Oxidation of non-phenolic β-O-aryl-lignin model dimers catalysed by lignin peroxidase. Comparison with the oxidation induced by potassium 12-tungstocobalt(III)ate

Article abstract of DOI:10.1039/b101362i

The H₂O₂-promoted oxidations of the non-phenolic β-O-aryl-lignin model dimers 1-(3,4-dimethoxyphenyl)-2-phenoxyethanol (1) and 2-(4-methoxyphenoxy)-1-phenylethanol (2) catalysed by LiP at pH = 4.0 have been studied. The oxidation of 1 mainly leads to the corresponding ketone, indicating that the prevailing reaction of the intermediate radical cation 1 ^+. is Cα-H deprotonation. The oxidation of 2 forms 2-(4-methoxyphenoxy)2-phenylethanol (7, an isomer of 2), 2-phenyl-1,4-dioxaspiro[4.5]deca-6,9-dien-8-one (8) and products coming from the cleavage of the C-C bond γ to the more electron rich ring. The formation of all these products can be rationalised by assuming that the main reaction of the intermediate 2^+. is a nucleophilic attack of the alcoholic OH group on the ring bearing the positive charge. This leads to a spirocyclohexadienyl radical, which either is then oxidised to the dioxaspirodecadienone 8 or undergoes ring opening to give an alkoxyl radical from which the isomer of 2 and the C-C bond cleavage products may form. Support for this mechanism has been provided by a study of the oxidation of 4-MeOC₆H₄OCH₂CD₂OH and by comparing the results with those obtained when the alkoxyl radical 4-MeOC₆H₄OCH₂CD₂O^. was generated from 4-MeOC₆H₄OCH₂CD₂OO_tBu. The oxidation of 1 induced by the genuine one-electron oxidant potassium 12-tungstocobalt(III)ate at pH = 4.0 confirms the results obtained with LiP. However, under the same conditions, no fragmentation products were observed in the oxidation of 2, probably due to a fast oxidation, by potassium 12-tungstocobalt(III)ate, of the spirocyclohexadienyl radical.

Full text of DOI:10.1039/b101362i

View Article Online / Journal Homepage / Table of Contents for this issue
Oxidation of non-phenolic ꢀ-O-aryl-lignin model dimers catalysed
by lignin peroxidase. Comparison with the oxidation induced by
potassium 12-tungstocobalt(III)ate
2
Enrico Baciocchi,*a Massimo Bietti,b Maria Francesca Gerini,c Osvaldo Lanzalungac
and Simona Mancinellic
a Dipartimento di Chimica, Università “La Sapienza”, Ple A. Moro, 5 I-00185 Rome, Italy.
E-mail: baciocchi@caspur.it; Fax: ϩ39 (06) 490421
b Dipartimento di Scienze e Tecnologie Chimiche, Università “Tor Vergata”,
Via della Ricerca Scientifica, I-00133 Rome, Italy
c Dipartimento di Chimica and Centro CNR di Studio sui Meccanismi di Reazione,
Università “La Sapienza”, Ple A. Moro, 5 I-00185 Rome, Italy
Received (in Cambridge, UK) 12th February 2001, Accepted 17th April 2001
First published as an Advance Article on the web 23rd May 2001
The H2O2-promoted oxidations of the non-phenolic β-O-aryl-lignin model dimers 1-(3,4-dimethoxyphenyl)-
2-phenoxyethanol (1) and 2-(4-methoxyphenoxy)-1-phenylethanol (2) catalysed by LiP at pH = 4.0 have been
studied. The oxidation of 1 mainly leads to the corresponding ketone, indicating that the prevailing reaction of
the intermediate radical cation 1ϩ is Cα–H deprotonation. The oxidation of 2 forms 2-(4-methoxyphenoxy)-
ؒ
2-phenylethanol (7, an isomer of 2), 2-phenyl-1,4-dioxaspiro[4.5]deca-6,9-dien-8-one (8) and products coming from
the cleavage of the C–C bond γ to the more electron rich ring. The formation of all these products can be rationalised
by assuming that the main reaction of the intermediate 2ϩ is a nucleophilic attack of the alcoholic OH group on
ؒ
the ring bearing the positive charge. This leads to a spirocyclohexadienyl radical, which either is then oxidised to
the dioxaspirodecadienone 8 or undergoes ring opening to give an alkoxyl radical from which the isomer of 2 and
the C–C bond cleavage products may form. Support for this mechanism has been provided by a study of the
oxidation of 4-MeOC6H4OCH2CD2OH and by comparing the results with those obtained when the alkoxyl radical
ؒ
4-MeOC6H4OCH2CD2O was generated from 4-MeOC6H4OCH2CD2OOt-Bu. The oxidation of 1 induced by the
genuine one-electron oxidant potassium 12-tungstocobalt()ate at pH = 4.0 confirms the results obtained with
LiP. However, under the same conditions, no fragmentation products were observed in the oxidation of 2,
probably due to a fast oxidation, by potassium 12-tungstocobalt()ate, of the spirocyclohexadienyl radical.
by oxidation of the native enzyme by H2O2. This electron trans-
fer leads to the formation of LiP Compound II (LiPII), [Por-
Introduction
Lignin, a three-dimensional phenolic polymer built from
Fe()᎐O], and the substrate radical cation. The latter should
᎐
phenylpropane units linked together by different bonds, is one
of the most abundant biopolymers on earth. The presence of
lignin mixed with hemicelluloses within the cellulosic fibre wall
of wood cells creates a naturally occurring composite material
which imparts strength and rigidity to trees and plants, provid-
ing protection from oxidative processes and from attacks by
microorganisms.1,2
then undergo non-enzymatic side-chain fragmentation reac-
tions10 (some exceptions are however possible11,12) involving the
cleavage of C–C and C–H β bonds (β with respect to the ring
bearing the SOMO), thus initiating the degradation of lignin.
A very simplified representation of this process is shown in
Scheme 1, where only two aromatic rings are present, linked
Studies concerning the oxidative degradation of lignin raise
continuous interest for the following two reasons. First, this
process can convert lignin into low molecular weight aromatic
compounds, thus making this polymer a renewable source for
the industrial preparation of a number of chemicals.3 Second,
and much more important, the selective degradation of lignin
and its removal from the carbohydrate component is a key
process in the pulp and paper industry.4
Recently, particular attention has been given to the possi-
bility of degrading lignin with fungi, which would represent
a process with low environmental impact and energy con-
sumption.5–8 Among the various fungi performing this task, the
white rot basidiomycetous Phanerochaete chrysosporium, which
secretes the ligninolytic enzyme lignin peroxidase (LiP), has
attracted the most interest.9 LiP is a ferric hemoprotein able to
catalyse the oxidative degradation of lignin by the transfer of
one electron from an aromatic ring of the substrate to LiP
Compound I (LiPI), the active species of the enzyme (an
Scheme 1
through a β-O-aryl bond,13 which is the most common linkage
in lignin.1
As it is of interest to understand the detailed mechanism by
which these fragmentations occur and also the structural fac-
tors that play a role in determining the type of bond that
undergoes the cleavage, we have now investigated the LiP-
promoted oxidation of the lignin model compounds 1 and 2,
both containing a β-O-aryl bond. These models possess differ-
iron()-oxo porphyrin radical cation [Porϩ Fe()᎐O]), formed
ؒ
᎐
1506
J. Chem. Soc., Perkin Trans. 2, 2001, 1506–1511
This journal is © The Royal Society of Chemistry 2001
DOI: 10.1039/b101362i

View Article Online
Table 1 Products and yields in the LiP–H2O2 and Co()W-catalysed
oxidation of 1 and 2
ent substitution patterns of the two aromatic rings and should
therefore form two radical cations that differ in the position of
the positive charge, which is located in the ring with two alkoxy
substituents.14 Thus, unambiguous information on how the
position of the positive charge in the radical cation influences
the cleavage of the C–C and C–H bonds can be obtained. The
Products (yield, %)a
Substrate
LiP–H2O2
Co()W
results of the enzymatic study have also been compared with
1
2
3 (31)
4 (3)
5 (32)
6 (23)
7 (12)
8 (18)
3 (26)
4 (2)
8 (5)
ϩ
those obtained when 1ϩ and 2 have been generated with the
genuine one-electron oxidant K5[Co()W12O40], from now on
simply indicated as Co()W.15,16
ؒ
ؒ
a Yields are given with respect to the substrate that is equimolar with the
oxidant, H2O2. Average of at least three determinations. The material
balance was >95% in all the experiments.
bond (Scheme 2, path a), leads to an α-hydroxybenzylic radical,
which, by oxidation followed by proton loss, produces the
observed ketone (path b). The cleavage of the β C–C bond leads
to 3,4-dimethoxybenzaldehyde (path c).20 From the relative
yields of ketone and aldehyde, it clearly appears that the main
pathway followed by 1ϩ is Cα–H deprotonation. C–C bond
ؒ
breaking is a minor route, probably due to the fact that such a
cleavage forms a primary carbon radical, which is much less
stable than the radical formed in the deprotonation pathway.21
A different situation arises in the LiP-catalysed oxidation of
2 where substantial amounts of C–C bond cleavage products
(benzaldehyde 5 and 4-methoxyphenol 6) are formed together
with the isomerization product 7 and the 2-phenyl-1,4-
dioxaspiro[4.5]deca-6,9-dien-8-one, 8.
Results
The reactions of 1 and 2 with LiP were carried out in an Ar-
saturated, 50 mM sodium tartrate buffer solution (pH = 4.0)
containing 5% CH3CN as cosolvent, at 25 ЊC. An equimolar
amount of H2O2 in the buffer solution was gradually added
over 1 h using an infusion pump. After the usual work-up,
1
products and yields were determined by GC, GC-MS and H
The C–C bond cleavage in the radical cation 2ϩ , formed by
ؒ
NMR. In the reaction of 1 the major product observed was the
corresponding ketone 3, accompanied by a small amount of
the Cα–Cβ bond cleavage product, 3,4-dimethoxybenzaldehyde
(4). In contrast, with 2, the Cα–Cβ bond cleavage products,
benzaldehyde (5) and 4-methoxyphenol (6), were formed
in substantial amounts along with 2-(4-methoxyphenoxy)-2-
phenylethanol (7, an isomer of 2)17 and 2-phenyl-1,4-dioxa-
spiro[4.5]deca-6,9-dien-8-one (8).18 In the absence of the oxidant
or in the absence of LiP no products were detected.
oxidation of 2 with LiPI, is very surprising as it involves a bond
which is γ with respect to the ring bearing the positive charge in
2ϩ . Accordingly, no interaction between this bond and the
ؒ
SOMO of the radical cation can be envisaged. As far as we
know, no example of the direct cleavage of a γ bond in alkyl-
aromatic radical cations has hitherto been reported.
A possible explanation, strongly supported by the formation
of products 7 and 8, is that for 2ϩ the most important reaction
ؒ
pathway is the intramolecular nucleophilic attack of the alco-
holic OH group on the oxygen-substituted position of the posi-
tively charged ring to form the spirocyclohexadienyl radical 9
(Scheme 3, path a), a possibility which has been previously pro-
posed by Gold and co-workers17 and by Gilbert and McCleland
in similar systems.22 As suggested by these authors, opening of
this spirocyclohexadienyl radical can involve both geminal C–O
bonds leading to the alkoxyl radicals 10 and 11 (Scheme 3,
paths b and c respectively). By hydrogen-atom abstraction,
10 can form the starting material 2 whereas 11 produces the
isomerized product 7 (Scheme 3, paths e and g respectively). In
addition, we now suggest that 10 and 11, in competition with
hydrogen-atom abstraction, can also undergo a β C–C bond
cleavage, a well known process for alkoxyl radicals.19 From 10,
benzaldehyde and the phenoxy-substituted carbon radical 12
may form (Scheme 3, path f), the latter being eventually con-
verted (by oxidation and hydrolysis) to 4-methoxyphenol. From
11, formaldehyde and the secondary carbon radical 13 can
be formed (Scheme 3, path h). Oxidation of 13 followed by
hydrolysis forms again benzaldehyde and 4-methoxyphenol. It
is very likely that compound 8 is also derived from 9, by oxid-
ation followed by water addition and methanol loss (Scheme 3,
path d).
The Co()W-induced oxidations of 1 and 2 were carried out
at pH = 4.0 (50 mM sodium tartrate buffer solution, containing
5% CH3CN as cosolvent) using equimolar amounts of oxidant
and substrate, under an argon atmosphere, at 70 ЊC and for 72
hours. In the reaction of 1 the main product found was the
corresponding ketone 3, together with a small amount of 3,4-
dimethoxybenzaldehyde 4, as observed in the LiP-catalysed
reaction. The oxidation of 2 showed low efficiency and led only
to 8, which was obtained with a 5% yield with respect to the
starting material.18 In the absence of the oxidant no products
were detected. All of the results are collected in Table 1.
In order to test the mechanism presented in Scheme 3 we
have also studied the oxidation with LiP of the isomerized
product 7 and of the partially deuterated compound 14. The
results are shown in Table 2.
Discussion
It is reasonable to attribute the formation of 3 and 4 in the LiP-
catalysed oxidation of 1 to the decay of the radical cation 1ϩ
ؒ
derived from the one-electron oxidation of 1 by LiPI. Once
formed, 1ϩ undergoes one of the most typical reactions of
ؒ
alkylaromatic radical cations, i.e. the cleavage of a bond β to
the ring bearing the positive charge.19 The cleavage of the Cα–H
J. Chem. Soc., Perkin Trans. 2, 2001, 1506–1511
1507

View Article Online
Scheme 2 Fragmentation pathways for 1ϩ
.
ؒ
Scheme 3 Proposed mechanism for the decomposition pathways of 2ϩ
.
ؒ
Table 2 Products and yields in the LiP-catalysed oxidation of 7 and 14
group leads to the same spirocyclohexadienyl radical formed
from 2ϩ . The study of the oxidation of 14 also provided results
ؒ
Substrate
Products (yield, %)a
in line with the proposed mechanism. Accordingly, the oxid-
ation of 14 led to its isomer 15, where the CH2 and CD2 groups
have inverted their relative position, and the dioxaspirodeca-
dienone 16.18 In order to find further evidence for the inter-
mediacy of the alkoxyl radical in the isomerization process, the
peroxide 17 was synthesised and its photolysis was carried out
(see Experimental). It was observed that such a reaction pro-
duces 14 and its isomer 15. Since the photolysis of 17 leads to
the alkoxyl radical 18, by O–O bond cleavage, this result clearly
shows that 18 is an intermediate en route to 14 and 15 (Scheme
4). From 18 no β fragmentation products are obtained, probably
since the β fragmentation of 18 leads to formaldehyde and the
primary carbon radical 12 and it is therefore much less favoured
with respect to that of 10 and 11.
7
2 (11)
5 (16)
6 (14)
8 (11)
15 (30)
16 (28)
14
a Yields are given with respect to the substrate that is equimolar with the
oxidant, H2O2. Average of at least three determinations. The material
balance was >95% in all the experiments.
There is, however, another mechanistic possibility whereby
an alkoxyl radical could be formed from 2ϩ without the inter-
ؒ
We can see that the LiP-catalysed oxidation of 7 leads to its
isomer 2 which is formed together with the same products (the
dioxaspirodecadienone 8, benzaldehyde 5 and 4-methoxy-
phenol 6) observed when 2 was the starting substrate. This
result is fully consistent with the mechanism shown in Scheme 3
vention of the spirocyclohexadienyl intermediate. This is based
on our recent observations that arylalkanol radical cations
exhibit oxygen acidity up to when the OH group is in the γ-
position.23,24 Thus, the alkoxyl radical 10 might also be formed
by OH deprotonation in 2ϩ as shown in Scheme 5. However, we
ؒ
as in 7ϩ , the intramolecular nucleophilic attack by the OH
feel that this possibility is unlikely as the radical cations gener-
ؒ
1508
J. Chem. Soc., Perkin Trans. 2, 2001, 1506–1511

View Article Online
Chemical and enzymatic oxidations seem instead to behave
differently in the case of 2 (Table 2). Accordingly, the oxidation
of 2 by Co()W leads exclusively to the formation of the dioxa-
spirodecadienone 8 and does not produce the isomerization or
fragmentation products observed with LiP. However, we sug-
gest that this difference is only apparent because in the oxid-
ation promoted by Co()W the spirocyclohexadienyl radical,
once formed, preferentially undergoes oxidation to 8, rather
than ring opening to form an alkoxyl radical (Scheme 3). Thus,
no isomerization and fragmentation products are formed.
Accordingly, the strong capacity of Co()W to oxidise carbon
radicals is well known.25
Conclusion
The results reported in this paper have shown that the oxidation
of the lignin model compound 1 catalysed by LiP leads to a
radical cation that preferentially undergoes deprotonation at
the Cα–H bond. Cleavage of the β carbon bond is a minor
route. Similar results have been obtained by using the bona fide
one-electron oxidant Co()W.
A somewhat different situation is observed when 2 is the
substrate. In this case, the LiP-promoted oxidation forms sub-
stantial amounts of the C–C bond cleavage products together
with the isomerization product 7 and dioxaspirodecadienone 8.
Evidence has been obtained that all these products are derived
from the spirocyclohexadienyl radical 9, formed by an intra-
molecular nucleophilic attack of the alcoholic OH group at the
positively charged ring in the radical cation. Radical 9 can then
undergo ring opening to form an alkoxyl radical from which the
isomer 7 and the products of C–C bond cleavage are formed.
Evidence for the intermediacy of an alkoxyl radical has been
obtained. Oxidation of 9 can also occur to produce the dioxa-
spirodecadienone 8. The latter reaction is the only one observed
Scheme 4 Photolysis of peroxide 17.
when Co()W is used as oxidant.
ϩ
The different behaviours of 1ϩ and 2 with respect to the
ؒ
ؒ
fragmentation pathway can be rationalised on the basis of the
different structural features of the two radical cations. In 1ϩ
ؒ
the easily scissile C–H and C–C β bonds are present and accord-
ingly the radical cation decays by cleavage of these bonds. No
such bonds are present in 2ϩ in which, however, intramolecular
ؒ
nucleophilic attack of the alcoholic OH group on the oxygen-
substituted position of the positively charged aromatic ring is
possible, leading to a spirocyclohexadienyl radical from which
all the observed products are formed.
Scheme 5 Fragmentation reaction induced by OH deprotonation of 2ϩ
.
ؒ
ally display OH acidity in basic media, whereas LiP-catalysed
Experimental
reactions take place at pH 4. Some specific interaction between
an amino acid residue of the enzyme with the OH group in 2ϩ
ؒ
Instrumentation
might be envisaged, but at present no evidence in this respect is
available.
1H NMR spectra were recorded on a Bruker AC300P spec-
trometer in CDCl3. GC-MS analyses were performed on an
HP5890 GC (OV1 capillary column, 12 m × 0.2 mm) coupled
with an HP5970 MSD. GC analyses were carried out on a
Varian 3400 GC (OV1 capillary column, 25 m × 0.2 mm).
UV–Vis measurements were performed on a Perkin Elmer
Lambda 18 spectrophotometer. Melting points were recorded
using a Büchi 510 melting point apparatus.
In summary, the mechanism illustrated in Scheme 3 appears
to satisfactorily account for the formation of 7 and 8 from 2ϩ
,
ؒ
as well as for that of the products derived from C–C bond
cleavage. If this mechanism is correct, an interesting conclusion
is that the possibility of forming a spirocyclohexadienyl radical
in β-O-aryl substructures seems to be an important factor with
respect to oxidative C–C bond cleavage in lignin.
The oxidation of 1 with Co()W, a well recognised one-
electron oxidant,15,16 has given results very similar to those
observed in the LiP-catalysed oxidation, Table 1. The ketone 3
is the predominant product, accompanied by small amounts of
3,4-dimethoxybenzaldehyde. Clearly, Cα–H deprotonation is
Substrates and reagents
All the reagents and solvents were of the highest purity avail-
able and used without further purification (unless otherwise
specified). The concentration of H2O2 (Carlo Erba Reagents)
was determined by titration with permanganate.26 Co()W was
prepared using the literature procedure15 with some modi-
fications.16 LiP was prepared and purified as described in the
literature.27 The concentration of the enzyme solution was deter-
mined spectrophotometrically [ε(409 nm) = 169 mMϪ1 cmϪ1].28
the main reaction of 1ϩ , whereas the cleavage of the β C–C
ؒ
bond is a minor route, as observed in the LiP-promoted reac-
tion. This suggests that the behaviour of 1ϩ is independent of
ؒ
its mode of generation (enzymatic or chemical) and fully sup-
ports the intervention of a free radical cation in the LiP-
catalysed oxidation.
J. Chem. Soc., Perkin Trans. 2, 2001, 1506–1511
1509

View Article Online
1-(3,4-Dimethoxyphenyl)-2-phenoxyethanol (1)
The reaction mixture was extracted with dichloromethane, the
organic layer was washed with water, with NaHCO3 saturated
solution and then dried over anhydrous MgSO4. The product 2
(8.8 g, 36.1 mmol, yield 91%) was purified by column chrom-
atography on silica gel using dichloromethane as eluent. Mp
59–61 ЊC (from dichloromethane). 1H NMR (CDCl3): δ 3.77 (3
H, s, OCH3), 3.92–4.09 (2 H, m, CH2), 5.08–5.12 (1 H, m, CH),
6.81–6.89 (4 H, m, Ar), 7.31–7.48 (5 H, m, Ar). MS m/z (%): 244
(20), 124 (100), 109 (40), 77 (30).
2-Bromo-1-(3,4-dimethoxyphenyl)ethanone. To a stirred solu-
tion of 3,4-dimethoxyacetophenone (Aldrich) (5.0 g, 27.8
mmol) in chloroform (20 mL) a solution of bromine (C. Erba)
(1.4 mL, 27.5 mmol) in chloroform (7 mL) was added. After
refluxing the solution for 3 hours, the solvent was removed
under reduced pressure. Recrystallization from methanol gave
1
3.2 g (12.3 mmol) of the brominated derivative, 44% yield. H
NMR (CDCl3): δ 3.93 (3 H, s, OCH3), 3.95 (3 H, s, OCH3), 4.43
(2 H, s, CH2), 6.88–6.93 (1 H, m, Ar), 7.52–7.53 (1 H, m, Ar),
7.58–7.63 (1 H, m, Ar).
2-(4-Methoxyphenoxy)-2-phenylethanol (7)
To a stirred solution of 4-methoxyphenol (Aldrich) (4.7 g,
37.9 mmol) and anhydrous K2CO3 (8.0 g, 58.0 mmol) in EtOH
(60 mL) was added styrene oxide (Aldrich) (5.0 g, 41.7 mmol).
After the solution had been refluxed for 2 hours water (20 mL)
was added and the mixture was extracted with ether. The
organic layer was washed with NaOH (1 mM) and dried over
anhydrous Na2SO4. The product was purified by column
chromatography on silica gel using ethyl acetate–petroleum
ether in an elution gradient of 1 : 50–1 : 5 (3.9 g, 16.0 mmol,
yield 38%). Mp 85–87 ЊC (from ethyl acetate). 1H NMR
(CDCl3): δ 3.71 (3 H, s, OCH3), 3.75–3.96 (2 H, m, CH2), 5.14–
5.19 (1 H, m, CH), 6.70–6.83 (4 H, m, Ar), 7.30–7.37 (5 H, m,
Ar). MS m/z (%): 244 (6), 124 (100), 109 (36), 77 (12).
1-(3,4-Dimethoxyphenyl)-2-phenoxyethanone. To a stirred
solution of phenol (C. Erba) (1.2 g, 12.7 mmol) in acetone
(25 mL) K2CO3 (Fluka) (1.8 g, 13.0 mmol) and 2-bromo-1-(3,4-
dimethoxyphenyl)ethanone (3.2 g, 12.3 mmol) were added.
After 1 hour the inorganic salts were filtered off and washed
with acetone. The combined filtrates were concentrated under
reduced pressure. The residue was dissolved with dichloro-
methane, the solution was washed with water, then with brine
and dried over anhydrous MgSO4. The product was purified by
column chromatography on silica gel using ethyl acetate–
petroleum ether 1 : 3 as eluent, 77% yield. 1H NMR (CDCl3): δ
3.91 (3 H, s, OCH3), 3.93 (3 H, s, OCH3), 5.21 (2 H, s, CH2),
6.87–6.91 (1 H, m, Ar), 6.92–7.00 (3 H, m, Ar), 7.22–7.31 (2 H,
m, Ar), 7.55–7.56 (1 H, m, Ar), 7.62–7.67 (1 H, m, Ar).
2-(4-Methoxyphenoxy)[1,1-2H2]ethanol (14)
1-(3,4-Dimethoxyphenyl)-2-phenoxyethanol (1). To a stirred
solution of NaBH4 (Aldrich) (1.1 g, 29 mmol) in methanol
(10 mL) a solution of 1-(3,4-dimethoxyphenyl)-2-phenoxy-
ethanone (2.6 g, 9.5 mmol) in methanol (10 mL) was added.
After 2 hours 30 mL of water–acetic acid (5 : 1) were added.
The reaction mixture was extracted with dichloromethane, the
organic layer was washed with water, with NaHCO3 solu-
tion and then dried over anhydrous MgSO4. The product was
purified by column chromatography on silica gel using dichloro-
methane as eluent, 92% yield. Mp 95–97 ЊC (from dichloro-
methane). 1H NMR (CDCl3): δ 2.77 (1 H, s, OH), 3.89 (3 H, s,
OCH3), 3.91 (3 H, s, OCH3), 3.96–4.13 (2 H, m, CH2), 5.05–5.11
(1 H, m, CH), 6.86–7.02 (5 H, m, Ar), 7.24–7.34 (3 H, m, Ar).
MS m/z (%): 274 (10), 167 (100), 139 (65), 77 (30).
(4-Methoxyphenoxy)acetic acid. Bromoacetic acid (Aldrich)
(10.5 g, 74.4 mmol) was added to a stirred solution of 4-
methoxyphenol (Aldrich) (6.0 g, 48.3 mmol), NaOH (5.0 g,
125 mmol) and Bu4NI (1.0 g, 2.7 mmol) in H2O (100 mL) and
benzene (50 mL). After reflux for 6 hours the solvent was par-
tially removed under reduced pressure. After the addition of a
saturated solution of Na2CO3 the reaction mixture was
extracted with chloroform, then a dilute solution of HCl was
added to the aqueous phase and the mixture was extracted with
chloroform. The organic layers were dried over anhydrous
Na2SO4. After removal of the solvent under reduced pressure
4.5 g of (4-methoxyphenoxy)acetic acid were obtained
(24.7 mmol, yield 51%). 1H NMR (CDCl3): δ 3.78 (3 H, s,
OCH3), 4.64 (2 H, s, CH2), 6.86–6.87 (4 H, m, Ar).
2-(4-Methoxyphenoxy)-1-phenylethanol (2)
2-(4-Methoxyphenoxy)[1,1-2H2]ethanol (14). At 0 ЊC a sus-
pension of 2.1 g of LiAlD4 (Aldrich) (50.0 mmol) in 15 mL of
anhydrous THF was gradually added to a stirred solution of
(4-methoxyphenoxy)acetic acid (4.5 g, 24.7 mmol) in 10 mL of
anhydrous THF under an argon atmosphere. The reaction mix-
ture was equilibrated at room temperature and, after 3 hours,
50 mL of H2O were added to remove the unreacted LiAlD4.
The precipitate was removed by filtration and the solvent was
evaporated under reduced pressure. A solution of Na2CO3 was
addedtotheresidueandthemixturewasextractedwithdichloro-
methane to give 3.7 g of product (21.7 mmol, yield 88%). Mp
2-Bromo-1-phenylethanone. 2-Bromo-1-phenylethanone was
prepared as previously described for 2-bromo-1-(3,4-dimeth-
oxyphenyl)ethanone. 1H NMR (CDCl3): δ 3.74 (3 H, s, OCH3),
4.41 (2 H, s, CH2), 6.88–6.92 (2 H, m, Ar), 7.47–7.65 (3 H, m,
Ar).
2-(4-Methoxyphenoxy)-1-phenylethanone. To a stirred solu-
tion of 4-methoxyphenol (Aldrich) (6.9 g, 55.6 mmol) in
acetone (100 mL), NaOH (2.2 g, 55.0 mmol) and 2-bromo-
1-phenylethanone (11.0 g, 55.1 mmol) were added. After the
solution had been refluxed for 1 hour the solvent was removed
under reduced pressure. The residue was dissolved with dichloro-
methane, and the solution was washed with water, then with
NaOH (1 mM) and dried over anhydrous MgSO4. 2-(4-
Methoxyphenoxy)-1-phenylethanone (9.6 g, 39.6 mmol, 72%
yield) was purified by column chromatography on silica gel
using dichloromethane–petroleum ether in an elution gradient
1
69–72 ЊC (from dichloromethane). H NMR (CDCl3): δ 2.15
(1 H, s, OH), 3.77 (3 H, s, OCH3), 4.02 (2 H, s, CH2), 6.84–6.85
(4 H, m, Ar). MS m/z (%): 170 (55), 124 (100), 110 (20), 109 (85).
2-(4-Methoxyphenoxy)[1,1-2H2]ethyl tert-butyl peroxide (17)
2-(4-Methoxyphenoxy)bromo[1,1-2H2]ethane. To
a stirred
1
of 1 : 20–1 : 1. H NMR (CDCl3): δ 3.76 (3 H, s, OCH3), 5.23
solution of 2-(4-methoxyphenoxy)[1,1-2H2]ethanol (14)
(170 mg, 1.0 mmol) in anhydrous ether (3 mL) was added a
solution of PBr3 (Aldrich) (270 mg, 1.0 mmol) in anhydrous
ether (1 mL) dropwise at 0 ЊC. After 6 hours the reaction mix-
ture was extracted with water and dried over anhydrous
Na2SO4. The product (0.4 mmol, yield 40%) was purified by
column chromatography on Florisil using n-hexane as eluent.
1H NMR (CDCl3): δ 3.76 (3 H, s, OCH3), 4.22 (2H, s, CH2),
6.79–6.90 (4 H, m, Ar).
(2 H, s, CH2), 6.81–6.84 (2 H, m, Ar), 6.88–6.92 (2 H, m, Ar),
7.47–7.65 (3 H, m, Ar), 7.98–8.01 (2 H, m, Ar).
2-(4-Methoxyphenoxy)-1-phenylethanol (2). To a stirred solu-
tion of NaBH4 (Aldrich) (4.5 g, 120 mmol) in methanol
(60 mL)
a
solution of 2-(4-methoxyphenoxy)-1-phenyl-
ethanone (9.6 g, 39.6 mmol) in methanol (100 mL) was added.
After 2 hours 300 mL of water–acetic acid 5 : 1 were added.
1510
J. Chem. Soc., Perkin Trans. 2, 2001, 1506–1511

View Article Online
2-(4-Methoxyphenoxy)[1,1-2H2]ethyl tert-butyl peroxide (17).
Compound 17 was prepared by reacting 2-(4-methoxy-
phenoxy)bromo[1,1-2H2]ethane with t-BuOOH as described in
the literature.29 Column chromatography on Florisil using
n-hexane as eluent gave the pure product (yield 47%). 1H NMR
(CDCl3): δ 1.25 (9 H, s, C(CH3)3), 3.77 (3 H, s, OCH3), 4.13
(2 H, s, CH2), 6.79–6.90 (4 H, m, Ar).
5 H. Janshekar and A. Fiechter, Adv. Biochem. Eng./Biotechnol., 1983,
27, 119.
6 K. E. L. Eriksson, R. A. Blanchette and P. Ander, Microbial and
Enzymatic Degradation of Wood and Wood Components, Springer-
Verlag, Berlin, 1990.
7 J. F. Levy, Cellulose Sources and Exploitation, eds. J. F. Kennedy,
G. O. Phillips and P. A. Williams, Ellis Horwood, New York, 1990,
pp. 397–408.
8 S. L. Pedlar and W. B. Betts, Cellulose Sources and Exploitation,
eds. J. F. Kennedy, G. O. Phillips and P. A. Williams, Ellis Horwood,
New York, 1990, pp. 435–442.
Enzymatic oxidation
9 P. J. Kersten, S. K. Stephens and T. K. Kirk, Biotechnology in the
Pulp and Paper Manufacture, eds. T. K. Kirk and H. M. Chang,
Butterworth-Heinemann, Stoneham, MA, 1990, pp. 457–463.
10 For a general overview see: H. E. Schoemaker, Recl. Trav. Chim.
Pays-Bas, 1990, 109, 255; G. Labat and B. Meunier, Bull. Soc. Chim.
Fr., 1990, 127, 553; H. B. Dunford, Heme Peroxidases, H. B.
Dunford, Wiley-VCH, New York, 1999, pp. 281–295.
11 E. Baciocchi, M. F. Gerini, P. J. Harvey, O. Lanzalunga and S.
Mancinelli, Eur. J. Biochem., 2000, 267, 2705.
The oxidant H2O2 (20 µmol) was added, over a period of 1 h by
an infusion pump, to a magnetically stirred argon-degassed
solution of the substrate (20 µmol) and LiP (0.63 units,
1.0 nmol) in 10 mL of a 50 mM sodium tartrate buffered solu-
tion, pH 4, with 5% of acetonitrile as cosolvent, at 25 ЊC. The
products of the reaction were extracted with CH2Cl2 and dried
over anhydrous Na2SO4.
12 E. Baciocchi, M. F. Gerini, O. Lanzalunga, A. Lapi, S. Mancinelli
and P. Mencarelli, Chem. Commun., 2000, 393.
Chemical oxidation
13 A β-O-aryl bond indicates the bond between the carbon β to an
aromatic ring and the oxygen of another aromatic ring.
14 It is reasonably assumed that no significant interaction occurs
between the two phenyl rings.
15 L. Eberson, J. Am. Chem. Soc., 1983, 105, 3192.
16 E. Baciocchi, M. Crescenzi, E. Fasella and M. Mattioli, J. Org.
Chem., 1992, 57, 4684.
The oxidant Co()W (20 µmol) and the substrate (20 µmol)
were magnetically stirred in 10 mL of an argon-degassed
50 mM sodium tartrate buffered solution, pH 4, with 5% of
acetonitrile as cosolvent, at 70 ЊC. After 3 days the products
of the reaction were extracted with CH2Cl2 and dried over
anhydrous Na2SO4.
17 In the literature the hypothesis of an isomerization process for the
radical cation of lignin model compounds has already been
proposed: K. Miki, V. Renganathan and M. H. Gold, Biochemistry,
1986, 25, 4790; K. Miki, V. Renganathan and M. H. Gold, FEBS
Lett., 1986, 203, 235; K. Miki, R. Kondo, V. Renganathan, M. B.
Mayfield and M. H. Gold, Biochemistry, 1988, 27, 4787.
18 Compounds 8 and 16 have been identified by comparison of their
Product analysis
Reaction products were identified by GC, GC-MS and 1H
NMR by comparison with authentic specimens. Yields were
determined by GC and 1H NMR (using 4-methoxyaceto-
phenone as the internal standard) with respect to the starting
material. A good material balance (>95%) was observed in all
the experiments.
1
GC-MS and H NMR spectra with those reported in the literature
for the parent compound 1,4-dioxaspiro[4.5]deca-6,9-dien-8-one:
J. E. Heller, A. S. Dreiding, B. R. O’Connor, H. E. Simmons,
G. L. Buchanan, R. A. Raphael and R. Taylor, Helv. Chim. Acta,
1973, 56, 272.
Photochemical reaction
19 E. Baciocchi, M. Bietti and O. Lanzalunga, Acc. Chem. Res., 2000,
33, 243.
The peroxide 17 (40 µmol), in 5 mL of argon-degassed 50 mM
sodium tartrate buffered solution, pH 4, with 5% of acetonitrile
as cosolvent, was irradiated in a Rayonet photoreactor (16 × 21
W sunlight phosphor lamps, centre of emission 300 nm) at
25 ЊC for 1 hour. The reaction products were extracted with
CH2Cl2, dried over anhydrous Na2SO4 and analysed by 1H
NMR.
20 The Cα–Cβ bond cleavage is, in this case, homolytic. Its occurrence
is shown by the formation of PhOCH2CH2OPh observed among the
products of the decay of 1ϩ when generated in γ-radiolysis
ؒ
experiments. E. Baciocchi, M. F. Gerini and S. Steenken,
unpublished work.
21 The primary carbon radical formed in Scheme 2, path c should
eventually produce phenol, which, however, has not been observed
among the reaction products, probably since it is formed in very
small amounts and it is easily oxidizable.
Acknowledgements
22 Intramolecular nucleophilic attack of the alcoholic OH group on
the positively charged ring of 3-phenylpropan-1-ol to form a
spirocyclohexadienyl radical was also proposed by Gilbert and
McCleland. B. C. Gilbert and C. W. McCleland, J. Chem. Soc.,
Perkin Trans. 2, 1989, 1545.
23 E. Baciocchi, M. Bietti and S. Steenken, Chem. Eur. J., 1999, 5, 1785.
24 E. Baciocchi, M. Bietti, L. Manduchi and S. Steenken, J. Am. Chem.
Soc., 1999, 121, 6624.
This work was carried out in the framework of the EU project
Towards Efficient Oxygen Delignification (Contract No. QLK5-
CT-1999-01277). The authors also thank Dr Patricia J. Harvey
for providing a sample of LiP.
25 E. Baciocchi, M. Bietti and S. Steenken, J. Chem. Soc., Perkin Trans.
2, 1996, 1261.
26 H. A. Flascka, A. J. Barnard, Jr. and P. E. Sturrock, Quantitative
Analytical Chemistry, Harper & Row, New York, 1969, vol. 2,
pp. 149–167.
References
1 K. V. Sarkanen, Lignins: Occurrence, Formation, Structure and
Reactions, eds. K. V. Sarkanen and C. H. Ludwig, Wiley-Interscience,
New York, 1971, pp. 95–195.
2 D. S. Argyropoulos and S. B. Menachem, Biotechnology in the Pulp
and Paper Industry, ed. K. E. L. Eriksson, Springer-Verlag, Berlin,
Heidelberg, 1997, pp. 127–158.
3 H.-R. Biørsvik and F. Minisci, Org. Process Res. Dev., 1999, 3, 330.
4 J. C. Roberts, The Chemistry of Paper, The Royal Society of
Chemistry, Cambridge, UK, 1996.
27 M. Tien and T. K. Kirk, Methods Enzymol., 1988, 161, 238.
28 M. Tien, T. K. Kirk, C. Bull and J. A. Fee, J. Biol. Chem., 1986, 261,
1687.
29 J. Moulines, M. J. Bougeois, M. Campagnole, A. M. Lamidey,
B. Maillard and E. Montaudon, Synth. Commun., 1990, 20,
349.
J. Chem. Soc., Perkin Trans. 2, 2001, 1506–1511
1511