Registration Dossier

Data platform availability banner - registered substances factsheets

Please be aware that this old REACH registration data factsheet is no longer maintained; it remains frozen as of 19th May 2023.

The new ECHA CHEM database has been released by ECHA, and it now contains all REACH registration data. There are more details on the transition of ECHA's published data to ECHA CHEM here.

Diss Factsheets

Environmental fate & pathways

Biodegradation in water and sediment: simulation tests

Currently viewing:

Administrative data

Link to relevant study record(s)

Description of key information

Note: A full degradation assessment for water and sediment is presented in the Overall Endpoint summary of Biodegradation. 

Aquatic compartment

The leading HHCB DT50 value in this WoE is the half-life of 4.2 days at 20°C from the Schaefer et al. (2005) study. After temperature correction this results in a DT50 value of 7 days (4.2 days*20/12°C). This lab study covers most of the OECD TG 309 criteria. 

The leading DT50 for HHCB-lactone is 27 days at 20°C as found in the lab study of Schaefer et al. (2005). After temperature correction this results in a DT50 value of 45 days (27 days*20/12°C).

Sediment compartment

The DT50 of 79 days at 12°C obtained in a limited documented study will be used for the risk assessment as the DT50 is considered too conservative.

The identity of the degradation products are anticipated to be similar to water: HHCB-lactone is formed and further degraded to HHCB-hydroxylated carboxylic acid and other hydroxylated HHCB products.

Key value for chemical safety assessment

Half-life in freshwater:
7 d
at the temperature of:
12 °C
Half-life in freshwater sediment:
79 d
at the temperature of:
12 °C

Additional information

(Bio)degradation of HHCB in the Aquatic Compartment: Results of the WoE

Several river water simulation and monitoring studies are available to assess the dissipation and degradation of HHCB and its primary metabolite, HHCB-lactone in water. The WoE presents sufficient data to cover the information from an OECD TG 309 test, considering half-life of parent and key metabolite. Three key studies are selected for the WoE, while other studies will be briefly summarized.


This first study in this WoE is holding the most weight considering HHCB’s biodegradability in the aquatic compartment. It is a river water die-away study analogous to the OECD 314E (Schaefer et al. 2005, RIFM). This study simulated the point at which waste-water effluent containing HHCB mixes with river water. In this study 10 mg/l 14C-HHCB was dosed into batch reactors containing river water amended with 10 mg/L suspended solids sourced from treated effluent. The total concentration of suspended solids from river-water and treated effluent amounted to 23 mg/L and is considered environmentally relevant. The batch reactors were incubated at 20°C for duration of 28 days.

Results: During the study HHCB rapidly degraded with a measured DT50 of 4.2 days. Loss of HHCB due to volatilization was relatively small with no significant volatilization until day 7 amounting to ~15% by day 28. Loss of HHCB to solids was found to be negligible too. Mass balance and abiotic controls provide evidence that the observed disappearance of parent HHCB is attributable to primary biodegradation.

It was also shown that HHCB was rapidly transformed into metabolites measured by thin layer chromatography (Schaefer et al., 2005). The primary metabolite appeared on day 1 and reached a maximum concentration of 26% of the initial radioactivity on day 4 followed by a 50% decrease in concentration from day 4 to 28. The primary metabolite has an estimated DT50 of 27 days.

Both parent HHCB and its primary metabolite have retention factors (Rf) of 0.58-0.72 and 0.44-0.51, respectively. The parent and this metabolite disappeared as more polar metabolic products were formed with Rf values ranging from <0.00-0.36. For those metabolites that were found above 10% initial radioactivity by termination of the test their Rf values ranged from <0.00-0.14 indicating that they were highly polar.

Based on supporting studies that used known standards of HHCB degradation products, similar analytical platforms and the measured Rf values, the primary metabolite is expected to be HHCB-Lactone and the secondary metabolites that are highly polar are expected to be HHCB-hydroxycarboxylic acid derivatives (Federle et al. 2002; and Schaefer et al. 2009).

The polar metabolites that were present above 10% initial radioactivity upon test termination have Rf values much smaller than the HHCB and HHCB-Lactone, the latter 2 having measured log Kow of 5.9 and 4.9, respectively, while the more polar metabolite have log Kow values of <=2.5 (Schaefer et al. 2009).

In a second study in this WoE: Federle et al. (2002 and 2003, internal P&G report, performed a similar non-GLP activated sludge die away study to understand wastewater removal of radiolabeled HHCB similar to that of Schaefer et. al. (2005). The concentration used is 0.5 µg/l.

Results: High recovery was noted with little loss via volatilization and NER formation. The parent molecule rapidly biotransformed into several more polar metabolites that were further characterized using HPLC. The parent molecule’s biotransformation half-life was determined to be 23 hours (probably at room temperature) when fit with a two compartment first order decay model. Less than 10% of the initial radioactivity remained as the parent compound at day 27 and RAD TLC indicated that the polar metabolites co-chromatographed with known oxidation product of HHCB, HHCB-lactone and its hydrolysis product: HHCB-hydrocarboxylic acid. Most of the radioactivity: 65%, had a calculated log Kow of -0.18, while 25% had a log Kow of 2.2 and 10% had a log Kow of 3.3 (Total 100%).

The third study in this WoE: Schwientek et al. (2016) presented a mass-balance study to calculate in-stream removal of HHCB and HHCB-lactone along a 4 km river segment in the Steinlach River, Germany. The river segment studied received inputs of HHCB and HHCB-lactone from WWTP discharge and hence can represent an effluent impact stream. The HHCB concentration in the influent was 72 ng/l and HHCB-lactone concentration was 331 ng/l, indicating that lactone was formed before entering the WWTP. Sampling was done in July at air temperature of 28°C but the river water temperature was not indicated but assumed to be lower than the air temperature. The average annual air temperature was 8°C.

Results: The concentration of HHCB and HHCB-lactone decreased to 49 and 266 ng/l respectively compared to the concentration measured in the effluent. The net removal of HHCB and HHCB-lactone amounted to 32 and 17.7%, respectively, after 3.5 hours. HHCB-lactone removal could be as high as 25% depending on whether the degraded feedstock of parent HHCB was considered additive to the HHCB-lactone pool. An increase of the ratio of HHCB-lactone/HHCB along the river segment indicates transformation of HHCB into HHCB-lactone. Disappearance of lactone indicates further biotransformation though volatilization and photo-degradation cannot be completely ruled out. Sorption to solids as a removal mechanism was considered negligible because of the low level of solids present in the river system (2-3 mg/L) and removal of the investigated compounds did not correlate with octanol-water partition coefficients.

Contrary to the rapid degradation observed in a number of monitoring studies, Bester et al. (2004) estimated an in-stream DT50 for HHCB and HHCB-lactone to be 67 and > 40 days, respectively. However, due to methodological constraints of the monitoring study performed and significant limitations to the author’s interpretation the calculations were not considered further.

Conclusion for HHCB fate in the aquatic compartment

The leading HHCB DT50 value in this WoE is the half-life of 4.2 days at 20oC from the Schaefer et al. (2005) study. This lab study covers most of the OECD TG 309 criteria. It is considered sufficiently conservative because the other used studies showed lower DT50's: Federle et al. (2002 and 2003) 21 hours and Schwientek et al. (2016) 3.5 hours.

The leading DT50 for HHCB-lactone is 27 days at 20°C as found in the lab study of Schaefer et al. (2005).

HHCB-hydroxycarboxylic acid is formed by further ester cleavage of the HHCB-lactone and additional oxidation. The half-life is not known but retention times measured show that log Kow is <=2.5 as indicated from the Federle et al (2002) presentation.

For risk and persistency assessment the DT50 values of HHCB and the first metabolite need to be temperature corrected. For HHCB this results in a DT50 value of 7 days (4.2 days*20/12°C). For HHCB-lactone this results in 45 days (27days*20/12°C). DT50 of HHCB-hydroxy-carboxylic acid is not known and is present in > 10%. This results in:

- The DT50 values of HHCB is 7 days at 12°C, based on the WoE.

- The DT50 value of HHCB-Lactone is 45 days at 12°C.

- The DT50 of HHCB-hydroxy-carboxylic acid is not known


Other (supporting) studies available for aquatic compartment are presented below


Barber et al. 2013: The in-stream attenuation of HHCB was determined in two effluent dominated streams in Colorado and Iowa, USA. Briefly, HHCB was measured at two WWTP effluent discharge points and at subsequent downstream sites in Boulder Creek, CO and Fourmile Creek, Iowa. In Boulder Creek from the discharge point to the downstream site suspended solids (SS) concentrations increased from 4 to 13 mg/L. In Fourmile Creek from the discharge point to the downstream site SS concentrations increased from 5-34 mg/L. The concentrations of HHCB from the effluent discharge point to the downstream site decreased from 1.8 to 0.6 and 3.3 to 0.7 µg/L for Boulder Creek and Fourmile Creek, respectively. The authors however noted that HHCB persisted downstream but the data suggests significant attenuation. Assuming first order process, DT50 for in-stream removal could be 1.95 hours and 1.25 hours for Boulder Creek and Fourmile Creek, respectively. However, in both cases the SS concentration significantly increased. Adsorption could be the main process but also biotransformation is likely a mechanism for removal.

Applicability for risk assessment: The study is a field study, the DT50s can be degradation or dissipation, it supports the DT50 of the key study and therefore the information is used as supporting information.

Schaefer et al. (2009): In a benchtop wastewater simulation study HHCB degraded to HHCB-Lactone and more polar metabolites as indicated by the inclusion of known HHCB-Lactone and hydroxycarboxylic acid standards. As abiotic controls did not elicit a similar response transformation of HHCB and HHCB-Lactone can be attributed to biodegradation.

Applicability for the risk assessment: This Schaefer et al. (2009) study is used as key information too for the assessment of HHCB’s metabolites.

Several field studies report HHCB-Lactone concentrations higher in effluent than influent concentrations providing evidence of the conversion of HHCB to HHCB-Lactone in situ (Bester et al. 2004; Horii et al. 2007; and Reiner et al. 2007). The detection of HHCB-Lactone in wastewater influent suggests that (bio)degradation of HHCB to lactone occurs in the sewer system prior to entering the wastewater treatment plant (Berset et al. 2004).

Applicability for the risk assessment: These studies support the degradation and the key metabolite of HHCB.

Horii et al. (2007) determined the removal of HHCB in two US WWTPs using mass balance approaches. In the two sites evaluated, HHCB-lactone was concomitantly measured with HHCB in both influents and effluents. In both sites HHCB-lactone was found to be produced which corresponded to a decrease in HHCB. In one site however, the net gain only accounted for ~16% of the net loss of HHCB. The total mass lost could not be explained by sorption to solids alone, suggesting that both HHCB and lactone can undergo biodegradation into products not measured in this study. Volatilization however was not ruled out for the observed loss. In total HHCB was removed 90-98%.

Applicability for the risk assessment: Study indicates that HHCB first degradation product is HHCB-lactone. DT50’s could not be derived.

Reiner et al. (2007) saw a similar trend with HHCB where HHCB decreased when comparing the influent to the effluent which corresponded in an increase of HHCB lactone. In this study, the average concentration of HHCB-lactone increased from influent to effluent across the two WWTPs amounting to a 48 and 69% increase. More than half of the HHCB entering the WWTPs was not detected in the treated wastewater or the sludge. Hence, the authors suggest biological transformation, chemical transformation, and/or volatilization as possible mechanisms of loss. Total removal of HHCB across the two plants assessed averaged 63%.

Applicability for the risk assessment: Study indicates that HHCB first degradation product is HHCB-lactone. DT50’s could not be derived.

Bester et al. (2004a) showed in a similar study using mass balance approach, HHCB removal in a large German WWTP to better understand the potential HHCB removal mechanisms and significance of biodegradation. They found that HHCB was removed ~65% when comparing the effluent to the influent and found an increase in the concentration of HHCB-lactone (its key oxidation product) in the effluent vs. the influent. Comparing the unaccounted for deficit of HHCB it was assumed that about 7% of HHCB was transformed into HHCB-lactone at steady state.

In a follow-up study, Bester et al. (2004b) assessed the enantiomeric pattern of HHCB-lactone in several WWTPs along the river Ruhr in Germany, to confirm whether the lactone they measured was from biotransformation of HHCB. The author concluded that in some, but not all, WWTPs biotransformation of HHCB to HHCB-lactone is probable.

Applicability for the risk assessment: Study indicates that HHCB first degradation product is HHCB-lactone. DT50’s could not be derived.

Simonich et al. (2002) assessed removal of HHCB in U.S. and EU WWTP with varying configurations of treatment technologies. HHCB was measured in influent and effluent of various wastewater treatment processes to understand the importance of biotransformation. HHCB nor its metabolites were measured in sludge. When physical processes were evaluated such as primary gravitational settling HHCB was measured to be removed 29%, which can be related to its adsorption potential (log Koc is 4.2). When wastewater treatment plants were assessed that have a combined settling and activated sludge process removal increased to 87%, indicating that 58% loss is due to biotransformation and/or volatilization. This was based on continuous sampling of influent and effluent.

The studies presented below indicate limited degradation of HHCB

Yang et al. (2005) reported wastewater removal efficiencies for a tertiary treatment plant operating in Ontario, Canada. When assessed quarterly throughout the year (February, April, July and October), with sampling during weekends, HHCB had an average annual removal of 72%. Yang et al. concluded that the removal was largely due to sorption to solids. Anaerobic digestion was investigated in their study and using mass balance approach, no biodegradation was detected.

Applicability for the risk assessment: The (absence of) degradation and adsorption cannot be distinguished. Also, DT50 values cannot be calculated.

Zeng et. al. (2007) assessed the removal of HHCB in a municipal wastewater treatment in Guangdong, China. The plant utilizing secondary treatment technology had high removal efficiency for HHCB ranging from 87-96%. This author also concluded the primary route for removal to be via sorption to sludge as the dissolved concentration of HHCB appeared rather stable.

Applicability for the risk assessment: The (absence of) degradation and adsorption cannot be distinguished. Also, DT50 values cannot be calculated.

Garciano et al. (2003a) investigated the removal of HHCB in four WWTPs in the Netherlands and came to a similar conclusion based on taking freely dissolved HHCB; degradation is not considered a major removal process. Interestingly, Garciano et al. (2003b) also previously published a study indicating a biodegradation half-life of 46 hours for HHCB with freely dissolved chemical and hydraulic retention times important for estimating total mass removed via biodegradation. This would correspond between 6.9-12.6% removal via biodegradation for their assessment on the four WWTP in Netherlands.

Applicability for the risk assessment: The (absence of) degradation and adsorption cannot be distinguished. Also, DT50 values cannot be calculated.

Zhou et al. (2008) assessed the removal of HHCB in three WWTPs in Beijing, China. Removal ranged from 41.7-71.1%, depending on the WWTP assessed but no evidence was provided with respect to biodegradation. The author attributed removal to sludge as dominant mechanism. The above studies cited however did not directly assess known metabolites of HHCB hence conclusions on biodegradation as a removal mechanism are limited.

Applicability for the risk assessment: The (absence of) degradation and adsorption cannot be distinguished. Also, DT50 values cannot be calculated.

Mass-balance calculations in field studies presented above in some instances yield an unaccounted fraction of HHCB and HHCB-Lactone, which can potentially be explained by the formation of metabolites other than HHCB-Lactone (Horii et al. 2007; and Reiner et al. 2007). Inexplicable net losses of HHCB and HHCB-Lactone in field studies cannot be definitively explained, due to experimental constraints or by formation of additional water-soluble metabolites. In general, the reported conversion of HHCB to HHCB-Lactone in monitoring studies varies significantly. This variation can be attributed to the complexity of field studies, methodology as well as inherent temporal and biological variation within wastewater treatment plants. The findings of lab-based studies are consistently supported by field observations indicating HHCB is inherently biodegradable by wastewater microbial communities as is presented above. Wastewater effluent contaminated surface waters and sludge amended soils therefore should possess a potential to degrade HHCB.

Degradation of HHCB in the Sediment compartment: available results

There are only limited studies on HHCB degradation in the sediment, meaning that there is limited information to fulfil the requirements for an OECD TG 308. An adequate WoE is therefore not possible with less than three studies. Despite these limitations it is still possible to derive a DT50 for this compartment using the available sediment information but also information from the aquatic and soil compartments.

Two studies were identified that can be used for the HHCB degradation in sediment.

The first study is a laboratory simulation study that illustrates the biodegradation potential of HHCB in aquatic river sediment (Envirogen et al. 1998). Using a study design similar to the OECD TG 307, 14C-HHCB was dosed with 500 µg in 50 g soil (10 mg/kg soil) into aerobic river sediment and incubated at 22-23°C over the course of one year. Transformation products were indicated.

Results and Discussion: At the end of the exposure (one year) 4% of the dosed 14C-HHCB remained relating to HHCB. The other part could be related to more polar metabolites formed as a product of primary degradation. No significant mineralization was reported. A DT50 of 79 days was calculated for HHCB. While this test was incubated at a temperature higher than what might be expected in aquatic sediments, the study used an unrealistically high test concentration (i.e. 10 mg/kg sediment), while 1 mg/kg bw is the maximum in the OECD T 308. Lower concentrations are more favorable for degradation and therefore it might be expected that the use of this conservative test concentration balanced out the impact from the optimal incubation temperature. The 79 days can therefore be used for the risk assessment.

The second study identified assessed the biodegradation potential of HHCB exposed to environmentally relevant fungal species isolated from aquatic sediments (Martin et al. 2007).

Results and discussion: In this study, HHCB was rapidly transformed by Myrioconium sp. exhibiting a DT50 value of 9.2 days under optimized conditions. HHCB removal in inactivated cultures was negligible indicating minimal effects of photo degradation and volatilization thereby suggesting the mechanism of removal to be biodegradation.

In enzymatic assays assessing HHCB degradation the concentrations of HHCB-lactone increased to approximately 780, 353, and 198%, relative to respective controls containing heat-inactivated enzymes, upon treatment with laccases from T. versicolor, Myrioconium sp., and C. aquatica, respectively. All fungal strains reported to rapidly degrade HHCB whether it be from sediment or soil are reported to express the enzyme laccase; an enzyme identified to degrade HHCB, other fragrances and lignin (Martin et al. 2007; Envirogen et al. 1997; and Vallecillos et al. 2017). This study suggests that the pathway for biodegradation is similar to what has been observed in soils, river water and wastewater.

Final conclusion for HHCB’s fate in sediment: A limited documented study is present showing a DT50 of 79-days in a lab study at 22°C but this is considered too high due to too high concentrations used (10 mg/kg soil). In addition, aquatic and soil studies indicate DT50s between 4 and 35-days and therefore this DT50 of 79-days is considered too conservative. In view of the too high concentration which will have limited the biodegradation and too high temperature which will have enhanced the biodegradation, the result of 79-days will be used for the risk assessment, without temperature correction: In conclusion:

- The DT50 of 79 days at 12°C will be used for the risk assessment.

The identity of the degradation products are anticipated to be similar to water: HHCB-lactone is formed and further degraded to HHCB-hydroxylated carboxylic acid and other hydroxylated HHCB products. These will be addressed in the B-assessment.

Other (supporting) studies related to sediment degradation are summarized below

Peck et al. (2006) measured the concentrations of HHCB in sediment cores from Lake Erie and Lake Ontario. Measured concentrations of HHCB followed trends in fragrance consumption volumes, over the course of several decades with concentrations of HHCB increasing in accordance with greater consumption of fragrance. The data illustrated measurable concentrations of HHCB, and other musks, preserved in sediment samples dating back several decades. The author notes that the variation of HHCB concentrations in top layers of sediments and the presence of HHCB in cores dating back prior to the use of HHCB could have been caused by experimental artifacts such as superficial mixing and laboratory contamination. The data might suggest HHCB can persist in anaerobic aquatic sediments over long periods of time.

Applicability for the risk assessment: It can be seen that HHCB remains in aquatic sediments but the study is not relevant for risk assessment.

Peng et al. (2019): Removal of HHCB from spiked sediment was assessed in a subtropical freshwater microcosm. Briefly, HHCB was spiked into uncontaminated sediment sampled from a reservoir in China. Over the course of a 28-day period, no biodegradation or loss was observed. The authors concluded HHCB to be not readily biodegradable.

Applicability for the risk assessment: It can be seen that HHCB remains in aquatic sediments but the study as such is not relevant for risk assessment.

Reference list

Artola-Garicano, Elsa, et al. "Sorption kinetics and microbial biodegradation activity of hydrophobic chemicals in sewage sludge: model and measurements based on free concentrations." Environmental science & technology 1 (2003): 116-122.

Artola-Garicano, Elsa, et al. "Removal of two polycyclic musks in sewage treatment plants: freely dissolved and total concentrations." Environmental science & technology 14 (2003): 3111-3116.

Barber, Larry B., et al. "Persistence and potential effects of complex organic contaminant mixtures in wastewater-impacted streams." Environmental science & technology 5 (2013): 2177-2188.

Berset, Jean-Daniel, et al. "Considerations about the enantioselective transformation of polycyclic musks in wastewater, treated wastewater and sewage sludge and analysis of their fate in a sequencing batch reactor plant." Chemosphere 8 (2004): 987-996.

Bester, Kai. "Polycyclic musks in the Ruhr catchment area—transport, discharges of waste water, and transformations of HHCB, AHTN and HHCB-lactone." Journal of Environmental Monitoring 1 (2005): 43-51.

Bester, Kai. "Retention characteristics and balance assessment for two polycyclic musk fragrances (HHCB and AHTN) in a typical German sewage treatment plant." Chemosphere 8 (2004): 863-870.

Envirogen (1998). Fate of HHCB in Soil Microcosms. Envirogen, Inc. Princeton Research Centre, report submitted to International Flavors and Fragrances, Lawrenceville, NJ.

Federle, T. W., N. R. Itrich, D. M. Lee, and D. Langworthy. 2002. Recent Advances in the Environmental Fate of Fragrance Ingredients. Presentation and poster of Proctor & Gamble presented at SETAC 23rd Annual Meeting, Salt Lake City, USA (as cited in EC, 2008).

Federle, T. W., N. R. Itrich 2003. Biodegradation of Galaxolide (HHCB) in Activated Sludge. Notebooks: ITS-257, ITL-417. Proctor and Gamble Central Product Safety Division/Product Safety and Regulatory Affairs Environmental Science Department, OH

Horii, Y., J. L. Reiner, B. G. Loganathan, K. Senthil Kumar, K. Sajwan, and K. Kannan. 2007. Occurrence and Fate of Polycyclic Musks in Wastewater Treatment Plants in Kentucky and Georgia, Chemosphere 68(11), 2011-2020.

Martin, Claudia, et al. "Biotransformation of the polycyclic musks HHCB and AHTN and metabolite formation by fungi occurring in freshwater environments." Environmental science & technology 15 (2007): 5395-5402.

Peck, Aaron M., Emily K. Linebaugh, and Keri C. Hornbuckle. "Synthetic musk fragrances in Lake Erie and Lake Ontario sediment cores." Environmental science & technology 18 (2006): 5629-5635.

Peng, Feng-Jiao, et al. "Fate and effects of sediment-associated polycyclic musk HHCB in subtropical freshwater microcosms." Ecotoxicology and environmental safety 169 (2019): 902-910.

Reiner, J. L., J. D. Berset, and K. Kannan. "Mass flow of polycyclic musks in two wastewater treatment plants." Archives of Environmental Contamination and Toxicology 4 (2007): 451-457.

Schaefer, E. C., 2005. 14C-HHCB: Dieaway of a Semi-Volatile Organic Compound in River Water. Project No. 558E-109A. Wildlife International, Ltd. Report submitted to the Research Institute for Fragrance Materials, Inc.

Schaefer, E. C., 2009. 14C-HHCB: Removal and/or Biodegradation of Semi-Volatile Organic Compound in an Activated Sludge Simulation System. Project No. 558E-112. Wildlife International, Ltd. Report submitted to the Research Institute for Fragrance Materials, Inc.

Schwientek, M., Guillet, G., Ruegnur, H., Kuch, B., Grathwhole, P. "A high-precision sampling scheme to assess persistence and transport characteristics of micropollutants in rivers." Science of the Total Environment 540 (2016): 444-454.

Simonich, Staci L., et al. "Removal of fragrance materials during US and European wastewater treatment." Environmental science & technology 13 (2002): 2839-2847.

Vallecillos, Laura, et al. "Degradation of synthetic fragrances by laccase-mediated system." Journal of hazardous materials 334 (2017): 233-243.

Yang, Jian-Jun, and Chris D. Metcalfe. "Fate of synthetic musks in a domestic wastewater treatment plant and in an agricultural field amended with biosolids." Science of the Total Environment 1-3 (2006): 149-165.

Zeng, Xiangying, et al. "Preliminary study on the occurrence and distribution of polycyclic musks in a wastewater treatment plant in Guandong, China." Chemosphere 8 (2007): 1305-1311.

Zhou, HaiDong, et al. "Distribution and elimination of polycyclic musks in three sewage treatment plants of Beijing, China." Journal of Environmental Sciences 5 (2009): 561-567.