Identification and Responsibility of 2,4,6-Tribromoanisole in Musty, Corked Odors in Wine

Article abstract of DOI:10.1021/jf030632f

In this work, gas phase chromatography analysis coupled with selective selected ion monitoring (SIM) identified 2,4,6-tribromoanisole (TBA) in wines found on tasting to have significant musty or corked character, although they did not contain noteworthy quantities of chloroanisoles or chlorophenols, the contaminants generally reported to cause this type of defect. The perception thresholds were studied, together with contamination conditions during winemaking, storage, and bottle-aging. A musty off-odor was perceptible on smelling wine containing as little as 4 ng L-1 TBA, and spoilage may be detected by retro-olfaction at even lower concentrations. TBA, produced by O-methylation of its direct precursor, 2,4,6-tribromophenol, generally comes from sources in the winery environment. This paper is the first to identify the sources of a large number of cases of wines polluted during storage in premises where the atmosphere was contaminated with TBA used recently to treat wood, or originating from much older structural elements of the winery, or from used wooden containers. In certain cases, although the initial source had been eliminated, residual pollution adsorbed on walls could be sufficient to make a building unsuitable for storing wooden barrels and plastics, as well as corks, which have been found to be particularly susceptible to contamination by the TBA in the winery atmosphere.

Full text of DOI:10.1021/jf030632f

J. Agric. Food Chem. 2004, 52, 1255−1262 1255
Identification and Responsibility of 2,4,6-Tribromoanisole in
Musty, Corked Odors in Wine
PASCAL CHATONNET,* SANDRA BONNET, STEÄ PHANE BOUTOU, AND
MARIE-DOMINIQUE LABADIE
Laboratory EXCELL, Parc Innolin, 10 rue du Golf, 33700 Merignac, France
In this work, gas phase chromatography analysis coupled with selective selected ion monitoring (SIM)
identified 2,4,6-tribromoanisole (TBA) in wines found on tasting to have significant “musty or corked”
character, although they did not contain noteworthy quantities of chloroanisoles or chlorophenols,
the contaminants generally reported to cause this type of defect. The perception thresholds were
studied, together with contamination conditions during winemaking, storage, and bottle-aging. A
“musty” off-odor was perceptible on smelling wine containing as little as 4 ng L-1 TBA, and spoilage
may be detected by retro-olfaction at even lower concentrations. TBA, produced by O-methylation of
its direct precursor, 2,4,6-tribromophenol, generally comes from sources in the winery environment.
This paper is the first to identify the sources of a large number of cases of wines polluted during
storage in premises where the atmosphere was contaminated with TBA used recently to treat wood,
or originating from much older structural elements of the winery, or from used wooden containers. In
certain cases, although the initial source had been eliminated, residual pollution adsorbed on walls
could be sufficient to make a building unsuitable for storing wooden barrels and plastics, as well as
corks, which have been found to be particularly susceptible to contamination by the TBA in the winery
atmosphere.
KEYWORDS: 2,4,6-Tribromoanisole; 2,4,6-tribromophenol; musty odor; wine; wood preservation;
atmosphere
INTRODUCTION
scribed in detail (5). Excessive use of hypochlorite-based
disinfectants may also lead to the contamination of organic
materials in contact with wine or the cellar atmosphere after
the formation of 2,4,6-trichlorophenol (TCP) (6), and its
breakdown by certain filamentous fungi (7, 8). TCA has a
relatively low olfactory detection threshold in water [300 pg
L-1 (9) and 30 pg L-1 (10)]. Depending on the type of wine,
changes in odor and, above all, retro-olfaction were significant
between 1.5 and 3 ng L-1 (11). TeCA is less odoriferous [4 ng
L-1 in water (9)], but noticeable spoilage has been reported at
concentrations above 10 ng L-1 in still wines and 5 ng L-1 in
sparkling wines (12).
However, some wines have exactly the same type of tasting
fault but do not contain sufficient quantities of chloroanisoles
to account for this “musty” character. Soleas et al. (13) reported
that only a small fraction (27%) of a large number of wine
samples analyzed contained a significant quantity of TCA (>2
ng L-1). However, these authors did not assay TeCA, so it is
not possible to draw any conclusions concerning the involvement
of other chloroanisoles.
“Musty” odors are some of the most unpleasant organoleptic
defects in wines and attract strong criticism from connoisseurs
and consumers alike. Several molecules have been identified
as giving wines this unpleasant smell. Among these, chloro-
anisoles, especially 2,4,6-trichloroanisole (TCA) and 2,3,4,6-
tetrachloroanisole (TeCA), are the most frequently identifiable
in wines criticized on tasting as “musty” or “corked”. The term
“corked”, meaning that spoilage is due to the natural cork
traditionally used as a bottle stopper, is frequently inappropriate.
In fact, although corks, made from the bark of cork oak Quercus
suber, may release TCA into wines if the quality of the raw
material or the manufacturing process is unsatisfactory (1-3),
other sources of chloroanisole pollution have also been identi-
fied. Pentachloroanisole (PCA), which has relatively little odor,
and especially 2,3,4,6-tetrachloroanisole (TeCA) are produced
by the biochemical breakdown of certain pesticides containing
2,3,4,6-tetrachlorophenol (TeCP) or pentachlorophenol (PCP),
with TeCP as an impurity. These compounds may contaminate
wine that has not been in contact with cork (4, 5). The
breakdown mechanisms of these precursors and the conditions
for distant contamination via the atmosphere have been de-
Exhaustive identification and assay in our laboratory of the
various chloroanisoles in a large number of wines found to have
off-odors on tasting made it possible to identify situations
leading to a high frequency of spoilage when neither TCA nor
TeCA was detectable at sufficiently high levels to account for
the defect.
* To whom correspondence should be addressed. Fax: +33 557 920
215. E-mail: pchatonnet@libertyurf.fr.
10.1021/jf030632f CCC: $27.50 © 2004 American Chemical Society
Published on Web 02/14/2004

1256 J. Agric. Food Chem., Vol. 52, No. 5, 2004
Chatonnet et al.
Table 1. Characteristics of the Halophenol and Haloanisole Assays by GC/MS-SIM
matrix
wines (ng L-1
TCA
TCP
TeCA
TeCP
PCA
PCP
TBA
TBP
TeBP
PBP
)
detection limit
quantification limit
uncertainty ±
0.95
2.95
3.90
1.25
4.80
5.20
0.85
3.30
3.20
2.10
6.00
5.00
0.95
5.00
4.40
3.60
8.75
4.90
0.90
2.00
3.50
1.50
3.50
5.50
a
a
a
a
a
a
solid materialsb (ng g-1
detection limit
)
0.45
2.40
2.00
2.45
7.45
3.50
0.90
3.20
1.60
0.55
4.10
1.60
1.95
5.20
2.20
1.40
5.90
3.80
0.40
2.00
2.00
2.55
7.55
3.60
a
a
a
a
a
a
quantification limit
uncertainty ±
a Not studied. b And control of atmospheres by static trapping.
the validation protocol for a standard method described in the OIV
repository (15) and the French standards for intralaboratory valida-
tion: NF ISO 5725-1 and NF V 03-110 (16a,b), as described previously
by Chatonnet et al. (14) (Table 1).
In this work, we identify 2,4,6-tribromoanisole (TBA) for
the first time in wines with a significant “musty” character,
which contained insufficient quantities of chloroanisoles to
produce this defect. The perception threshold and conditions
for contamination during winemaking, storage, and bottle-aging
are specified.
The wine (200 mL) was centrifuged (15 min at 5000g) and put into
a 200 mL borosilicate glass graduated vial deactivated by silanizing
with hexamethyldisilazane, and then exactly 1 mL of an internal
standard solution in absolute ethanol containing 0.1 mg L-1 TBP-d2 as
well as 0.05 mg L-1 TCA-OCD3, TeCA-OCD3, PCA-OCD3, and TBA-
d5 was added. The contents of the vial were homogenized and
transferred to a 250 mL Erlenmeyer flask, and then ∼1 mL of 1/3
sulfuric acid was added using a graduated pipet to acidify the medium,
as well as a Teflon stir bar (6 g, 25 mm) previously rinsed with ethanol.
Then 10 mL of a 1:1 dichloromethane/n-pentane blend (Pestipur, SDS)
was added and stirred for 5 min (250 rpm). The contents of the
Erlenmeyer flask were then static settled in a separating funnel rinsed
with dichloromethane. The organic phase was collected carefully in a
sealed 50 mL flask. The solution was extracted twice more using 5
mL of solvent. The organic phases were combined, the emulsion broken
by asymmetrical agitation, and the aqueous phase eliminated by means
of a disposable Pasteur pipet. The organic phase was dried on anhydrous
sodium sulfate (Aldrich) and then transferred into a 100 mL Zymark
concentrating tube; the flask was rinsed carefully twice with 1 mL of
dichloromethane/pentane. The solution was concentrated using a
Zymark Turbovap II, operating at 25 ( 2 °C in a steam of nitrogen
(quality I, Air Products) at 1 bar. It was concentrated automatically to
0.5 mL in ∼15 min. The sample was transferred to a 500 µL dispos-
able injection flask and kept at -20 °C in a dark place prior to
analysis.
The dry materials (corks, barrel wood, wood from roof timbers,
cardboard, chipboard. etc.) were shredded in a granulating mill with a
stainless steel bowl. The mill was decontaminated between samples
by soaking it in a detergent solution (5% Decon 90) with added ethanol
(50% vol) and dried overnight in an oven at 105 °C. The material was
grated onto a sheet of aluminum foil.
A material sample weighing 0.5-5 g depending on the contamination
level ((0.01 g) was put into an 150 mL Erlenmeyer flask stopped with
emery and 100 µL of internal standard (2,4,6-tribromophenol-d2 at 1.180
mg L-1 in absolute ethanol), 70 mL of dichloromethane, and 0.1 mL
of glacial acetic acid (Rectapur, Prolabo) were added to facilitate
extraction of the halophenols. It was stirred for 120 min at room
temperature with a stir bar (25 mm, PFTE, 6 g, 250 rpm). The extract
was purified by continuous extraction with ethanol in a Soxhlet for 12
h, rapidly filtered through cellulose paper to eliminate any cork dust,
concentrated to 0.5 mL, and then transferred to a flask for storage as
described above.
EXPERIMENTAL PROCEDURES
Materials and abbreviations: 2,4,6-trichoroanisole (TCA) [CAS
Registry No. 87-40-1], 99%, Aldrich; 2,4,6-trichlorophenol (TCP) [88-
06-2], >98%, Aldrich; 2,3,4,5-tetrachloroanisole [938-86-3], 98% (used
as equivalent of 2,3,4,6-TeCA but with exactly the same retention time
under the analysis conditions used); 2,3,4,6-tetrachlorophenol (TeCP)
[58-90-2], 95%, Lancaster; 2,3,4,5,6-pentachloroanisole (PCA) [1825-
21-4], 98%, Aldrich; pentachlorophenol (PCP) [87-86-5], 98%, Aldrich;
2,4,6-tribromoanisole (TBA) [607-99-8], >99%, Aldrich; 2,4,6-tribro-
moanisole-d5, 99%, CDN Isotopes Inc.; 2,4,6-tribromophenol (TBP)
[118-79-6], 98%, Aldrich; 2,4,6-tribromophenol-d2, 98%, CDN Isotopes
Inc.; pentabromophenol (PBP) [608-71-9], Aldrich; deuterated io-
domethane CD3I [865-50-9], >99.5%, isotope enrichment, CDN
Isotopes Inc.
Synthesis of Deuterated Chloroanisole Analogues. A blend of pure
chlorophenol (TCP, TeCP, or PCP) (6 mmol), anhydrous potassium
carbonate (5 mmol), and deuterated iodomethane was prepared in 10
mL of dimethyl sulfoxide (Pestipur, SDS). It was stirred for 3 h at
room temperature The mixture was then poured into 50 mL of dilute
hydrochloric acid N. The solution was extracted three times with 50
mL of diethyl ether (Pestipur, SDS). The ether phases were static-
stettled, combined, and then rinsed with 250 mL of purified distilled
water (MilliQ), sodium hydroxide N, and then water again. The mixture
was dried on anhydrous sodium sulfate (Rectapur, Prolabo), and the
ether was slowly evaporated at 38 °C in a Kuderna-Danish concentrator
equipped with a 10-plate reflux column and a nickel helix. The solid
residue was recrystallized in ethanol. The deuterated chloroanisole on
the methoxy group was purified by adsorption chromatography on a
50 × 0.5 cm flash chromatography column filled with activated silica
(activity V). The synthesis product was rinsed with n-pentane (SDS,
Pestipur > 99.5%) and eluted (5 mL/min) with a 96:4 blend of pentane/
diethyl ether (SDS, Pestipur > 99%).
The solvent was evaporated in a cold vacuum rotary evaporator,
and the purity of the solid diluted in n-pentane was measured by gas
phase chromatography (DB5-MS J&W column, 30 m × 0.25 mm, phase
thickness ) 0.25 µm) coupled with a mass spectrometer (HP 5973-II)
operating in electron impact (70 keV, 150 °C) and scan mode (40-
400 uma). The mass spectrum obtained from the analogues was as
expected, showing that the purity of the 2,4,6-TCA-OCD3, 2,3,4,6-
TeCA-OCD3, and PCA-OCD3 synthesized was >99% of the total signal
(scan mode).
Simultaneous Assay of Chlorophenols and Chloroanisoles in
Wines and Dry Materials. The chlorophenols and chloroanisoles in
the wines and dry materials were assayed simultaneously using a method
derived from that previously described by Chatonnet et al. (14). The
halophenol assay was carried out using 2,4,6-tribromophenol-d2 (TBP-
d2) as an internal standard, and the haloanisoles were assayed using
the deuterated chloroanisole analogues -OCD3 and 2,4,6-TBA-d5 for
2,4,6-tribromoanisole. The proposed analysis method was assessed using
Atmospheric Quality Control by Static Trapping on an Inert
Adsorbent. The technique used was described by Chatonnet et al. (5)
and Chatonnet and Labadie (12) for trapping atmospheric chloroanisoles
and chlorophenols in wineries. The following system was placed in
the atmosphere to be monitored: 25 g of a blended adsorbent compound
containing bentonite, silica gel, and Tenax (85:10:5), purified by
continuous methanol extraction in a Soxhlet for 24 h and dried in an
oven at 105 °C prior to use. The blend was spread in a thin layer on
a 30 × 30 cm sheet of aluminum foil and left in the atmosphere for
120 h to trap any contaminants presents in the premises under
investigation.

Musty Corked Odors in Wine
J. Agric. Food Chem., Vol. 52, No. 5, 2004 1257
A cellulose paper scoop was then used to take a 5 g sample of
adsorbent. The blend of internal standards (100 µL) was added to the
scoop, which was installed in a 75 mL Soxhlet continuous extractor.
The evaporator flask was filled with 70 mL of dichloromethane, and
0.1 mL of glacial acetic acid was added. The condenser was fed with
water at 12 ( 5 °C. The flask was heated to produce a reflux of
condensed solvent at a rate of ∼3 mL/min for 120 min. The extractor
was then removed from the heating system and cooled. The solvent
was transferred from the extractor and flask directly into a 100 mL
concentration tube, and the flask was rinsed out twice with ∼2 mL of
hexane. The solution was concentrated using a Zymark Turbovap II,
operating at 25 ( 2 °C in a steam of nitrogen (quality I, Air Products)
at 1 bar. It was concentrated automatically to 0.5 mL in ∼15 min. The
sample was transferred to a 500 µL disposable injection flask and kept
at -20 °C in a dark place prior to analysis. The quantity of contaminants
assayable in air is conventionally expressed in nanograms per gram of
adsorbent.
Table 2. Concentrations of Halophenols and Haloanisoles in Various
Samples of Red Wines Suspected of Musty, Corky Off-Flavorsa
wine
sample
(ng L-1
) TCA TCP TeCA TeCP PCA PCP TBA TBP TeBP PBP
1
2
3
4
5
6
7
8
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
tr
11.5 nd
nd
nd
tr
16.8 12.9 68.5 nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
tr
nd
tr
nd
nd
nd
nd
tr
nd
nd
nd
nd 113.5 34.4 17.8 nd
tr
22.4
13.2 nd
19.5 2.1
33.7 3.9
8.3 12.1
tr
nd
8.3 nd
5.4 nd
71.0
7.5
6.9
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
23.9
nd
nd
nd
tr
tr
tr
tr
tr
tr
nd
tr
nd
tr
nd
tr
675.4 2.2 252.8 nd
147.6 12.0 67.0 nd
48.0 0.0 68.3 nd
nd
nd
9
nd
nd
2.5 11.1 nd
tr 7.2 nd
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
6.5 nd
7.6 nd
8.5 nd
7.1 nd
5.4 nd
40.2 12.3 37.1 nd
28.5 26.7 31.3 nd
nd 13.1 13.4 nd
nd
tr
tr
9.1 14.6 nd
26.1 32.4 nd
17.9 36.1 nd
Analysis by Gas Phase Chromatography and Selected Ion
Monitoring (GC-MS/SIM). A 1 µL (extracted from dry materials or
trapping residues) or 2 µL (wine) sample was injected into an HP5-
MS (5% phenyl-methyl-siloxanes) capillary column 30 m × 0.25 mm
(phase thickness ) 0.25 µm) installed on an HP 6890 chromatograph
equipped with an injector operating in splitless mode (250 °C, initial
pressure ) 7.1 psi, pulsed splitless 25 psi, pulse time ) 1.5 min, bleed
) 50 mL/min, bleed time ) 1.5 min) and an Agilent 5183 4711 insert
using an automatic Combi PAL injector (10 µL syringe, CTC
Analyticals Inc.). The balance gas (helium N55, Air Product) was used
at constant flow rate (initial pressure at column head ) 7.1 psi, flow
rate ) 1.0 mL/min, linear velocity ) 36 cm s-1). The temperature was
programmed from 40 °C (initial isotherm ) 3 min) to 110 °C at a rate
of 25 °C/min, then up to 230 °C at a rate of 5 °C/min, and up to 310
°C at a rate of 25 °C/min (final isotherm ) 5 min). Analysis system
inertness was checked by weekly injection of a column test (pentade-
cane, decylamine, 3-octanol, 2,4-dichlorophenol). Analysis of chlo-
rophenols requires a perfectly inert system, assessed by the retention
time, surface area, and width of the decylamine and 2.4-dichlorophenol
peaks, as well as changes in the response factors of the chlorophenols
in relation to the internal standard.
tr
2.9 tr
nd tr
nd
2.6 17.1 nd
nd
nd
nd
nd
tr
tr
nd
8.3 nd
3.8
7.5 nd
3.6 nd
12.3 nd
29.8 3.5 37.4 nd
45.2 6.3 18.8 nd
tr
tr
21.5 nd
72.0 nd
14.3 nd
tr
nd
nd
nd
10.7 nd
tr
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
12.9 13.7 43.5 nd
nd 16.8 104.1 nd
87.7 12.8 78.2 nd
nd 37.9 108.1 nd
23.7 28.3 100.8 nd
tr
nd
tr
nd
nd
nd
27.0 nd
nd
nd
nd
nd
59.2
nd
nd
nd
11.7 nd
64.6 nd
3.2 nd
11.0 nd
122.9 38.0 4.0 30.1 nd
tr
nd
4.0
9.6 nd
tr
nd
nd
nd
81.2 6.5 392.6 nd
nd 6.3 44.9 nd
a nd, <DL; tr, >DL but <QL.
trained tasters using a paired test to determine whether samples are
perceptibly different in comparison with the reference range of TBA
in wine.
In a second stage, a range of TBA concentrations (0, 2, 4, 8, and 16
ng L-1) was prepared in a dilute alcohol solution at 12% vol ethanol
with 5 g/L tartaric acid neutralized to pH 3.65 with sodium hydroxide
N, used as a model solution designed to simulate wine, as well as in
the same standard wine used above. The olfactory perception threshold
was estimated by a blind duo-trio test in which the range of
concentrations was presented in increasing order to six trained, but not
expert, tasters, according to the procedure described by Boidron et al.
(17). The threshold determined in model dilute alcohol solution, known
as the perception threshold, and the covering threshold determined in
wine corresponded to the detection threshold obtained by 50% of the
tasters.
An HP5973 quadrupole mass detector operating in electron impact
mode was used for detection (source temperature ) 250 °C, quadrupole
) 150 °C, constant ionization potential ) 70 keV, electron multiplier
) 1500 V) and fragmentometry mode on selected ions characteristic
of each molecule (TCA, 197, 210, 212; TCP, 196, 198; TeCA, 244,
246; TeCP, 230, 232; PCA, 278, 280; PCP, 264, 266; TBA, 344, 346,
329; TBP, 328, 330, 332; TeBP, 408, 410, 411; PBP, 486, 488, 490;
dwell time ) 100 ms). The following ratios were used for quantifica-
tion: m/z 212/213 (TCA/TCA-OCD3); 196/336 (TCP/TBP-d2); 246/
249 (TeCA/TeCA-OCD3); 232/336 (TeCP/TBP-d2); 280/283 (PCA/
PCA-OCD3); 266/336 (PCP/TBP-d2); 346/349 (TBA/TBA-d2), and 328/
336 (TBP/TBP-d2). In the absence of a reference standard, TeBP and
PBP were respectively estimated using the same response factor as
TeCP and PCP. Table 1 presents the detection and quantification
thresholds and uncertainties associated with the various compounds
studied using this method.
RESULTS AND DISCUSSION
Identifying and Isolating 2,4,6-Tribromoanisole in Wines
Found To Have a “Musty” Character on Tasting. Table 2
presents the assay results for several chloroanisoles and chlo-
rophenols, reported to be standard contaminants in wines with
a musty character, together with TBP and TBA. With the
exception of a few samples that contained small amounts of
TCA, the wines selected in this study contained little (<3 ng
L-1) or no chloroanisoles (<1 ng L-1 detection threshold). It
was observed that the relative intensity of the musty defect on
an arbitrary scale varying from 0 to 5 was significantly
correlated (p < 0.05) with the TBA content of the wine, rather
than with the other chloroanisoles studied (Figure 1). In cases
of high-level TBA contamination (>20 ng L-1), besides an
intense musty odor, the wines also had a “phenolic” or “iodized”
character of varying intensity. This defect was mainly perceived
on the rear palate when the wine was tasted and was particularly
persistent on the aftertaste. The odor is similar to that of TBP,
The system was calibrated using a range of known concentrations,
prepared from pure standard products at concentrations of 0, 2, 5, 10,
20, and 50 ng L-1 diluted in uncontaminated wine, and 0, 1, 5, 10, 20,
50, and 100 ng g-1 impregnated in cork, wood, or adsorbent material
previously extracted by continuous extraction (24 h) in a Soxhlet by
methanol (dried for 1 h in an oven at 24 °C and then stored in glass).
The pure halophenols were stored in Teflon flasks and haloanisoles in
borosilicate glass. The range of standard substances was injected every
15 samples, every 24 h, and at least once per week, to ensure that the
analysis system was always properly calibrated.
Sensory Analysis. Initially, a range of concentrations of TBA was
prepared at 0, 2, 4, 8, 16, and 32 ng L-1 in a standard red wine (red
wine, 11.85% vol, pH 3.68) to provide reference values for olfactory
intensity (0 ) undetected, up to 5 in increments of 1) in the red wine
thought to be contaminated that were investigated in this study and
training of the tasters. The wines were tasted by two series of three

1258 J. Agric. Food Chem., Vol. 52, No. 5, 2004
Chatonnet et al.
Figure 3. Estimation of the perception threshold of 2,4,6-tribromoanisole
(TBA) in a model hydroalcoholic solution simulating wine (O) and in a
standard red wine (0).
Figure 1. Relationship between the concentration of haloanisoles in various
suspected red wines and the intensity of the musty odor measured by
tasting.
into TCA by O-methylation (31). Wylie (32) also reported the
presence of TBP and TBA in pears, although these products
had not been used in the orchard. Indirect pollution of
pharmaceuticals by polyethylene stoppers contaminated through
contact with wooden pallets impregnated with TBP has also
been reported by Hoffman and Sponholz (33). Finally, trace
amounts of TBA are known to cause this type of organoleptic
defect in drinking water (34-36).
TBA has been reported to have extremely low perception
thresholds in water. Saxby et al. (37) measured a threshold of
8 pg L-1; Whitfield et al. (30), 20 pg L-1; and Malleret and
Bruchet (35), 30 pg L-1. These thresholds make TBA one of
the most powerful contaminants identified, with a similar
pollution capacity to that of TCA which also has, according to
Griffiths (10), a perception threshold in water in the vicinity of
30 pg L-1. The thresholds estimated according to 50% of the
tasters (trained but not expert) in an acidic dilute alcohol solution
(pH 3.5-4.0) containing 10-15% ethanol volume, that is, a
model medium (Pt) or a standard wine (Rt), were considerably
higher. The perception threshold PtR:50% was estimated at 3.4
ng L-1 and the covering threshold RtR:50% at 7.9 ng L-1 (Figure
3). These results were compatible with tasting notes for the
suspect wines studied (Table 2), although TBA could affect
the quality and character of wines even below its detection or
identification thresholds.
TBP may be formed chemically in wastewater treated with
chlorine in the presence of bromide ions and traces of organic
phenols. Chlorine may oxidize bromide ions, releasing bromine,
which may then combine rapidly with a number of organic
pollutants, producing not only chlorophenols but also bro-
mophenols and bromochlorophenols (38). In the presence of
sodium hypochlorite and UV radiation, the bromides present
in salting pans may form 2-bromo-2-methylphenol, which is
capable of giving off the powerful “phenolic” or “chemical”
odor sometimes identified in some Gouda cheeses (39). No
attempt was made to identify this compound in this research,
but it could explain the iodized and phenol character found in
certain wines in addition to musty off-odors.
Figure 2. Relationship between the 2,4,6-tribromoanisole (TBA) and 2,4,6-
tribromophenol (TBP) contents in various contaminated red wines.
but the intensity of the defect is not linked in any obvious way
with the concentrations assayed in wines. Similarly, the TBA
content of a wine is not significantly correlated with the
assayable TBP level. We suppose that the conversion from TBP
does not take place in wine but results from outside pollution
(Figure 2). Unlike pentabromophenol, tetrabromobiphenol A
(TBBPA) and polybrominated diphenyl ethers (PBDEs), widely
used molecules known as brominated flame retardants (BFR),
which may be endocrine disruptors (18), with regard to their
toxicological properties TBA and TBP (19) do not seem to have
any severe toxicological effects at the concentrations measured
in the wines studied.
TBA has been identified as a trace contaminant in marine
fauna and certain marine sediments (20, 21), as well as in the
atmosphere (22). TBA is a derivative of its direct precursor,
TBP, via O-methylation by bacterial microorganisms (23). TBP
is naturally present in many marine organisms (24-28) and, in
particular, certain brown algae, which may synthesize TBP de
novo from bromides in seawater by means of specific enzymes
(29).
TBA is known to cause powerful musty and earthy flavors
in fruit packed in cases contaminated with TBP that has been
broken down into TBA by Paecelomyces Variotii (30), a
filamentous fungus also known to be capable of converting TCP
Source of TBA Contamination in Wines. Analysis of
several samples of red wines from different wineries with known
winemaking and storage conditions indicated several ways in
which wines could be contaminated.

Musty Corked Odors in Wine
J. Agric. Food Chem., Vol. 52, No. 5, 2004 1259
In winery A (Table 5), wines kept either in unpolluted
stainless steel vats or in barrels where pollutants had ac-
cumulated presented high levels of contamination. In this case,
relatively nonvolatile TBP from a primary source of contamina-
tion was broken down microbiologically into TBA, a more
volatile, unpleasant-smelling compound, easily transported
through the air. Wine handled in this atmosphere may dissolve
contaminants present in the gas phase or be contaminated
directly via contact with easily contaminated porous materials
(barrel wood) as initially demonstrated for chloroansioles (5).
Table 3. Control of Various Cellar Atmospheres by Static Trapping for
the Presence of Halophenols and Anisolesa
atmosphere
(ng g-1 ad-
sorbant)
TCA TCP TeCA TeCP PCA PCP TBA TBP TeBP PBP
1
2
3
4
5
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
54
64
2
nd
nd
5
15 179 26
nd
1
48
1
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
28 72
26 52
210
393
30
10
0
0
2
2
a nd, <DL; tr, >DL but <QL.
In winery B (Table 6), the original source of contamination
(old wood vats) had been eliminated several years earlier and
the barrels remaining in the winery were uncontaminated on
arrival. It was, however, still possible to detect significant
quantities of TBA in the atmosphere. Residual pollutants,
adsorbed into the microporous winery structure covered with
clay bricks, could be gradually released via the considerable
surface in contact with the air.
As in the case of chloroanisole derivatives of PCP and TeCP,
which are capable of polluting wines directly or at a distance
via the atmosphere (5, 12), TBA, apparently produced by the
microbiological breakdown of TBP, has been detected in the
atmosphere in several wineries. Atmospheres contaminated by
PCP and TeCP derivatives did not contain any TBP derivatives
and, conversely, TeBP and PBP were never identified in the
cases studied here (Table 3).
The physicochemical characteristics of these contaminants
(Table 4) explain that, although the halophenols and haloani-
soles considered have high boiling points, they may be relatively
easily found in the atmosphere in buildings containing sources
of emission of these molecules. All of the contaminants under
consideration are similarly hydrophobic (log P). The low
saturation vapor pressure of TeCA explains why it is easy to
detect this molecule in places where there is TeCP in the
atmosphere [wood treated with industrial PCP contains 10-
15% TeCP as an impurity (5)]. TCP may sublimate at room
temperature, but it is highly soluble in water and its low Henry
constant restricts its presence in the gas phase. TBA has a log
P similar to that of TCA, but it is less soluble in water and has
a lower saturation vapor pressure but a higher Henry constant.
These characteristics explain the ease of detecting TBA in the
atmosphere and its perception threshold equivalent to that of
TCA in water but higher in matrices with a higher ethanol
content.
In winery C (Table 7), wines aged in barrels were strongly
contaminated with TBA and TBP. Pollutant concentrations in
the wine and barrel wood increased regularly with the age of
the barrels, that is, the length of time they had been in the
contaminated atmosphere of the winery. However, TBA pol-
lution in the wine must also come from the silicone bungs used
to seal the barrels as they were in direct contact with the wine.
As is the case with chloroanisoles, plastic materials in general
and silicone elastomer resins in particular readily adsorb TBA
and, to a lesser extent, TBP, which is just a little bit less
hydrophobic but much less volatile (lower saturation vapor
pressure and much higher Henry constant). Polyethylene- or
polyester-based winemaking equipment, vulcanized rubber
gaskets, and the silicone bungs used in barrels readily fix
pollutants from the air and release them into wine over time.
Analysis of the materials in contact with the atmosphere in this
winery showed that the wooden roof timbers had been massively
impregnated with TBP, which had gradually broken down into
TBA due to the action of microflora in the atmosphere. Some
Table 4. Physicochemical Characteristics of Halophenols and Anisoles from PhysProp Database Syrres Inc. (45)
molecule [CAS
Registry No.]
boiling point,
°C (760 mmHg)
melting point,
°C (760 mmHg)
vapor pressure,
10-3 mmHg
log P
octanol/water
water solubility,
mg L-1
Henry’s law constant,
3 -1
10-6 atm (mol‚m )
TCA [87-40-1]
TCP [88-06-2]
TBA [607-99-8]
TBP [118-79-6]
TeCA [938-86-3]
TeCP [58-90-2]
PCA [1825-21-4]
PCP [87-86-5]
241
246
298
286
61
69
88
95
84
70
108
174
0.228
2.500
0.644
0.303
3.190
0.666
0.592
0.110
4.11
3.69
4.74
4.18
4.65
4.09
5.29
5.12
10.0
800.0
1.0
70.0
1.4
23.0
0.4
14.0
130.000
2.600
20.200
0.035
96.100
8.840
1930.000
0.025
232
309
a
Table 5. Concentration of Halophenols and Anisoles in the Atmosphere and Various Materials Sampled in Cellar A
material
TCA
TCP
TeCA
TeCP
PCA
PCP
TBA
TBP
TeBP
PBP
cellar atmosphere (ng g-1 adsorbant)
tank cellar
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
46
162
13
4
nd
nd
nd
nd
barrel cellar
wines stored in the cellar (ng L-1
)
stored in stainless steel tanks
stored in 12-month-old barrels
nd
nd
nd
3
tr
nd
8
7
tr
tr
675
148
2
12
253
67
nd
nd
nd
nd
potential sources of contaminationb (ng g-1
12-month-old oak wood barrels
)
nd
nd
nd
nd
nd
tr
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
tr
6
nd
nd
143
224
nd
902
1158
nd
nd
nd
nd
nd
nd
nd
nd
nd
24-month-old oak wood barrels
new wood frameworks of the tank cellar
old wood frameworks of the tank cellar
33
835
a nd, <DL; tr, >DL but <QL. b External part of the material on 0−3 mm thickness.

1260 J. Agric. Food Chem., Vol. 52, No. 5, 2004
Table 6. Concentration of Halophenols and Anisoles in the Atmosphere and Various Materials Sampled in Cellar B
Chatonnet et al.
a
material
TCA
TCP
TeCA
TeCP
PCA
PCP
TBA
TBP
TeBP
PBP
cellar atmospheres (ng g-1 adsorbant)
underground cellar 1
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
tr
15
122
tr
15
nd
nd
nd
nd
underground cellar 2
wines stored in the cellar (ng L-1
)
stored in new barrel cellars 1 + 2
nd
5
nd
8
nd
tr
13
78
nd
nd
residual sources of contaminationb (ng g-1
ground (earth) cellars 1 + 2
concrete wall cellars 1 + 2
brick wall cellar 1
)
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
tr
nd
nd
nd
tr
2
nd
tr
tr
nd
nd
nd
nd
nd
nd
nd
nd
brick wall cellar 2
15
tr
a nd, <DL; tr, >DL but <QL. b External part of the material on 0−3 mm thickness.
Table 7. Concentration of Halophenols and Anisoles in the Atmosphere and Various Materials Sampled in Cellar C
a
material
TCA
nd
TCP
tr
TeCA
nd
TeCP
tr
PCA
nd
PCP
nd
TBA
190
TBP
38
TeBP
nd
PBP
nd
cellar atmosphere (g g-1 adsorbant)
wines stored in the cellar (ng L-1
stored in new barrels
)
nd
nd
nd
nd
4
12
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
13
38
28
78
108
101
nd
nd
nd
nd
nd
nd
stored in 12-month-old barrels
strored in 24-month-old barrels
potential sources of contaminationb (ng g-1
new oak wood barrels (9 months in the cellar)
12-month-old oak wood barrels
24-month-old oak wood barrels
36-month-old oak wood barrels
barrel racks
)
nd
nd
nd
nd
nd
nd
nd
nd
tr
tr
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
tr
tr
6
tr
tr
nd
22
143
224
1973
2185
1337
240
902
1158
2943
13127
57
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
silicon bungs
wood framework of the cellar
superficial part (0−3 mm)
inner part (3−15 mm)
varnish
white paint
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
234
201
tr
82363
963
39
nd
nd
nd
nd
nd
nd
nd
nd
1002
4933
a nd, <DL; tr, >DL but <QL. b External part of the material on 0−3 mm thickness.
Table 8. Detection of Halophenols and Anisoles in Used and Unused Stoppersa
material
TCA
TCP
TeCA
TeCP
PCA
PCP
TBA
TBP
TeBP
PBP
atmosphere of storage (ng g-1 adsorbant)
new natural cork stopper (ng per cork)
new PET synthetic cork (ng per cork)
wine in bottle with contamination
nd
16
nd
nd
13
nd
tr
3
51
tr
nd
106
tr
15
107
5
24
872
14
9
39
2
33
294
nd
nd
nd
nd
nd
nd
wine (ng L-1
used corka (ng per cork)
)
nd
27
12
8
nd
nd
nd
nd
2
5
17
48
13
275
68
117
nd
nd
nd
nd
a nd, <DL; tr, >DL but <QL. b One-third from the external part of the stopper removed to avoid the risk of a contamination from the atmosphere.
of the paints assayed in the same winery also contain TBP as
flame retardant or/and fungicide.
TBP Sources in Winery Atmospheres. The cases of wines
contaminated with TBA and TBP described in this work were
obviously due to wooden or wood-based materials impregnated
or surface-treated with TBP, similar to the situations described
for PCP in direct (40) or indirect contact with wine (4, 5).
Neither PBP nor TeBP, both potential precursors of TBP, was
found in the wines and materials studied, or they were present
in only nonquantifiable trace amounts. TBP and other bro-
mophenolic derivatives are widely used on wood as antifungal
agents (41, 46) and flame retardants, as well as on many plastics
(33, 42, 43).
Conclusions. TBA, produced by the microbiological break-
down of TBP, may cause serious organoleptic defects. Like TCA
and TeCA, low concentrations of TBA may give wines pungent
musty odors. In view of the perception thresholds and physi-
cochemical characteristics of this molecule, it may be considered
similar to TCA in its capacity to spoil wine. The identification
of trace amounts of TBA in certain wines may thus explain
cases of reported spoilage when TCA or TeCA was not present
in significant quantities.
Analysis of wines in corked bottles contaminated with TBA
and possibly TCA is more difficult to explain (Table 8). At
this stage, we do not know whether the wine spoilage was due
to pollution from corks stored in a contaminated atmosphere
before use (on the cork manufacturer’s or user’s premises) or,
in a scenario similar to that described for TCA, attributable to
contamination by TBA formed in situ in the cork. Quite
obviously, the discovery of large quantities of pollutants in
extruded polyethylene (PET) corks stored in a contaminated
atmosphere showed that, as was the case of the silicone bungs
mentioned above, airborne pollution of stoppers over large
distances via the atmosphere and through packaging materials
was entirely possible. This result should be compared with
observations made by Whitfield et al. (30) for TBA and TCA
with PET packaging films and by Hoffman and Sponholz (33)
with PET screw caps stored on wooden pallets treated with TBP.
However, the extent to which pollutants adsorbed on PET may
later migrate into wine has not yet been determined.

Musty Corked Odors in Wine
J. Agric. Food Chem., Vol. 52, No. 5, 2004 1261
(12) Chatonnet, P.; Labadie, D. Contamination des locaux vinicoles
et alte´ration des vins - controˆle de la qualite´ de l’atmosphe`re
des chais. ReV. Œnologues 1995, 74, 19-21.
(13) Soleas, G. J.; Yan, J.; Seaver, T.; Goldberg, S. M. Method for
the gas chromatographic assay with mass selective detection of
trichloro compounds in corks and wines applied to elucidate the
potential cause of cork taint. J. Agric. Food Chem. 2002, 50,
1032-1039.
(14) Chatonnet, P.; Boutou, S.; Labadie, M. D. Dosage simultane´ des
chlorophe´nols et des chloroanisoles dans les vins et les obtura-
teurs en lie`ge ou a` base de lie`gesApplication a` la de´termination
de l’origine de la pollution des vins en bouteille. J. Int Sci. Vigne
Vin 2003, 37, 181-193.
(15) OIV (Office International de la Vigne et du Vin). Recueil des
Me´thodes Internationales d’analyse des Vins et des Mouˆts;
OIV: Paris, France, 2000.
TBA and TBP seem mainly to derive from environmental
pollution in wineries. This paper is the first to identify the
sources of pollution in a large number of cases when wines
became polluted during storage in premises where the atmo-
sphere was contaminated with TBA coming from TBP used
recently to treat wood or originating from much older structural
elements of the winert or from used wooden containers. In
certain cases, although the initial source had been eliminated,
residual pollution adsorbed on walls could be sufficient to make
a building unsuitable for storing wines or sensitive materials
intended for direct contact with wine. Both plastics and corks
are highly susceptible to TBA contamination. Atmospheric
quality in storage areas is thus vital to ensure the chemical and
organoleptic inertia of materials that come into contact with
wine.
(16) (a) AFNOR. Exactitude (justesse et fide´lite´) des re´sultats et
me´thodes de mesuresPartie 1: Principes ge´ne´raux et de´finitions.
Norme NF ISO 5725-1, indice de classement X 06-041-1;
AFNOR: Paris, France, 1998. (b) AFNOR. Proce´dure de
validation intra-laboratoire d’une me´thode alternative par rapport
a` une me´thode de re´fe´rence. Norme NF V 03-110; AFNOR:
Paris, France, 1998.
(17) Boidron, J. N.; Chatonnet, P.; Pons, M. Influence du bois sur
certaines substances odorantes des vins. Connaiss. Vigne Vin
1988, 22, 275-294.
(18) Ilonka, A.; Meerts, T. M.; Van Zanden, J. J.; Luijks, E. A. C.;
Van Leuwen-Bol, I.; Mersh, E.; Jakobson, E.; Bergman, A.;
Brouwer, A. Potent competitive interactions of some brominated
flame retardants and related compounds with human transthyretin
in vitro. Toxicol. Sci. 2000, 56, 95-104.
Unlike PCP, the use of which is restricted and even prohibited
in many parts of the world, especially Europe (44), there are
apparently no restrictions on the use of TBP. Furthermore, the
fact that TBP derivatives are widely used as flame retardants
in a wide range of materials raises the risk of widespread
environmental pollution. The increasing use of recycled wood-
and polymer plastic-based materials, which may previously have
been impregnated with TBP or its derivatives, may represent a
source of latent pollution likely to contaminate not only sensitive
food products but also the water table as a result of irresponsible
storage. The extensive use of TBP and its derivatives thus
deserves further investigation.
LITERATURE CITED
(19) Merck Index, 13th ed.; Merck Research Laboratories: White-
house Station, NJ, 2001.
(1) Tanner, H.; Zannier, C.; Wu¨rdig, G. Zur analytischen Differen-
zierung von Muffton und Korgeshmackes in Weinen. Schweiz.
Z. Obst-Weinbau 1981, 117 752-757.
(2) Buser, H. R.; Zanier, C.; Tanner, H. Identification of 2,4,6-
trichloroanisole as a potent compound causing cork taint. J.
Agric. Food Chem. 1982, 30, 359-362.
(3) Tanner, H.; Zannier, C. Unber die Bestimmung der chloroanisole
in wein und Korkstopfen. Schweiz. Z. Obst-Weinbau 1983, 119,
468-473.
(4) Dubois, P.; Rigaud, J. A propos des gouˆts de bouchons. Vigne
Vins 1981, 301, 48-49.
(5) Chatonnet, P.; Guimberteau, G.; Dubourdieu, D.; Boidron, J. N.
Nature et Origine des odeurs de “moisi” dans les caves. Influence
sur la contamination des vins. J. Int. Sci. Vigne Vin 1994, 28
131-151.
(6) Burtschell, R. H.; Rosen, A. A.; Middleton, F. M.; Ettinger, M.
B. Chlorine derivatives of phenol causing taste and odour. J.
Am. Water Assoc. 1959, 205-214.
(7) Maujean, A.; Millery, H.; Lemaresquier, P. Explications bio-
chimiques de la confusion entre gouˆt de bouchon et gouˆt de
moisi. ReV. Fr. Oenol. 1985, 99, 55-62.
(8) Alvarez-Rodriguez, M. L.; Lopez-Ocana, L.; Lopez-Coronado,
J. M.; Rodriguez, E.; Martinez, M. J.; Carriba, G.; Coque, J. J.
Cork taint of wine: Role of the filamentous fungi isolated from
cork in the synthesis of 2,4,6-trichloroanisole by O-methylation
of 2,4,6-trichlorophenol. Appl. EnViron. Microbiol. 2002, 68
5860-5869.
(9) Curtis, R. F.; Land, D. G.; Griffiths, N. M.; Gee, M.; Robinson,
D.; Peel, J. L. Chloroanisoles as a cause of musty taint in chicken
and their microbiological formation from chlorophe´nols. J. Sci.
Food Agric. 1974, 25, 811-828.
(10) Griffiths, N. M. Sensory properties of chloroanisoles. Chem.
Senses FlaVour 1974, 1, 187-195.
(11) Duerr, P. Wine quality evaluation. In Proceedings of the
International Symposium on Cool Climate Viticulture and
Enology; Heatherbell, A., Lombard, P. D., Bodyfelt, F. W., Price,
S. F, Eds.; Oregon State University: Corvallis, OR, 1985; pp
257-266.
(20) Miyazaki, T.; Kaneko, S.; Horii, S.; Yamagishi, T. Identification
of polyhagenated anisoles and phenols in oysters collected from
Tokyo Bay. Bull. EnViron. Contam. Toxicol. 1981, 26, 577-
584.
(21) Watanabe, I.; Kashimoto, T.; Tatsukawa, R. Polybrominated
anisoles in marine fish, shellfish and sediments in Japan. Arch.
EnViron. Contam. Toxicol. 1983, 12, 615-619.
(22) Wittlinger, R.; Ballschmiter, K. Studies of the global baseline
pollution. XIII C6-C14 organohalogens in the lower troposphere
of the southern Indian Ocean. Fresenius’ J. Anal. Chem. 1990,
336, 193-200.
(23) Allard, A. S.; Remberger, M.; Neilson, A. H. Bacterial O-
methylation of halogen-substituted phenols. Appl. EnViron.
Microbiol. 1987, 53, 839-845.
(24) Whitfield, F.; Shaw, K. J.; Svoronos, D. The volatile aroma
compounds of Australian marine algae. In Proceedings of 12th
International Congress on Fragrances and Essential Oils; 1992;
pp 365-372.
(25) Whitfield, F.; Shaw, K. J.; Walker, D. I. The source of 2,6-
dibromophenol: cause of an iodoform taint in Australian prawns.
Water Sci. Technol. 1992, 25, 131-138.
(26) Whitfield, F.; Nguyen, T. H. L.; Shaw, K. J.; Last, J. H.; Tindale,
C. R. 2,6-dibromophenol: the cause of an iodoform-like off-
flavour in some Australian crustacea. J. Sci. Food Agric. 1988,
46, 29-42.
(27) Boyle, J. L.; Lindsay, R. C.; Suiber, R. D. A. Bromophenol
distribution in salmon and selected seafoods of fresh and
saltwater origin. J. Food Sci. 1992, 57, 918-922.
(28) Whitfield, F. B.; Helidoniotis, F.; Svoronos, D.; Shaw, K. J.;
Ford, G. L. The source of bromophenols in some species of
Australian ocean fish. Water Sci. Technol. 1995, 31, 113-120.
(29) Flodin, S.; Whitfield, K. J. Biosynthesis of bromophenols in
marine algae. Water Sci. Technol. 1999, 40, 53-58.
(30) Whitfield, F. B.; Hill, J. L.; Shaw, K. J. 2,4,6-tribromoanisole:
a potential cause of mustiness in packaged food. J. Agric. Food
Chem. 1997, 45, 889-893.

1262 J. Agric. Food Chem., Vol. 52, No. 5, 2004
Chatonnet et al.
(31) Tindale, C. R.; Whitfield, F. B.; Levingston, S. D.; Nguyen, T.
H. L. Fungi isolated from packaging materials: their role in the
production of trichloroanisole. J. Agric Food Chem. 1989, 49,
437-447.
(32) Wylie, H. Trace LeVel Pesticides Analysis by GC/MS Using
Large Volume Injection; Agilent Technologies Application Note;
Palo Alto, CA, Sept 1997; 11 pp.
(33) Hoffmann, A.; Sponholz, W. Direct thermal analysis of solids-A
fast method for the determination of halogenated phenols and
anisoles in work. Am. Lab. 1997, 22-23.
(34) Lahoussine, V. Ge´ne´ration de gouˆts de moisi en re´seaux de
distribution d’eau potable. In Re´sume d’e´tude Alimentation en
eau Potable; No. 01 AEP7; Agence de l’eau Seine-Normandie:
France, 2000.
(35) Malleret, L.; Bruchet, A. Application of large volume injection
GC/MS to the picogram analysis of chlorinated and brominated
anisoles in “earthy-musty” off-favor water samples. Water Sci.
Technol. 2001, 1, 1-8.
(36) Malleret, L.; Bruchet, A. A taste and odour episode caused by
2,4,6-tribromoanisole. J. Am. Water Works Assoc. 2002, 94, 7.
(37) Saxby, M. J.; Reid, W. J.; Wragg, G S. Index of Chemical Taints;
Leatherhead Food RA: Leatherhead, U.K., 1992.
(38) Patnaik, P.; Powers, E.; Khoury, J. N. Pathways of Phenol-
Chlorine Reactions in the Presence of Bromide Ions in Waste-
waters; American Laboratory News, Application Note, Aug 26-
28, 2002.
(39) Mills, O. E.; Gregory, S. P.; Visser, F. R.; Broome, A. J.
Chemical taint in rindless Gouda cheese. J. Agric. Food Chem.
1997, 45, 487-492.
(40) Wu¨rdig, G. Vorsicht bei der Anwendung von Fassaussendesin-
fektionmitteln. Weinwirtschaft 1975, 111, 1250-1251.
(41) Pulido, M. L.; Ayzaguer, J. M. Synergistic microbicidal com-
binations of 2-thiocyanomethylthio)benzothiazole and a triha-
logenated phenols. U.S. Patent 4,983,618, 1991.
(42) Nishibori, S.; Kondo, H. Halogen-containing aromatic diester
flame retardants for organic polymers. Jpn. Patent JP 05295;
1993; p 163.
(43) Takashi, K.; Sato, J. Fire-resistant thermoplastc resin composi-
tions, fireproofing agents and manufacture of fire proofing agents.
Jpn. Patent JP 06157873, 1994.
(44) CEE. Directive du conseil du 21 mars 1991. 91/173/CEE, du
5/04/1991. J. Off. Eur. Communities 1991, L 85/34.
(45) Syrres. Syracuse Research Corp. SRC environmental science,
PhysProp Database, 2003; www.esc.syress.com.
(46) Towa Mokuzai Co. Ltd. Preservation of wood. Japan Patent,
JP57131506, 1982.
Received for review August 29, 2003. Revised manuscript received
November 24, 2003. Accepted December 23, 2003.
JF030632F