Generation and reactivity of ketyl radicals with lignin related structures. On the importance of the ketyl pathway in the photoyellowing of lignin containing pulps and papers

Article abstract of DOI:10.1021/jo047826u

(Chemical Equation Presented) Ketyl radicals with lignin related structures have been generated by means of radiation chemical and photochemical techniques. In the former studies ketyl radicals are produced by reaction of α-carbonyl-β-aryl ether lignin models with the solvated electron produced by pulse radiolysis of an aqueous solution at pH 6.0. The UV-vis spectra of ketyl radicals are characterized by three main absorption bands. The shape and position of these bands slightly change when the spectra are recorded in alkaline solution (pH 11.0) being now assigned to the ketyl radical anions and a pK_a = 9.5 is determined for the 1-(3,4,5-trimethoxyphenyl)-2- phenoxyethanol-1-yl radical. Decay rates of ketyl radicals are found to be dose dependent and, at low doses, lie in the range (1.7-2.7) × 10³ s^-1. In the presence of oxygen a fast decay of the ketyl radicals is observed (k₂ = 1.8-2.7 × 10⁹ M^-1 s ^-1) that is accompanied by the formation of stable products, i.e., the starting ketones. In the photochemical studies ketyl radicals have been produced by charge-transfer (CT) photoactivation of the electron donor-acceptor salts of methyl viologen (MV²⁺) with α-hydroxy-α- phenoxymethylaryl acetates. This process leads to the instantaneous formation of the reduced acceptor (methyl viologen radical cation, MV^+?), as is clearly shown in a laser flash photolysis experiment by the two absorption bands centered at 390 and 605 nm, and an acyloxyl radical [ArC(CO ₂^?)(OH)CH₂(OC₆H₅)], which undergoes a very fast decarboxylation with formation of the ketyl radicals. Steady-state photoirradiation of the CT ion pairs indicates that 1-aryl-2-phenoxyethanones are formed as primary photoproducts by oxidation of ketyl radicals by MV²⁺ (under argon) or by molecular oxygen. Small amounts of acetophenones are formed by further photolysis of 1-aryl-2-phenoxyethanones and not by β-fragmentation of the ketyl radicals. The high reactivity of ketyl radicals with oxygen coupled with the low rates of β-fragmentation of the same species have an important bearing in the context of the photoyellowing of lignin containing pulps and papers.

Full text of DOI:10.1021/jo047826u

Generation and Reactivity of Ketyl Radicals with Lignin Related
Structures. On the Importance of the Ketyl Pathway in the
Photoyellowing of Lignin Containing Pulps and Papers
Claudia Fabbri,† Massimo Bietti,‡ and Osvaldo Lanzalunga*,†
Dipartimento di Chimica, Universita` degli Studi di Roma “La Sapienza” and Istituto CNR di
Metodologie Chimiche (IMC-CNR), Sezione Meccanismi di Reazione, P. le A. Moro, 5 I-00185 Rome, Italy,
and Dipartimento di Scienze e Tecnologie Chimiche, Universita` “Tor Vergata”,
Via della Ricerca Scientifica, I-00133 Rome, Italy
osvaldo.lanzalunga@uniroma1.it
Received December 10, 2004
Ketyl radicals with lignin related structures have been generated by means of radiation chemical
and photochemical techniques. In the former studies ketyl radicals are produced by reaction of
R-carbonyl-â-aryl ether lignin models with the solvated electron produced by pulse radiolysis of an
aqueous solution at pH 6.0. The UV-vis spectra of ketyl radicals are characterized by three main
absorption bands. The shape and position of these bands slightly change when the spectra are
recorded in alkaline solution (pH 11.0) being now assigned to the ketyl radical anions and a pKa )
9.5 is determined for the 1-(3,4,5-trimethoxyphenyl)-2-phenoxyethanol-1-yl radical. Decay rates of
ketyl radicals are found to be dose dependent and, at low doses, lie in the range (1.7-2.7) × 103
s-1. In the presence of oxygen a fast decay of the ketyl radicals is observed (k2 ) 1.8-2.7 × 109 M-1
s-1) that is accompanied by the formation of stable products, i.e., the starting ketones. In the
photochemical studies ketyl radicals have been produced by charge-transfer (CT) photoactivation
of the electron donor-acceptor salts of methyl viologen (MV2+) with R-hydroxy-R-phenoxymethyl-
aryl acetates. This process leads to the instantaneous formation of the reduced acceptor (methyl
viologen radical cation, MV+•), as is clearly shown in a laser flash photolysis experiment by the
•
two absorption bands centered at 390 and 605 nm, and an acyloxyl radical [ArC(CO2 )(OH)CH2-
(OC6H5)], which undergoes a very fast decarboxylation with formation of the ketyl radicals. Steady-
state photoirradiation of the CT ion pairs indicates that 1-aryl-2-phenoxyethanones are formed as
primary photoproducts by oxidation of ketyl radicals by MV2+ (under argon) or by molecular oxygen.
Small amounts of acetophenones are formed by further photolysis of 1-aryl-2-phenoxyethanones
and not by â-fragmentation of the ketyl radicals. The high reactivity of ketyl radicals with oxygen
coupled with the low rates of â-fragmentation of the same species have an important bearing in
the context of the photoyellowing of lignin containing pulps and papers.
Introduction
mainly due to the interaction of lignin functional groups
with the UV portion of sunlight.1 Photoyellowing has an
industrial and environmental relevance and in the last
When exposed to light, lignin containing pulps and
papers undergo color reversion (photoyellowing) that is
† Universita` degli Studi di Roma “La Sapienza” and Istituto CNR
di Metodologie Chimiche (IMC-CNR).
(1) Heitner, C.; Scaiano, J. C., Eds. The Photochemistry of Ligno-
cellulosic Materials; ACS Symp. Ser. No. 531; American Chemical
Society: Washington, DC, 1993.
‡ Universita` “Tor Vergata”.
10.1021/jo047826u CCC: $30.25 © 2005 American Chemical Society
Published on Web 03/05/2005
2720
J. Org. Chem. 2005, 70, 2720-2728

Generation and Reactivity of Ketyl Radicals
SCHEME 1
twenty years a considerable effort has been made to
understand the mechanistic aspects of this process.1-4
Such a knowledge is of fundamental importance for the
prevention or reduction of photoyellowing by means of
chemical modification of functional groups, use of addi-
tives such as antioxidants, reducing agents, or UV
screens.2,5-14 Three main reaction pathways have been
identified for the photoyellowing process: (1) formation
of phenoxyl radicals by direct excitation or hydrogen atom
transfer reactions of phenolic groups and further oxida-
tion of phenoxyl radicals to colored products (phenol
pathway, Scheme 1, path a);2,15-18 (2) â-cleavage of
R-aryloxy substituted aromatic ketones to give a pair of
phenoxyl and phenacyl radicals which are precursors of
colored products (phenacyl pathway, Scheme 1, path
b);2,19-25 and (3) â-cleavage of ketyl radicals (Scheme 1,
path c) formed either by photoreduction of aromatic
ketones (path e) or by hydrogen abstraction from aryl-
glycerol â-aryl ethers (path d) with formation of phenoxyl
radicals and enols which tautomerize to give aromatic
ketones (ketyl pathway).2,24,26-28
The importance of the ketyl pathway is due to the great
abundance of arylglycerol â-aryl ether structures in lignin
(accounting for more than 30% of the phenylpropane
units).29 Several studies have been carried out to assess
(2) Leary, G. J. J. Pulp Paper Sci. 1994, 20, J154-J160.
(3) Davidson, R. S. J. Photochem. Photobiol. B: Biol. 1996, 33, 3-25.
(4) Lanzalunga, O.; Bietti, M. J. Photochem. Photobiol. B: Biol.
2000, 56, 85-108.
(5) Francis, R. C.; Dence, C. W.; Alexander, T. C.; Agnemo, R.;
Omori, S. Tappi J. 1991, 127-133.
(6) Castellan, A.; Noutary, C.; Davidson, R. S. J. Photochem.
Photobiol. A: Chem. 1994, 84, 311-316.
(15) Neumann, M. G.; Machado, A. E. H. J. Photochem. Photobiol.
B: Biol. 1989, 3, 473-481.
(16) Castellan, A.; Colombo, N.; Nourmamode, A.; Zhu, J. H.;
Lachenal, D.; Davidson, R. S.; Dunn, L. J. Wood Chem. Technol. 1990,
10, 461-493.
(17) Shkrob, I. A.; Depew, M. C.; Wan, J. K. S. Res. Chem. Intermed.
1992, 17, 271-285.
(7) Davidson, R. S.; Dunn, L.; Castellan, A.; Nourmamode, A. J.
Photochem. Photobiol. A: Chem. 1995, 86, 275-282.
(8) Trichet, V.; Grelier, S.; Castellan, A.; Choudhury, H.; Davidson,
R. S. J. Photochem. Photobiol. A: Chem. 1996, 95, 181-188.
(9) Cook, C. M.; Ragauskas, A. J. J. Photochem. Photobiol. A: Chem.
1997, 104, 217-224.
(18) Shukla, D.; Shepp, N. P.; Mathivanan, N.; Johnston, L. J. Can.
J. Chem. 1997, 75, 1820-1829.
(19) Vanucci, C.; Fornier de Violet, P.; Bouas-Laurent, H.; Castellan,
A. J. Photochem. Photobiol. A: Chem. 1988, 41, 251-265.
(20) Castellan, A.; Colombo, N.; Cucuphat, C.; Fornier de Violet, P.
Holzforschung 1989, 43, 179-185.
(10) Hu, T. Q.; James, B. R.; Wang, Y. J. Pulp Paper Sci. 1999, 25,
312-317.
(21) Netto-Ferreira, J. C.; Avellar I. G. J.; Scaiano, J. C. J. Org.
Chem. 1990, 55, 89-92.
(11) McGarry, P.; Heitner, C.; Schmidt, J.; Seltzer, R.; Cunkle, G.;
Wolf, J.-P. J. Pulp Paper Sci. 2000, 26, 59-66.
(22) Schmidt, J. A.; Berinstain, A. B.; de Rege, F.; Heitner, C.;
Johnston, L. J.; Scaiano, J. C. Can. J. Chem. 1991, 69, 104-107.
(23) Palm, W.-U., Dreeskamp, H.; Bouas-Laurent, H.; Castellan, A.
Ber. Bunsen-Ges. Phys. Chem. 1992, 96, 50-61.
(24) Scaiano, J. C.; Whittlesey, M. K.; Berinstain, A. B.; Malenfant,
P. R. L.; Schuler, R. H. Chem. Mater. 1994, 6, 836-843.
(25) Argyropoulos, D. S.; Sun, Y. Photochem. Photobiol. 1996, 64,
510-517.
(12) Evans, P. D.; Owen, N. L.; Schmid, S.; Webster, R. D. Polym.
Degrad. Stab. 2002, 76, 291-303.
(13) McGarry, P.; Heitner, C.; Schmidt, J.; Rodenhiser, A.; St. John
Manley, R.; Cunkle, G.; Thompson, T. J. Photochem. Photobiol. A:
Chem. 2002, 151, 145-155.
(14) Pu, Y.; Anderson, S.; Lucia, L.; Ragauskas, A. J. J. Photochem.
Photobiol. A: Chem. 2004, 163, 215-221.
J. Org. Chem, Vol. 70, No. 7, 2005 2721

Fabbri et al.
the role of the ketyl pathway in the light-induced
yellowing of lignin containing pulps and papers.24,26,27,30
In this context it is of fundamental importance to
determine the rate of fragmentation of ketyl radicals
since the ketyl pathway would play a significant role only
if the fragmentation process is relatively fast. On the
other side, if the lifetime of the ketyl radical is long
enough then alternative pathways become available to
compete with the â-fragmentation31 such as reaction with
molecular oxygen leading to R-carbonyl-â-aryl ether
structures (Scheme 1, path f). In such a case the result
would be an increase of the contribution of the phenacyl
pathway to photoyellowing.32 In an early study, the ketyl
radical generated by hydrogen atom transfer from 1-phen-
yl-2-phenoxyethanol to the tert-butoxyl radical was found
to fragment with a relatively high rate constant (k > 2
× 106 s-1).26 An even higher limit (>5 × 107 s-1) was given
for the rate of fragmentation of the ketyl radical of R-(p-
methoxyphenoxy)-p-methoxyacetophenone.24 Moreover,
product studies of the reaction of tert-butoxyl radical with
1-phenyl-2-phenoxyethanol revealed the presence of the
products of photofragmentation: phenol and aceto-
phenone.26 Later on, in a related study, the involvement
of the ketyl pathway in the photoyellowing process has
been questioned since it was observed that R-phenoxy-
acetophenone was the primary photoproduct formed by
reaction of the ketyl radical of 1-phenyl-2-phenoxyethanol
with oxygen (Scheme 1, path f), acetophenone and phenol
being formed as secondary photoproducts, and a much
lower rate constant, around 10 s-1, was estimated for the
fragmentation of 1-phenyl-2-phenoxyethanol-1-yl radi-
cal.30 The relatively long lifetime of this radical was
confirmed by kinetic studies, using hydrogen atom ab-
straction from thiophenol and Arrhenius expression for
the â-scission.33
produced by pulse radiolysis of an aqueous solution (eq
1) as reported by Scaiano for the generation of the
ketyl radical of R-(p-methoxyphenoxy)-p-methoxyaceto-
phenone.24 The radical anion thus produced by reduction
of the ketone in neutral or acidic conditions can be rapidly
protonated by the solvent to form the ketyl radical (eq
2).
In the photochemical studies, ketyl radicals have been
produced by charge-transfer photoactivation of the elec-
tron donor-acceptor salts of methyl viologen (MV2+) with
carboxylate donors, R-hydroxy-R-phenoxymethyl-aryl ac-
etates 8-10.34 Photoexcitation of these CT complexes
However, if the ketyl radicals are characterized by low
fragmentation rates it is quite surprising that these
species have not been detected and characterized, so far,
by using time-resolved techniques. We felt that further
studies were needed on this important topic and to this
purpose we carried out both time-resolved and product
studies by generating ketyl radicals with lignin related
structures by means of radiation chemical and photo-
chemical techniques.
In the radiation chemical studies ketyl radicals have
been generated by reaction of R-carbonyl-â-aryl ether
lignin models 1-7, which differ in the number and/or
position of methoxy substituents, and the presence or
absence of a γ methyl group, with the solvated electron
leads to the instantaneous formation of the reduced
acceptor (methyl viologen radical cation, MV+•) and an
•
acyloxyl radical [ArC(CO2 )(OH)CH2(OC6H5)] (eq 4), which
undergoes a fast decarboxylation35,36 with formation of a
ketyl radical (eq 5). Both time-resolved spectroscopic
(laser flash photolysis) and product studies (steady-state
irradiation) have been carried out.
(26) Scaiano, J. C.; Netto-Ferreira, J. C.; Wintgens, V. J. Photochem.
Photobiol. A: Chem. 1991, 59, 265-268.
(27) Schmidt, J. A.; Heitner, C. J. Wood Chem. Technol. 1993, 13,
309-325.
(28) Ruggiero, R.; Machado, A. E. H.; Castellan, A.; Grelier, S. J.
Photochem. Photobiol. A: Chem. 1997, 110, 91-97.
(29) Argyropoulos, D. S.; Menachem, S. B. In Biotechnology in the
Pulp and Paper Industry; Eriksson, K. E. L., Ed.; Springer-Verlag:
Berlin, Heidelberg, 1997; pp 127-158.
(30) Huang, Y.; Page´, D.; Wayner D. D. M.; Mulder, P. Can. J. Chem.
1995, 73, 2079-2085.
Results
(31) In a diffusionally restricted lignin matrix competitive bimo-
lecular degradation pathways may occur with difficulty so that even
modest scission rates would result in ketyl radical fragmentation.
(32) Omori, S.; Francis, R. C.; Dence, C. W. Holzforschung 1991,
45, 93-100.
Pulse Radiolysis Study. Ketyl radicals were gener-
ated in aqueous solution containing 2-methyl-2-propanol
(33) Kandanarachchi, P. H.; Autrey, T.; Franz, J. A. J. Org. Chem.
2002, 67, 7937-7945.
(34) Bockman, T. M.; Hubig, S. M.; Kochi, J. K. J. Org. Chem. 1997,
62, 2210-2221.
2722 J. Org. Chem., Vol. 70, No. 7, 2005

Generation and Reactivity of Ketyl Radicals
TABLE 1. Spectral Data for the Ketyl Radicals and Radical Anions Generated by Pulse Radiolysis of Aqueous
Solutions at pH ∼6.0 and ∼11.0 Containing the Ketones 1-7
pH 6.0
pH 11.0
ketone
λmax/nm
380, 440
330, 430, 560
340, 490, 590
350, 450, 560
330, 370, 450
330, 430, 540
340, 490, 580
ꢀmax/M-1 cm-1
λmax/nm
ꢀmax/M-1 cm-1
1
2
3
4
5
6
7
1000, 600
330, 390, 460
330, 540
340, 600
350, 460, 540
330, 460
340, 520
340, 580
1700, 700, 800
2800, 1300
2900, 1200
2000, 1200, 1000
3000, 900
4700, 1500
4400, 1400
2700, 700, 1200
2400, 900, 1000
1600, 1400, 700
1700, 1000, 700
3900, 800, 1300
3800, 1200, 1200
(1 M) by reaction of R-carbonyl-â-aryl ether lignin models
1-7 with the solvated electrons (eqs 1 and 2). The
hydrated electron (eaq-) reacts with the ketone leading
to the ketyl radical anion (eq 1), with diffusion controlled
rates.37 2-Methyl-2-propanol was added to scavenge the
This spectrum can be reasonably assigned to the ketyl
radical 2H• and is comparable with the spectrum of the
ketyl radical, produced in the same way by pulse radi-
olysis of 4-methoxyacetophenone, which displayed three
absorption bands centered at 310, 410, and 520 nm.39 The
three absorption bands are all quenched by oxygen, with
the same rate, as expected for a carbon centered radical.
An analogous behavior was observed with substrates 1,
3-7. The spectral data of the ketyl radicals are reported
in Table 1. The UV-vis spectra of ketyl radicals 1H•, 3H•,
and 5H•-7H• are reported in the Supporting Information.
The comparison of the absorption spectra of 1H• and 5H•,
2H• and 6H•, and 3H• and 7H• shows that the presence
of the γ-methyl group has no significant effect on the
position and relative intensities of the absorption bands.
The comparison of the absorption spectra of 1H•, 2H•,
3H•, and 4H•, and of 5H•, 6H•, and 7H• shows instead
that the nature of the ring substituent has a significant
effect on the position and relative intensities of the
absorption bands.
hydroxyl radical •OH (eq 7, k ) 6 × 108 M-1s-1 38
)
and to
increase the solubility of the ketone. Experiments were
carried out in neutral (pH 6.0) and basic (pH 11.0)
conditions under argon, air or N2/O2 9:1 (v/v). As an
example, in Figure 1 are displayed the time-resolved
When the spectrum was recorded in alkaline solution
(see Figure 1, open circles), the shape and position of the
absorption bands slightly changed. The bands should now
be assigned to 2-• on the basis of a pKa ≈ 9-10 that can
be estimated for the acid-base equilibrium between 2H•
and 2-• (eq 2). An accurate determination of the pKa value
for 2H•, by measuring the absorption at a suitable
wavelength as a function of pH, was not possible due to
the small difference in the absorption between the ketyl
radical and the ketyl radical anion. A pKa value of 9.5
was instead determined for 4H•, and in this case the
difference in the absorption between 4H• and 4-• at 450
nm is sufficiently large. The UV-vis spectra of 4H• and
FIGURE 1. Time-resolved absorption spectra observed on
4
-• are reported in Figure 2. In the inset, the plot of ∆OD,
-
reaction of eaq with 2 (0.2 mM) at T ) 25 °C, recorded after
monitored at 450 nm vs pH is also shown.
pulse radiolysis of an Ar-saturated aqueous solution at pH 6.0
(filled circles) and at pH 11 (empty circles), containing 1 M
2-methyl-2-propanol at 20 µs after the 1 µs, 10-MeV electron
pulse.
The spectral data for 1-•-7-• are collected in Table 1.
The UV-vis spectra of 1-•, 3-•, and 5-•-7-• are reported
in the Supporting Information.
In neutral solution, the rate of decay of 1H•-7H• was
measured spectrophotometrically by following the change
in optical density at the wavelengths corresponding to
the UV and visible absorption maxima (see Table 1).
Under these conditions the ketyl radicals were found to
decay by a second-order reaction and the same value of
the decay rate was determined for all the absorption
bands, fully supporting their assignment to the same
absorption spectra observed on reaction of eaq- with 2 in
argon saturated aqueous solutions at pH 6.0 and 11.0,
recorded 20 µs after the electron pulse.
At pH 6.0 (filled circles) the spectrum shows three
absorption bands centered around 330, 430, and 560 nm.
(35) This process is too fast (k > 107 s-1) to be followed in the
nanosecond time scale.34,36
(36) Hilborn, J. W.; Pincock, J. A. J. Am. Chem. Soc. 1991, 113, 2683.
(37) Ross, A. B.; Mallard, W. G.; Helman, W. P.; Bielski, B. H. J.;
Buxton, G. V.; Cabeli, D. A.; Greenstock, C. L.; Huie, R. E.; Neta, P.
NRDL/NIST Solution Kinetics Database, Ver 1, 1.0 ed. NIST, U.S.
Department of Commerce: Gaithersburg, MD, 1992.
(39) The absorption spectrum of the ketyl radical 4-MeOC6H4C•(OH)-
CH3 was recorded 20 µs after pulse radiolysis (3 MeV van de Graaf
accelerator, 300 ns electron pulse, 5 Gy/pulse) of an argon saturated
aqueous solution (pH 5.5) containing 4-methoxyacetophenone (0.05
mM) and 2-methyl-2-propanol (0.1 M).
(38) Buxton, G. V.; Greenstock, C. L.; Helman, W. P.; Ross, A. B. J.
Phys. Chem. Ref. Data 1988, 17, 513-886.
J. Org. Chem, Vol. 70, No. 7, 2005 2723

Fabbri et al.
FIGURE 2. Time-resolved absorption spectra observed on
FIGURE 3. Decay and following buildup of the absorption at
350 nm observed after pulse radiolysis of an aqueous solution
of 3 (0.2 mM) at T ) 25 °C, saturated with a 9:1 N2/O2 mixture
at pH 6.0, containing 1 M 2-methyl-2-propanol after the 1 µs,
10-MeV electron pulse.
-
reaction of eaq with 4 (0.2 mM) at T ) 25 °C, recorded after
pulse radiolysis of an Ar-saturated aqueous solution at pH 6
(empty circles) and at pH 11 (filled circles), containing 1 M
2-methyl-2-propanol at 20 µs after the 1 µs, 10-MeV electron
pulse. Inset: Plot of ∆OD monitored at 450 nm vs pH. From
the curve fit pKa ) 9.5.
salt of the carboxylate anions 8-10 (generated in situ
by dissolution of the free carboxylic acid in aqueous
sodium bicarbonate) were saturated with argon or oxygen
and then irradiated for 1 h in a photoreactor equipped
with 10 × 14 W black phosphorus lamps with emission
at λMAX 360 nm. A deep blue coloration of the solution
characteristic of MV+• developed in the steady-state
photoirradiation under argon.42 After extraction with
chloroform, reaction products were analyzed by 1H NMR.
The results are reported in Table 2. The mass balance
was always satisfactory, accounting for more than 85%
of the starting material.
transient species. The rate of decay was found to be
highly dose-dependent and at the lowest doses that
allowed a reliable measurement of the decay rates (ca.
3.5 Gy/pulse), the decays better fitted first-order kinetics.
The rate constants for the decay of 1H•-7H• determined
at low doses under argon were very similar and com-
prised in the range (1.7-2.7) × 103 s-1
.
When the solutions were air saturated a very fast
decay of the ketyl radical bands was observed. Under
these conditions it was not possible to measure the rate
of reaction of ketyl radicals with molecular oxygen. To
this purpose the solutions were saturated with O2 (10%
(v/v) in N2). The rate of the reaction of 1H•-7H• with
oxygen was measured spectrophotometrically by follow-
ing the change in optical density at the wavelengths
corresponding to the UV and visible absorption maxima
reported in Table 1. The second-order rate constants have
been calculated by dividing the pseudo-first-order rate
constants kobs for the concentration of oxygen (0.15 mM).40
The rates of reaction of 1H•-7H• with molecular oxygen
are very similar and lie in the range (1.8-2.7) × 109
The photoirradiation of the [MV2+/8] ion pair, under
argon, leads to the formation of 1-phenyl-2-phenoxy
ethanone (1) as the major photoproduct. Acetophenone
was formed in very small amounts (<1% referred to the
starting material). Higher yields of 1 were observed when
the solution was saturated with oxygen. Similar results
were also obtained in the photoirradiation of the [MV2+
/
9] ion pair. The major product, 1-(4-methoxyphenyl)-2-
phenoxy ethanone (2), is accompanied by small amounts
of 4-methoxyacetophenone, 2-(2-hydroxyphenyl)-4-meth-
oxyacetophenone, and 2-(4-hydroxyphenyl)-4-methoxy-
acetophenone. Also in this case a significant increase of
the reaction yields was observed under oxygen. When the
reaction time was reduced to 15 min, only traces of
4-methoxyacetophenone were observed. Photoirradiation
of the [MV2+/10] ion pair again led to 1-(3,4-dimeth-
oxyphenyl)-2-phenoxy ethanone (3) as the major photo-
product accompanied by minor amounts of 2-(2-hy-
droxyphenyl)-3,4-dimethoxyacetophenone and 2-(4-hy-
droxyphenyl)-3,4-dimethoxyacetophenone and of 3,4-
dimethoxyacetophenone. As in the photoirradiation of the
[MV2+/9] ion pair, higher yields of products were observed
M-1s-1
.
The decay of the ketyl radicals 3H•, 4H•, and 7H• is
accompanied by the formation of stable products, likely
the starting ketones, which absorb around 350 nm. In
Figure 3 it can be noticed that the buildup of the ketone
3 is slower than the decay of 3H• and is somewhat
delayed. This is due to the fact that the reaction of the
ketyl radicals with O2 leads to the formation of R-hy-
droxyperoxyl radicals (eq 8) from which the ketones are
formed by release of hydroperoxyl radical HOO• (eq 9).41
(40) Murov, S. L.; Carmicael, I.; Hug, G. L. Handbook of Photo-
chemistry, 2nd ed.; Marcel Dekker: New York, 1993.
(41) The Chemistry of Free Radicals: Peroxyl Radicals. Alfassi, Z.
V., Ed.; John Wiley & Sons: New York, 1997. von Sonntag, C. The
Chemical Basis of Radiation Biology; Taylor & Francis: London, UK,
1987.
Steady-State Photoirradiation. Aqueous solutions
of methyl viologen triflate [MV(OTf)2] and of the sodium
(42) Watanabe, T.; Honda, K. J. Phys. Chem. 1982, 86, 2617.
Bockman, T. M.; Kochi, J. K. J. Org. Chem. 1990, 55, 4127.
2724 J. Org. Chem., Vol. 70, No. 7, 2005

Generation and Reactivity of Ketyl Radicals
TABLE 2. Products and Yields in the Photoirradiation of MV2+/8-10 Ion Pairs
a Yields are referred to the initial amount of the substrate. Average of three determinations, error (5%.
when the solution was saturated with oxygen moreover
the relative amounts of minor photoproducts was reduced
by decreasing the irradiation time to 15 min. In a control
experiment it was observed that photoirradiation of
1-(3,4-dimethoxyphenyl)-2-phenoxy ethanone (3) in H2O/
CH3CN 3:2, under the same experimental conditions used
for the irradiation of the [MV2+/10] ion pair, led to the
formation of 3,4-dimethoxyacetophenone, 2-(2-hy-
droxyphenyl)-3,4-dimethoxyacetophenone, and 2-(4-hy-
droxyphenyl)-3,4-dimethoxyacetophenone.
Laser Flash Photolysis Study. Time-resolved laser
flash photolysis studies have been carried out only with
the carboxylate 9, which gave the higher yields of
photoproducts. In Figure 4 is shown the transient spec-
trum (filled circles) observed 200 ns after 355 nm laser
flash photolysis of an argon saturated aqueous solution
of MV(OTf)2 (20 mM) and 9 (20 mM). The spectrum
showed two bands: a strong, sharp band with λMAX ) 390
nm and a broad absorption band in the visible region that
can be assigned to the reduced form of methyl viologen
(MV+•), the absorption spectrum of which has two maxima
(390 and 605 nm) in aqueous solution.42 Analysis of the
spectral evolution in the former 2 µs indicates that the
bands at 390 and 600 nm increase slightly in intensity
(Figure 4, insets a and c), while, on the same time scale,
the absorption in the 400-550 nm region decreases in
intensity (Figure 4, inset b). It is likely that the latter
absorption is due to the presence of the ketyl radical 2H•,
which is formed according to eq 5. The absorption band
of this species is not distinguishable in the spectrum
reported in Figure 4 due to its weak absorption, which
is significantly smaller than that of MV+•.42
The decrease in absorbance in the 400-550 nm region
and the increase in absorbance at 390 and 600 nm are
likely due to the relatively fast oxidation of 2H• by MV2+
(eq 10), which produces an additional amount of MV+•.34,43
2H• should be characterized by a very low oxidation
potential44 and can be easily oxidized by MV2+ 42
.
Discussion
Reaction of R-carbonyl-â-aryl ether lignin models 1-7
with the solvated electrons produced by pulse radiolysis
FIGURE 4. Time-resolved absorption spectra recorded 200
ns (filled circles) following the 8-ns laser excitation (355 nm)
of an Ar-saturated aqueous solution of the [MV2+/9] ion pair.
Insets: (a) Buildup of the absorption at 390 nm. (b) Decay of
the absorption at 480 nm. (c) Buildup of the absorption at 600
nm.
(43) Barnett, J. R.; Hopkins, A. S.; Ledwith, A. J. Chem. Soc., Perkin
Trans. 2 1973, 80.
(44) An E1/2 value of -0.46 V vs SCE in CH3CN was obtained by
photomodulated voltammetry for the ketyl radical 4-CH3C6H4C•(OH)-
CH3.45
J. Org. Chem, Vol. 70, No. 7, 2005 2725

Fabbri et al.
of aqueous solutions allowed the spectral characterization
of ketyl radicals 1H•-7H•. The assignment of the absorp-
tion spectra observed after the pulse to the ketyl radicals
is supported by the fact that a similar spectrum is
produced in the same way by pulse radiolysis of 4-meth-
oxyacetophenone. Moreover, in accordance with the as-
signment of the absorption bands to a carbon centered
radical, they are all quenched by oxygen. Finally, when
the spectra were recorded in alkaline solution, the shape
and position of the absorption bands slightly changed
being now assigned to the ketyl radical anions. By
measuring the absorption at 450 nm as a function of pH,
the pKa for the acid-base equilibrium between 4H• and
4-• has been determined. The value obtained (9.5) is
typical for R-hydroxyalkyl radicals of related struc-
tures.45,46
7 by release of hydroperoxyl radicals HOO• (eq 9).41 The
second-order rate constants for the reaction of ketyl
radicals with molecular oxygen are typical for reactions
of carbon-centered radicals with oxygen48 and are very
close together. Determination of the rate of reaction of
ketyl radicals with molecular oxygen (eq 8) is an impor-
tant result of this study, which has a bearing in the
context of the photoyellowing process. The high reactivity
of ketyl radicals with lignin-related structures 1H•-7H•
with oxygen coupled with the low rates of â-fragmenta-
tion of the same species suggests that R-aryloxy substi-
tuted aromatic ketones should be formed preferentially
with respect to the products of fragmentation (aromatic
ketones and phenoxyl radicals). This hypothesis is fully
supported by the results of the steady-state photoirra-
diation (vide infra).
It is interesting to note that our observation of the
transient absorption spectra of ketyl radicals 1H•-7H•
is in contrast with the results of Scaiano and co-workers,
who were unable to observe the ketyl radical of R-(p-
methoxyphenoxy)-p-methoxyacetophenone.24 The only
transient observed after the pulse was the 4-methoxy-
phenoxyl radical and this result was interpreted with a
â-fragmentation process of the ketyl radical too fast to
be followed spectrophotometrically (k > 5 × 107 s-1). The
rate of decay of the ketyl radicals 1H•-7H•, measured
spectrophotometrically by us under acidic conditions,
following the change in optical density at the wavelengths
corresponding to the UV and visible absorption maxima
(330-380 nm, 430-490 nm, and 540-590 nm), are
indeed more than 4 orders of magnitude lower than 5 ×
107 s-1. Ketyl radicals were found to decay by a second-
order reaction and the rate of decay was found to be
highly dose dependent. At lowest doses that allowed a
reliable determination of the rate constants (ca. 3.5 Gy/
pulse) the decay rates for the ketyl radicals 1H•-7H•
were found to obey a first-order process. The values are
very similar and comprised in the range 1.7 × 103 to 2.7
× 103 s-1, thus the decay of the ketyl radicals does not
change significantly by increasing the number of aryl
methoxy substituents or in the presence of a γ-methyl
substituent.47 Since the rate of decay of the ketyl radicals
measured by us should be considered an upper limit for
the rate of â-fragmentation, our results are more in
accordance with the low fragmentation rate determined
by Mulder and co-workers for the fragmentation of
1-phenyl-2-phenoxyethanol-1-yl radical.30
The results of the steady-state photoirradiation and
laser flash photolysis experiments can be rationalized on
the basis of reactions 3-5 and 10. Photoexcitation of the
CT ion pairs [MV2+/8-10] leads to the instantaneous
formation of the reduced acceptor (methyl viologen radi-
cal cation, MV+•) and an acyloxyl radical [ArC(CO2 )(OH)-
•
CH2(OC6H5)] (eq 4), which undergoes a very fast decar-
boxylation with formation of ketyl radicals (eq 5).34 The
formation of MV+• is clearly visible by the deep blue
coloration that developed in the steady-state photoirra-
diation under argon and, in the LFP experiment with the
[MV2+/9] ion pair, by the observation of the two absorp-
tion bands of MV+• centered at 390 and 605 nm.42 The
absorption of the ketyl radical 2H• is not distinguishable
in the spectrum due to its relative low absorption as
compared to that of MV+•. Analysis of the spectral
evolution in the former 2 µs indicates that an additional
small amount of MV+• is formed after the laser pulse by
oxidation of the ketyl radical 2H• induced by MV2+ (eq
10). The occurrence of reaction 10 is also supported by
the formation of 1-(4-methoxyphenyl)-2-phenoxy-
ethanone, the product of oxidation of 2H•, in the steady-
state photoirradiation of the CT ion pair [MV2+/9].
The analysis of the results of steady-state photoirra-
diation of the CT ion pairs [MV2+/8-10] clearly indicates
that 1-aryl-2-phenoxyethanones 1-3 are formed as pri-
mary photoproducts. The small amounts of aceto-
phenones are likely formed by further photolysis of
1-aryl-2-phenoxyethanones and not by â-fragmentation
of the ketyl radicals. Accordingly, under oxygen atmo-
sphere, it was observed that the yields of acetophenones
relative to those of 1-aryl-2-phenoxyethanones decreased
significantly by reducing the irradiation time from 60 to
15 min. This hypothesis is confirmed by the observation
of the R-carbonyl â-1 products, 2-(2-hydroxyphenyl)-4-
methoxyacetophenone and 2-(4-hydroxyphenyl)-4-meth-
oxyacetophenone, in the steady-state irradiation of the
[MV2+/9] ion pair and 2-(2-hydroxyphenyl)-3,4-dimethoxy-
acetophenone and 2-(4-hydroxyphenyl)-3,4-dimethoxy-
Analysis of the time-resolved absorption spectra of 3H•,
4H•, and 7H• under oxygen showed that the decay of the
ketyl radicals is accompanied by the formation of stable
products, i.e., the starting ketones 3, 4, and 7. It was also
observed that the buildup of the ketones was slower than
the decay of the ketyl radicals and somewhat delayed.
Thus, in this time scale, it was possible to follow the
reaction of the ketyl radicals 3H•, 4H•, and 7H• with O2
leading to the formation of the R-hydroxyperoxyl radicals
(eq 8) and the subsequent formation of ketones 3, 4, and
acetophenone in the steady-state irradiation of the [MV2+
/
10] ion pair. These products are formed according to the
mechanistic pathways already described in the photo-
irradiation of 3,4-dimethoxy-R-(2′-methoxyphenoxy)ac-
etophenone (guaiacylacetoveratrone) in ethanol/water
(45) Lund, T.; Wayner, D. D. M.; Jonsson, M.; Larsen, A. G.;
Daasbjerg, K. J. Am. Chem. Soc. 2001, 123, 12590-12595.
(46) Bard, A. J.; Faulkner, L. R. Electrochemical Methods; John
Wiley: New York, 1980.
(47) While substituent effects on the phenyl ring R to the radical
center should not significantly affect the rate of scission of the ketyl
radical, a substitution to the 2-position may enhance the rate of
â-scission.33
(48) Maillard, B.; Ingold, K. U.; Scaiano, J. C. J. Am. Chem. Soc.
1983, 105, 5095-5099.
2726 J. Org. Chem., Vol. 70, No. 7, 2005

Generation and Reactivity of Ketyl Radicals
SCHEME 2
mixture.49 When excited, 1-aryl-2-phenoxyethanones un-
dergo an intramolecular â-cleavage leading to a phenoxyl
and phenacyl radical pair. In cage recombination of the
radical pair leads to a cyclohexadienone coupling product
that, via a prototropic rearrangement, is converted into
the R-carbonyl â-1 compounds (Scheme 2).4,19,20,51
The amount of secondary photoproducts increases by
increasing the number of aryl methoxy substituents, in
the order 8 < 9 < 10, in accordance with the fact that
the absorption of light emitted from the lamps by
products 1-aryl-2-phenoxyethanones increases in the
order 1 < 2 < 3.
Significantly higher yields of photoproducts were ob-
served when the solutions of the [MV2+/8-10] ion pairs
were saturated with oxygen. In argon degassed solutions,
photogenerated MV+• absorbs the light emitted by the
lamps and an “inner filter” effect results which prevents
efficient photolysis.34 In the presence of oxygen, MV+•
does not accumulate since it is rapidly oxidized to MV2+
(eq 11) and the solutions does not develop the charac-
teristic blue coloration of the reduced viologen.52
the first time by using time-resolved techniques. By
combining all the information thus obtained with the
different precursors and methods of generation of the
ketyl radicals, we gave a better insight into some of the
unsolved mechanistic aspects associated with the pho-
toyellowing of lignin containing pulps and papers. Low
fragmentation rates of ketyl radicals appear to be a
rather general process. The implication for the lignin
photochemistry is that a ketyl radical of this type will
not cleave to produce aromatic ketones and phenoxyl
radicals fast and efficiently so that there is the possibility
of scavenging them, i.e., by reaction with molecular
oxygen, before the cleavage occurs. Thus, only a minor
role is played by the ketyl pathway in the photoyellowing
process. A major role would instead be played by the
â-cleavage of R-aryloxy-substituted aromatic ketones
(phenacyl pathway) that are formed by the reaction of
ketyl radicals with O2.
Experimental Section
Materials. All the reagents and solvents were of the highest
commercial quality available and were used without further
purification (unless otherwise specified). Milli-Q-filtered water
was used for all solutions. Methyl viologen ditriflate was
prepared according to a procedure reported in the literature.34
MV+• + O2 f MV2+ + O2
(11)
-•
Since no fragmentation products have been observed
as primary photoproducts in the steady-state irradiation
either under argon or in the presence of oxygen, it derives
that â-fragmentation of the ketyl radicals 1H•-3H• is
significantly slower than their oxidation induced by MV2+
(eq 10) or by molecular oxygen (eq 8).
Pulse Radiolysis Experiments. The pulse radiolysis
experiments were performed using a 10 MeV electron linear
accelerator, which supplied 0.2-2 µs pulses with doses be-
tween 2 and 20 Gy. Dosimetry was performed with N2O-
saturated 10-2
M aqueous KSCN solutions, for which
G[(SCN)2-•] ) 6.0 × 10-7 mol J-1 and ꢀ[(SCN)2-•] ) 7600 M-1
cm-1 at 480 nm.53 Experiments were performed at room
temperature, using aqueous solutions containing the substrate
(0.2 mM) and 2-methyl-2-propanol (1 M). Solutions were
saturated with argon, oxygen, or a 9:1 N2/O2 mixture. Experi-
ments were performed at pH 6 (5 mM Na2HPO4 adjusted with
0.1 M HClO4) and at pH 11 (5 mM Na2B4O7‚10H2O adjusted
with 0.1 M NaOH). A flow system was employed in all the
experiments. The rate constants were obtained by averaging
Conclusions
Detection and spectral characterization of ketyl radi-
cals with lignin related structures was accomplished for
(49) In polar solvents â-cleavage should occur from both singlet and
triplet states.22,50
(50) Wan, J. K. S.; Tse, M. Y.; Depew, M. C. Res. Chem. Intermed.
1992, 17, 59-75.
(51) Castellan, A.; Zhu, J. H.; Colombo, N.; Nourmamode, A.;
Davidson, R. S.; Dunn, L. J. Photochem. Photobiol. A 1991, 58, 263-
273.
(52) Rodgers, M. A. J. Radiat. Phys. Chem. 1984, 23, 245.
Tsukuhara, K.; Wilkins, R. G. J. Am. Chem. Soc. 1985, 107, 2632.
(53) Schuler, R. H.; Hartzell, A. L.; Behar, B. J. Phys. Chem. 1981,
85, 192.
J. Org. Chem, Vol. 70, No. 7, 2005 2727

Fabbri et al.
3 to 5 values, each consisting of the average of 3 to 10 shots,
and were reproducible to within 5%.
Laser Flash Photolysis Studies. Laser flash photolysis
experiments were carried out by irradiating a 3 mL Suprasil
quartz cell (10 mm × 10 mm) containing an argon-saturated
aqueous solution of the sodium salt of 9 (20 mM), prepared as
described in the steady-state photoirradiation, and of methyl
viologen ditriflate (20 mM). The experiments were carried out
at T ) 25 ( 0.5 °C under magnetic stirring.
The pKa value for the acid-base equilibria between the ketyl
radical 4H• and the ketyl radical anion 4-• was determined
by plotting ∆A as a function of pH at 450 nm (where the
difference in absorption between radical cation and radical
zwitterion is sufficiently large) in the pH range 7-12. The ∆A
vs pH curve was fitted to the following equation: ∆A ) [∆A0
+ ∆A1 × 10(pH-pKa)]/1 + 10(pH-pKa)
.
Acknowledgment. Thanks are due to the Ministero
dell’Istruzione, dell’Universita` e della Ricerca (MIUR)
and the Consiglio Nazionale delle Ricerche (CNR) for
financial support. Pulse radiolysis experiments were
performed at the Free Radical Research Facility of
Daresbury, Cheshire, UK, with the support of the
European Commission through the Access to Large-
Scale Facilities activity of the TMR Program. The
authors also thank Prof. Suppiah Navaratnam and Dr.
Ruth Edge for technical assistance.
Preparation of Samples for Steady-State Photooxida-
tion and Laser Flash Photolysis. Aqueous solutions of the
sodium salt of the R-hydroxycarboxylates of substrates 8-10
were prepared by addying 50, 20, or 10 µmol of the free acid
and 55, 22, or 11 µmol of NaHCO3, respectively, to 1 mL of
Milli-Q-filtered water. Different concentrations of the three
substrates were chosen in order to have an absorbance of the
charge-transfer complexes at 350 nm of ca. 1. The solutions
were sonicated until complete dissolution then filtered and
transferred to a Pyrex test tube. An equimolar amount of
methyl viologen ditriflate with respect to the substrate was
then added. The solutions were saturated with argon or with
oxygen and then photoirradiated for 15 or 60 min. At the end
the internal standard (phenylacetic acid) was added. The
solutions were acidified, extracted with chloroform, and dried
over anhydrous Na2SO4 and the solvent was evaporated. The
Supporting Information Available: Instrumentation;
syntheses of substrates 1-10. The UV-vis absorption spectra
of ketyl radicals 1H•, 3H•-7H• and UV-vis absorption spectra
of ketyl radical anions 1-•, 3-•-7-•. This material is available
free of charge via the Internet at http://pubs.acs.org.
1
products were analyzed by H NMR.
JO047826U
2728 J. Org. Chem., Vol. 70, No. 7, 2005