An overview of hazard and risk assessment of the OECD high production volume chemical category—Long chain alcohols [C6–C22] (LCOH)

ARTICLE IN PRESS

Ecotoxicology and Environmental Safety ] (]]]]) ]]]–]]]

Contents lists available at ScienceDirect

Ecotoxicology and Environmental Safety

0147-65

doi:10.1

$ Fun

(Interna

Associa$$ A

accorda

subjects

et al. (2

(2008).� Corr

Institut

Denmar

E-m

Pleaschem

journal homepage: www.elsevier.com/locate/ecoenv

An overview of hazard and risk assessment of the OECD high productionvolume chemical category—Long chain alcohols [C6–C22] (LCOH),$,$$

Hans Sanderson a,�, Scott E. Belanger b, Peter R. Fisk c, Christoph Schafers d, Gauke Veenstra e, AllenM. Nielsen f, Yutaka Kasai g, Andreas Willing h, Scott D. Dyer b, Kathleen Stanton a, Richard Sedlak a

a The Soap and Detergent Association, Washington, District of Columbia, DC 20005, USAb The Procter & Gamble Company, Central Product Safety, Miami Valley Innovation Center, P.O. Box 538707, Cincinnati, OH 45253-8707, USAc Peter Fisk Associates, 39 Bennell’s Avenue, Tankerton, Whitstable, Kent, UK CT5 2HP, UKd Fraunhofer-Institute for Molecular Biology and Applied Ecology (IME), P.O. Box 1260, Schmallenberg 57377, Germanye Shell International B.V., P.O. Box 162, 2501 AN The Hague, The Netherlandsf Sasol North America, Research and Development Department, Westlake LA, USAg Kao Corporation, 2-1-3 Bunka, Sumida-ku, Tokyo 131-8501, Japanh Cognis GmbH, Henkelstrasse 67, D-40551 Dusseldorf, Germany

a r t i c l e i n f o

Article history:

Received 1 April 2008

Received in revised form

10 October 2008

Accepted 11 October 2008

Keywords:

Risk assessment

Long chain alcohols

LCOH

Chemical category

High production volume chemical

Human health

International chemical safety programs

13/$ - see front matter & 2008 Published by E

016/j.ecoenv.2008.10.006

ding sources and experimental guidelines: This

tional Council of Chemical Associations)/SD

tion) Aliphatic Alcohols Consortium.

ny studies referred to in this overview

nce to national and/or international guidelin

and animal welfare. For further details see

008); Schafers et al. (2008); Veenstra et al.

esponding author. Current address: Nation

e of Denmark, P.O. Box 358, Frederiksborg

k. Fax: +145 4630 1114.

ail address: [email protected] (H. Sanderson).

e cite this article as: Sanderson, H.ical category—Long chain alcohols

a b s t r a c t

This review summarizes the findings of the assessment report for the category, long chain alcohols

(LCOH) with a carbon chain length range of C6–C22 covering 30 substances, and 41.5 million tonnes/

year consumed globally. The category was evaluated under the Organization for Economic Co-operation

and Development (OECD) high production volume chemicals program in 2006. The main findings of the

assessment include: (1) no unacceptable human or environmental risks were identified; (2) these

materials are rapidly and readily biodegradable; (3) a parabolic relationship was demonstrated between

carbon chain length and acute and chronic aquatic toxicity; (4) category-specific (quantitative)

structure-activity relationships were developed enabling prediction of properties across the entire

category; (5) LCOH occur naturally in the environment in an equilibrium between synthesis and

degradation; (6) industry coming together and sharing resources results in minimizing the need for

additional animal tests, produces cost savings, and increases scientific quality of the assessment.

& 2008 Published by Elsevier Inc.

1. Introduction

The aim of this paper is to summarize and introduce theassessment of the long chained alcohols (LCOH) category, and alsoto give a brief review of ongoing national and internationalassessment frameworks addressing non-assessed or high produc-tion volume (HPV) chemicals. Most HPV chemicals have been onthe market for decades but rarely have comprehensive data sets

lsevier Inc.

work was funded by the ICCA

A (The Soap and Detergent

article were conducted in

es for protection of human

papers in this issue by: Fisk

(2008); and Belanger et al.

al Environmental Research

vej 399, DK-4000 Roskilde,

, et al., An overview of haz[C6–C22] (LCOH),. Ecotoxico

on their physicochemical and toxicological properties beenpublicly available. Therefore HPV chemicals are under increasingregulatory scrutiny globally. In 1990, member countries of theOrganisation for Economic Co-operation and Development(OECD) decided to undertake the investigation of HPV chemicalsin a co-operative way. These HPV chemicals include all chemicalsreported to be produced or imported at levels greater than1000 tonnes/year in at least one member country or in theEuropean Union region. The most recent OECD HPV Chemicals Listcompiled in 2004 contains 4843 substances based on submissionsof nine national inventories and that of the European Union. TheOECD HPV program proceeds by the agreement that membercountries will co-operatively select the chemicals to be investi-gated, collect and characterize effects and exposure informationfrom government and public sources and encourage industry toprovide information from their files, complete the agreed dossierfor the Screening Information Data Set (SIDS), and make an initialassessment of the potential hazard of each chemical investigated.When a full SIDS dossier on a chemical is available, an initialassessment of the information is undertaken and conclusions are

ard and risk assessment of the OECD high production volumel. Environ. Saf. (2008), doi:10.1016/j.ecoenv.2008.10.006

ARTICLE IN PRESS

H. Sanderson et al. / Ecotoxicology and Environmental Safety ] (]]]]) ]]]–]]]2

drawn on the potential hazard(s) and exposure information to putthe hazard information into context (e.g., based on use in theSponsor country). Since 1999, the work in OECD has concentratedon data gathering, testing, and initial hazard assessment. Detailedexposure information gathering and assessment of risk is nolonger required as part of the SIDS initial assessment, but can becarried out in follow-up at the national (or regional) level, asappropriate, following national (or regional) priority setting aspost-SIDS work. Detailed international assessment of risks tohuman health and/or the environment is also no longer carriedout under the SIDS initial assessments. In the policy bodies ofOECD, member countries discuss and agree on any follow-upactions on chemicals for which further work is recommended.Finalized SIDS dossiers and initial assessment reports are madeavailable worldwide through UNEP Chemicals website (http://www.chem.unep.ch/). Protocols were established for closeco-operation with the industry in the various stages of theProgramme, which is undertaken in co-ordination with national,regional and other international existing chemicals programmes(OECD, 2004).

The global chemical industry, through the InternationalCouncil of Chemical Associations (ICCA), launched a globalinitiative on HPV chemicals in 1998 to expedite the OECD HPVprogram. Through this commitment, the chemical industry hasundertaken to provide, as a first step, harmonized data sets on theintrinsic hazards of and initial hazard assessments for approxi-mately 1000 HPV substances by the end of 2004. The informationconsisting of a SIDS Dossier, a SIDS Initial Assessment Report(SIAR); and the SIDS Initial Assessment Profile (SIAP) aresubmitted to the OECD for international agreement as part of itsrefocused HPV Chemicals Programme. The cost of generating dataand the work to draft the assessments will be borne byindustry—and shared, whenever possible, by companies ininternational consortia. The main features of the ICCA HPVchemicals initiative are for voluntary action by the world chemicalindustry to speed up the process under existing regional and/orglobal programmes with a clear target date, provide globallyharmonized, internationally agreed data sets and initial hazardassessments under the refocused HPV Chemicals Programme ofthe OECD, and the elimination of duplication of testing andassessment efforts. The main expected benefits of these actionsare to restore public confidence in chemicals and to foster apositive reputation of the chemical industry on a global basis, toestablishment of a sound scientific basis for any subsequentregional, national, or global risk assessment need, to minimize thecost for the industry and to reduction in the number of animalsneeded for testing (CEFIC, 2007).

There are, of course, other ongoing or planed national andinternational HPV initiatives with different regulatory objectives,but a common feature among them is the desire for increasedtransparency regarding the properties of chemicals. The OECDdata sets complement these other initiatives, some of which aredescribed in the following paragraphs.

The United States Environment Protection Agency (US EPA)initiated their HPV Challenge program in 1999, challengingindustry to provide data on some 2860 HPVs. This was doneunder the chemical rights-to-know program (http://www.epa.gov/chemrtk/). Of these substances, US EPA found that 43% haveno publicly available data on basic toxicity, and only 7% have a fullset of basic test data publicly available. EPA also found that only55% of the chemicals reported in the Toxics Release Inventory hadfull basic toxicity testing data publicly available. Only about onequarter of chemicals in consumer products had basic testinginformation publicly available. This lack of publicly availabletoxicity data compromised, in the Administration’s view, thepublic’s right to know about chemicals in their homes, their

Please cite this article as: Sanderson, H., et al., An overview of hchemical category—Long chain alcohols [C6–C22] (LCOH),. Ecotoxic

workplaces, and the products they buy (Goldman, 1998). Hencethe US EPA HPV Challenge Program. The results are made publiclyavailable via the US EPA HPV Information System (HPVIS) (http://www.epa.gov/hpvis/).

The Canadian Environment Protection Act (CEPA) was revised in1999 and required that all existing chemicals on the market andidentified the Domestic Substances List (DSL) would need reviewbased on their properties and likelihood of exposure to humans orthe environment. The compounds would then be categorized basedon their persistence (P), bioaccumulation (B), and toxic (T) (PBT)properties and likelihood of exposure. The DSL includes �23,000substances that were in Canadian commerce, used for manufactur-ing purposes, or manufactured in or imported into Canada in aquantity of 100 kg or more in any calendar year between January 1,1984 and December 31, 1986. The aim here is to prioritize whichchemicals to categorize for and risk assessment (CEPA, 1999).

The European Union (EU) Existing Chemicals Program, in-itiated in the mid-1990s mandated industry to provide allavailable data for EU priority chemicals in two priority phases.The first phase included compounds on the EU market41000 tonnes/year with known hazardous properties, and thesecond phase included all other compounds 41000 tonnes/year.Subsequently the authorities were to prioritise the substancesaccording to their environmental relevance (exposure) and hazardproperties. Starting with the substances of highest concern,comprehensive risk assessments were to be conducted. If relevantdata gaps were identified, industry had to conduct and providedata from additional studies. However, the progress of theprogram was rather slow so that after 10 years only for a minorfraction of the substances at 41000 tonnes/year final risk wereassessment available. As a consequence of this the EU issued awhite-paper on the future chemicals policy (EU, 2001a, b), andinitiated the European Registration Evaluation Authorization ofChemicals (REACH) process in 2001 (EU, 2001b). This policy willaddress registration of compounds marketed before 1981 involumes greater than 1 tonne (�30,000 compounds). Priority isgiven to HPV chemicals under REACH (41000 tonnes/year). Thedeadline for registration of chemicals used at these tonnages is3 years after full implementation of REACH (�2011). For feasibilitypurposes (minimizing animal tests, time, costs), assessment ofchemical categories as well as use of read-across and quantitativestructure-activity relationships ((Q)SARs) play important roles inthe implementation of REACH. The bottom-line of REACH hasbeen summarized as: No data–No market.

In 2005, the Japanese government started its voluntary HPVprogram for substances produced or imported in Japan in volumesgreater than 1000 tonnes/year. The program focuses on HPVchemicals on the Japanese market which are not evaluated byany other HPV program, such as the OECD and US challengeprograms. A total of 652 HPV chemicals appear on the Japanesemarket today, 140 of which are not covered by any of the otherinternational HPV programs and have no data specific for Japan.Hence, these substances need assessment under the Japanese HPVchallenge program. Roughly half of these have voluntarily beensponsored by the Japanese industry and are currently beingassessed. The required endpoints are the OECD SIDS data package.The initial assessment phase of the program will be completed byMarch of 2009 (Japan MOE, 2005).

In concert with these national and international chemicalmanagement programs the United Nations Economic and SocialCouncil adopted the Globally Harmonized System (GHS) forChemicals Classification and Labeling in 2003 (UN, 2003), whichwill allow the intrinsic hazard properties to be translated intohazard classes that are readily interpretable worldwide for hazardcommunication. Thus, HPV data sets support national andregional efforts to improve the safe handling and use of chemicals.

azard and risk assessment of the OECD high production volumeol. Environ. Saf. (2008), doi:10.1016/j.ecoenv.2008.10.006

ARTICLE IN PRESS

H. Sanderson et al. / Ecotoxicology and Environmental Safety ] (]]]]) ]]]–]]] 3

The purpose of this special issue of Ecotoxicology and Environ-

mental Safety is to review the SIAR for one of the best documentedand largest categories of substances evaluated under the OECDHPV programme in terms of number of chemicals (30) and globalproduction volume (41.5 million metric tonnes/year). The cate-gory is the LCOH with a carbon chain length range C6–C22. A globalAliphatic Alcohols Consortium was formed and managed by TheUS Soap and Detergent Association in 1999 (consortia are definedunder HPV programs as a group of companies working together tomutually provide data and expertise from their respectivecompanies for the purpose of HPV submissions). The Consortiumis comprised of 13 international companies from three differentcontinents, representing both up- and downstream users of thesubstances. The United Kingdom was the sponsor country andShell International was the lead company. The issue addressproperties of LCOH covering: (1) description of the physicochem-ical properties and acute aquatic toxicity (Fisk et al., 2008, thisissue); (2) chronic toxicity (Schafers et al., 2008, this issue);mammalian toxicity (Veenstra et al., 2008, this issue); andenvironmental risk assessment (Belanger et al., 2008, this issue).The aim of this review is to summarize, with an emphasis on theenvironmental compartment, the findings of the SIAR (OECD,2006), which will be outlined in greater detail in the followingpapers.

2. Materials and methods

Any studies referred to in this overview article were conducted in accordance

to national and/or international guidelines for protection of human subjects and

animal welfare.

2.1. Chemicals

This case-study addresses a category of LCOH [C6–C22]. It covers 30 LCOH (pure

substances and commercially available complex mixtures) CAS number entries.

The commercially available products generally include several LCOH components,

with a range of carbon chain lengths and various compositions and structures. The

composition depends on the production method and the related feedstocks. Most

of the alcohols have linear carbon chains but certain manufacturing processes

create linear and essentially linear structures. Long chain alcohols are manufac-

tured by a number of processes, but these can be divided into two general

categories by feedstock. (1) Oleochemical—includes coconut, palm, kernel oil, and

tallow fat or other triglycerides. (2) Various processes use petrochemical—the

most commonly being olefins (alpha and internal) and ethylene. Some commer-

cially available products are blends of two or more specific chain length alcohols to

produce mixtures. Moreover, the commercial industrial processes used to produce

alcohols in some cases necessarily result in a spread of carbon number, and some

alkyl chain branching. In addition, a limited number of unsaturated substances

very similar to the saturated analogues are included. The total annual global usage

in 2004 was 1,580,429 metric tons. A significant portion of these substances are

used in personal and household care products ultimately disposed of down-the-

drain, and enter the environment at low levels via wastewater treatment plant

effluent.

2.2. Chemical category rationale

Key attributes that the category members share are: comparable structural

features, similar metabolic pathways, a common mode of ecotoxicological action,

Table 1Ranges of measured physicochemical and biological properties.

End point Melting point

(1C)

Boiling point (1C) Density, (g/cm3)

Range �47.5–72.5a 158–4400a 0.8–0.85a

a C6–C22.b C6–C20.c C18–C6.d C6–C16.

Please cite this article as: Sanderson, H., et al., An overview of hazchemical category—Long chain alcohols [C6–C22] (LCOH),. Ecotoxico

and common levels and mode of human health-related effects. Grouping of the

LCOH into a common category is possible because the group is a homologous

series of structures that impart the predictable pattern of properties. In addition,

certain branched and unsaturated structures are considered to have such similar

properties that their inclusion in the category is justified. Commercial products

contain a range of alcohols, including unsaturated alcohol components, essentially

linear (mono-alkyl branched) components, and linear alcohols. All components of

all commercial products relevant to this category are primary alcohol structures.

This allows multi-component reaction products to be considered within the

category by the application of validated models of exposure and effects, based on

the detailed knowledge of the composition. For the environmental end-point

assessment, this has been done by read-across and modelling. The read-across

approach was applied to biodegradability where sufficient data existed to allow

interpretation of degradation patterns across the entire category to fill data gaps

directly. For algae, read-across-based expert judgement was applied taking into

account measured and predicted effects in daphnids and fish for the substance of

interest. Modelling of the ideal solubility of the components of the substances was

developed, allowing component and total solubility at any loading rate to be

calculated. By using knowledge of the properties of each component, ecotox-

icological effects have been predicted.

The mammalian biotransformation of LCOH involves an oxidation step of the

alcohol function to the corresponding aliphatic carboxylic acid, with the aldehyde

being a transient intermediate. These carboxylic acids (i.e., fatty acids) are

subsequently broken down by stepwise removal of one or several C2 units from the

aliphatic carbon chain through b-oxidation. The stepwise breakdown of LCOH

results in common intermediate metabolites with shorter chain lengths which

provides additional justification that the alcohols under consideration can be

regarded as a single category. The observation also explains the similarity in

toxicological profile for systemic effects. LCOH are generally metabolized in a

highly efficient manner and limited potential exists for retention or bioaccumula-

tion for the parent alcohols and their biotransformation products.

3. Results

3.1. Physicochemical properties

While (Q)SAR techniques are developed using reported andmeasured data, they may also be confirmed and useful acrossmany other structural types. Their success in predicting propertiesfor category members for which measured data exist suggests thatthe members do not possess any particularly unusual features.The physicochemical properties vary across members of thecategory (Fisk et al., 2008, this issue; OECD, 2006) (Table 1).

3.2. Biodegradation

LCOH with chain lengths up to C18 (docosanol) are readilybiodegradable in tests that conform most closely to ready testbiodegradability methods (OECD 301 series). At carbon chainlengths greater than C14, most tests showed that pass levels forready biodegradation were reached within the 10-day window.Chain lengths of C16�18 achieved ready test pass levels, althoughnot within the 10-day window. The one test on a single carbonchain length greater than C18 showed degradation of 37% mostlikely due to the interplay of de-sorption, solubility, andbiodegradbility. These rates are in accordance with field data formeasured concentrations in wastewater treatment plant influentand effluent showing greater than 99% removal for carbon

Log Kow Water solubility,

(mg/L)

Vapour pressure,

(hPa)

BCF

2.03–47b 0.001–5900c 8.2�10�8–1.22a 7–46,000d

ard and risk assessment of the OECD high production volumel. Environ. Saf. (2008), doi:10.1016/j.ecoenv.2008.10.006

ARTICLE IN PRESS

H. Sanderson et al. / Ecotoxicology and Environmental Safety ] (]]]]) ]]]–]]]4

numbers 12–18. This summary of degradation is applicable toboth linear and essentially linear components of substances in thecategory. Therefore, the whole category is considered to show veryhigh levels of biodegradability. The substances are susceptible toatmospheric degradation by hydroxyl radicals, with half-livesranging between ca. 10 and 30 h (based on measured andestimated rate constants, for a hydroxyl radical concentration of5�105 molecules/cm3). Predictions from the SRC BIOWIN v4.00program (part of the EPI Suite v3.12) supports the conclusion ofrapid degradation for the linear alcohols, but cannot be usedquantitatively (Fisk et al., 2008, this issue).

Moreover, Federle and Itrich (2006) have studied the fate of freeand linear alcohol ethoxylate-derived fatty alcohols in activatedsludge. Radiolabelled (14C) C12, C14, and C16 alcohols were used. Thestudy was a batch-mode activated sludge die-away system, wheredisappearance of parent, formation and disappearance of metabo-lites, uptake into biomass and mineralization to 14C CO2 weremonitored over time. The activated sludge from a municipalwastewater treatment plant was obtained, and used at 2500 mg/L.The degradation of LCOH involved two principle pathways, whichwere oxidation to a fatty acid, which was then b-oxidized to CO2, andO-oxidation of the methyl group to yield dioic acids, which thenundergo b-oxidation. In conclusion, Federle and Itrich (2006) foundthat LCOH are extensively and rapidly mineralized to carbon dioxideand water, with half-lives in sewage treatment being less than 1 min.

3.3. Environmental exposure

The published and grey literature on the environmentaloccurrence, fate, and behaviour of LCOH has been reviewed(Mudge et al., 2008). The principal focus of that review was on thenatural production of alcohols, which occurs in all livingorganisms from bacteria to man, and the profiles and concentra-tions of these compounds in water, soils, and sediments. Themajor production mechanism is from the reduction of fatty acids,through aldehyde intermediates, to fatty alcohols and in manyorganisms to esters with fatty acids to form waxes (Metz et al.,2000). Due to the nature of the synthetic pathway using acetyl-CoA, most long chain alcohols are of an even numbered chainlength. Terrestrial plants utilize fatty alcohols as a waxy coatingand these compounds are dominated by long chain moieties, withchain lengths from C22 to C32. In contrast, marine organismssynthesize smaller compounds with peak chain lengths ofC14–C16. Bacteria also produce fatty alcohols but these can alsobe odd chain lengths and contain branching. The alcohols areubiquitous and occur in most environments around the world,including the deep ocean and in sediment cores. The rates ofproduction of LCOH from natural waxes and fatty acids inenvironmental conditions are not known (Mudge et al., 2008).These include release rates and mass contributions from microbialsenescence, household consumption of animal and vegetablematter, etc. For example, Leeming et al. (1994) established thatLCOH (C14�32) are measurable components of human faeces inwastewater. The sum of C14–C18 chain lengths ranged from 217 to1825mg/g in human waste sent to sewage treatment. It is clearthat measurements of long chain alcohol in environmentalmatrices will reflect the combination of both natural andanthropogenic sources. The concentration of individual freealcohols in the environment ranges from low values in old(7000 years) deep cores from the open ocean floor (undetectableto 12 ng/g dry weight for C16) to high values near natural sourcesand especially in suspended particulate matter (2.7 mg/g dryweight for C16) (Mudge et al., 2008).

Several methods that measure LCOH in environmental ma-trices are available. Dunphy et al. (2001) devised and executed a

Please cite this article as: Sanderson, H., et al., An overview of hchemical category—Long chain alcohols [C6–C22] (LCOH),. Ecotoxic

method whereby alcohols present in an environmental samplecan be detected at extremely low concentrations, often less than10 ng/L. The method involves extraction of wastewater effluentand associated solids followed by derivatization with 2-fluoro-N-methylpyridinium p-toluenesulfonate to a permanent cation forquantitation by HPLC/MS.

Influent levels of LCOH have been reported for 12 wastewatertreatment plants across the United States (MRI, 2004; Morrall etal., 2006). Average influent concentrations for C12–C15 rangedfrom 64.0 (C13) to 117.5 (C12)mg/L. The sum of C12–15 LCOHaveraged 394.5mg/L across all influents that were sampled. Fortreatment plant effluents monitored within the US, the weighedaverage concentrations were ordered as C12 (0.255mg/L)4C144C154C13 (0.035mg/L). For treatment plant effluents mon-itored in Canada, the average concentrations were greatest for C15

(0.619mg/L)4C144C184C124C13 (0.209mg/L). For treatment ofplant effluents monitored within Europe, the average concentra-tions were greatest for C12 (0.281mg/L)4C144C154C13 (0.165mg/L).The overall trend appears to be that effluents have higherconcentrations of longer LCOH than shorter chain lengths (OECD,2006; Belanger et al., 2008, this issue). This is also consistent withthe expected chain length distributions found in the watersassociation with primarily natural sources of LCOH (USEPA, 1997).For example, activated sludge (AS) treatment accounts for 80.6% oftotal US wastewater flow versus 7.1% for trickling filters. There-fore, individual measurements can be weighed to achieve anational average concentration. For the US, Canada, and Europe,the average total long chain alcohol concentrations (C12–C15) ineffluent are 0.572, 1.711, and 0.910mg/L, respectively. The 90thpercentile (all measurements at this concentration or lower) forindividually monitored effluent measurements worldwide, notaccounting for treatment type and flow, is 1.979mg/L and theglobal average for the three regional measurements is 1.064mg/L.These values include both free and bound alcohol to wastewatersolids. For the US, Canada, and Europe, the average total LCOHconcentrations (C12–C15) following adjustment due to sorption are0.417, 1.487, and 0.654mg/L, respectively, yielding a global averageof 0.739mg/L. Note that the focus from a toxicological and riskassessment point of view is on the chain lengths oC15, as thelonger chain lengths are not bioavailable (OECD, 2006; Belanger etal., 2008, this issue).

Dyer et al. (2006) recently documented the appropriateness ofadapting the Dunphy et al. (2001) analytical method for measur-ing alcohol ethoxylate in coarse sediments. The method wasapplied at three sites of varying sediment composition. Furtherrefinements to the methods were instituted to potentiallymeasure free LCOH and alcohol ethoxylates in pore water, surfacewaters, and chemical sorbed to coarse and fine sediments. LCOHwere ubiquitous and primarily associated with fine particulatematter in river sediments. Measurements by chain length andlocation were variable with the highest measurements (up to12mg/g) recorded far downstream of sewage treatment plantinputs (above that recorded in the mixing zones and dischargeproper). Levels of alcohols upstream of sewage inputs highlyoverlapped those in discharge and mixing zone samples (ca.0.1–1mg/g) (Dyer et al., 2006). These observations are indicative ofand consistent with the widespread natural presence of alcoholsin sediments reviewed by Mudge et al. (2008).

3.4. Environmental toxicology

3.4.1. Acute

Alcohols, with the exception of some propargylic alcohols(Veith et al., 1989) which are excluded from this category, act bynon-polar narcosis (Lipnick et al., 1985). As chain length increases,

azard and risk assessment of the OECD high production volumeol. Environ. Saf. (2008), doi:10.1016/j.ecoenv.2008.10.006

ARTICLE IN PRESS

H. Sanderson et al. / Ecotoxicology and Environmental Safety ] (]]]]) ]]]–]]] 5

hydrophobicity increases resulting in greater toxicity, and inparallel, solubility decreases. At a critical point, solubility becomeslower than expected toxicity and longer chain lengths show noacute toxicity. Chronic effects for such substances are also known;data indicate that effects are anticipated up to C15. For alcoholswith carbon numbers higher than C15, there are significantexperimental difficulties in producing, maintaining and quantify-ing exposures of the test substance. Even so, it is unlikely that theywould exhibit chronic toxicity because the relationship betweencarbon number and chronic toxicity suggests that the solubilityof the alcohol would limit the bioavailable dissolved fraction tosub-toxic concentrations (Schafers et al., 2008, this issue). In thisassessment, trends between aquatic toxicity and carbon chainlength are based on normal (non-propargylic) alcohols.

Acute toxicity is predicted well by category-specific (Q)SARs.Their success in predicting toxicity for category members suggeststhat the members do not possess any particularly unique features.Substances comprising a range of carbon chain lengths can bedealt with by appropriate addition of their individual toxic unitcontributions to the whole (Dyer et al., 2006). Effect concentra-tions vary across members of the category (Table 2).

No effects up to the limit of water solubility for single chainlengths 4C13–14 and for some multi-component substances areobserved. Moreover, no chronic effects are expected for singlechain lengths 4C15 up to limit of aqueous solubility. For Daphnia

magna which has the steepest dose–response relative to chainlength and is generally recognized as the most sensitive of thethree trophic levels tested, the subsequent acute (Q)SAR devel-oped (r2

¼ 0.98) is thus (Fisk et al., 2008, this issue):

EC50 Daphnia magnaðmmol=LÞ ¼ 1:92� 0:83 Log Kow (1)

hron

ic T

oxic

ity E

ffect

or W

ater

Sol

ubili

ty (µ

mol

)

0

1

2

3 Reproduction EC10

Reproduction NOEC

Water solubility

3.4.2. Chronic daphnia magna reproduction

Theory and practice in aquatic toxicology are established fortesting that occurs at or below the level of solubility. Atconcentration above the limit of solubility, physical effects enterinto observed responses of the organism, but do not reflect theinfluence of the chemical entering the body, target tissues or cells(ECETOC, 1996; Rufli et al., 1998). A Daphnia magna chronictoxicity value for C8 exists in the open literature (Kuhn et al.,1989). Further studies have been conducted for C10, 12, 14, 15.Studies with C14 and C15 were especially difficult as the predictedwater solubilities for these alcohols are very low and very close toexpected chronic effect concentrations. Thus, interpretation of theresponses can potentially be confounded due to a combination ofboth physical effects (e.g., entrapment of particles in feedingstructures, oil droplets and micro-emulsions coating organismsurfaces), and toxicity. It is a reality that separating these physicaleffects and those responses associated with chemical uptake orecotoxicity is not possible. However, it is possible to evaluatewhether test observations adhere to theory and thus allow theresults to assist in the inference of solubility being exceeded ornot. An example of this would be the expectation that a

Table 2Representative ranges of measured ecotoxicological effect concentrations (mg/L).

Fish acute LC50 D. magna acute

EC50

D. magna

chronic NOEC

Algae growth

rate

480–97,000a 800–200,000b 10–1000c 100–80,000d

a C14-C6.b C11-C6.c C14-C8.d C10�16-C6.

Please cite this article as: Sanderson, H., et al., An overview of hazchemical category—Long chain alcohols [C6–C22] (LCOH),. Ecotoxico

monotonic increase in toxicity would be observed as hydropho-bicity of a chemical series increases. Great care was taken withanalytical preparations for the chronic 21-day Daphnia magna

tests. Measurements of solubilities, particularly for alcohols ofhigher chain length (4C13), become increasingly difficult toconduct and increasingly variable. Predicted solubilities thenbecome useful to reduce the importance of variability in the dataand its interpretation, and to provide grounds for comparisonacross all compounds. Due to the extensive and rapid biodegrada-tion of alcohols during the conduct of aquatic toxicity tests,extreme care was taken to minimize the loss of test substancesduring the tests. Thus, slight modifications of the OECD 211 testguideline were introduced. The vessels containing the D. magna

were closed to reduce entry of bacteria from the atmosphere, andgently aerating the test vessels top layer to prevent unacceptablylow dissolved oxygen levels due to degradative losses of alcohol.Fig. 1 represents the measured chronic reproduction toxicityvalues as a function of hydrophobicity (i.e., log Kow). The datareveal a deflection in the toxicity–hydrophobicity (carbon chainlength-dependent) relationship with C14 as the most toxic chainlength.

The resulting chronic reproduction (Q)SAR (r2¼ 0.96) with a

cut-off at C14 is thus (OECD, 2006; Schafers et al., 2008, this issue):

Log NOECDaphnia magnaðmmol=LÞ ¼ 4:28� 1:03 Log Kow (2)

Emphasis in the risk assessment was placed on 21-d EC10values as opposed to NOECs due to the greater technicaldefensibility of EC10s to represent the exposure–response rela-tionship (OECD, 1998).

3.5. Mammalian toxicity

A review of the toxicological database for the category of theLCOH demonstrates that these materials are of a low order oftoxicity upon single or repeated exposure. Overall, the data showan inverse relationship between chain length and toxicity. Theshorter chain alcohols tend to induce more pronounced effectswhen compared to materials with a longer chain length. This isillustrated most clearly by the degree of local irritation in studiesinvolving single or repeat administration. LCOH have no skinsensitisation potential, are not mutagenic and have not shown any

3 4 5 6 7

log

C

-2

-1

log Kow (µmol)

Fig. 1. Chronic toxicity of long chain alcohols to Daphnia magna. Both 21-d NOEC

and EC10 for reproduction are indicated relative to water solubility. Brackets

indicate the range of measured exposure concentration in chronic toxicity studies

described in Schafers et al., 2008, this issue). Deflection in the structure-activity

relationship occurs as effects are observed at the limit of water solubility with

maximum toxicity observed at the C14 chain length.

ard and risk assessment of the OECD high production volumel. Environ. Saf. (2008), doi:10.1016/j.ecoenv.2008.10.006

ARTICLE IN PRESS

H. Sanderson et al. / Ecotoxicology and Environmental Safety ] (]]]]) ]]]–]]]6

adverse effects on fertility, development, and reproduction. Thekey human health hazards identified for this category are theirritative properties for skin and eye of the alcohols with chainlengths of C11 or below. These hazards are well characterized anddo not to lead tissue destruction or irreversible changes. Theyshould nevertheless be noted by chemical safety professionals andusers. On the basis that a clear relationship exists between chainlength and toxicological properties, substances with chain lengthsexceeding the upper range tested can be expected to possesstoxicological properties similar to those tested. The hydroxylgroup in alcohols confers upon the hydrocarbon chain aconsiderable degree of polarity, and hence affinity for water. It issusceptible to oxidation by metabolic processes. Linear oressentially linear hydrocarbon chains are also readily oxidisedmetabolically. No highly branched structures are included in thecategory reviewed in this paper. Substances that contain a numberof homologous components can be expected to behave in a wayconsistent with the carbon number distribution present (Fisket al., 2008, this issue; Veenstra et al., 2008, this issue; OECD,2006).

The typical no observed adverse effect levels (NOAELs)recorded for this category range between ca. 200 mg/kg bodyweight (BW)/day and 1000 mg/kg BW/day in the rat upon sub-chronic administration via the diet. The dermal NOAEL is42000 mg/kg BW/day. The maximum consumer exposure(dermal) is via body moisturisers (28 mg/kg BW/day). Therepresentable dermal absorption percentage is 32%, yielding aworst-case margin of exposure (MOE) ¼ 2000/28 mg/kg BW/day*32% absorbed equals a MOE factor of 333 (Veenstra et al., 2008,this issue; OECD, 2006). The exposure assessment was conductedaccording to the methods outlined by Sanderson et al. (2006) andSDA (2005).

4. Discussion

The recommendation from the OECD SIDS Initial AssessmentMeeting (SIAM) in April of 2006 regarding human health hazardsof LCOH was that the key hazards identified for the category arethe local irritative properties for skin and eye of alcohols withchain lengths of C11 or below. The category is thus of low priorityfor further work (OECD, 2006; Veenstra et al., 2008, this issue).

The recommendation regarding the environmental hazardsconcluded that all of the category members are rapidly biode-gradable, especially at environmentally relevant concentrations.Alcohols are metabolized and/or biotransformed in living organ-isms. This biotransformation suggests that bioaccumulationpotentials based on octanol–water partition coefficients may beoverestimates. Measured BCF data on a related alcohols categorysupports the concept that the bioaccumulation potential of thesesubstances will be lower than estimated from log Kow. Sixteen outof the 30 compounds had an acute aquatic toxicity below 1 mg/L,which leads to a default recommendation for further work by theOECD, if the compound is not an intermediate, regardless of anyother property associated to the compound (e.g. rapid biodegra-dation). Hence, member states, and others, are invited to conductan exposure assessment and, if necessary, a risk assessment(OECD, 2006). The papers provided in this journal issue providethat risk assessment.

LCOH tend to sorb to sediments after release to the environ-ment via wastewater treatment effluent. Trickling filter treatmentfacilities have the least effective removal rate of alcohols (98.8%)and also the relatively highest effluent concentrations (maximumtotal (C12–15) measured ¼ 4.92mg/L). Hence, the expected worst-case exposure scenario would be the sediments downstream atrickling filter treatment plant. Dyer et al. (2006) reported a

Please cite this article as: Sanderson, H., et al., An overview of hchemical category—Long chain alcohols [C6–C22] (LCOH),. Ecotoxic

sediment-dependent organism PEC/PNEC ¼ 0.03–0.07 for thecombined mixture of LCOH C12–18 downstream of a trickling filterplant (total sediment concentration ¼ 0.546mg/L), indicative oflow environmental risk (Belanger et al., 2008, this issue). In thisrelation it is important to recall Federle and Itrich (2006) resultson the rapid degradation of LCOH, with half-lives in sewagetreatment being less than 1 min. The conditions downstream thetrickling filter plant will also incur rapid degradation in situ and itis thus unlikely that LCOH will accumulate to levels that wouldcause physical stress to biota in or above the sediment undernormal environmental conditions. The focus on the C12�18

homologue range was chosen because this is range of greatestcommercial interest to the detergent industry and this rangeoverlaps the compounds that were recommended for further workby the OECD. Hazard screening is certainly an important first tierin evaluating the safety of chemicals. However, regulatorydecision-making and risk management based on hazard informa-tion only, and ignoring exposure and risk aspects, is relying oninformation fraught with significant scientific and extrapolationuncertainties. To a large extent, these uncertainties are the verysame uncertainties a sound and science-based risk assessmentaim to elucidate.

5. Conclusion

The main findings of the HPV LCOH category case-study are:

�

azaol.

No unacceptable human or environmental risks were identified.

equilibrium between synthesis and degradation by naturalprocesses.

(hydrophobicity) and toxicity as a demonstrated in chronictesting of the very rapidly biodegradable compounds in thiscategory—very close to the compounds’ limit of watersolubility.

enabling prediction of key properties across the entire category(C6–22).

HPV activities are that industry can come together and forminternational consortia to share resources resulting in minimizingthe need for additional animal tests, producing cost savings, andincreasing transparency and the scientific quality of the workeven for large and very complex categories of chemicals.

Acknowledgments

This work was funded by the Global ICCA Aliphatic AlcoholsConsortium. The Consortium appreciates the collaboration withDaniel Merckel (United Kingdom Environment Agency) and IanIndans (United Kingdom Human Safety Executive). Any studiesreferred to in this research article were conducted in accordanceto national and/or international guidelines for protection ofhuman subjects and animal welfare.

References

Belanger, S.E., Sanderson, H., Fisk, P.R., Schafers, C., Mudge, S.M., Willing, A., Kasai,Y., Nielsen, A.M. Dyer, S.D., Toy, R., 2008. Assessment of the environmental riskof long chain aliphatic alcohols. Ecotoxicol. Environ. Saf., this issue.

CEFIC, 2007. /http://www.cefic.be/activities/hse/mgt/hpv/hpvinit.htmS.CEPA, 1999. /http://www.ec.gc.ca/substances/ese/eng/dsl/dslprog.cfmSDunphy, J.C., Pressler, D.G., Morrall, S.W., Evans, K.A., Robaugh, D.A., Fujimoto, G.,

Negahban, A., 2001. Derivatization LC/MS for the simultaneous determination

rd and risk assessment of the OECD high production volumeEnviron. Saf. (2008), doi:10.1016/j.ecoenv.2008.10.006

ARTICLE IN PRESS

H. Sanderson et al. / Ecotoxicology and Environmental Safety ] (]]]]) ]]]–]]] 7

of fatty alcohol and alcohol ethoxylate surfactants in water and wastewatersamples. Environ. Sci. Technol. 35, 1223–1230.

Dyer, S.D., Sanderson, H., Waite, S.W., Evans, K.A., van Compernolle, R., DeCarvalho,A.J., Hooton, D.J., Price, B.B., Nielsen, A.M., Sherren, A.J., 2006. Assessment ofalcohol ethoxylate surfactants and fatty alcohols mixtures in river sedimentsand prospective risk assessment. Environ. Monitor. Assess. 120, 45–63.

ECETOC (European Centre for Toxicology and Ecotoxicology of Chemicals), 1996.Aquatic toxicity of sparingly soluble, volatile and unstable substances.Monograph 26, ECETOC Brussels, Belgium.

EU, 2001a. /http://ec.europa.eu/environment/chemicals/pdf/0188_en.pdfS.EU, 2001b. /http://europa.eu.int/comm/environment/chemicals/reach.htmS.Federle, T.W., Itrich, N.R., 2006. Fate of free and linear alcohol and alcohol

ethoxylate and derived fatty alcohol in activated sludge. Ecotoxicol. Environ.Saf. 64, 30–41.

Fisk, P.R., Wildey, R., Girling A.E., Sanderson, H., Belanger, S.E., Veenstra, G., Nielsen,A.M., Kasai, Y., Willing, A., Dyer, S.D., 2008. Environmental properties of longchain alcohols part I: physicochemical, environmental fate and acute aquatictoxicity properties. Ecotoxicol. Environ. Saf., this issue.

Goldman, 1998. /http://www.epa.gov/opptintr/library/pubs/archive/oppts_speeches/toxdoc3f.htmS.

Japan MOE, 2005. /http://www.env.go.jp/en/press/2005/0601b.htmlS.Kuhn, R., Pattard, M., Pernak, K.-D., Winter, A., 1989. Results of the harmful effects

of water pollutants to Daphnia magna in the 21 day reproduction test. WaterRes. 23, 501–510.

Leeming, R., Nichols, P., Nespolo, S., 1994. Distinguishing human and non-humansources of sewage using faecal sterols I: a survey of animal faeces from theSydney catchment. Report 94-AWT1, Sydney, Australia, AWT Science andEnvironment, Water Board, Sydney, 21pp.

Lipnick, R.L., Johnson, D.E., Gilford, J.H., Bickings, C.K., Newsome, L.D., 1985.Comparison of fish toxicity screening data for 55 alcohols with the quantitativestructure-activity relationship predictions of minimum toxicity for nonreactivenonelectrolyte organic compounds. Environ. Toxicol. Chem. 4, 281–296.

Metz, J.G., Pollard, R.M., Anderson, L., Hayes, T.R., Lassner, M.W., 2000. Purificationof a jojoba embryo fatty acyl-coenzyme A reductase and expression of its cDNAin high erucic acid rapeseed. Plant Physiol. 122, 635–644.

Morrall, S.W., Eckhoff, W.S., Evans, A., Cano, M.L., Dunphy, J.C., McAvoy, D.C., 2006.Removal of alcohol ethoxylates and environmental exposure determination inthe United States. Ecotoxicol. Environ. Saf. 64, 3–13.

MRI (Midwest Research Institute), 2004. MRI Project No. 310428. Analysis ofinterstitial water and sediment for surfactants by liquid chromatography/massspectrometry (LC/MS). Final Report to The Soap and Detergent Association,May 5, 2004. 127pp.

Please cite this article as: Sanderson, H., et al., An overview of hazchemical category—Long chain alcohols [C6–C22] (LCOH),. Ecotoxico

Mudge, S.M., Belanger, S.E., Nielsen, A.M., 2008. Fatty Alcohols—Anthropogenicand Natural Occurrence in the Environment. Royal Society of Chemistry,London, UK.

OECD (Organization for Economic Co-Operation and Development), 1998. Report ofthe OECD Workshop on Statistical Analysis of Aquatic Toxicity Data. OECDEnvironmental Health and Safety Publication Series on Testing and AssessmentNo. 10, Paris, France, 133pp.

OECD, 2004. /http://www.oecd.org/document/21/0,2340,en_2649_34379_1939669_1_1_1_1,00.htmlS.

OECD (Organisation for Economic Cooperation and Development) 2006. SIDSInitial Assessment Report for SIAM 22. Paris, France, 18–21 April 2006. TOME1: SIAR. Category: Long Chain Alcohols, 294pp. /http://www.oecd.org/document/63/0,2340,en_2649_34379_1897983_1_1_1_37465,00.htmlS.

Rufli, H., Fisk, P.R., Girling, A.E., King, J.M.H., Lange, R., Lejeune, X., Stelter, N.,Stevens, C., Suteau, P., Tapp, J., Thus, J., Versteeg, D.J., Niessen, H.J., 1998. Aquatictoxicity of sparingly soluble, volatile, and unstable substances and interpreta-tion and use of data. Ecotoxicol. Environ. Saf. 39, 72–77.

Sanderson, H., Counts, J., Greggs, W., Stanton, K., Sedlak, R., 2006. Exposure andprioritization—human screening data and methods for high productionvolume (HPV) chemicals in consumer products: amine oxides a case-study.Risk Anal. 26, 1637–1657.

Schafers, C., Boshof, U., Jurling, H, .Belanger, S.E., Sanderson, H., Dyer, S.D., Nielsen,A.M., Willing, A., Gamon, K., Kasai, Y. Fisk, P.R., Girling, A.E., 2008. Environ-mental properties of long chain alcohols. Part 2: structure-activity relationshipfor chronic aquatic toxicity of long-chained alcohols. Ecotoxicol. Environ. Saf.,this issue.

The Soap and Detergent Association (SDA), 2005. /http://cleaning101.com/files/Exposure_and_Risk_Screening_Methods_for_Consumer_Product_Ingredients.pdfS.

UN, 2003. /http://www.unece.org/trans/danger/publi/ghs/ghs_welcome_e.htmlS.USEPA, 1997. Clean Water Needs Survey Report to Congress: Assessment of

Needs for Publicly Owned Wastewater Treatment Facilities, Correction ofCombined Sewer Overflows, and Management of Storm Water and Non-PointSource Pollution in the United States. USEPA, Washington, DC, EPA/832/R-97/003, USA.

Veenstra, G., Webb, C., Sanderson, H., Belanger, S.E., Fisk, P., Nielsen, A.M., Kasai, Y.,Willing, A., Dyer, S.D., Stanton, K., Sedlak, R., 2008. Human health riskassessment for long chained aliphatic alcohols. Ecotoxociol. Environ. Saf., thisissue.

Veith, G.D., Lipnick, R.L., Russom, C.L., 1989. The toxicity of acetylenic alcohols tothe fathead minnow Pimephales promelas: narcosis and proelectrophileactivation. Xenobiology 19, 555–565.

ard and risk assessment of the OECD high production volumel. Environ. Saf. (2008), doi:10.1016/j.ecoenv.2008.10.006