the Environment, Forest and Landscape (SAEFL), the Swiss Federal Office of Energy (SFOE), and the Office for Waste, Water, Energy and Air (AWEL), Zürich, for their financial support.


    REFERENCES
 TOP
 ABSTRACT
 INTRODUCTION
 DATA SELECTION
 DATA PROCESSING
 STATISTICS
 CRITICAL ASSESSMENT OF SAMPLE...
 DATA EVALUATION
 PERSISTENT ORGANIC POLLUTANTS...
 OTHER ORGANOCHLORINES
 OTHER COMPOUNDS
 CONCLUSIONS
 REFERENCES
 


Related articles in JEQ:

This Issue in Journal of Environmental Quality

JEQ 2005 34: ix. [Full Text]  




This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Related articles in JEQ
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Web of Science (12)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Brändli, R. C.
Right arrow Articles by Tarradellas, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Brändli, R. C.
Right arrow Articles by Tarradellas, J.
Agricola
Right arrow Articles by Brändli, R. C.
Right arrow Articles by Tarradellas, J.
Related Collections
Right arrow Municipal Wastes
Right arrow Ecological Risk Assessment
Right arrow Organic Compounds
Right arrow Soil Pollution
Right arrow Municipal Waste


HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Journal of Natural Resources
and Life Sciences Education
Vadose Zone Journal
Soil Science Society of America Journal Journal of Plant Registrations The Plant Genome