Regioselective synthesis of a branched isomer of nonylphenol, 4-(3′,6′-dimethyl-3′-heptyl)phenol, and determination of its important environmental properties

Article abstract of DOI:10.1002/1521-3765(20011119)7:22<4790::AID-CHEM4790>3.0.CO;2-L

A method for the synthesis of a pure nonylphenol isomer, 4-(3′,6′-dimethyl-3′-heptyl)phenol, by Friedel-Crafts reaction between anisole and 3-bromo-3,6-dimethylheptane that gives a 47.3% overall yield is reported. The reactions were followed by GC-MS, and

Full text of DOI:10.1002/1521-3765(20011119)7:22<4790::AID-CHEM4790>3.0.CO;2-L

FULL PAPER
Regioselective Synthesis of a Branched Isomer of Nonylphenol,
4-(3',6'-Dimethyl-3'-heptyl)phenol, and Determination of its Important
Environmental Properties
Joseph O. Lalah,[a, c] Karl-W. Schramm,*[a] Dieter Lenoir,[a] Bernhard Henkelmann,[a]
Norbert Hertkorn,[a] Georg Matuschek,[a] Antonius Kettrup,[a] and Klaus Günther[b]
Abstract: A method for the synthesis of a pure nonylphenol isomer, 4-(3',6'-di-
methyl-3'-heptyl)phenol, by Friedel ± Crafts reaction between anisole and 3-bromo-
3,6-dimethylheptane that gives a 47.3% overall yield is reported. The reactions were
followed by GC-MS, and the chemical structures are in agreement with the NMR and
IR spectra. The log Kow value for this compound, its water solubility, vapor pressure,
and Henryꢀs Law constant were also determined. These physicochemical properties
were required for prediction of the compoundꢀs behavior in aquatic ecosystems.
Keywords: environmental chemistry
´
Friedel ± Crafts alkylation
´
nonylphenol isomer
design
´
synthesis
dimethyl-2'-heptyl)phenolÐexist in the product mixture in
higher concentrations than all the other components.[3] These
isomers have also been detected by GC-MS and some of them
separated by HPLC in environmental studies.[1, 4, 5] In the
laboratory, microsynthesis attempts to obtain a single isomer
of ring-labeled 14C-NP by using ring-labeled 14C-phenol and
non-1-ene gave three major isomers: 4-(2'-nonyl)phenol,
4-(3'-nonyl)phenol, and 4-(4'-nonyl)phenol.[2, 4] The isomer-
ization occurs due to the relative instability of the formal
primary carbocation formed from nonene in the presence of
the acid catalyst, which then alkylates the phenol preferen-
tially at the para position.[9] The only other reported
laboratory synthesis attempt involved preparation of a radio-
labeled technical 14C-NP mixture by means of an alkylation
reaction between 14C-phenol and technical nonene.[4]
The synthesis of pure isomers of NP is of interest because of
the known estrogenicity and persistence of this compound in
aquatic environments.[4±6] It has not, however, been estab-
lished which of the isomers are responsible for these estro-
genic effects, since none of the pure single isomers has been
synthesized before to enable such studies. However, a lot of
interest is focused mainly on the branched isomers, due to
their high concentrations in the mixture and the belief that
they may be more estrogenic than the straight-chain isomers.
We have synthesized the branched isomer 4-(3',6'-dimethyl-3'-
heptyl)phenol (7), both as a normal standard and in its 14C-
ring-labeled form for use in toxicity and metabolism studies.
The complex technical mixture of p-nonylphenol isomers has
been well studied and its fundamental physicochemical
properties, useful for prediction of its ecotoxicological behav-
ior, have been reported.[5, 23±26] In order to predict the
environmental behavior and toxicity of the single isomer that
Introduction
Nonylphenols (NPs) are an important class of isomeric
compounds, available as a mixture of more than 20 isomeric
compounds and used for the production of nonylphenol
polyethoxylate surfactants.[1] The synthesis of single isomers
of nonylphenol has not been reported. The commercial
preparation of NPs is performed through Friedel ± Crafts
alkylation of phenol with technical nonene, a complex
mixture consisting predominantly of nine-carbon olefins and
referred to as propylene trimer.[1, 2] The alkylation reaction
results in a very complex mixture of 90 ± 93% para-substi-
tuted nonylphenols (both straight- and branched-chain iso-
mers) and minor quantities of ortho isomers. High resolution
chromatography on a 100 m capillary column GC with 400000
theoretical plates has shown that the branched isomersÐsuch
as 4-(3',6'-dimethyl-3'-heptyl)phenol (7), 4-(3',6'-dimethyl-4'-
heptyl)phenol, 4-(4',6'-dimethyl-4'-heptyl)phenol, 4-(2'-meth-
yl-4'-octyl)phenol, 4-(2'-methyl-2'-octyl)phenol, and 4-(4',6'-
[a] Dr. J. O. Lalah, Priv.-Doz. Dr. K.-W. Schramm, Prof. Dr. D. Lenoir,
Dipl.-Ing. B. Henkelmann, Dr. N. Hertkorn, Dr. G. Matuschek,
Prof. Dr. A. Kettrup
Institut für Ökologische Chemie
GSF-Forschungszentrum für Umwelt und Gesundheit
Ingolstädter Landstrasse 1, 85764 Neuherberg, München (Germany)
Fax : (‡49)89-3187-3371
E-mail: schramm@gsf.de
[b] Prof. Dr. K. Günther
Institut für Angewandte Physikalische Chemie
Forschungszentrum Jülich, 52425 Jülich (Germany)
[c] Dr. J. O. Lalah
Department of Chemistry, Maseno University
P.O. Box 333, Maseno (Kenya)
4790
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0722-4790 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 22

4790±4795
we had synthesized, we determined some of its major
physicochemical properties, including its octanol ± water par-
tition coefficient (logKow), water solubility (Sw), vapor pres-
sure (VP), and its Henryꢀs Law constant (H). These param-
eters were necessary in helping us design experi-
ments to study the distribution, fate, and effects on molluscs
(Lymnaea stagnalis L.) of this isomer in aquatic environ-
ments.
been used to measure the vapor pressures of organic
compounds, including the gas-saturation method and the
effusion method, both of which are recommended by the
OECD.[21, 22] Capillary gas chromatography has also been used
to determine the vapor pressures of moderately polar and
nonpolar compounds, for which the GC-determined vapor
pressures correlate well with liquid-phase vapor pressures at
given temperatures.[24] In our determinations, we used the
effusion method, as described in detail in the Experimental
Section.
The water solubility of a compound determines its mobility
between environmental compartments and influences its
further distribution from the point of discharge to rivers and
oceans. Water solubilities of organic pollutants also correlate
well with their bioaccumulation potential through a single-
variable linear-regression relationship, by which the often
very high bioconcentration factors (BCFs) in aquatic organ-
isms of organic chemicals with low water solubilities
Results and Discussion
We have synthesized the branched nonylphenol isomer,
4-(3',6'-dimethyl-3'-heptyl)phenol (7) and determined its
important environmental properties, that is water solubility,
logKow, vapor pressure, and Henryꢀs Law constant, which are
required to be able to predict its distribution and fate in
aquatic ecosystems.[15±18,20±26] The synthesis was accomplished
by means of a Friedel ± Crafts reaction between anisole (5)
and the tertiary alkyl bromide 3-bromo-3,6-dimethyl heptane,
followed by cleavage of the methoxy bond with a Lewis acid,
BBr3 (Scheme 1). The tertiary alkyl bromide, not commer-
cially available, was synthesized by Grignard reaction be-
tween butan-2-one (2) and 1-bromo-3-methylbutane, fol-
lowed by hydrolysis with water to yield 3-hydroxy-3,6-
dimethylheptane (3), which was then heated under reflux in
48% HBr to yield the required tertiary alkyl bromide 4. In our
reaction, the Grignard reagent 1 was prepared in situ by
dropwise addition of an equimolar amount of 1-bromo-3-
methylbutane to Mg in anhydrous diethyl ether while
stirring.[7] Due to the high reactivity of this tertiary alkyl
bromide, it was not necessary to initiate the reaction by
addition of iodine as is normally done in Grignard reactions.
The reactants were cooled to control the rate of the reaction
and then heated under reflux at 408C for 30 minutes before
addition of butan-2-one. There was only a minimal possibility
of side reactions with this tertiary alkyl bromide at the b-
position to give alkenes, and cleavage of the Mg R bond to
give alkanes was also minimal. Hydrolysis was achieved by the
addition of crushed ice, which produced excess HBr and
precipitated Mg(OH)2, followed by addition of 10% aq.
1
(<50 mgL ), such as the organochlorine pesticides and
dioxins, may be predicted.[23] The bioaccumulation and
retention potentials of organic chemicals are therefore
important criteria for their ecotoxicological evaluation both
in terrestrial and in aquatic organisms, especially in cases in
which the acute toxicity of the chemical is low and the
resultant physiological effects on the organisms are not
noticeable until chronic effects are manifested in them.[24]
The BCF also determines the distribution of chemicals in
various target organs in the body of an organism, for example
in the adipose tissue, muscle, liver, brain, and kidney, and is
therefore relevant in studies of phenolic compounds, the
toxicities of which have been reported to increase propor-
tionately with their lipophilicities.[23, 24] Except for chemicals
with very high molecular weights (MWs) and those that are
readily metabolized by target organisms, the tendency of
organic compounds to bioconcentrate has also been shown to
be strongly related to their lipophilicities, expressed in the
form of logKow values.[26] The logKow value of a compound
therefore gives a good indication of its bioaccumulation
potential and is important for the design of toxicity and
metabolism studies in aquatic ecosystems. We used the Flask
and Shake Flask methods for water solubility and logKow
determinations, respectively, as explained in the experimental
section below.
The vapor pressure (VP) of a chemical also influences its
distribution in the environment. In conjunction with external
factors such as wind and tem-
perature, the vapor pressure of
a compound dictates its volatil-
ity in a given environment. The
vapor-pressure value can be
used to calculate the Henryꢀs
Law constant (H) for the com-
pound, as the ratio of its vapor
pressure and its water solubility.
The Henryꢀs Law constant of a
compound therefore deter-
mines its vapor exchange across
air± water interfaces and so is a
good indicator of its persistence
in aquatic environments.[21, 22]
Several physical methods have
Scheme 1. Reaction scheme for the regioselective synthesis of 4-(3',6'-dimethyl-3'-heptyl)phenol (7). i) 10%
NH4Cl; ii) HBr (48%), 1 h, reflux; iii) AlCl3, hydrolysis; iv) BBr3/CH2Cl2, 1.5 h, 08C.
Chem. Eur. J. 2001, 7, No. 22
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0722-4791 $ 17.50+.50/0
4791

K.-W. Schramm et al.
FULL PAPER
NH4Cl until all the precipitate redissolved. The reaction
proceeded easily to yield 71.3% of 3,6-dimethylheptan-3-ol
(3) after fractionation by distillation under vacuum (78 ± 828C
at 25 mmHg). The identity of this alcohol was also confirmed
by GC-MS and NMR.
Several procedures for the cleavage of aryl alkyl ethers that
avoid simultaneous cleavage of the phenyl alkyl bonds have
been reported. They include the use of organosilicon com-
pounds such as iodomethyl silane, Lewis acids such as AlCl3
and AlBr3, hydrogen halides, dilute hydrochloric acid, di-
methyl boron bromide, and boron tribromide (BBr3).[10±14] The
reaction conditions with these reagents also prevent any
possibility of attack on the hydroxyl group. Boron tribromide,
for example, has been used effectively for de-ethylation of
aryl alkyl ethers and can cleave linear alkyls of chain lengths
of up to C10, to yield mixtures of products including phenols.[10]
In our microsynthesis, after attempts with BI3 (stirring for
15 minutes in CH2Cl2 at 08C, GC-MS analysis of reaction
mixture) had failed to give a reaction, we changed to BBr3,
which we found to be more reactive and a lot cheaper. The
micro reaction was performed by addition of a measured
volume of BBr3 to the reactants in CH2Cl2 and stirring at 08C
for 1.5 hours, followed by hydrolysis with water for 1 hour at
the same temperature. We found that both SN1 and SN2
reactions were competing in such a way that re-formation of
large amounts of the alkyl bromide was favored if the reaction
was allowed to run for longer than 1.5 hours. A good yield
(81%) of crude liquid product was obtained after separation
of the organic phase from the aqueous phase and drying the
product in a rotary evaporator under a vacuum. Unreacted
compounds and other impurities were removed by TLC
separation (TLC plates: silica gel 60F254 (20 Â 20 cm), solvent:
5% diethyl ether in n-hexane, UV 254; Rf for 6: 0.54, Rf for 7:
0.14). The clean product 7 was then analyzed by GC-MS
(MW 220), IR, and NMR. It had a refractive index of 1.5068
(21.158C). By HPLC, it was more polar (Rt: 3.1 min) than the
technical mixture (Rt: 5.4 min).
A number of methods have been reported for nucleophilic
substitution of hydroxyl groups by bromine ions, including
sulfonation or phosphorylation followed by substitution of the
alkyl sulfate/phosphate by bromide.[8, 9] Sulfonation prevents
the formation of rearrangement products, making the reaction
suitable for primary, secondary, and tertiary alkyl halides,
although tertiary sulfonates are extremely reactive. Sulfona-
tion followed by bromination with PBr3 has also been used
with, for example, pentan-3-ol, although two products, 2- and
3-bromopentane, were obtained in this case.[8] Primary,
secondary, and tertiary alcohols can also be halogenated by
treatment with the appropriate NaX, KX, or NH4X in
polyhydrogen-pyridine solution.[9] The method we used was
based on a simple reported halogenation reaction with HX.[7, 8]
With this method, HBr and HI react easily for all primary,
secondary, and tertiary alcohols, the reactions being per-
formed by refluxing the alcohol in the hydrogen halide
solution for 6 hours. In this case, however, side reactions such
as isomerization and formation of alkenes are also to be
expected. We were able to carry out this substitution by
refluxing the tertiary alcohol 3 in 48% HBr for one hour
before chilling the reaction by the addition of ice-cold water,
to give an 80.6% yield of crude product. Fractionation of the
crude product was done by distillation under vacuum (83 ±
918C, 23 mmHg) to yield the pure alkyl bromide product
(47.9%), confirmed by GC-MS and NMR.
A Friedel ± Crafts reaction between anisole (5) and the
tertiary alkyl bromide 4 was achieved by AlCl3-catalyzed
alkylation as reported.[7] This reaction was performed on a
micro scale in order to establish a procedure for synthesis of
the 14C-ring-radiolabeled 4-(3',6'-dimethyl-3'-heptyl)phenol
isomer required for toxicity and metabolism studies. In
Friedel ± Crafts reactions, the polarization of the R X bond
increases from primary to secondary to tertiary alkyl bromide,
and electrophilic activity therefore also increases in the same
order. The alcohol requires at least equimolar quantities of
Lewis acid and a water-free environment. Excess anisole was
used in our reaction to get a monoalkylated product and avoid
further alkylation on the benzene ring. Electrophilic substi-
tution reactions at other positions in the benzene ring were
avoided by working at low temperatures; this ensured
predominant substitution at the para position. Our product
isomer was later compared with a similar standard isomer
obtained by a modified macroscale synthesis based on the
procedures followed in this microsynthesis.[27] The identity of
the liquid product, 1-(3',6'-dimethyl-3'-heptyl)-4-methoxyben-
zene, was confirmed by GC-MS (MW 234) and NMR, and had
a refractive index of 1.4909 (21.158C); the compound was also
characterized by IR. There were traces (approximately 1%)
of the ortho isomer apparent in the NMR spectra. The
methoxy bond in the resultant ether was then cleaved to give
the required branched p-nonylphenol isomer. This cleavage
was also done on a micro scale, as explained in the procedure
given in the Experimental Section.
Table 1 shows some of the determined physicochemical
properties of the branched isomer of p-nonylphenol, com-
pared with those reported for the technical mixture.
The structure ± property relationships that apply to hydro-
phobic organic compounds such as chlorobenzenes and
alkylbenzenes have been found to apply to alkylphenols as
well, but the relationships are more scattered for alkylphenols
because of their relatively higher polarity. The correlations
developed for nonpolar organic chemicals cannot therefore
be applied directly to alkylphenols. The physicochemical
properties of the synthesized isomer were found to be
different from those of the technical mixture. From high-
resolution gas chromatography and GC-MS data of the
resolved isomers in the technical isomeric nonylphenol, the
proportion of this isomer in the mixture can be estimated as
less than 10%, and so its physicochemical properties would
not be expected to be very similar to those of the technical
compound mixture.[1, 3] The logKow value in particular was
lower, although this low value was consistent with the high
water solubility of the compound (ten times that of the
technical compound). Its logKow value is closer to that of the
branched compound 5-tert-butyl-2-methylphenol (water sol-
ubility: 410 at 258C, logKow: 2.62 at 258C), which has a similar
structure. The tertiary structure of the isomer molecule is
almost symmetrical, and so the molar volume of the molecule
would be expected to be slightly smaller than those of less
branched and straight-chain isomers of p-nonylphenol mole-
4792
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0722-4792 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 22

Branched Nonylphenol
4790±4795
Table 1. The determined physicochemical properties of the isomer 4-(3',6'-dimethyl-3'-heptyl)phenol, compared with those of other similar compounds,
obtained from the literature.[a]
1
1
Compound
Sw [mgL
]
VP [Pa]
H [Pam3 mol
]
logKow
2 [c]
2 [d]
2 [c]
2 [d]
2 [d]
2 [c]
4-(3',6'-dimethyl-3'-heptyl)phenol
53.69 (pH 5)[c]
1.93 Â 10
3.57 Â 10
2.07 Â 10
3.77 Â 10
2.27 Â 10
42.93[f]
±
7.91 Â 10
1.89[c]
3.4[b][18]
4.48[e][17]
1 [c]
nonylphenol
5.43[e][18]
6.35[d][18]
8.39 Â 10
13.1 Â 10
1.27 Â 10
5.0 Â 10
±
1.35[d]
0.43[d]
1 [d]
2 [d]
2 [d]
(technical mixture)
pentachlorophenol
phenol
4-octylphenol
2-methyl-5-tert-butylphenol
2,3,5-trimethylphenol
14[d] (pH 5.1)
88360 (pH 5.1)[d]
12.6[e]
5.05[d] (pH 5.1)
1.5[d]
4.12[e]
2.62[b, d]
2.86[d]
410[d]
762[d]
3.69[d]
2.43[d]
[a] Data from the literature were obtained from Wan-Ying Shiu et al.[25] unless indicated otherwise. [b] calcd. [c] 208C. [d] 258C. [e] 20.58C. [f] 24.858C.
Note: Experimentally determined logKow value was lower than that obtained by calculation (3.4) with the formula logKow
ˆ
0.747 ‡ 0.73.[25]
cules. The smaller molar volume would consequently be
Conclusion
expected to affect the water solubility and vapor pressure of
the single branched isomer. The tertiary structure of the single
branched isomer also enhances the polarity of the phenolic
bond and makes it easier for the molecule to form hydrogen
bonds with water molecules; this would also enhance its water
solubility relative to that of a straight-chain isomer. The vapor
pressure of the synthesized single branched isomer was,
however, found to be similar to that of one of the components
of the technical p-nonylphenol mixture resolved by capillary
GC as reported by Bidleman and Renberg, who found vapor
All the reaction steps and procedures were found to be
reliable and reproducible. The microsynthesis of pure 14C-
ring-labeled 4-(3',6'-dimethyl-3'-heptyl)phenol (NP) from 14C-
ring-labeled anisole was also achieved in good yield by
following the established microsynthesis reaction procedures
outlined in this paper. This microscale synthesis method is
therefore considered to be a very important synthesis method
both for native and for 14C-radiolabeled forms of the isomer.
1
The water solubility value (53.69 mgL at 208C) and the
2
pressures ranging between 5.7 Â 10 and 17.4 Â 10 2 Pa for
logKow value (1.89) found for this isomer were, however,
significantly different from those reported in the literature for
the technical mixture (water solubility: 5.43 mgL 1 at 20.58C,
logKow: 4.48).[17, 18] On the basis of the vapor pressure values,
the Henryꢀs Law constants, and the air± water partition
coefficients, the branched isomer would be expected to be
more persistent than the technical p-nonylphenol mixture in
aqueous media, although it should tend to show relatively
lower bioaccumulation in sediments and biota, if possible
chemical degradation processes in aquatic environments are
discounted.
the various resolved isomers at 258C.[24] From its logKow value,
the synthesized branched isomer would be expected to have a
slightly lower bioaccumulation potential than technical p-
nonylphenol in aquatic organisms, given the same species of
organisms and exposure concentrations.
From these results, we can conclude that, like other
phenolic compounds, this branched isomer would mainly
tend to be associated with sediments and soils, and its
persistence in the environment would be mainly governed
by its degradation in those media. Its volatilization loss from
water and soil would be relatively less than that for technical
p-nonylphenol, because of its lower air± water ratio partition
5
coefficient of 3.25 Â 10 at 208C, which is comparable to
Experimental Section
those of pentachlorophenol and phenol. Because of its higher
water solubility and lower vapor pressure, the isomer should
be less persistent than the technical p-nonylphenol mixture in
aqueous media. At normal aquatic conditions (pH 7 ± 8),
negligible dissociation of the p-nonylphenol isomer is to be
expected, on the assumption that it has a pKa value close to
the 10.8 reported for the technical p-nonylphenol mixture.[28]
Under these conditions, if the compound were to be dis-
charged through air, it would rapidly be removed and be
deposited in water and soil. Little vapor loss would occur
when discharged into water, but appreciable amounts would
be deposited onto sediment. The air± water partition coef-
ficient for the branched isomer was found to be ten times
lower than that for the technical mixture (3.44 Â
Anisole (density 0.995, 99% pure by GC), Mg turnings (98% pure), AlCl3
(99% pure), and butan-2-one (density 0.805, 99% pure by spectrophoto-
metric methods) were obtained from Aldrich, while 1-bromo-3-methyl-
butane (density 1.208, 95% pure by GC), anhydrous diethyl ether (99.5%
pure by GC), 48% hydrobromic acid, and BBr3 were obtained from Fluka.
GC-MS determination was performed on GC HP5890 Series II and MS
Finnigan SSQ7000 machines; column: J&W DB-5ms, 60 m, 0.25 mm i.d.,
1
film thickness 0.10 mm. Temperature program: 608C (1.5 min); 7.58Cmin
to 2608C (12 min), injection temperature: 2508C, 1 mL splitless injection:
carrier gas: helium, pressure: 25 psi. MS conditions: full scan from mass 50
to 500, cycle time 0.7 sec, Tˆ 1508C. HPLC was performed on a Merck
Hitachi System (L-6200A pump, L-4000 UV detector and D-2500
Chromato integrator), column: RP18, 10 cm long, 5 mm i.d., solvent:
90% methanol in water, flow rate: 1.5 mLmin 1, isocratic conditions.
NMR spectra were measured with a Bruker DMX500 NMR spectrometer,
with a 5 mm inverse geometry probehead (908(1H) ˆ 9.3 ms, 908(13C) ˆ
9.8 ms) in CDCl3 (d ˆ 7.25, 77.0) at 303 K. Phase-sensitive 1H,13C HSQC,
HSQC-TOCSY (mixing time: 70 ms) and absolute value DQ-COSY NMR
spectra were measured by using Bruker standard software. The assignment
of proton resonances of compounds 6 and 7 was deduced from the HSQC-
1
10 4 Pam3 mol 1 K at 208C); this indicates that the isomer
would be less rapidly volatilized from water and therefore
more persistent in aqueous environments than the technical p-
nonylphenol mixture.
Chem. Eur. J. 2001, 7, No. 22
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0722-4793 $ 17.50+.50/0
4793

K.-W. Schramm et al.
FULL PAPER
NMR spectra, assisted by calculations of proton and carbon chemical shifts
with the ACD/Labs (Pegnitz, Germany) HNMR and CNMR predictor
program, version 4.5. IR was performed on a Perkin Elmer Model 552,
from films.
hydrolysis was left to continue for one hour, after which the two phases
were separated. The aqueous phase was washed with CH2Cl2 (20 mL), and
the combined organic phase was dried overnight over Na2SO4, filtered, and
then concentrated to yield 70.2 mg of crude product 7. This crude product
was confirmed by GC-MS and then purified by preparative TLC (TLC
plate: silica gel, 20 cm  20 cm; Rf: 0.14 (diethyl ether/n-hexane 10:190)).
The pure compound was then analyzed by GC-MS and NMR and
characterized. Refractive index ˆ 1.5068 (21.158C); 1H NMR: d ˆ 7.14
(AA'XX', J ˆ 8.8 Hz, 2H; H2/6), 6.77 (AA'XX', J ˆ 8.8 Hz, 2H; H3/5); 1.70
(m, H2'), 1.67 (m, H4'), 1.56 (m, H4'), 1.49 (m, H2'), 1.43 (m, H6'), 1.23 (s,
3H; H38); 1.04 (tq, J ˆ 11.3, 6.6, 4.6 Hz, 1H; H5'); 0.98 (m, 1H; H5'); 0.84,
0.82 (d, J ˆ 6.7 Hz, 6H, H68/H7'); 0.69 (t, J ˆ 7.4 Hz, 3 H ; H1'); 13C NMR:
d ˆ 152.88 (C1), 140.34 (C4), 127.64 (C3/5) 114.66 (C2/6), 40.59 (C2'), 40.27
(C3') 35.68 (C4'), 33.22 (C5'), 28.65 (C6'), 23.60 (C38), 22.63, 22,61 (C68,
Synthesis of 3,6-dimethylheptan-3-ol (3) from butan-2-one (2) and
1-bromo-3-methylbutane (1): Grignard reaction: Magnesium (12.2 g,
0.502 mol) was added to anhydrous diethyl ether (50 mL) in a reaction
flask. 1-Bromo-3-methylbutane (60 mL, 0.48 mol) was then slowly added to
the stirred mixture from a dropping funnel over one hour. The reaction was
brought to completion by warming on a water bath at 408C for 30 minutes
until all the Mg had disappeared. Butanone (36 mL, 0.4 mol, density 0.805
gL 1) in anhydrous diethyl ether (65 mL) was then added to this reaction
mixture, while still stirring and warming at 408C. After addition was
complete, the reaction was allowed to continue for 2 hours. The reaction
mixture was then cooled in an ice bath and hydrolyzed by the addition of
crushed ice (50 g). Aqueous NH4Cl (10%, 150 mL) was then added until
the precipitate redissolved. The ether phase was then separated, the
aqueous slurry phase was extracted twice with diethyl ether (100 mL), and
the organic phases were combined. The ether phase was then washed with
10% KHCO3 and dried overnight over Na2SO4, filtered, and evaporated in
a rotary evaporator to yield 76.7% of the crude alcohol (55.6 g). This crude
product was fractionated by distillation under vacuum (25 mmHg, 87 ±
888C) to give pure product (49.6 g, 68.4%). The identity of the product
was confirmed by NMR.
C7'), 8.60 (C1'); IR: nÄ ˆ 3344, 3031, 2959, 2927, 2870, 1612, 1596, 1513, 1466,
1
1380, 1236, 1181, 827 cm
.
Determination of the logKow value of 7: The isomer standard 4-(3',6'-
dimethyl-3'-heptyl)phenol obtained from a modified macroscale synthesis
procedure based on the above microsynthesis procedures was used for the
determination of its important physicochemical parameters.[27] The deter-
mination of logKow was performed by the Shake Flask method, as described
in the OECD methods of testing chemicals.[16] From the formula (logKow
ˆ
0.747  logSw ‡ 0.73, where Sw is the water solubility in mol L 1), the
logKow value of the single pure isomer of nonylphenol was estimated to lie
Synthesis of 3-bromo-3,6-dimethylheptane (4) from 3,6-dimethylheptan-3-
ol (3): 48% HBr (40 mL, 1.5 mol) was added, with stirring, to 3,6-
dimethylheptan-3-ol (47 g, 0.227 mol). The reaction mixture was heated
under reflux for 1 hour and then chilled by the addition of ice-cold water.
Separation of the two layers was performed rapidly. The alkyl bromide
layer was washed twice with ice-cold water (30 mL), and then twice with
ice-cold 40% methanol (30 mL). The separated organic phase was
neutralized by shaking in a separating funnel with 10% KHCO3 (30 mL)
and then rinsed with ice-cold water (30 mL). The product was dried
in the range, from
2
to ‡4, recommended for the Shake Flask
method.[15, 18, 19] The tests were performed at pH 5, conditions under which
it was assumed that the isomer molecules would exist almost completely in
nonionized form, based on the pKa value of 10.8 stated for the p-
nonylphenol technical mixture.[25] Analytical grade octan-1-ol and Milli-
pore water (obtained from a Milli-Q Plus deionizer fitted with a RIOS
system for removal of dissolved organic matter) were used. The Millipore
1
1
water had an electrical conductivity of 18.2 Mcm and TOC < 30 mgL
(ppb). Saturated stock solutions of octan-1-ol and water were made by
shaking equal volumes of octan-1-ol and water in a 500 mL conical flask on
a mechanical shaker for 24 hours and then allowing the solvent mixture to
stand for six hours in a separating funnel to separate the two phases
completely. 30 mL centrifuge tubes with Teflon caps were used for the tests,
which were performed in duplicate with octan-1-ol/water ratios of 1:5, 1:2,
1:1, 2:1, and 2:1 with a total volume of 30 mL in each centrifuge tube. The
nonylphenol isomer standard (0.0294 mg) containing 14C-labeled non-
ylphenol (0.595 mg) was dissolved in octan-1-ol (60 mL) in a 250 mL conical
flask, and aliquots of 5 mL, 10 mL, 15 mL, 20 mL, and 25 mL were taken
into the centrifuge tubes. Appropriate volumes of water were then added to
each centrifuge tube to make up the volume to 30 mL. The total
radioactivity in the 60 mL stock solution was 578581 dpm (disintegrations
per minute). The tubes were hand-shaken continuously for 30 minutes and
then placed in a centrifuge (Heraus Biofuge 22R, 4000 rpm for 2 min.) to
separate the octanol/water phases. Aliquots (1 mL) were taken in duplicate
from both phases in each centrifuge tube, by Gilson standard pipette, into
Ultima Gold scintillation cocktail (10 mL) for determination of the amount
of radioactivity. The dpm values obtained from the scintillation counter
were converted into mg values to determine the concentrations of the test
compound in each tube, from which the ratios of concentrations of the
compound in octan-1-ol and water were calculated. A logKow value of
1.89 Æ 0.0056 was obtained.
overnight over Na2SO4. After distillation,
a number of by-products
including alkenes were distilled off under vacuum at 83 ± 1048C
(45 mmHg). The identity of the remaining colorless liquid product was
confirmed by NMR.
Microsynthesis of 1-(3',6'-dimethyl-3-heptyl)-4-methoxybenzene (6) by
Friedel ± Crafts alkylation with AlCl3: Anisole (130 mg, 1.203 mmol) and
AlCl3 (3.18 mg, 0.0234 mmol) were added with stirring to 3-bromo-3,6-
dimethylheptane (49.2 g, 0.0238 mmol) in n-hexane (30 mL). The mixture
was heated under reflux below 208C overnight until no more HBr gas
evolution was observed. Crushed ice was then added to the reaction
mixture to hydrolyze the components, and the organic phase was separated,
washed with distilled water (20 mL) and then with 10% NaOH (10 mL)
solution, and then rinsed again with distilled water (20 mL) until neutral.
The organic phase was then dried overnight over Na2SO4. The solvent was
removed in a rotary evaporator to yield crude 1-(3',6'-dimethyl-3'-heptyl)-
4-methoxybenzene (6) (90.6 mg), together with traces of the ortho isomer
(1% by NMR). The crude product was also analyzed by GC-MS (MW 234)
and then used in its crude form for the second step of the microsynthesis as
detailed below. TLC-purified compound 6 was characterized by IR and
NMR. Refractive index ˆ 1.4909 (21.158C); 1H NMR: d ˆ 7.18 (AA'XX',
J ˆ 9.0 Hz, 2H; H2/6,), 6.83 (AA'XX', J ˆ 8.9 Hz, 2H; H3/5,), 3.79 (s, 3H;
OCH3), 1.69 (m, H2'), 1.66 (m, H4'), 1.55 (m, H4'), 1.49 (m, H2'), 1.40 (hept,
H6'), 1.23 (s, 3H; H38'), 1.01 (tq, J ˆ 11.3, 6.6, 4.7 Hz 1H; H5',), 0.85 (m,
1H; H5'), 0.83, 0.82 (d, J ˆ 6.7 Hz, 6H; H68/H7'), 0.67 (t, J ˆ 7.5 Hz, 3H;
H1'); 13C NMR: d ˆ 157.09 (C1), 140.08 (C4), 127.41 (C3/5), 113.18 (C2/6),
55.10 (OCH3), 40.56 (C2'), 40.23 (C3'), 35.68 (C4'), 33.24 (C5'), 28.67 (C6'),
23.64 (C38), 22.64, 22.62 (C6'/C7'), 8.62 (C1'); IR: nÄ ˆ 3037, 2958, 2870,
Determination of the water solubility (Sw) of 7: From the equation
logKow
least 44.3 mgL
ˆ
0.747logS ‡ 0.73, the water solubility was estimated to be at
.
1 [15, 18, 19] The experimental value was determined by taking,
in duplicate, excess amounts of standard nonylphenol (29.1 mg) in CH2Cl2
(15 mL). Aliquots of this solution were then added to diatomaceous earth
(600 mg) in glass centrifuge tubes with Teflon caps as follows: 5 mL
(containing 9.70 mg of compound) in tube 1, 4.8 mL (9.31 mg of com-
pound) in tube 2, and 5 mL (9.70 mg of compound) in tube 3. The
solvent was then left to evaporate completely from each tube, leaving
the nonylphenol absorbed onto the dry diatomaceous earth. Distilled
water (25 mL, pH 5) was then added to each of the tubes, which were
incubated at 308C under constant stirring for 96 hours. The tubes were
then removed and left to equilibrate at 208C prior to centrifugation at
4000 rpm for 10 min. The water samples were then analyzed by
2833, 1611, 1581, 1513, 1465, 1378, 1365, 1298, 1250, 1184, 1040, 826, 780,
1
757 cm
.
Cleavage of the methyl aryl ether 6 with BBr3 to yield 4-(3',6'-dimethyl-3'-
heptyl)phenol (7): Cleavage of the methyl aryl ether with BI3 failed to give
any detectable amounts of the desired product, and so a change was made
to BBr3. The crude product 6 from the above microsynthesis step was
completely dried in a rotary evaporator and then dissolved in CH2Cl2
(15 mL) in a two-necked reaction flask (with stirrer and reflux). This was
cooled and stirred in an ice/water bath, and BBr3 (5 ± 6 drops, 184 mL,
1.936 mmol) was added to the reactants by pipette. The reactants were then
stirred for 1.5 hours at 08C before addition of distilled water (10 mL). The
1
HPLC, which gave an average water solubility value of 53.69 mgL
(0.244 molm 3) at 208C.
4794
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0722-4794 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 22

Branched Nonylphenol
4790±4795
Determination of vapor pressures (VP) and Henryꢀs Law constants (H) of
compound 7 and of 4-nonylphenol technical mixture: Vapor pressures were
determined by the effusion method.[20, 21] This method is also recommended
by the OECD for the determination of VPs of substances of low
volatility.[22]
[6] R. Ekelund, A. Bergman, A. Granmo, M. Berggren, Env. Poll. 1990,
64, 107 ± 120.
[7] J. A. Barth, Organikum: Organisch-Chemisches Grundpraktikum,
20th ed., Johann Ambrosius Barth, Heidelberg 1996, pp. 353 ± 360.
[8] J. March, Advanced Organic Chemistry: Reactions, Mechanisms and
Structure, 4th ed., Wiley, New York, 1992, pp. 431 ± 460.
[9] G. A. Olah, J. T. Welch, Y. D. Vankar, M. Nojima, I. Kerekes, J. A.
Olah, J. Org. Chem. 1979, 44, 3872 ± 3881.
[10] M. V. Bhatt, S. U. Mulkarni, Synthesis 1983, 249 ± 270.
[11] B. Jursic, J. Chem. Research (S) 1989, 284 ± 285.
[12] H. Niwa, T. Hida, K. Yamada, Tetrahedron Lett. 1981, 22, 4239 ± 4240.
[13] Y. Guindon, C. Yoakim, H. E. Morton, Tetrahedron Lett. 1983, 24,
2969 ± 2972.
[14] J. B. Press, Synthetic Comm. 1979, 9, 407 ± 410.
[15] Guidelines for the Testing of Chemicals, Vol. 1, 105, OECD, Paris,
1993.
[16] Guidelines for the Testing of Chemicals, Vol. 1, 107, OECD, Paris, 1993.
[17] M. Ahel, W. Giger, Chemosphere 1993, 26, 1471 ± 1478.
[18] M. Ahel, W. Giger, Chemosphere 1993, 26, 1461 ± 1470.
[19] C. T. Chiou, D. W. Schmedding, M. Manes, Environ. Sci. Technol.
1982, 16, 4 ± 10.
The sample was stored in an evaporation stove, acting as an effusion cell
under high vacuum conditions. The VP was calculated from the effusion
rate of the substance. The effusion rate was determined from the
condensation of molecules from the beam on a cooled plate of a highly
sensitive microbalance in a fixed time interval. The VP was measured at six
different temperatures between 23.00 and 48.508C, by using triplicate
points for the condensation, as well as impulse methods. From the
regression line, the VP of compound 7 was then calculated to be 1.93 Â
10 2 Pa at 208C and 3.57 Â 10 2 Pa at 258C. The Henryꢀs Law constant for
this compound was calculated from the relationship H ˆ VP/Sw to be 7.91 Â
10 2 Pam3 mol 1 at 208C. The VP of 4-nonylphenol (technical mixture) was
also determined to be 2.07 Â 10 2 Pa at 208C and 3.77 Â 10 2 Pa at 258C;
2
this gave H values of 83.87 Â 10 and 130.61 Â 10 2 Pam3 mol 1, respec-
tively. The air± water partition coefficient values given by H/RT were
1
4
3.25 Â 10 5 Pam3 mol 1 K (208C) for the isomer and 3.44 Â 10 and
5.27 Â 10 4 Pam3 mol 1 K 1 (208C and 258C, respectively) for the technical
nonylphenol mixture.
[20] M. A. V. R. Da Silva, M. J. S. Monte, Thermochim. Acta 1990, 171,
169 ± 183.
[21] J. Niederfellner, D. Lenoir, G. Matuschek, F. Rehfeldt, H. Utschik, R.
Brüggemann, Quant. Struct.-Act. Relat. 1997, 16, 38 ± 48.
[22] Guidelines for the Testing of Chemicals, Vol. 1, 104, OECD, Paris,
1993.
[23] H. Geyer, A. G. Kraus, W. Klein, E. Richter, F. Korte, Chemosphere
1980, 9, 277 ± 291.
[24] T. F. Bidleman, L. Renberg, Chemosphere 1985, 14, 1475 ± 1481.
[25] W.-Y. Shiu, K.-C. Ma, D. Varhanickova, D. MacKay, Chemosphere
1994, 29, 1155 ± 1224.
Acknowledgement
We wish to express our sincere gratitude to the Alexander von Humboldt
Foundation in Bonn for a Georg Forster Research Fellowship grant to J. O.
Lalah. We also wish to thank Prof. Dr. P. Lemmen of Munich Technical
University for performing the IR and refractive index tests.
[26] B. G. Oliver, A. J. Niimi, Environ. Sci. Technol. 1985, 19, 842 ± 849.
[27] J. O. Lalah, D. Lenoir, B. Henkelmann, G. Hertkorn, K. Günther,
K-W. Schramm, A. Kettrup, unpublished results.
[1] B. D. Bhatt, J. V. Prasad, G. Kalpana, S. Ali, J. Chrom. Sci. 1992, 30,
203 ± 210.
[28] O. Hutzinger, K.-W. Schramm, A. Fischer, K.-U. Goss, J. Klasmeier,
Umweltchemikalien/Schadstoffwirkung Forschungsbericht 10602067,
Umweltbundesamtes, Bayreuth Universität, 1991, pp. 289.
[2] R. Vittorio, S. Piergenido, J. Chrom. 1978, 153, 181 ± 182.
[3] T. F. Wheeler, J. R. Heim, M. R. LaTorre, A. B. Janes, J. Chrom. Sci.
1997, 35, 19 ± 30.
[4] A. C. Mehldal , K. Nithipatikom, J. J. Lech, Xenobiotica 1996, 26,
1167 ± 1180.
[5] W. Giger, E. Stephanous, C. Schaffner, Chemosphere 1981, 10, 1253 ±
1263.
Received: May 3, 2001 [F3233]
Chem. Eur. J. 2001, 7, No. 22
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0722-4795 $ 17.50+.50/0
4795