analysis, occurrence and fate of mtbe in the aquatic environment over the past decade

Trends Trends in Analytical Chemistry, Vol. 25, No. 10, 2006

Analysis, occurrence and fate ofMTBE in the aquatic environmentover the past decadeMonica Rosell, Sılvia Lacorte, Damia Barcelo

In the past decade, it became progressively more evident that fuel oxygenate methyl tertiary butyl ether (MTBE) is nearly

ubiquitous in the worldwide environment. The frequency of detection of MTBE rivals other volatile organic compounds (VOCs)

that have been produced and used for a much longer time. Its mere presence in water bodies used as drinking-water reservoirs

(rivers, lakes or groundwater tables) has aroused concern about its potential sources, persistence and possible adverse effects

(aesthetic or toxic implications) for end users and aquatic life. This article aims to provide an updated overview of the analytical

techniques applied for current environmental concentrations, the occurrence of MTBE as a pollutant in the different aquatic

compartments, the relevance of diffuse and point sources and the different options for remediation of MTBE-contaminated sites.

ª 2006 Elsevier Ltd. All rights reserved.

Keywords: Diffuse source; Environmental; Gas chromatography; GC; Methyl tertiary butyl ether; MTBE; Point source; Water

Abbreviations: AED, Atomic emission detection; ATD, Automated thermal desorption sampler; BP, British Petroleum; BTEX, Benzene, toluene,

ethylbenzene and xylenes; CAA, Clean Air Act; CSIA, Compound-specific stable isotope analysis; DAI, Direct aqueous injection; DIPE, Diisopropyl

ether; ETBE, Ethyl tertiary butyl ether; EU, European Union; FID, Flame ionization detector; GC, Gas chromatography; GW, Groundwater; HS,

Headspace; IS, Internal standard; LUST, Leaking underground storage tank; MS, Mass spectrometry; MTBE, Methyl tertiary butyl ether; P&T, Purge

and trap; PID, Photoionization detector; P-THREE, ‘‘Removal of Persistent Polar Pollutants through improved treatment of wastewater effluents’’;

RFG, Reformulated gasoline; RON, Research octane number; SMCL, Secondary maximum contaminant level; SPME, Solid-phase microextraction;

t1/2, Half-life time; TAME, Tertiary amyl methyl ether; TBA, Tertiary butyl alcohol; TBF, Tertiary butyl formate; tert-, Tertiary; USEPA, US

Environmental Protection Agency; VOC, Volatile organic compound; WATCH, ‘‘Water catchment Areas: Tools for management and Control of

Hazardous compounds’’; WHO, World Health Organization; WWTP, Wastewater-treatment plant.

Monica Rosell,

Sılvia Lacorte*, Damia Barcelo

Department of Environmental

Chemistry, IIQAB-CSIC. Jordi

Girona 18-26

E-08034 Barcelona

Catalonia, Spain

*Corresponding author.

Tel.: +34 93 400 61 69;

Fax: +34 93 204 59 04;

E-mail: [email protected]

1016

1. Introduction

Since the late 1970s, large amounts(about 20 million tonnes) of methyl ter-tiary (tert-) butyl ether (MTBE) have beenproduced worldwide each year. Of thischemical production, 98% is used as anadditive in petrol. MTBE consumptionworldwide has been dominated by USA (asshown in Fig. 1) mainly to meet the oxy-gen requirements mandated in 1990 byClean Air Act (CAA) Amendments inareas where certain air-quality standards(related to CO or O3) have not been at-tained. By contrast, MTBE was incorpo-rated in European gasoline as octaneenhancer to replace banned tetraalkyl leadcompounds and as a result of increasingrestrictions on aromatics content.

Table 1 compares the chemical andphysical properties of MTBE with those of

0165-9936/$ - see front matter ª 2006 Elsev

other common fuel oxygenates and aro-matic hydrocarbons. In general, alcoholsand ethers have higher water solubilities,lower Henry’s Law constants and lowersorption constants than aromatics.Among fuel additives, MTBE is the etherwith more extreme values that favored itshigher mobility (nearly as fast as that ofgroundwater) and the difficulty of remov-ing it from water by aeration or degrada-tion processes [1].

With such production, use and proper-ties, it is not surprising that MTBE releasedto the environment has adversely affectedthe quality of water. Its responsibility fortaste and odor problems in drinking wateris well documented, as are concerns aboutpossible adverse effects on human health.

In view of this alarm, several environ-mental, health and government institutionshave prepared their own risk-assessment

ier Ltd. All rights reserved. doi:10.1016/j.trac.2006.06.011

mailto:[email protected]

US61%

Canada & Mexico4%

South America4%

Eastern Europe3%

Western Europe12%

Middle East4%

Asia12%

Figure 1. MTBE global use in 1999. The total worldwide annualproduction was about 21 million tonnes (Data from: MTBE/Oxygenates Clean Fuels newsletter (2000), DeWitt & CompanyInc., Houston, Texas, USA).

Trends in Analytical Chemistry, Vol. 25, No. 10, 2006 Trends

studies, and new MTBE regulation is required in Europe[2]. Since 1997, the US Environmental Protection Agency(USEPA) has had an MTBE drinking water advisory at20–40 lg/L based on aesthetic (taste and odor) criteria [3].For some time, it was expected that USEPA would adopt afederal secondary maximum contaminant level (SMCL),probably at 15 lg/L, for MTBE, in line with lower consumeracceptance [4].

However, the World Health Organization (WHO)decided not to establish a health-based guideline valuefor MTBE because any such value based on any adverseeffects would be significantly higher than the concen-tration at which it would be detected by odor [5].

MTBE-toxicity effects on freshwater and marineorganisms have been found at concentrations (mg/L)that hardly ever happen in the environment [6,7].However, the presence of MTBE can substantiallyenhance the toxicity of other pollutants, such as pesti-cides, which are often present in the same waters [8].

For all these reasons and its omnipresence in watersamples all over the world during the past decade,international attention has focused on the environmen-tal behavior of MTBE. Determination of MTBE levels anddistribution in natural waters (groundwater and surface)and soil became a challenging task because conventionalanalytical methods, such as liquid-liquid extraction(LLE), were not feasible. Moreover, the concentrations ofMTBE in water differ by several orders of magnitudebetween environmental background (ng/L) and sitesaffected by point sources (mg/L), which therefore requiredifferent analytical strategies.

This article aims to provide an updated overview of theanalytical techniques, environmental occurrence andfate of MTBE in studies over the past decade and tocompare them with some of our results from completedEU projects:

� WATCH (‘‘Water catchment Areas: Tools for manage-ment and Control of Hazardous compounds’’); and,� P-THREE (‘‘Removal of Persistent Polar Pollutants

through improved treatment of wastewater effluents’’).

2. Analysis

As volatile organic compounds (VOCs), fuel oxygenatesare almost exclusively analyzed by gas chromatography(GC) and mass spectrometry (MS) detection, which is inall cases the best option due to its selectivity and sensi-tivity. Other detectors, such as flame ionization (FID),photoionization (PID) and atomic emission (AED), can beused for screening purposes, although they are not sosensitive.

The most critical step in the trace analysis of thesepolar VOCs is definitely enrichment. Comprehensive re-views from Schmidt et al. [9,10] and recently from Ati-enza et al. [11] discussed advantages and disadvantagesin using several enrichment and injection techniques,including direct aqueous injection (DAI), membrane-introduction mass spectrometry (MIMS), headspaceanalysis (HS), purge and trap (P&T), solid-phasemicroextraction (SPME) by direct immersion or head-space and the emerging tool in environmental sciences,compound-specific stable isotope analysis (CSIA). Bothsets of authors agreed that the choice of an appropriatemethod depends on the matrix to be investigated, theconcentration ranges to be analyzed, the available lab-oratory equipment and the need for compliance withregulations. P&T and SPME are recommended for lowerconcentrations and obtained the best values for accuracyin a MTBE inter-laboratory comparison from 20 Euro-pean laboratories [12].

Whereas USEPA and other multi-compound methodsrely on the use of one or two generic internal standards(ISs), commonly fluorobenzene, the specific analysis ofMTBE can benefit from the use of deuterated standards,such as MTBE-d3, if MS detection is used [13–16].Recently, Tanabe et al. [17] proposed to employMTBE-d12 instead of MTBE-d3 because the presence ofcarbon bisulfide in the samples can interfere withquantification of m/z 76. Tert-amyl methyl ether(TAME) was also suggested as a cheaper IS when itspresence is not expected in the environmental samples[14].

3. Source characterization (gasoline composition)

Detailed knowledge of the oxygenate type and fraction ingasoline is essential in any attempts to estimate thepotential local or regional environmental impacts ofusing oxygenated fuel [18].

http://www.elsevier.com/locate/trac 1017

Table 1. Physico-chemical properties of MTBE, its main degradation products and other common gasoline additives/octane enhancers

Gasoline additive/substance Abbreviation CAS n� Molecularweight(g/mol)

BlendingRON

Boilingpoint(�C)

Solubilityin water(mg/L)

Henry’sLaw Constant(atm Æ m3)/(g Æ mol)

Vaporpressure(mm Hg at 25�C)

Log Kow

ETHER OXYGENATESMethyl tert-butyl ether MTBE 1634-04-4 88 116a–118b 55 51,000 5.87 · 10�4 250.00 0.94Ethyl tert-butyl ether ETBE 637-92-3 102 118a,b 73 12,000 1.39 · 10�3 124.00 1.92tert-Amyl methyl ether TAME 994-05-8 102 109b–111a 86 2,640 2.68 · 10�3 75.20 1.92Diisopropyl ether DIPE 108-20-3 102 nd 69 8,800 2.28 · 10�3 149.00 1.52

ALCOHOL OXYGENATESMethanol MeOH 67-56-1 32 125b–133a 65 complete 4.55 · 10�6 127.00 �0.77Ethanol EtOH 64-17-5 46 129a–130b 78 complete 5.00 · 10�6 59.30 �0.31

DEGRADATION PRODUCTStert-Butyl alcohol TBA 75-65-0 74 105b 82 complete 9.05 · 10�6 40.70 0.35tert-Butyl formate TBF 762-75-4 102 nd 83 11,200 6.90 · 10�4 86.40 1.19

AROMATIC HYDROCARBONSBenzene B 71-43-2 78 98a 80 1,790 5.55 · 10�3 94.80 2.13Toluene T 108-88-3 106 124a 111 526 6.64 · 10�3 28.40 2.73Ethylbenzene E 100-41-4 106 124a 136 169 7.88 · 10�3 9.60 3.15m-Xylene 108-38-3 106 162a 138 161 7.18 · 10�3 8.29 3.20p-Xylene X 106-42-3 106 155a 139 162 6.90 · 10�3 8.84 3.15o-Xylene 95-47-6 106 126a 144 178 5.18 · 10�3 6.61 3.12

nd: no data available.All data at 25�C, obtained from Syracuse Research Corporation PhysProp Database (free access under www.syrres.com/esc/physdemo.htm) except the Research Octane Number (RON) valuesobtained from:a Department of Information and Computing Sciences (University of Utrecht, The Netherlands): http://www.cs.uu.nl/wais/html/na-dir/autos/gasoline-faq/part2.html.b European Fuel Oxygenates Association (EFOA): www.efoa.org/EFOA_Pages/02_What/02b_Propertie.html.

Tren

ds

Tren

ds

inA

nalytical

Chem

istry,V

ol.

25,

No.

10,

2006

1018

http

://ww

w.elsevier.co

m/lo

cate/trac

http://www.syrres.com/esc/physdemo.htm

http://www.cs.uu.nl/wais/html/na-dir/autos/gasoline-faq/part2.html

http://www.efoa.org/EFOA_Pages/02_What/02b_Propertie.html


Reformulated gasoline (RFG) represented almost 30%of all gasoline sold in the USA. The CAA required 2.0%(w/w) minimum oxygen content, which was mainlyaccomplished by an MTBE content of about 11% (v/v),but also by a mixture of other oxygenates, such asTAME, ethyl tert-butyl ether (ETBE) or diisopropyl ether(DIPE).

In 2000, in a study conducted within the State of NewHampshire, TAME was found in 88% of all gasolinesamples at a mean volume of 1.2% and ETBE in 51% ofthe samples at 0.5% [19]. However, in recent years,MTBE has been phased out in many US states andsubstituted by ethanol.

By contrast, the EU allows a maximum oxygen con-tent of 2.7% (w/w) and up to 15% ethers with P 5carbon atoms (v/v) [20], so the average content of MTBE

Toluene; 8.59

o- xylene; 3

Ethers; 8.8

Benzene

m+p- xylene; 5.62

Ethylbenzene; 2.97

Toluene; 15.31

o- xylen

Ethers; 10

Ben

m+p- xylene; 4.52

Ethylbenzene; 2.40

A

B

Figure 2. Evolution in the composition of British Petroleum (BP) gasoline inand (B) May 2003). The additive concentrations were expressed as percen2003/17/CE levels: up to 1% benzene, 42% total aromatics and 15% ethe

in European gasoline is quite low (about 2%), but greatdifferences can be found between countries and gasolinegrades [21] (e.g., Achten et al. [22] measured averageMTBE content in German regular (0.4%), Eurosuper(0.4–4.2%), super premium unleaded (9.8%), andOptimax (11.9%) gasoline by GC-FID methodology).

Fast methods to determine MTBE in gasoline roughlyare usually based on MS detection of m/z 73 when non-separative techniques are applied (direct injection orheadspace generation) [23]. However, GC separation ordilution of the matrix is necessary to avoid backgroundinterferences, which often resulted in overestimations[23,24] (e.g., TAME has also m/z 73).

In 2003, we employed consecutive dilutions inmethanol and organic-free water and a previously per-formed P&T-GC-MS method [15] to analyze different

MTBE; 7.35

.86

DIPE; 0.31ETBE; 1.22

7

; 1.07

MTBE; 1.99

ETBE; 8.97

e; 3.13

.96

zene; 0.98

Spain during 2003 (e.g., gasoline RON grade 97 in (A) January 2003tage in volume % (v/v) and complied with maximum EU Directivers.



gasoline Research Octane Number (RON) grades sold byBritish Petroleum (BP) in Barcelona, Spain. First resultsin January 2003 showed MTBE contents from 1%(RON95) to 7% (RON97 and 98), but, in March 2003,we detected lower amounts – 0.06–4% – accompaniedwith higher ETBE and toluene percentages (see Fig. 2).These discrepancies were later explained as a progressivechange of gasoline composition.

In Spain, as other European countries, MTBE issubstituted by ETBE due to tax incentives for using bio-mass-derived ethanol, which is synthesized to producethe ethyl ether group of ETBE [25], as the EuropeanCommission is expected to forbid MTBE during the yearahead.

MTBE has also received much attention as a chemicalrequiring serious investigation in Japan [17], where theoil industry is currently exploring the possibility ofblending ETBE into gasoline to mitigate CO2 emissionsfrom the road-transport sector [26].

Accidental spills and leakage from corroded tanks atgasoline stations and refineries are the main sources ofMTBE being released into the environment. From anRFG release, the MTBE solubility was estimated to be oneorder of magnitude lower than from the pure compound,but still 200–1600 times higher than the solubilities ofany BTEX [27]. Given the varying levels of MTBE indifferent types of gasoline and countries, the risk thatMTBE poses to the environment should include risksfrom specific pollution episodes as, in time, these spreadcontamination in water. The occurrence of MTBE hasbeen the focus on studies in USA and Europe (mostlycarried out in Germany). MTBE levels in backgroundenvironmental samples are generally below 2 lg/L;when higher concentrations are found, this is an indi-cation of an unknown point source.

4. Environmental fate

4.1. PrecipitationPartitioning between air and water is normally assumedto be the primary process affecting the occurrence ofVOCs in precipitation samples. Due to MTBE character-istics, storm-water run-off and atmospheric transport arelow contributors to the water concentrations of thispollutant, as shown in several occurrence, transport andmodeling studies. In air, MTBE degradation is expectedto be fast (half-life, t1/2, 3–6 days) depending mainly onthe hydroxyl radical (OH) concentration, which is con-sidered much more of a determinant than photolysis orreaction with ozone or other radicals [1]. In all cases,tert-butyl formate (TBF) was observed to be the majordegradation product.

In 1991–95, Delzer et al. [28] detected MTBE in 7% ofUS municipal storm-water samples (up to 8.7 lg/L) so itwas the seventh most frequently found VOC. The

1020 http://www.elsevier.com/locate/trac

reporting level for MTBE in that study was 1 lg/L, so itwas likely that lower detection limits would haveresulted in higher occurrence.

That was the challenge of Achten et al. [13], whodeveloped HS-SPME-GC-MS to detect MTBE in waterfrom 0.01 lg/L. Following that, a sampling campaignwas carried out in winter 2000–01 at several loca-tions in Germany. Rainwater collectors were placed onthe top of buildings to avoid direct vehicle emissions.MTBE varied according to spatial distribution (higheroccurrence in urban (86%) than in rural (18%) pre-cipitation samples) and climatic conditions (detectableat temperatures lower than 10–15�C) [29]. However,the highest values (0.03–0.085 lg/L), detected in thecentre of the city of Frankfurt am Main, were twoorders of magnitude lower than values formerly mea-sured in USA. These differences were mainly explainedby the lower and constant year-round MTBE percent-age in gasoline in Germany. In addition, the analysisof urban run-off and corresponding rain-water samplesrevealed that about 20% of MTBE originated fromatmospheric (air and precipitation) transport, whereasabout 80% may be attributed to direct uptake fromvehicle emissions and leakage near the road duringprecipitation.

Extending this investigation and applying the sameanalytical method, Kolb and Puttmann [30] measuredMTBE levels in snow samples (up to 0.6 lg/L) at thesame locations during the following two winter seasons.Since only 4 g of snow were required for the analysis, thecollection time could be kept short to avoid post-depo-sitional processes and the melting snow was transferredas soon as possible to vials to minimize volatile losses.Comparison with the previous rainwater samples indi-cated atmospheric transfer of MTBE from urban to ruralareas preferentially in winter due to lower atmosphericdegradation rates and suggested that snow more effec-tively scavenged MTBE from the atmosphere than rain.

Within the framework of the WATCH project, a two-year monitoring program was carried out in the vicinityof an airport located in the Southern Iberian Peninsula.MTBE and related compounds were analyzed in sevensampling campaigns (April 2002–August 2004) tocheck their occurrence in different environmental waterbodies. A total of 25 run-off rainwater samples from theairport runways showed a mean MTBE value of 0.15 lg/Lwith higher values, up to 1.40 lg/L, in summer periodswhen the density of passengers was assumed to begreater. This seasonal trend was also observed for BTEX,especially in July 2003 with levels in the range0.9–26 lg/L while, in January 2003, the maximum was0.5 lg/L. Although tert-butyl alcohol (TBA) presentedmore variable behavior, higher values were found inboth October campaigns (2002 and 2003) just after themore active periods and they may have originated fromatmospheric degradation of MTBE and TBF.


4.2. GroundwaterInfiltration of precipitation and dispersion from urbanatmosphere can act as a non-point source in trans-porting MTBE and other VOCs into shallow ground-water [31]. One of the first studies that pointed out thepotential occurrence and persistence of MTBE ingroundwater tables was a survey conducted in 1993–94 as part of the US Geological Survey’s NationalWater-Quality Assessment [32]. Among 60 VOCs ana-lyzed by P&T-GC-MS, MTBE (at a reporting level of0.2 lg/L) was the second most frequently detectedchemical (after chloroform) in shallow ambientgroundwater samples collected in urban areas. Re-cently, Moran et al. [33] statistically examined an up-dated data compilation from the first 10-year cycle ofthis study. MTBE showed a total detection frequency of7.6% (higher than trichloroethylene at 4.5%, whichhas been produced for much longer) and a medianconcentration around 0.3 lg/L. Only 0.3% groundwa-ter samples exceeded the MTBE lower limit of theUSEPA drinking water advisory (20 lg/L). The proba-bility of detecting MTBE in groundwater was stronglyassociated with urban land-use, population density, useof MTBE in gasoline, and recharge rates. Other etheroxygenates, such as TAME or DIPE, were less fre-quently detected (0.25% and 0.19%, respectively) andETBE has not been found so far.

In Germany, a similar groundwater-monitoring pro-gram was undertaken in 1999–2000, but the use ofP&T-GC with ion-trap MS allowed a lower limit ofdetermination (0.05 lg/L) [34]. The study concludedthat MTBE was regularly present (almost 50%) ingroundwater under urban areas although the medianconcentration was low (0.17 lg/L).

However, the most significant episodes of ground-water contamination by MTBE resulted from pointsources, such as accidental spills during transport andmanipulation of gasoline or leaking underground stor-age tanks (LUSTs) in petrol stations or refineries. Oncethere, MTBE moves at velocities similar to localgroundwater, biodegrades slowly (abiotic processes areconsidered negligible) and has low sorption. In the USA,the city of Santa Monica lost 50% of its total watersupply in 1996 as a result of high (up to 230 mg/L)groundwater contamination by MTBE from LUSTs [35].The annual cost for water replacement was estimatedaround $4 m and culminated in the ban of MTBE inCalifornia. Johnson et al. [27] estimated the t1/2 ofMTBE in LUST sites to be at least 2 years, but 10 yearsmight be necessary to reduce concentrations belowclean-up levels in the USA. Later on, Shih et al. [36]evaluated the impact of fuel hydrocarbons and oxy-genates over 868 LUST sites in Los Angeles, California.MTBE was detected in 83% of them with a medianconcentration of 1.2 mg/L (benzene and TBA showedsimilar findings).

A wide range of MTBE concentrations (0.120–830 mg/L) have been also reported in pollutedgroundwater tables in European countries [21]. Similarresults were obtained from our study sites in Spain,Austria and Germany by applying a P&T-GC-MSmethodology, as shown in Table 2. After four years of agasoline release in Girona (Spain), MTBE levels werestill higher than the USEPA drinking water advisory(40 lg/L) [15], and, in a spill in Dusseldorf, the con-centration of MTBE did not appreciably decrease duringa two-year monitoring program and reached maximumspot values above the toxicity level (350 lg/L) sug-gested in Denmark [37]. This last study also revealedhigh variations in MTBE and TBA concentrations in thevertical profile, thus the need for multilevel wells toimprove risk assessment.

4.3. Surface water4.3.1. Rivers. The multifunctional use of rivers (sourceof drinking water, sewage disposal or ship carrier) hasaroused concern about the potential sources, persistenceand removal rates of MTBE before the water arrives atend users. The t1/2 of MTBE in rivers is highly variable(from seconds to months) mainly affected by the vola-tilization processes that depend on water velocity, depth,temperature and wind speed.

MTBE concentrations in German rivers were higher aturban agglomerations (maximum of 2.36 lg/L) com-pared to rural areas [38]. These results correlated withpreviously analyzed precipitation samples [29]. Similarfindings were obtained in another sampling programalong the river Rhine carried out by Baus et al. [39].Although concentrations of MTBE tend to be balanced inthe course of the river (by dilution and evaporation),illegal releases from tanker ships during washing andindustrial discharges have been reported as major MTBEinputs that generate periodic ‘‘waves’’ of the pollutant(e.g., 14 lg/L were detected by chance) in time andspace [39]. In addition, levels at the bank side whereindustries are located are higher than in mid-river [40].

In addition, in Japan, MTBE levels were analyzed insome rivers using an improved P&T-GC-MS (purgingtemperature optimized at 40�C, with cryo-focusing at�180�C) allowing for a low limit of detection (0.003 lg/L)[17]. Increases in MTBE were observed in going fromupstream to the river mouth as were higher levels inwinter than in summer, and these observations wereconsistent with other studies.

4.3.2. Lakes. The discovery of MTBE in lakes used forrecreational boating and reservoirs has raised concernsover the potential impact on the quality of drinkingwater from such water bodies. Multiple-use lakes in USA,such as Donner Lake, located in the Sierra NevadaMountains, California [41], or Cranberry Lake in NewJersey [42], have been analyzed mainly by P&T-GC-MS.


Table 2. Overview of maximum reported concentrations (expressed in lg/L) of MTBE and related compounds in different environmental water bodies by different analytical methods over thepast decade in comparison with EU project (WATCH and P-THREE) results

Location Country Sample type Analytical method MTBE ETBE TAME DIPE TBA BTEX Ref.

PrecipitationSeveral locations Germany Precipitation HS-SPME-GC/MS 0.09 na na na na na [29]Frankfurt/M. Germany Road runoff water HS-SPME-GC/MS 1.17 na na na na na [29]Airport vicinity Southern Iberian Peninsula Platform runoff water P&T-GC/MS 1.40 0.15 0.50 0.02 0.83 35 WATCHSeveral locations US Storm water P&T-GC/MS 8.70 na na na na 15 [28]Several locations Germany Snow HS-SPME-GC/MS 0.60 na na na na na [30]

GroundwaterPetrol station, Salzburg Austria Groundwater P&T-GC/MS 3.32 0.04 nd 0.01 0.41 0.45 WATCHPetrol station, Girona Spain Groundwater P&T-GC/MS 48 nd nd 0.03 8.86 1.43 [15]a

Petrol station, Dusseldorf Germany Groundwater P&T-GC/MS 645 nd 0.08 0.17 440 0.2 [37]a

Petrol station Germany Groundwater P&T-GC/ion-trap MS 730 na na na na na [34]Refinery site, Tarragona Spain Groundwater P&T-GC/MS 666 0.68 nd 1.53 62 4,121 [15]a

Refinery site, East Germany Germany Groundwater P&T-GC/MS 215,000 nd nd nd 37,000 920 WATCHSeveral GW sites (maximum in UK) EU countries Groundwater several 830,000 na na na na na [21]Santa Monica, CA US Groundwater unknown 230,000 na na na na na [35]LUST sites in Los Angeles, CA US Groundwater P&T-GC/MS 1.6 · 107 7,500 12,000 4,700 4.4 · 106 4.2 · 107 [36]Niigata Prefecture Japan Groundwater P&T-GC/MS 5.90 na na na na na [17]

Surface waterRiver Rhine (in Dusseldorf) Germany River P&T-GC/MS 0.12 nd nd 0.08 0.51 0.1 WATCHRiver Rhine (in Koln) Germany River P&T-GC/MS 0.15 <0.01 <0.01 <0.01 0.4 <0.01 P-THREERiver Rhine (median) Germany River HS-SPME-GC/MS 0.25 na na na na na [38]River Rhine (median) Germany River P&T-GC/ion-trap MS 0.26 na na na na na [39]Several rivers Germany River HS-SPME-GC/MS 2.36 na na na na na [38]Several rivers Germany River P&T-GC/ion-trap MS 14 na na na na na [39]Rivers in northern Italy Italy River HS-SPME-GC/MS 0.15 na na na na na [56]Niigata Prefecture rivers Japan River P&T-GC/MS 5.30 na na na na na [17]San Gabriel river, CA US Stream P&T-GC/MS 52 na na na na na [46]Tegeler See Germany Lake P&T-GC/MS 0.16 <0.01 <0.01 <0.01 0.21 <0.01 P-THREELake Zurich Switzerland Lake HS-SPME-GC/FID 1.40 na na na na 3.90 [43]Donner Lake, CA US Lake P&T-GC/MS 12 na na na na na [41]Cranberry Lake, NJ US Lake P&T-GC/MS & FID 31 na na na na na [42]

Tren

ds

Tren

ds

inA

nalytical

Chem

istry,V

ol.

25,

No.

10,

2006

1022

http

://ww

w.elsevier.co

m/lo

cate/trac

Table 2 (continued)

Location Country Sample type Analytical method MTBE ETBE TAME DIPE TBA BTEX Ref.

Wastewater3 Catalonian WWTP Spain Influent P&T-GC/MS 0.40 0.04 0.04 0.02 200 30 P-THREE

Effluent P&T-GC/MS 6.34 1.32 <0.01 nd 1.79 2.50 P-THREE3 German WWTP Germany Influent P&T-GC/MS 0.18 <0.01 <0.01 nd 1.62 0.75 P-THREE

Effluent P&T-GC/MS 0.17 <0.01 <0.01 nd 0.66 <0.01 P-THREE1 Austrian WWTP Austria Influent P&T-GC/MS 121 nd nd <0.5 215 705 P-THREE

Effluent P&T-GC/MS 5.60 nd nd 5.43 0.39 0.20 P-THREE1 Belgian WWTP Belgium Influent P&T-GC/MS 0.11 <0.01 <0.01 nd 0.95 0.01 P-THREE

Effluent P&T-GC/MS 0.08 <0.01 <0.01 nd 0.51 <0.01 P-THREENiigata Prefecture (n = 2) Japan Influent P&T-GC/MS 0.03 na na na na na [17]

Effluent P&T-GC/MS 0.02 na na na na na [17]Frankfurt/M-Niederrad & Sindlingen Germany Influent HS-SPME-GC/MS 1.27 na na na na na [38]Southern California US Effluent P&T-GC/MS 123 na na na na na [46]

Drinking waterUnknown Italy Mineral water HS-SPME-GC/MS <0.01 na na na na na [56]Unknown Italy Tap water HS-SPME-GC/MS 0.40 na na na na na [56]Frankfurt/M. Germany Tap water HS-SPME-GC/MS 0.07 na na na na na [38]Leuna/Spergau Germany Tap water HS-SPME-GC/MS 0.70 na na na na na [55]Big German city (1) Germany Tap water P&T-GC/MS 0.09 <0.01 <0.01 <0.01 nd <0.01 P-THREEBig German city (2) Germany Tap water P&T-GC/MS 0.01 <0.01 <0.01 <0.01 nd <0.01 P-THREESmall Belgian city (rural area) Belgium Tap water P&T-GC/MS 0.01 nd <0.01 <0.01 nd <0.02 P-THREEUnknown The Netherlands Drinking water well P&T-ATD-GC/MS 2.90 na na na na na [53]Santa Monica, CA US Production well unknown 610 na na na na na [35]

Sea waterAirport vicinity Southern Iberian Peninsula Coastal water P&T-GC/MS 40 0.09 0.19 0.02 12 55 WATCHAlmeria/Malaga Spain Coastal water P&T-GC/AED & MS 1,842 na na na 600 na [48]Tamar Estuary (harbors/marinas) UK Coastal water HS-SPME-GC/MS 0.19 na na na na na [49]Marina del Rey harbor, CA US Coastal water direct-SPME-GC/MS 18 na na na na na [47]Mission Bay, CA US Coastal water P&T-GC/MS 34 na na na na 1.9 [46]Santa Monica Bay (Chevron), CA US Refinery discharge P&T-GC/MS 1,878 na na na na na [46]

nd: not detected, na: not analyzed or no data available.a These publications were also part of WATCH project.

Tren

ds

inA

nalytical

Chem

istry,V

ol.

25,

No.

10,

2006

Tren

ds

http

://ww

w.elsevier.c

om

/locate/trac

1023


In Europe, Lake Zurich, which supplies drinking water tothe largest Swiss city [43], was investigated by applyingHS-SPME-GC-FID. In general, detected levels of MTBE inthese studies were in the range 0.03–31 lg/L, similar torivers.

The use of motorized watercraft was, in all cases,the major contribution as a source of MTBE, whereasneither highway run-off nor precipitation contributedsignificantly.

Spatial and temporal variations of MTBE concentra-tions in the lakes were observed associated with thermalstratification during the boating season (summer), whichslowed the exchange and transport of MTBE. Schmidtet al. [43] concluded that they did not expect any risk tothe drinking-water supply from such lakes if water wasextracted from well below the thermocline; but, in orderto further reduce emissions of unburned fuel into surfacewater, restrictions of highly emitting two-stroke enginestypes used in motorboats should be considered.

4.3.3. Wastewater. Most abiotic elimination tech-niques, normally used in wastewater-treatment plants(WWTPs), such as ozonation or adsorption on granularactivated carbon, are not very effective for removingMTBE or its main degradation product, TBA [39,44].These limitations may generate additional problems forwater suppliers and regulators, since TBA may be con-sidered even more toxic than its parent compound [45].

Achten et al. [38] estimated that roughly 30–35% ofMTBE was eliminated in two German sewage plants.This value was slightly lower than the EU risk-assessment calculation (43%) [2], the difference beingmainly attributed to evaporation and dilution muchmore than adsorption to the sludge or biodegradationprocesses (considered negligible). In fact, the influent ofthe sewage plant, which collected mostly industrialdischarges, was characterized by receiving someexceptionally high MTBE concentrations (e.g., 1.27 lg/L),and spot samples during these events showed higheramounts in the effluent than in the influent [38].

Within the framework of P-THREE project, two sam-pling campaigns were carried out in February and May2003 with the aim of screening the presence and theremoval of different organic pollutants in eight EuropeanWWTPs and some related tap waters from cities close by(two big German cities with more than 100,000inhabitants and one small rural city in Belgium). Despitethe limited number of samples, MTBE was detected in 15of the 16 wastewater samples at median values of0.12 lg/L and 0.08 lg/L for influent and effluent waters,respectively. This data demonstrated no evident removalof the compound, which accorded with a study carriedout in Japan [17]. These estimates excluded the highestvalues of MTBE, TBA and aromatic hydrocarbons, whichwere detected in WWTPs in Austria, probably due to theproximity of a refinery (see Fig. 4 and Table 2).


Although refinery effluents generally contained thehighest MTBE concentrations, discharges from WWTPsaccounted for the greatest proportion (78%) of the dailymass emitted to bays and coastal waters in southernCalifornia [46].

4.4. Sea waterLimited data is available on the extent of MTBE con-tamination in coastal waters, as well as on the persis-tence of the pollutant in the marine environment toassess potential toxic effects on marine life.

Brown et al. [46] calculated that large point sources(WWTPs and refineries) throughout southern Californiadischarged 214 kg/day of MTBE to coastal waters, ofwhich 98% arrived in Santa Monica Bay, whereas theinput from streams was considered trivial (<0.5%).Marinas and areas used intensively for recreationalboating had the highest average MTBE concentration(8.8 lg/L).

Later, Zuccarello et al. [47] focused on one of thesezones, the harbor of Marina del Rey, where personalwatercraft were allowed. As expected, the highest con-centration of MTBE (18 lg/L) was found at the boat-launching ramp and the lowest (0.2 lg/L) near theharbor entrance (2.3 km away). Despite the volatility ofMTBE, similar concentrations along the depth profile(0–6 m) suggested that vertical mixing in the watercolumn was more efficient than volatilization.

For the first time in Europe, Mezcua et al. [48]determined MTBE and TBA in coastal water samplesfrom various marinas in the south of Spain (Almeria andMalaga), involving P&T-GC and comparing two detectorsAED and MS. AED was not sensitive enough for currentenvironmental concentrations (MTBE detection limit of10 lg/L), but validated alarm points. GC-MS detectedMTBE in all the samples at levels generally in the range0.033–2.20 lg/L, but occasionally higher (up to 1.842mg/L) in the vicinity of gasoline stations or boat-launching facilities.

Much lower levels were measured by Guitart et al.[49] with HS-SPME-GC-MS in pre-selected potentialcontaminated harbors and marinas throughout TamarEstuary in UK. However, the elevated levels (up to0.19 lg/L) were generally associated with motor vehicleand boating activities. Run-off from road and rail bridgeswas identified as major inputs of MTBE in the lowerestuary.

From our study in the southern Iberian Peninsula,four points along the coast were sampled at low andhigh tide through seven campaigns to obtain morerepresentative data. The median value of MTBE from atotal of 38 samples was 0.37 lg/L and a comparablelevel (0.23 lg/L) was found for TBA. Lower amounts ofBTEX were usually detected (0.09 lg/L as median).Exceptionally, high values of all gasoline additives (checkTable 2) were detected at one point in July 2003,

Run-off (n=25) Domestic waste (n=42) Coastal (n=48)

0.0

0.5

1.0

1.5

38

39

40

Water bodies

MT

BE

con

c. [µ

g/L]

Apr.02 Jul.02 Oct.02 Jan.03 Jul.03 Aug.04

0.0

0.5

1.0

1.5

10

20

30

40

50

MT

BE

con

c. [µ

g/L]

Sampling dates

B

A

Figure 3. Plot of MTBE concentrations found in the vicinity of an airport (A) at different water bodies and (B) detailed for coastal water samples(n = 8) in seven sampling campaigns. For each variable, the box has lines at the lower quartile (25%), median quartile (50%), and upper quartile(75%) values. The whiskers are the lines extending from each end of the box to show the extent of the data up to 1.5 times the inter-quartile range(IQR). Mean value is marked (j) and outliers ( ). Each sample (n) was analyzed in triplicate and the average value was considered for calcu-lations. Non-detected levels were expressed as half the instrumental limit of detection (5 · 10�4 lg/L).


probably associated with recreational boating activities.In fact, slightly higher values of MTBE were foundduring summer compared to winter or spring periods (asshown in Fig. 3B).

5. Human exposure via drinking water

Several studies have tried to estimate the human uptakewhen MTBE-contaminated water is used for drinking,preparing food or showering (e.g., 1% of drinking watersupplies in the USA contains MTBE above 20 lg/L [35]and it was estimated that, via potable water, 5% of theUS population may be exposed to levels higher than2 lg/L of MTBE [50]).

Williams [51] reported results from a 1995–2000survey of MTBE in drinking waters in California, whichis supposed to be the state where MTBE has had thegreatest impact, and where it was detected in about1.3% of all drinking-water samples and 27% of themabove state’s primary health-based standard of 13 lg/L.

In Europe, some studies have been carried out to checkthe presence of MTBE in drinking water and corre-sponding sources. In UK, Dottritge et al. [52] reporteddetectable concentrations (>0.1 lg/L) at 13% of studiedlocations. However, MTBE levels were predominantlylow (<1 lg/L) and the study concluded that the presenceof less than 1% MTBE (v/v) on average in British gaso-line was not a major threat to public water supplies inEngland and Wales.


RT: 0.00 - 49.99

0 5 10 15 20 25 30 35 40 45Time (min)

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

Re

lativ

e A

bun

da

nce

RT: 11.36 - 15.25

11.5 12.0 12.5 13.0 13.5 14.0 14.5 1

Time (min)

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

Rel

ativ

e A

bund

ance

1

2

4

IS2

5

6

78

9

IS3

RT: 0.00 - 49.99

0 5 10 15 20 25 30 35 40 45

Time (min)

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

Re

lativ

e A

bun

da

nce

RT: 11.36 - 15.25

11.5 12.0 12.5 13.0 13.5 14.0 14.5 1

Time (min)

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

Rel

ativ

e A

bund

ance

1

2

4

IS2

5

6

78

9

IS3

RT: 0.00 - 49.99

0 5 10 15 20 25 30 35 40 45Time (min)

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

Re

lativ

e A

bun

da

nce

2

3

IS2

54

IS3

RT: 0.00 - 49.99

0 5 10 15 20 25 30 35 40 45

Time (min)

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

Re

lativ

e A

bun

da

nce

2

3

IS2

54

IS3

A

B

Figure 4. Total ion chromatograms (TIC, 108) of the (A) influent and (B) effluent from a wastewater treatment plant (WWTP) in Austria analyzed inFebruary 2003 by P&T-GC/MS. The high concentrations of gasoline additives detected were assumed to originate from a nearby refinery. Com-pound identification numbers: 1 = TBA (m/z = 59), 2 = MTBE-d3 (IS1, m/z = 76) + MTBE (m/z = 73), 3 = DIPE, 4 = Benzene, IS2 = Fluorobenzene,5 = Toluene, 6 = Ethylbenzene, 7 = m + p-Xylene, 8 = o-Xylene, 9 = Dicyclopentadiene (DCPD), IS3 = 1,2-dichlorobenzene-d4.




A similar survey was carried out in The Netherlandsin 2001. Morgenstern et al. [53] developed an off-lineP&T coupled to a GC-MS equipped with an automatedthermal desorption (ATD) sampler that enabled theanalysis of at least 40 samples per day and an MTBEquantification limit of 0.02 lg/L. MTBE concentrationswere in the range <0.01–0.42 lg/L in Dutch drinkingwater sources with a median below 0.01 lg/L. Thehighest value, 2.9 lg/L, was associated with point-source contamination of groundwater.

In Germany, about 15% of the drinking water used isproduced by riverbank filtration or artificial infiltration, soits quality directly depends on the state of the rivers(primarily the Rhine and the Elbe). Some studies havepointed out that MTBE is not totally removed by this sand-filtration technique and at least 40% of the pollutantpasses unchanged through the subsoil [39]. MTBE wasfound at an average concentration of 0.09 lg/L inrecovered well water and riverbank filtered waters, and ata maximum level of 0.07 lg/L in tap water from themetropolitan area of Frankfurt [54]. A comparable value(0.09 lg/L) was detected in tap water collected in Berlin ina P-THREE survey. Recently, Kolb et al. [55] found oneorder of magnitude MTBE higher concentration in tapwater from Leuna and Spergau (Saxony-Anhalt), whichwas probably influenced by the well-known gasoline-contaminated aquifer at the Leuna chemical industrialzone nearby.

Some data are also available from Italy, where MTBEwas not detected in 12 commercial mineral watersamples (i.e. <0.01 lg/L), whereas five tap waters fromdifferent groundwater sources had MTBE in the range0.05–0.40 lg/L, all measured by HS-SPME-GC-MS [56].

In conclusion, it has been demonstrated that, in gen-eral, no aesthetic implications (taste and odor) or healthrisks are likely to be associated with chronic andsub-chronic human exposure to MTBE in tap water.However, for point sources, risk-assessment studies areneeded, especially because consumers may find unac-ceptable the mere presence of gasoline components indrinking-water supplies.

6. Remediation actions

Although the procedures are often difficult and timeconsuming, several methods have been proposed forremoving MTBE from contaminated sites, including:� physical removal (e.g., granular activated carbon, soil

vapor extraction, air-stripping, selected zeolites, ultra-sonic irradiation combined with ozonation or ozone/hydrogen peroxide treatment; and,� biological treatment by means of microbial consortia

or plants (phytoremediation).To start with, MTBE was classified as resistant to

biodegradation processes because its removal was much

slower than that of conventional gasoline hydrocarbons[1]. However, during the past decade, the potential ofmicrobial and fungi communities to degrade MTBE hasbeen demonstrated under oxic and nearly all anoxicconditions, as summarized in several reviews [57–59].When MTBE and TBA removal by conventional tech-nologies is not easily achieved [39,44], new, simpler, lessexpensive alternatives (e.g., ex-situ reactors, naturalattenuation and bioaugmentation) are envisaged andcan be successfully applied for remediation of MTBE-contaminated aquifers [60,61].

7. Data treatment and modeling studies

The behavior of MTBE through the different environ-mental compartments can be investigated using model-ing approaches. However, models developed so far havediffered in their predictions of relative MTBE concentra-tions for relevant environmental compartments and ofseasonal variations in concentration; further, they havehardly considered the formation of transformationproducts [62]. Moreover, limitations in pollutantenvironmental data or key physico-chemical parametersoften make it difficult to validate the predictions of themodels.

Achten et al. [63] simulated a German environmentusing the equilibrium criterion (EQC) model. MTBE con-centrations of 0.02 lg/L in surface water and 0.17 lg/m3

in air were estimated from the year-round scenario at10�C. Lower MTBE concentrations in atmospheric andaqueous compartments in summer were explained byhigher degradation rates at higher temperatures.

Arp et al. [62] recently performed more accurateanalysis, taking into account the two major degradationproducts of MTBE, TBA and TBF, and it was used topredict their concentrations in various environmentalcompartments in Europe. Water and air concentrationsof MTBE predicted from this innovative multi-speciestransformation model were considered to be in goodagreement with measurements of environmentalsamples (e.g., the predicted average MTBE concentrationin surface water (0.25 lg/L) at 10�C correspondedexactly with the median found in river Rhine [38]).MTBE concentrations were found to be strongly influ-enced by temperature (in water and air) or hydroxyl(OH) radical levels (only in air). However, the lack ofbackground information about MTBE in soils in Europeand the scarce data on degradation products preventedfurther validation of the model.

8. Future perspective

The place of MTBE as the top gasoline additive seemsclose to expiry, but how long it will be detected in the



environment? Have the responsible authorities learnedfrom past errors?

In the absence of completely new design andconstruction of underground storage-tank systems, theextent of potential human and environmental exposureshould be an important criterion in determining theamount of information needed before making anenvironmental policy decision.

Alternatives to MTBE, such as ETBE, are quite similarin structure. Although ETBE has been studied less, pre-liminary results from using a Level III fugacity approachmodel [64] showed that, despite the differences in theirpartitioning properties (Table 1), in general, both ethershave similar behavior in the environment when theemission rates being evaluated are the same. However,ETBE emissions would be expected to be less evaporativedue to the relatively lower vapor pressure of ETBE.

In water, the taste and odor thresholds of ETBE (47and 13 lg/L, respectively) are almost identical to thosefor MTBE [35], so, due to tax incentives for using bio-mass-derived ethanol, ETBE will be the next fuel-derivedcontaminant to emerge in the future. At least, ETBE hasa higher Henry’s Law constant than MTBE (up to 2–3times higher), indicating that an air-stripping removaltechnique would be at least slightly more effective thanfor MTBE [65] and ETBE biodegradation has been alsodemonstrated with several strains [66]. But TBA is alsothe main degradation product of ETBE. Site-groundwaterconcentrations and plume-length data have alreadyindicated TBA contamination from ETBE to be on a scalesimilar to that from MTBE in LUST sites [36]. Since TBAcan be stoichiometrically formed from MTBE, ETBE andTBF degradation and may be considered as resistant totreatment as MTBE, TBA concentrations in water bodiescould pose the greatest problem in the future. So, due tothe widespread use of ETBE, further investigation ofETBE and TBA will be required very soon.

Acknowledgements

Some results shown in the present review came fromEU projects [WATCH (EVK1–CT–2000–00059) andP-THREE (EVK1-2001-00283)] funded by the EUEnvironment and Sustainable Development sub-programand from the Spanish Ministerio de Educacion y CienciaProject EVITA (CTM2004-06255-CO3-01). M. Rosellacknowledges a grant from Department of Universities,Research and Information Society, La Generalitat deCatalunya (2005FIR 00348).

References

[1] P.J. Squillace, J.F. Pankow, N.E. Korte, J.S. Zogorski, Environ.

Toxicol. Chem. 16 (1997) 1836.


[2] Finnish Environment Institute, EUR 20417 EN - European Union

Risk Assessment Report, tert-butyl methyl ether, Office for Official

Publications of the European Communities, Luxembourg, 2002.

[3] US Environmental Protection Agency, Drinking water advisory:

Consumer acceptability advice and health effects analysis on

methyl tertiary-butyl ether (MTBE), Office of Water, Washington,

DC, USA, EPA-822-F-97-009, 1997.

[4] A.J. Stocking, I.H. Suffet, M.J. McGuire, M.C. Kavanaugh, J. Am.

Water Works Assoc. 93 (2001) 95.

[5] World Health Organization, Methyl Tertiary-Butyl Ether (MTBE)

in Drinking-water, background document for development of

WHO Guidelines for Drinking-water Quality, WHO/SDE/WSH/

05.08/122, 2005.

[6] I. Werner, C.S. Koger, L.A. Deanovic, D.E. Hinton, Environ. Pollut.

111 (2001) 83.

[7] G.A. Rausina, D.C.L. Wong, W.R. Arnold, E.R. Mancini, A.E. Steen,

Chemosphere 47 (2002) 525.

[8] M.D. Hernando, M. Ejerhoon, A.R. Fernandez-Alba, Y. Chisti,

Water Res. 37 (2003) 4091.

[9] T.C. Schmidt, H.-A. Duong, M. Berg, S.B. Haderlein, Analyst

(Cambridge, UK) 126 (2001) 405.

[10] T.C. Schmidt, Trends Anal. Chem. 22 (2003) 776.

[11] J. Atienza, P. Aragon, M.A. Herrero, R. Puchades, A. Maquieira,

Critical Rev. Anal. Chem. 35 (2005) 317.

[12] R. Schuhmacher, M. Fuhrer, W. Kandler, C. Stadlmann, R. Krska,

Anal. Bioanal. Chem. 377 (2003) 1140.

[13] C. Achten, A. Kolb, W. Puttmann, Fresenius’ J. Anal. Chem. 371

(2001) 519.

[14] L. Zwank, T.C. Schmidt, S.B. Haderlein, M. Berg, Environ. Sci.

Technol. 36 (2002) 2054.

[15] M. Rosell, S. Lacorte, A. Ginebreda, D. Barcelo, J. Chromatogr., A

995 (2003) 171.

[16] P. Dewsbury, S.F. Thornton, D.N. Lerner, Environ. Sci. Technol.

37 (2003) 1392.

[17] A. Tanabe, Y. Tsuchida, T. Ibaraki, K. Kawata, A. Yasuhara, T.

Shibamoto, J. Chromatogr., A 1066 (2005) 159.

[18] D. Barcelo, M. Petrovic, Trends Anal. Chem. 24 (2005) 275.

[19] F.J. McGarry, Proc. 2nd Eur. Conf. MTBE, Barcelona, Spain, 2004,

p. 17.

[20] European Union, Off. J. EU L76 (2003) 10.

[21] T.C. Schmidt, E. Morgenroth, M. Schirmer, M. Effenberger,

S.B. Haderlein, in: A.F. Diaz, D.L. Drogos (Editors), Oxygen-

ates in Gasoline: Environmental Aspects, American Chemical

Society, Washington, DC, USA, 2002, p. 58.

[22] C. Achten, W. Puttmann, J. Chromatogr., A 910 (2001) 377.

[23] J.L. Perez-Pavon, M.D. Sanchez, C.G. Pinto, M.E.F. Laespada,

B.M. Cordero, J. Chromatogr., A 1048 (2004) 133.

[24] H. Kanai, V. Inouye, R. Goo, R. Chow, L. Yazawa, J. Maka, Anal.

Chem. 66 (1994) 924.

[25] European Union, Off. J. EU L123 (2003) 42.

[26] H. Hara, Representative of Japan Petroleum Energy Center (JPEC),

Brussels office, Personal communication, 2006.

[27] R. Johnson, J. Pankow, D. Bender, C. Price, J. Zogorski, Environ.

Sci. Technol. 34 (2000) 210A.

[28] G.C. Delzer, J.S. Zogorski, T.J. Lopes, Abstr. Papers Am. Chem. Soc.

213 (1997) 100.

[29] C. Achten, A. Kolb, W. Puttmann, Atmos. Environ. 35 (2001)

6337.

[30] A. Kolb, W. Puttmann, Atmos. Environ. 40 (2006) 76.

[31] J.F. Pankow, N.R. Thomson, R.L. Johnson, A.L. Baehr, J.S. Zogorski,

Environ. Sci. Technol. 31 (1997) 2821.

[32] P.J. Squillace, J.S. Zogorski, W.G. Wilber, C.V. Price, Environ. Sci.

Technol. 30 (1996) 1721.

[33] M.J. Moran, J.S. Zogorski, P.J. Squillace, Ground Water 43 (2005)

615.

[34] J. Klinger, C. Stieler, F. Sacher, H.J. Branch, J. Environ. Monit. 4

(2002) 276.


[35] US Environmental Protection Agency, Achieving Clean Air and

Clean Water: The Report of the Blue Ribbon Panel on Oxygenates

in Gasoline, EPA, Washington, DC, USA, EPA420-R-99-021,

1999.

[36] T. Shih, Y. Rong, T. Harmon, M. Suffet, Environ. Sci. Technol. 38

(2004) 42.

[37] M. Rosell, S. Lacorte, C. Forner, H.P. Rohns, R. Irmscher, D. Barcelo,

Environ. Toxicol. Chem. 24 (2005) 2785.

[38] C. Achten, A. Kolb, W. Puttmann, P. Seel, R. Gihr, Environ. Sci.

Technol. 36 (2002) 3652.

[39] C. Baus, H. Hung, F. Sacher, M. Fleig, H.J. Brauch, Acta

Hydrochim. Hydrobiol. 33 (2005) 118.

[40] C. Baus, F. Sacher, H.J. Brauch, Ozone Sci. Eng. 27 (2005) 27.

[41] J.E. Reuter, B.C. Allen, R.C. Richards, J.F. Pankow, C.R. Goldman,

R.L. Scholl, J.S. Seyfried, Environ. Sci. Technol. 32 (1998)

3666.

[42] L. Toran, C. Lipka, A. Baehr, T. Reilly, R. Baker, Water Res. 37

(2003) 3756.

[43] T.C. Schmidt, S.B. Haderlein, R. Pfister, R. Forster, Water Res. 38

(2004) 1520.

[44] R.A. Deeb, K.H. Chu, T. Shih, S. Linder, I. Suffet, M.C. Kavanaugh,

L. Alvarez-Cohen, Environ. Eng. Sci. 20 (2003) 433.

[45] J.D. Cirvello, A. Radovsky, J.E. Heath, D.R. Farnell, C. Lindamood,

Toxicol. Ind. Health 11 (1995) 151.

[46] J.S. Brown, S.M. Bay, D.J. Greenstein, W.R. Ray, Mar. Pollut. Bull.

42 (2001) 957.

[47] J.L. Zuccarello, J.A. Ganske, D.B. Green, Chemosphere 51 (2003)

805.

[48] M. Mezcua, A. Aguera, M.D. Hernando, L. Piedra, A.R. Fernandez-

Alba, J. Chromatogr., A 999 (2003) 81.

[49] C. Guitart, J.M. Bayona, J.W. Readman, Chemosphere 57 (2004)

429.

[50] B.R. Stern, R.G. Tardiff, Risk Anal. 17 (1997) 727.

[51] P.R.D. Williams, Environ. Forensics 2 (2001) 75.

[52] J. Dottridge, M. Hall, S. Firth, A Review of Current MTBE Usage

and Occurrence in Groundwater in England and Wales, Research

and Development Technical Report P406, Environment Agency,

Bristol, UK, 2000.

[53] P. Morgenstern, A.F.M. Versteegh, G.A.L. de Korte, R. Hoogerbrugge,

D. Mooibroek, A. Bannink, E.A. Hogendoorn, J. Environ. Monit. 5

(2003) 885.

[54] C. Achten, A. Kolb, W. Puttmann, Environ. Sci. Technol. 36

(2002) 3662.

[55] A. Kolb, W. Puttmann, Environ. Pollut. 140 (2006) 294.

[56] F. Piazza, A. Barbieri, F.S. Violante, A. Roda, Chemosphere 44

(2001) 539.

[57] R.A. Deeb, K.M. Scow, L. Alvarez-Cohen, Biodegradation 11

(2000) 171.

[58] F. Fayolle, J.P. Vandecasteele, F. Monot, Appl. Microbiol. Biotech-

nol. 56 (2001) 339.

[59] T.C. Schmidt, M. Schirmer, H. Weiß, S.B. Haderlein, J. Contam.

Hydrol. 70 (2004) 173.

[60] A.J. Stocking, R.A. Deeb, A.E. Flores, W. Stringfellow, J. Talley,

R. Brownell, M.C. Kavanaugh, Biodegradation 11 (2000) 187.

[61] E. Zanardini, C. Pisoni, G. Ranalli, M. Zucchi, C. Sorlini, Ann.

Microbiol. 52 (2002) 207.

[62] H.P.H. Arp, K. Fenner, T.C. Schmidt, Environ. Sci. Technol. 39

(2005) 3237.

[63] C. Achten, W. Puttmann, J. Klasmeier, J. Environ. Monit. 4

(2002) 747.

[64] S. Valtchev, M. Bittens, H.P.H. Arp, T.C. Schmidt, Proc. 2nd Eur.

Conf. MTBE, Barcelona, Spain, 2004, p. 23.

[65] US Environmental Protection Agency, Oxygenates in Water:

Critical Information and Research Needs, Office of Research and

Development, EPA, Washington, DC, USA, EPA/600/R-98/048,

1998.

[66] F. Fayolle, G. Hernandez, F. Le Roux, J.-P. Vandecasteele,

Biotechnol. Lett. 20 (1998) 283.


analysis, occurrence and fate of mtbe in the aquatic environment over the past decade

Documents