Model compound studies of the beta-O-4 linkage in lignin: absolute rate expressions for beta-scission of phenoxyl radical from 1-phenyl-2-phenoxyethanol-1-yl radical.

Article abstract of DOI:10.1021/jo025581k

Arrhenius rate expressions were determined for beta-scission of phenoxyl radical from 1-phenyl-2-phenoxyethanol-1-yl, PhC*(OH)CH2OPh (V). Ketyl radical V was competitively trapped by thiophenol to yield PhCH(OH)CH2OPh in competition with beta-scission to yield phenoxyl radical and acetophenone. A basis rate expression for hydrogen atom abstraction by sec-phenethyl alcohol, PhC*(OH)CH3, from thiophenol, log(k(abs)/M(-1) s(-1)) = (8.88 +/- 0.24) - (6.07 +/- 0.34)/theta, theta = 2.303RT, was determined by competing hydrogen atom abstraction with radical self-termination. Self-termination rates for PhC*(OH)CH3 were calculated using the Smoluchowski equation employing experimental diffusion coefficients of the parent alcohol, PhCH(OH)CH3, as a model for the radical. The hydrogen abstraction basis reaction was employed to determine the activation barrier for the beta-scission of phenoxyl from 1-phenyl-2-phenoxyethanol-1-yl (V): log(k beta)/s(-1)) = (12.85 +/- 0.22) - (15.06 +/- 0.38)/theta, k beta (298 K) ca. (64.0 s(-1) in benzene), and log(k beta /s(-1)) = (12.50 +/- 0.18) - (14.46 +/- 0.30)/theta, k beta (298 K) = 78.7 s(-1) in benzene containing 0.8 M 2-propanol. B3LYP/cc-PVTZ electronic structure calculations predict that intramolecular hydrogen bonding between the alpha-OH and the -OPh leaving group of ketyl radical (V) stabilizes both ground- and transition-state structures. The computed activation barrier, 14.9 kcal/mol, is in good agreement with the experimental activation barrier.

Full text of DOI:10.1021/jo025581k

Mod el Com p ou n d Stu d ies of th e â-O-4 Lin k a ge in Lign in : Absolu te
Ra te Exp r ession s for â-Scission of P h en oxyl Ra d ica l fr om
1-P h en yl-2-p h en oxyeth a n ol-1-yl Ra d ica l
Pramod H. Kandanarachchi, Tom Autrey, and J ames A. Franz*
Pacific Northwest National Laboratory, Fundamental Science Division, 902 Battelle Boulevard,
Richland, Washington 99352
james.franz@pnl.gov
Received February 2, 2002
Arrhenius rate expressions were determined for â-scission of phenoxyl radical from 1-phenyl-2-
phenoxyethanol-1-yl, PhC•(OH)CH2OPh (V). Ketyl radical V was competitively trapped by thiophenol
to yield PhCH(OH)CH2OPh in competition with â-scission to yield phenoxyl radical and acetophe-
none. A basis rate expression for hydrogen atom abstraction by sec-phenethyl alcohol, PhC•(OH)-
CH3, from thiophenol, log(kabs/M-1 s-1) ) (8.88 ( 0.24) - (6.07 ( 0.34)/θ, θ ) 2.303RT, was
determined by competing hydrogen atom abstraction with radical self-termination. Self-termination
rates for PhC•(OH)CH3 were calculated using the Smoluchowski equation employing experimental
diffusion coefficients of the parent alcohol, PhCH(OH)CH3, as a model for the radical. The hydrogen
abstraction basis reaction was employed to determine the activation barrier for the â-scission of
phenoxyl from 1-phenyl-2-phenoxyethanol-1-yl (V): log(kâ/s-1) ) (12.85 ( 0.22) - (15.06 ( 0.38)/θ,
kâ(298 K) ca. (64.0 s-1 in benzene), and log(kâ/s-1) ) (12.50 ( 0.18) - (14.46 ( 0.30)/θ, kâ(298 K) )
78.7 s-1 in benzene containing 0.8 M 2-propanol. B3LYP/cc-PVTZ electronic structure calculations
predict that intramolecular hydrogen bonding between the R-OH and the -OPh leaving group of
ketyl radical (V) stabilizes both ground- and transition-state structures. The computed activation
barrier, 14.9 kcal/mol, is in good agreement with the experimental activation barrier.
SCHEME 1
In tr od u ction
Lignin is one of the most abundant naturally occurring
biopolymers. The development of detailed knowledge of
the chemistry of lignin is a desirable undertaking since
large volumes of lignin produced as waste in the pulp
and paper industry provide a potentially important raw
material base for value-added aromatic compounds. The
structure of lignin is a complicated network of aromatic
ethers connected by several types of linkages in which
arylglycerol â-aryl ether or â-O-4 linkages predominate,1
Scheme 1. A fundamental understanding of the decom-
position pathways of lignin model compounds is impor-
tant in quantifying the thermal reaction pathways of
lignin-like structures under hydropyrolysis conditions,2
understanding the fate of lignin in the geosphere, and
understanding the practical consequences of lignin re-
activity, such as mechanisms of paper yellowing.3
chanical pulps due to the formation of light-absorbing
chromophores formed by the reactions of oxidatively
produced phenoxyl radicals. A variety of pathways lead-
ing to the formation of “yellowing precursors”, substituted
phenoxyl radicals, have been proposed.4-10 An important
(4) Huang, D.; Page, D.; Wayner, D. D.; Mulder, P. Can. J . Chem.
1995, 73, 2079.
(5) J ohnston, L. J .; Scaiano, J . C. Can. J . Chem. 1991, 69, 104.
(6) Schmidt, J . A.; Heitner, C. J . Wood. Chem. Technol. 1991, 11,
The radiative degradation of lignin is considered to be
responsible for photoyellowing of paper made from me-
397.
(1) (a) Pearl, I. A. The Chemistry of Lignin; Marcel Dekker: New
York, 1967. (b) Adler, E. Wood Sci. Technol. 1977, 11, 169.
(2) Britt, P. F.; Buchanan, A. C.; Cooney, M. J .; Martineau, D. R. J .
Org. Chem. 2000, 65, 1376.
(7) Schmidt, J . A.; Heitner, J . J . Wood. Chem. Technol. 1993, 13,
397.
(8) Shkrob, I. A.; Pepew, M. C.; Wan, J . K. S. Res. Chem. Intermed.
1992, 17, 271.
(3) (a) Dorrestigin, E.; Laarhoven, J . J .; Arends, I. W. C. E.; Mulder,
P. J . Anal. Appl. Pyrolysis 2000, 54, 153. (b) Dorrestijin, E.; Hemmink,
S.; Hultsman, G.; Monnier, L.; van Scheppingen, W.; Mulder, P. Eur.
J . Org. Chem. 1999, 1267, 7.
(9) Argyropoulos, D. S.; Heitner, C.; Schmidt, J . A. Res. Chem.
Intermed. 1995, 9, 263.
(10) Wan, J . K. S.; Tse, M. Y.; Depew, M. C. Res. Chem. Intermed.
1992, 17, 59.
10.1021/jo025581k This article not subject to U.S. Copyright. Published 2002 American Chemical Society
Published on Web 10/25/2002
J . Org. Chem. 2002, 67, 7937-7945 7937

Kandanarachchi et al.
SCHEME 2
variable for prediction of the course of lignin thermal or
photooxidative dissolution is the lifetime of the ketyl
radical II. In an early study, a lignin model ketyl radical
PhCH•(OH)CH2OPh was examined by laser flash pho-
tolysis techniques.11 The authors reported that the ketyl
radical undergoes rapid fragmentation with a lifetime (τ)
of about 0.5 µs at ambient temperature to yield phenoxyl
radical. This was a surprising result in view of the
activation barrier for scission of a related lignin model
radical, PhCH•CH(CH3)OPh, Ea ) 16 kcal/mol12 and τ >
1 s. In a recent study,4 the rapid fragmentation rate
reported in the transient absorption experiment was
questioned. A rough estimate of the expected barrier
based on thermochemical kinetic estimates was pub-
lished13 for a calculated lifetime on the order of seconds
at 298 K, but no actual kinetic measurements were
carried out. If the lifetime of the ketyl radical is seconds,
instead of microseconds, then alternate pathways (e.g.,
radical-radical disproportionation reactions) to yield an
aryl ketone (III) become available to compete with
â-scission.14 Formation of phenoxyl radicals could then
occur by direct photolysis of, e.g., III. The relative
importance of the thermal (Scheme 2, path A) versus
photochemical scission (path B) pathways leading to
phenoxyl radical intermediates generated from â-guia-
coxybenzyl alcohols depends on the lifetime of the ketyl
radical in the lignin matrix. If the lifetime of the phenoxyl
radical precursor is short as suggested by transient
absorption experiments, then â-scission of ketyl radicals
will lead directly to phenoxyl radicals. The phenoxyl
radicals undergo reactions yielding quinones and related
structures possessing light-absorbing chromophores. An
unambiguous measurement of the rate constants for
â-scission would allow the importance of the various
proposed pathways to be assessed, as well as providing
an important rate-describing step in the hydropyrolysis
of lignin.
Thus, we undertook a kinetic study of the scission of
the 1-phenyl-2-phenoxyethanol-1-yl radical (V). The Ar-
rhenius parameters for â-scission were determined in an
analogous manner to our previous quantitative study of
the â-scission of the phenoxyl radical from 1-phenyl-2-
phenoxypropyl radical (VI).12 In the present study, we
also apply ab initio electronic structure calculations to
investigate the â-scission energetics as well as the role
of intramolecular hydrogen bonding, depicted below, in
controlling the rate of scission through stabilization of
ground and/or transition states.
Resu lts a n d Discu ssion
The approach used in this work is analogous to
previous kinetic studies.12 The lifetime of the 1-
phenyl-2-phenoxyethanol-1-yl radical (V) was expected
to be sufficiently long to be trapped by a suitable
hydrogen atom donor, allowing relative rates to be
obtained from product studies. â-Scission yields the enol
of acetophenone, which rearranges to the observed
product, acetophenone, and phenoxyl radical, observed
as phenol, eq 1. The competing abstraction pathway
forms 1-phenyl-2-phenoxyethanol, eq 2. Under the condi-
tions of short extent of conversion of hydrogen donor, the
relation kâ (s-1) ∼ [DH] kabs (M-1 s-1) holds, and the
relative rate of â-scission to hydrogen abstraction can be
(11) Scaiano, J . C.; Netto-Ferreira, J . C.; Wintgens, V. J . Photochem.
Phtobiol. A: Chem. 1991, 59, 265.
(12) Autrey, T.; Alnajjar, M. S.; Nelson, D. A.; Franz, J . A. J . Org.
Chem. 1991, 56, 2197.
(13) Benson, S. W. Thermochemical Kinetics, 2nd ed.; Wiley: New
York, 1976.
(14) Reaction of the ketyl with oxygen by disporoportionation or
radical hydrogen transfer to another carbonyl compound.
7938 J . Org. Chem., Vol. 67, No. 23, 2002

Studies of the â-O-4 Linkage in Lignin
determined from the ratio of acetophenone to 1-phenyl-
2-phenoxyethanol, eq 3. Applying independently deter-
mined values of kabs to the relative rates, kâ/kabs, yields
values of kâ. The Arrhenius parameters describing the
â-scission of phenoxyl radical from the â-O-4 linkage of
the 1-phenyl-2-phenoxyethanol-1-yl radical (V) were thus
obtained.
acetophenone through a ketyl radical intermediate is
outlined in eqs 4-10.
PhCOCH3 + hν f [PhCOCH3]*1 f [PhCOCH3]*3 (4)
[PhCOCH3]*3 + CH3CH(OH)CH3 f
PhC•(OH)CH3 + CH3C•(OH)CH3 (5)
kâ
98
PhC•(OH)CH2OPh
PhCOCH3 + CH3C•(OH)CH3 f
PhC•(OH)CH3 + CH3COCH3 (6)
PhC(OH)dCH2 + PhO•(+DonorH f) HOPh (1)
kabs
PhCOCH3 + PhC•(OH)CH3 f
PhC•(OH)CH3 + PhCOCH3 (7)
PhC•(OH)CH2OPh + DonorH
8
PhCH(OH)CH2OPh (2)
PhC•(OH)CH3 + PhSH f PhCH(OH)CH3 + PhS• (8)
2PhC•(OH)CH3 f meso- + d,l-[PhCH(OH)CH3]2 (9)
2PhC•(OH)CH3 f PhCH(OH)CH3 + PhCOCH3 (10)
kâ/kabs[DH] ) [PhC(OH)dCH2]/
[PhCH(OH)CH2OPh] (3)
Deter m in a tion of Ra te Con sta n ts for Abstr a ction
of Hyd r ogen fr om Th iop h en ol by 1-P h en yleth a n ol-
1-yl Ra d ica l. A convenient means of determining the
basis reaction rate, eq 2 above, would have been to
compete the self-termination of the abstracting radical,
PhC•C(OH)CH2OPh, to form dimers, with the abstraction
process. However, self-termination of the ketyl radical
was found to proceed essentially exclusively by dispro-
portionation, with no more than a few percent of dimers
formed. It thus was impractical to compete self-termina-
tion with abstraction. The radical PhC•(OH)CH3, which
undergoes nearly exclusive dimerization, was thus em-
ployed, with the necessary assumption that the abstrac-
tion rate is identical to the ketyl radical V. The rate
constants of abstraction of hydrogen atom from thiophe-
nol by a number of radicals have been reported.15-17
These rate expressions for the transfer of hydrogen from
thiophenol serve as important rate standards for the
competitive measurements of intramolecular radical
reactions. No rate expressions for the reaction of thiophe-
nol with ketyl radicals have been reported. We measured
an absolute rate expression for hydrogen abstraction
using the competition between the abstraction of hydro-
gen by 1-phenylethanol-1-yl radical (substituted for V)
with the self-termination of 1-phenylethanol-1-yl radical
to yield the corresponding termination products, pinacol
coupling products meso- and d,l-[PhCH(OH)CH3]2 by
combination and sec-phenethyl alcohol and acetophenone
by disproportionation. Acetophenone was used as the
photoprecursor to generate 1-phenylethanol-1-yl radical
since it is photoreduced by aliphatic alcohols with quan-
tum yields of 0.5-0.7 yielding acetophenone pinacols.18-20
The reaction suite involved in the photoreduction of
Upon irradiation, the low-lying triplet state of aceto-
phenone, [PhCOCH3]*3, abstracts hydrogen from 2-pro-
panol to form two hemipinacol radicals, eq 5. Equations
6 and 7 depict the reduction of ground-state acetophenone
by transfer of a hydrogen atom from ketyl radicals. By
increasing the acceptor ketone concentration, the rate of
the transfer reaction (eq 6) can be adjusted to be faster
than coupling of ketyl radicals (eqs 9 and 10).21 The
absence of any cross-coupling or self-combination prod-
ucts of 2-propanol-1-yl radicals indicates that acetophe-
none ketyl radical (PhC•(OH)CH3) is the only radical in
appreciable concentration in solution. In the presence of
thiophenol, the acetophenone ketyl radical abstracts
hydrogen, eq 8, to form the corresponding alcohol with a
rate constant, kabs, in competition with radical termina-
tion, kt (eqs 9 and 10), by combination and dispropor-
tionation to form the dimer, 2,3-diphenyl-2,3-butanediol,
sec-phenethyl alcohol, and acetophenone. The thiophe-
noxyl radical, PhS•, undergoes cross- and symmetric
termination (not shown), with no consequence on the
measured product ratios at short reaction times. (Ac-
etophenone ketyl radical can both be oxidized by thiyl
radicals to form acetophenone and reduced by mercap-
tans to form sec-phenethyl alcohol.22 However, this side
reaction does not complicate the kinetic analysis to
determine the competitive rate constant, kabs/kt, although
it does significantly lower the mercaptan concentrations
that can be employed in kinetic experiments.) The
disproportionation of acetophenone ketyl radicals to yield
acetophenone and the corresponding alcohol, eq 10, does
not appear to be significant; only some 2% dispro-
(15) Franz, J . A.; Bushaw, B. A.; Alnajjar, M. S. J . Am. Chem. Soc.
1989, 111, 268.
(16) Franz, J . A.; Suleman, N. K.; Alnajjar, M. S. J . Org. Chem. 1986,
51, 19.
(17) Alnajjar, M. S.; Franz, J . A. J . Am. Chem. Soc. 1992, 114, 1052.
(18) Porter, G.; Suppan, P. Trans. Faraday. Soc. 1966, 62, 3375.
(19) Yang, N. C.; McClure, D. S.; Murov, S. L.; Houser, J . J .;
Dusenbery, R. J . Am. Chem. Soc. 1967, 89, 5466.
(20) (a) Beckett, A.; Osborne, A. D.; Porter, G. Trans Faraday Soc.
1964, 60, 873. (b) Cohen, S. G.; Green, B. J . Am. Chem. Soc. 1969, 91,
6824.
(21) The rate constant of exchange between acetone hemipinacol and
benzophenone was measured to be in the order of (1.5-7) × 104 M-1
s-1. (a) Wagner, P. J .; Zhang, Y.; Puchalski, A. E. J . Phys. Chem. 1993,
97, 13368. (b) Franzen, V. J ustus Liebigs Ann. Chem. 1960, 633, 1. (c)
Schuster, D. L.; Karp, P. B. J . Photochem. 1980, 12, 332. (d) Naguib,
Y. M. A.; Steel, C.; Cohen, S. G. J . Phys. Chem. 1988, 92, 6574. (e)
Demeter, A.; Laszlo, B.; Berces, T. Ber. Bunsen-Ges. Phys. Chem. 1988,
92, 1478.
(22) (a) Cohen, S. G.; Rose, A. W.; Stone, P. G.; Ehret, A. J . Am.
Chem. Soc. 1979, 101, 1827. (b) Cohen, S. G.; Laufer, D. A.; Sherman,
W. V. J . Am. Chem. Soc. 1964, 86, 3060.
J . Org. Chem, Vol. 67, No. 23, 2002 7939

Kandanarachchi et al.
portionation relative to coupling has been observed.19a,23
The combination-to-disproportionation ratio, kdis/kcom, for
1-phenylethanol-1-yl radical was determined by measur-
ing the dimer and sec-phenethyl alcohol concentrations
of the reaction mixtures from irradiation of the radical
precursor in the absence of thiophenol as a function of
temperature. The sec-phenethyl alcohol formed by the
disproportionation pathway ranged from 3% to 5% of
the dimer, 2,3-diphenyl-2,3-butanediol (30-110 °C), and
the average value of kdis/kcom (0.04) was used to correct
the sec-phenethyl alcohol concentration for dispropor-
tionation.
At short reaction times (1-5 s), i.e., less than 1%
consumption of acetophenone and less than 10% con-
sumption of thiophenol, the rate of formation of the sec-
phenethyl alcohol and the rate of formation of the dimer
in steady-state concentrations of ketyl radical are given
by eqs 11 and 12. The product alcohol produced by
reaction of the ketyl radical with the donor [PhCH(OH)-
CH3]tr and [PhCH(OH)CH3]dis terms in eq 11 and 12
refers to the sec-phenethyl alcohol (PhCH(OH)CH3) formed
by the hydrogen transfer reaction (eq 5) and that formed
by disproportionation of ketyl radicals (eq 10).
kabs if the self-termination rate constants are available
for the temperature range. The self-termination rate
constants for small, carbon-centered radicals can be
accurately calculated by using the Smoluchowski equa-
tion, eq 16. The diffusion coefficient, DAB, can either be
calculated by an empirical approach such as the Spernol-
Wirtz (SW) treatment or measured experimentally.25,26
Other terms include Avogadro’s number, N, the fraction
of cage encounter radical pairs that undergo reaction, σ,
taken to be equal to the statistical population of singlet
radical pairs, 1/4, and the collision diameter of the
diffusing species, F. Estimation of the latter value is
probably the greatest source of error in the Smoluchowski
equation, perhaps 15% for small radicals in nonassoci-
ating media. A detailed description of the use of diffusion
coefficients in the calculation of self-termination rate
constants is given in the Supporting Information (Table
1S). The diffusion coefficients were measured in benzene
and a benzene solution containing 0.8 M 2-propanol by
Taylor’s dispersion method using sec-phenethyl alcohol
as a model for the corresponding ketyl radical.27
2kt ) (8π/1000)σFDAB
N
(16)
There are negligibly small differences between the
termination rate constants for Ph(C•)(OH)CH3 in ben-
zene, calculated by the SW treatment (using diffusion
coefficients for the parent alcohol (PhCH(OH)CH3) esti-
mated from solvent viscosities), and those calculated
using experimentally measured diffusion coefficients
d[PhCH(OH)CH3]tr/dt )
kabs[PhSH]av[PhC•(OH)CH3] (11)
d([dimer] + [PhCH(OH)CH3]dis)/dt )
kt[PhC•(OH)CH3]2 (12)
(exp) in benzene: log(2kt/M-1 s-1
)
) 11.92-2.93/θ; log-
sw
(2kt/M-1 s-1
)
) 11.64 ( 0.16-2.76 ( 0.23/θ.28 On the
exp
Under steady-state conditions, dividing the integral of
eq 11 by the square root of the integral of eq 12 yields eq
13, which relates the rate constant of abstraction, kabs
other hand, in a mixed solvent system (e.g., 0.8 M
2-propanol in benzene), the SW approach is expected to
fail due to solvent-solvent and solvent-solute associa-
tion;29 thus, diffusion coefficients of sec-phenethyl alcohol
in this mixed solvent were experimentally measured by
Taylor’s dispersion method and used in the Smolu-
chowski equation to obtain self-termination rate con-
stants. The Arrhenius expression for self-termination,
log(2kt/M-1 s-1) ) 12.08 ( 0.14-3.16 ( 0.18/θ, was
,
with the termination rate constant, kt, the photolysis
time, ∆t, thiophenol, dimer, and sec-phenethyl alcohol
concentrations. The concentrations of reactants and
products were corrected for density changes of benzene
at the reaction temperatures.24 The concentrations of
PhCH(OH)CH3, eq 14, and the dimer, 2,3-diphenyl-2,3-
butanediol, eq 15, were measured in the acetophenone
solutions of 0.8 M 2-propanol in benzene in the presence
of thiophenol (0.002-0.008 M) upon irradiation under
steady-state conditions. The Arrhenius expression, log-
(kabs/(kt)1/2 ) (2.99 ( 0.24) - (4.49 ( 0.34)/θ, was
determined from the data obtained at the temperature
range 25-100 °C.
employed to determine the abstraction rate constant, kabs
,
eq 13, in the mixed solvent. The Arrhenius expression
determined for termination permits us to calculate an
Arrhenius expression for kabs from the temperature
dependent relative rate data. The activation parameters,
in the mixed solvent system (0.8 M 2-propanol in ben-
zene), log(kabs/M-1 s-1) ) (8.88 ( 0.24) - (6.07 ( 0.34)/θ,
θ ) 2.303RT kcal/mol, and the rate constant at 298 K,
2.7 × 104 M-1 s-1, are compared with those obtained
for R-(phenylthio)benzyl and other carbon-centered
radicals,15-17 Table 1.
kabs ) (kt)1/2[PhCH(OH)CH3]corr
/
[PhSH]av[dimer]corr1/2(∆t)1/2 (13)
[PhCH(OH)CH3]corr
)
(25) (a) Schuh, H. H.; Fischer, H. Helv. Chim. Acta 1978, 61, 2130.
(b) Ghai, R. K.; Dullien, F. A. L. J . Phys. Chem. 1974, 78, 2283. (c)
Spernol, A.; Wirtz, K. Z. Naturforsch. 1953, 8a, 522. (d) Edward, J . T.
J . Chem. Educ. 1970, 47 (4), 261.
(26) (a) Taylor, G. Proc. R. Soc. London A 1954, 235, 473. (b) Pratt,
K. C.; Wakeham, N. A. Proc. R. Soc. London A 1974, 336, 393. (c)
Huggenberger, C.; Lipscher, J .; Fischer, H. J . Phys. Chem. 1980, 84,
3467.
[PhCH(OH)CH3] - kdis/kcom[dimer] (14)
[dimer]corr ) [dimer] + [dimer]kdis/kcom (15)
Equations 13-15, where dimer ) meso- and d,l-2,3-
diphenylbutane-2,3-diol, allow for the determination of
(27) Autrey, T.; Kandanarachchi, P.; Franz, J . A. J . Phys. Chem. A
2001, 105, 5948.
(23) (a) Lewis, F. D.; Magyar, J . G. J . Org. Chem. 1972, 37, 2102.
(b) Wagner, P. J .; Truman, R. J .; Puchalski, A. E.; Wake, R. J . Am.
Chem. Soc. 1986, 108, 7727.
(24) Franz, J . A.; Linehan, J . C.; Birnbaum, J . C.; Hicks, K. W.;
Alnajjar, M. S. J . Am. Chem. Soc. 1999, 121, 9824.
(28) Figure 1S (Supporting Information) shows the variation of self-
termination rate constants for 1-phenylethanol-1-yl radical calculated
using sec-phenethyl alcohol as a radical model with temperature.
(29) SW only corrects for differences between solute and solvent size.
No electrostatic interactions are accounted for by this empirical recipe.
7940 J . Org. Chem., Vol. 67, No. 23, 2002

Studies of the â-O-4 Linkage in Lignin
TABLE 1. Ar r h en iu s P a r a m eter s for th e Abstr a ction of Hyd r ogen by 1-P h en yleth a n ol-1-yl, Alk yl, a n d Ben zylic
Ra d ica ls fr om Th iop h en ola
radical
log(A/M-1 s-1
)
Ea (kcal/mol)
kabs (25 °C) (×104 M-1 s-1
)
solvent
PhC•(OH)CH3
CH3CH2CH2CH2
CH3CH•CH3
(CH3)3C•
8.88 ( 0.22
9.41 ( 0.13
9.26 ( 0.19
9.26 ( 0.26
8.27 ( 0.14
8.60 ( 0.04
6.07 ( 0.34
1.74 ( 0.21
1.70 ( 0.21
1.50 ( 0.20
3.79 ( 0.24
6.64 ( 0.07
2.68
0.8 M 2-propanol in benzeneb
•
13 600
10 500
14 700
31.3
nonanec
nonanec
nonanec
hexaned
nonanee
•
PhCH2
PhCH•SPh
0.58
a
b
d
Errors are in 2σ. This study. c Reference 13. Reference 14. e Reference 15.
TABLE 2. P r od u ct Distr ibu tion s fr om th e P h otolysis of
1,2-Dip h en yl-2-h yd r oxy-3-p h en oxyp r op a n on e in Ben zen e
a t 100.0 °C
trol experiments. The concentration of 2-phenoxyaceto-
phenone in kinetic experiments was less than 5 × 10-4
M at the typical reaction time of 5 s. Consideration of its
relative absorptivity in the presence of the substrate at
10-2 M concentrations suggests a negligible percentage
of this ketone will have been photolyzed.
Examples of this category of ketone typically exhibit
very low quantum yields, about 0.004, in benzene.30 When
0.004 M solutions of 2-phenoxyacetophenone in benzene,
were irradiated in degassed Pyrex tubes for 10-30 s, no
reaction was detected, i.e., no acetophenone or phenol is
observed. Equations 17-22 describe the formation and
reaction of 1-phenyl-2-phenoxyethanol-1-yl radical upon
photolysis of 1,2-diphenyl-2-hydroxy-3-phenoxypropanone.
The reaction of O2, eq 20, is included for completeness,
although the experiments were carried out in rigorously
degassed media.
PhCOCH3
products
(PhCO)2
products
time
(s) PhCOH PhOH (mol × 104)a (mol × 104)a PhCOR1 R2OH
10
20
30
1.0
1.8
4.9
2.6
8.6
13
2.4
9.1
17
0.44
2.6
2.9
2.01
8.5
11
0.06
0.09
tr.
a
PhCOR1 ) PhCOCH2OPh; R2OH ) PhCH(OH)CH2OPh.
PhCOC(OH)PhCH2OPh
[PhCOC(OH)PhCH2OPh]* (17)
[PhCOC(OH)PhCH2OPh]* f
PhCO• + PhC•(OH)CH2OPh (18)
PhCO• + PhC•(OH)CH2OPh f
PhCOH + PhCOCH2OPh (19)
9
hν8
F IGURE 1. The variation of the product ratio [PhCH(OH)-
CH2OPh]/[PhCOCH3] with the concentration of thiophenol at
(O) 71.3 °C, (4) 91.2 °C, (b) 101.0 °C, ([) 118.7 °C, and
(0) 142.3 °C in the photolysis of 1,2-diphenyl-2-hydroxy-3-
phenoxypropanone (ROH ) PhCH(OH)CH2OPh).
â-Scission of P h en oxyl Ra d ica l fr om 1-P h en yl-2-
P h en oxyeth an ol-1-yl Radical. 1,2-Diphenyl-2-hydroxy-
3-phenoxypropanone, eq 17, was used as the photopre-
cursor for the ketyl radical in benzene and in benzene
containing 0.8 M 2-propanol (e.g., eq 17). Benzaldehyde,
phenol, acetophenone, 2-phenoxyacetophenone, 1,2-di-
phenylethane-1,2-dione, and 1-phenyl-2-phenoxyethanol
are formed as major products from the photolysis of 1,2-
diphenyl-2-hydroxy-3-phenoxypropanone, Table 2.
Figure 1 shows a linear increase in the ratio of reduced
hydrocarbon (1-phenyl-2-phenoxyethanol) relative to scis-
sion products (phenol and acetophenone) with increasing
thiophenol concentraton. While radical-radical termina-
tion dimers are observed (e.g., PhCOCOPh), no dimeric
products derived from the parent radical V, are observed.
The formation of 2-phenoxyacetophenone can be at-
tributed to the cage disproportionation of parent radical
V with a variety of partners, including benzoyl radical,
eq 19, and/or oxygen-assisted disproportionation of the
ketyl radical, eq 20, rather than to a self-termination
reaction of this radical since only trace amounts of the
alcohol, 1-phenyl-2-phenoxyethanol, are formed in the
absence of thiophenol.
PhC•(OH)CH2OPh + O2 f
•
PhCOCH2OPh + HO2 (20)
PhC•(OH)CH2OPh + PhSH f
PhCH(OH)CH2OPh + PhS• (21)
PhC•(OH)CH2OPh f
PhO• + Ph(OH)dCH2 (f PhCOCH3) (22)
In the presence of thiophenol, abstraction of hydrogen
atom (kabs), eq 21, competes with â-scission of the phe-
noxyl radicals, eq 22. The relative rate of abstraction to
â-scission is calculated using eq 23 by plotting the
product ratio [PhCH(OH)CH2OPh]/[PhCOCH3] against
thiophenol concentration, Figure 1. The consumption of
thiophenol in most cases is less than 2% and less than
10% when low thiophenol concentrations are used.
k
abs/kâ ) [PhCH(OH)CH2OPh]/[PHCOCH3][PhSH]
(23)
The possibility of the formation of acetophenone and
phenol by the subsequent photolysis of the primary
product 2-phenoxyacetophenone was addressed with con-
(30) (a) Scaiano, J . C.; Netto-Ferreira, J . C. J . Photochem. 1986, 32,
253. (b) Netto-Fereira, J . C.; Avellar, I. G. J .; Scaiano, J . C. J . Org.
Chem. 1990, 55, 89.
J . Org. Chem, Vol. 67, No. 23, 2002 7941

Kandanarachchi et al.
F IGURE 2. Arrhenius data for the scission of radical V in benzene (diamonds) and in benzene/0.8 M methanol (squares). A basis
rate expression to convert relative expressions (kâ/kabs) to scission rate constants (kâ) was measured for the radical Ph•(OH)CH3
in 0.8 M 2-propanol in benzene and applied to both solvents in the above plot. The agreement of the two rate expressions reflects
a parallel solvent effect for kabs and kâ.
F IGURE 3. BLYP/cc-PVTZ geometries and energies of the non-hydrogen-bonded ground state (nHBGS), H-bonded GS (HBGS),
non-H-bonded transition state (nHBTS), and the hydrogen bonded TS (HBTS). Hydrogen bonding stabilizes the TS 3.1 kcal/mol
and the GS by 1.6 kcal/mol. The zpe and thermally (enthalpy) corrected activation enthalpy for â-scission of the H-bonded complex
is 14.9 kcal/mol.
The relative rate constants, kabs/kâ, were measured in
benzene and in a benzene solution containing 0.8 M
2-propanol at 70-150 °C since the basis rates, kabs, were
measured in the mixed solvent system. The Arrhenius
of abstraction to scission is essentially identical in both
solvents. This observation appears to indicate either the
absence of a solvent effect on either â-scission or hydro-
gen abstraction due to hydrogen bonding interactions of
the ketyl radical with the solvent or a parallel solvent
effect on abstraction and scission. The relative rate
constants were converted to the absolute rate constants
of â-scission of PhC•(OH)CH2OPh (V) using Arrhenius
expression for the relative rate constants are (log(kabs
/
kâ) ) (8.99 ( 0.39)/θ - (3.97 ( 0.22) in benzene and
(log(kabs/kâ) ) (8.39 ( 0.30)/θ - (3.62 ( 0.18) in 0.8 M
2-propanol. It is interesting to note that the relative rate
7942 J . Org. Chem., Vol. 67, No. 23, 2002

Studies of the â-O-4 Linkage in Lignin
TABLE 3. Ar r h en iu s P a r a m eter s for â-Scission
TABLE 4. DF T Ca lcu la tion s of Hyd r ogen -Bon d ed a n d
Non -Hyd r ogen -Bon d ed Ketyl Ra d ica lsa
Rea ction s of Lign in Mod elsa
Ea
kâ
radical
log(A/M-1 s-1
)
(kcal/mol)
(25 °C) (s-1
)
PhC•(OH)CH2OPhb 12.85 ( 0.22 15.06 ( 0.38
PhC•(OH)CH2OPhc 12.50 ( 0.18 14.46 ( 0.30
PhC•(OH)CH2OPhd
64.0
78.7
>106
10.7
0.18
10-6
PhCH•CHCH3OPhe 13.45 ( 0.26 16.94 ( 0.52
PhCH•CHCH3OPhf 13.41 ( 0.30 19.30 ( 0.75
PhCH•CH2CH2Phg
14.8
28.3
a
b
Errors are in 2σ in the relative rate data. kâ ) kabs/krel
.
In
benzene, (log(kabs/kâ) ) (8.99 ( 0.39)/θ - (3.97 ( 0.22). c In 0.8 M
2-propanol solution in benzene, (log(kabs/kâ) ) (8.39 ( 0.30)/θ -
(3.62 ( 0.18). Reference 11. e Trans-â-scission, ref 12. f Cis-â-
d
scission, ref 12. g Poutsma, M. L.; Dyer, C. W. J . Org. Chem. 1982,
47, 4903.
parameters for hydrogen abstraction determined for
PhC•(OH)CH3 in 0.8 M 2-propanol in benzene.
Figure 3 displays results for the solvent mixture, 0.8
M 2-propanol in benzene (squares) and Arrhenius plots
for scission, kâ. The linear least-squares regression line
provides the activation parameters for the â-scission
pathway of this lignin model ketyl radical: log(kâ/s-1) )
(12.85 ( 0.22) - (15.06 ( 0.38)/θ in benzene and log(kâ/
s-1) ) (12.50 ( 0.18) - (14.46 ( 0.30)/θ in 0.8 M
2-propanol. A comparison of the rate expressions for
â-scission with other lignin model radicals is shown in
Table 3.
a
Geometries were optimized at the B3LYP/6-31G(d) and B3LYP/
cc-PVTZ levels of theory.
bonding ground state (HBGS) and non-hydrogen-bonding
ground state (nHBGS) and the hydrogen bonding transi-
tion state (HBTS) and non-hydrogen-bonding transition
state (nHBTS) are shown in Figure 2. Note that the non-
hydrogen-bonded isomer is a true conformational isomer,
as opposed to a geometry-constrained optimization. The
most notable feature revealed by the calculations is the
distortion of the transition state from the normal peri-
planar departure geometry of the phenoxyl radical from
the forming olefin. A dihedral angle ( 1-2-9-10) of
50.2° results from intramolecular hydrogen bonding
between H-17 and O-10, whereas for the non-hydrogen-
bonded transition structure, the dihedral is 83.7°.
At the B3LYP/cc-PVTZ level of theory, intramolecular
hydrogen bonding stabilizes the transition state by ca.
3.1 kcal/mol. The ground state is stabilized to a slightly
less extent, 1.6 kcal/mol. The slightly greater stabilization
of the TS may be due to an increase in the polarization
of the C-O bond during scission, thus favoring hydrogen
bonding to a greater extent. The activation barrier for
â-scission of the ketyl radical is 14.9 kcal/mol, in excellent
agreement with experiment. The entropy loss on freezing
the OH rotation in the hydrogen-bonded GS and TS is
the primary difference that would be expected to affect
the Arrhenius A factor in the gas-phase reaction, al-
though it is uncertain how this would be modified in a
hydroxylic solvent. The theoretical results point to a
compensating stabilization of both the TS and GS by
intramolecular hydrogen bonding.
A decrease of ca. 2-2.5 kcal/mol in the activation
barrier is observed compared to 1-phenyl-2-phenoxypro-
pyl radical.12 The rate constants at 25 °C, 64.0 s-1
(benzene) and 78.7 s-1 (2-propanol/benzene), suggest that
the transient species observed by Scaiano should be
attributed to some other prompt photoprocess, perhaps
involving the ketone PhCOCH2OPh.11 Although the rates
of scission of V appear similar in benzene and 0.8 M
2-propanol in benzene, in fact, there is an unknown
magnitude of error introduced by using the ketyl radical
Ph•(OH)CH3 to determine the basis rate for the rear-
rangement, since it cannot undergo intramolecular hy-
drogen bonding. The abstraction basis rate expression
(kabs/M-1 s-1) used to convert relative expressions (kâ/kabs
)
to scission rate constants (kâ) was measured for the
radical Ph•(OH)CH3 in 0.8 M 2-propanol in benzene and
applied to both solvents. The agreement of the two rate
expressions is due to either the lack of a significant
solvent effect or a parallel solvent effect for kabs and kâ.
Resolution of this question must await development of
an improved abstraction basis rate expression.
Electr on ic Str u ctu r e Ca lcu la tion s. The potential
surface of the scission of PhC•(OH)CH2OPh was exam-
ined at the B3LYP/6-31G(d) and B3LYP/cc-PVTZ levels
of theory31 to examine the role of intramolecular hydrogen
bonding (HB) in the activated complex leading to scission
of the phenoxyl radical, Table 4. Initial calculations were
carried out at the B3LYP/6-31G(d) level of theory. The
more extended cc-PVTZ basis set was found to give a
small (∼1 kcal/mol) improvement in relative energies and
was deemed adequate for obtaining relative energetics
at saddle points within 1-2 kcal/mol. Geometries were
optimized as ground state (GS) or transition state (TS)
as confirmed by vibrational calculations that revealed
zero or one negative eigenvalue, respectively. A compari-
son of the geometries and energies of the hydrogen
(31) Frisch, M. J .; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J . R.; Zakrzewski, V. G.; Montgomery, J . A.,
J r.; Stratmann, R. E.; Burant, J . C.; Dapprich, S.; Millam, J . M.;
Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J .;
Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo,
C.; Clifford, S.; Ochterski, J .; Petersson, G. A.; Ayala, P. Y.; Cui, Q.;
Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J . B.; Cioslowski, J .; Ortiz, J . V.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.;
Fox, D. J .; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.;
Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; J ohnson, B. G.; Chen,
W.; Wong, M. W.; Andres, J . L.; Head-Gordon, M.; Replogle, E. S.;
Pople, J . A. Gaussian 98, revision A.7; Gaussian, Inc.: Pittsburgh, PA,
1998.
J . Org. Chem, Vol. 67, No. 23, 2002 7943

Kandanarachchi et al.
them to the trifluoroacetate and trimethylsilyl derivatives,
respectively.
From this work, it is clear that the scission of the ketyl
radical V occurs by a conventional pathway, exhibiting
a lifetime of about 0.1 s at 25 °C. Unconventional
pathways, for example, the deprotonation of V, would
result in the formation of a radical anion with a micro-
second lifetime for C-O scission.32 The present results
rule out a radical anion decomposition mechanism. In a
diffusionally restricted lignin matrix, even the modest
scission rates found in this work would result in facile
phenoxyl radical scission, such that phenolic radicals will
provide an efficient yellowing process for paper. Factors
such as pH and solvent properties would enhance such
processes.33 While substituent effects on the phenyl ring
R to the radical center do not significantly effect the rate
of scission of phenoxyl radical from the corresponding
benzyl radical,34 more complex lignin models, e.g., with
a methyl ether function R-substituted to the radical
center instead of hydroxy, or a hydroxymethyl substitu-
ent at the 2-position, may further enhance the â-scission
pathway. Further work to examine more complex model
structures of lignin is in progress.
2-P h en oxya cetop h en on e was synthesized from 2-bromo-
acetophenone and phenol in the presence of potassium carbon-
ate according to the literature procedure2 in 84% yield: mp
71-72 °C (lit.25 mp 71-72 °C); 1H NMR (CDCl3) δ 5.26 (s, 2H),
6.90-7.01 (m, 3H), 7.25-7.31 (m, 2H), 7.47-7.51 (m, 2H),
7.58-7.61 (m, 1H), 7.98-8.01 (m, 2H); 13C NMR (CDCl3) δ
71.0, 115.1, 121.9, 128.4, 129.1, 129.8, 134.1, 158.3, 192.5; FTIR
(CH2Cl2) ν 3062, 1703, 1598, 1499, 1450, 1217, 1090 cm-1; GC-
MS m/z 212, 105, 77.
1-P h en yl-2-p h en oxyeth a n ol. A solution of 2-phenoxy-
acetophenone (5 mmol) in methanol (50 mL) was treated with
small portions of sodium borohydride (2.5 mmol) and stirred
for 1 h. Saturated ammonium chloride solution (100 mL)
followed by methylene chloride (200 mL) was added to the
reaction mixture. The organic layer was separated, washed
with water (2 × 100 mL), dried, concentrated, and recrystal-
lized from 95% ethanol: mp 60.5-61.5 °C; 1H NMR (CDCl3) δ
2.82 (s, 1H), 3.96-4.12 (m, 2H), 5.09-5.12 (d, 1H, J ) 8.7 Hz),
6.80-7.00 (m, 3H), 7.10-7.47 (m, 7H); 13C NMR (CDCl3) δ
72.8, 73.5, 114.9, 121.6, 126.5, 128.4, 128.8, 129.8, 139.9, 159.4;
FTIR (CH2Cl2) ν 3577, 2922, 2873, 1598, 1492, 1231, 1041
cm-1; GC-MS m/z 214, 108, 107, 94, 79, 77; HRMS calcd for
C
14H14O2 214.09938, found 214.09949.
1,2-Dip h en yl-2-h yd r oxy-3-p h en oxyp r op a n on e was pre-
Su m m a r y
pared using procedures similar to those reported in the
literature.35 Butyllithium in hexanes (1.6 M, 6 mL) was added
dropwise under nitrogen to a solution of 1-phenyl-2,6-dithiane
(7.5 mmol) at -78 °C in dry tetrahydrofuran (25 mL) freshly
distilled from sodium benzophenone ketyl. The solution was
stirred for 2.5 h at -50 to -70 °C, and a solution of
2-phenoxyacetophenone (7.5 mmol) in tetrahydrofuran (20 mL)
was added. Water (150 mL) was added and extracted to
methylene chloride (3 × 100 mL) after keeping the reaction
mixture in a freezer for 24-30 h under argon. The methylene
chloride extract was washed with 7% potassium hydroxide
solution and water (2 × 100 mL), dried over anhyd magnesium
sulfate, and concentrated to obtain the crude dithiane product
(1.7 g).
A solution of the dithiane product (0.77 g) in acetonitrile (3
mL) was added quickly to a well-stirred suspension of N-
chlorosuccinimide (1.02 g) and silver nitrate (1.46 g) in 80%
acetonitrile 20% water (25 mL). The reaction mixture was
treated with aqueous solutions of saturated sodium sulfite,
sodium carbonate, and sodium chloride (1 mL each) after
stirring for 5-10 min at 55 °C. A mixture of pentane and
dichloromethane (1:1; 20 mL) was added to the reaction
mixture and filtered through a pad of a filter aid, Celite. The
organic phase of the filtrate was dried over anhyd magnesium
sulfate. The crude product (0.5 g) was recrystallized three
times from 95% ethanol: mp 120-121 °C; 1H NMR (CDCl3) δ
4.10 (d, 1H, J ) 10.5 Hz), 4.30 (s, 1H), 4.95 (d, 1H, J ) 10.5
Hz), 6.85-8.05 (m, 15H); 13C NMR (CDCl3) δ 75.42, 82.53,
115.1, 121.7, 125.3, 127.1, 128.1, 128.3, 128.6, 128.9, 132.8,
135.0, 138.4, 158.3, 199.4; FTIR (CH2Cl2) ν 3556, 3063, 1675,
1597, 1499, 1450, 1238 cm-1; GC-MS m/z 213, 195, 120, 119,
105, 91, 77.
We have developed a new basis rate reaction, hydrogen
atom abstraction by an aryl ketyl radical from thiophenol,
andArrhenius expressions for the â-scission of phenoxyl
radical from lignin radical model compound 1-phenyl-2-
phenoxyethanol-1-yl. The observed scission rates are
consistent with a conventional scission pathway. Elec-
tronic structure calculations show that intramolecular
hydrogen-bond formation stabilizes both transition and
ground states. The combined experimental and compu-
tational results suggest that intramolecular hydrogen
bonding results in slightly greater stabilization of the TS
for scission, resulting in a slight net decrease in the
â-scission barrier for the lignin model radical.
Exp er im en ta l Section
Ma ter ia ls. Benzene (99.9%), 2-bromoacetophenone (98%),
phenol (99+%), dimethyl sulfate (99+%), 2-phenyl-1,3-dithiane
(97%), tetrahydrofuran (99.9%), styrene (99+%), N-methybis-
trifluoroacetamide (98%, derivatization grade), bis(trimethyl-
silyl)trifluoroacetamide (99+%), tetradecane (99+%), dodecane
(99+%), sec-phenethyl alcohol (98%), 2-propanol (99.5+%),
thiophenol (99+%), and butyllithium (2.5 M in hexane) were
purchased from Aldrich. Tetrahydrofuran was distilled from
sodium benzophenone ketyl under nitrogen. Styrene and
thiophenol were distilled at reduced pressure before use. 2,3-
Diphenyl-2,3-butanediol was purchased from Fluka and used
as received. Acetophenone (Eastman Organic Chemicals) and
benzene were recrystallized at low temperature before use.
P r od u ct Id en tifica tion . All products were identified by
the comparison of their GC and GC/MS spectra with authentic
samples. Gas chromatography was carried out using a HP-5
(cross linked 5% PHME siloxane) capillary GC column (15m
× 0.25 mm i.d.). 1-Phenyl-2-phenoxyethanol and sec-phenethyl
alcohol were identified and quantified by GC after converting
Kin et ic E xp er im en t s. R ela t ive R a t e Con st a n t s of
â-Scission to Abstr a ction of Hyd r ogen fr om Th iop h en ol
by 1-P h en yl-2-p h en oxyeth a n ol-1-yl Ra d ica l. A series of
solutions of 1,2-diphenyl-2-hydroxy-3-phenoxypropone (0.016-
0.02 M) and thiophenol (0.008-0.2 M) were prepared in
benzene or in 0.8 M 2-propanol solution in benzene in the
presence of an internal standard (tetradecane or diphenyl
ether; 0.003-0.007 M). Thiophenol was freshly distilled before
use. Aliquots of these solutions (50-100 µL) were added to
argon-purged Pyrex tubes, degassed by three freeze-thaw
cycles, and sealed. These solutions were photolyzed using a
(32) Tanner, D. D.; Chen, J . J .; Chen, L.; Luelo, C. J . Am. Chem.
Soc. 1991, 113 (21), 8074-81.
(33) Preliminary experiments performed on the photolysis of 2-
phenoxyacetophenone in ethanol showed 3-6 times increase in aceto-
phenone formation when 0.015 M sodium hydroxide is added to the
ethanolic solution. The importance of ketyl anion type intermediates
on fragmentation also should be addressed.
(35) (a) Seebach, D.; Corey, E. J . J . Org. Chem. 1975, 40, 231. (b)
Corey, E. J .; Erickson, B. W. J . Org. Chem. 1971, 36, 3553.
(34) Suleman, N. K.; Nelson, D. A. J . Org. Chem. 1989, 54, 503.
7944 J . Org. Chem., Vol. 67, No. 23, 2002

Studies of the â-O-4 Linkage in Lignin
1000 W xenon lamp for 3-10 s at 70-150 °C in a constant-
temperature bath, resulting in 0.5-10% reaction of thiophenol.
A derivatization agent, N-methylbistrifluoroacetamide (10-
15 µL), was added to the contents of the reaction mixture to
convert 1-phenyl-2-phenoxyethanol to its trfluoroacetate de-
rivative and analyzed by GC after keeping the contents of the
derivatized reaction mixture for 5-10 h at room temperature.
Degassed and sealed, but unphotolyzed, samples were ana-
lyzed in a same manner as control reactions. The relative rate
constant, kabs/kâ, at each temperature was determined by using
the rate equation, eq 18 (Figure 3). The thiophenol concentra-
tions were corrected for the density variations of benzene with
the temperature24 and the GC area ratios of acetophenone and
1-phenyl-2-phenoxyethanol were converted-to-concentration
ratios by using their molar response factors with respect to a
common standard.
P h otolysis of 2-P h en oxya cetop h en on e. One hundred
microliter samples of 2-phenoxyacetophenone solutions in
benzene (0.004 M) in the presence of an internal GC standard
(tetradecane; 0.004 M) were degassed by three freeze-thaw
cycles and sealed in Pyrex tubes. The sealed tubes were
irradiated for 10-30 s at 100.0 °C using a 1000 W xenon lamp.
The contents of the sealed tubes were analyzed by gas
chromatography.
( 0.025 cm; coil diameter 21 cm) held at constant temperature
in a Neslab RTE-211 temperature bath. A detailed description
of the apparatus was described elsewhere.12 The helium purged
solvent was pumped through the coil at a flow rate of 0.2-0.3
mL/min using Waters associate HPLC pump with Waters 410
differential refractometer or Waters 2410 refractive index
detector. The variance (σ2) of sec-phenethyl alcohol peak was
determined by the least-squares analysis of a Gaussian curve
of the refractive index data. The experimental diffusion
coefficient, D, is related to the internal radius of the coiled
tube r, the retention time t, and the variance σ2, eq 24.
D ) r2t/24σ2
(24)
Com bin a tion to Disp r op or tion a tion Ra tio of 1-P h en yl-
eth a n ol-1-yl Ra d ica l. Solutions of acetophenone (0.10 M) and
2-propanol (0.8 M) in benzene (50 mL) were degassed and
sealed in Pyrex tubes. These solutions were photolyzed using
a 1000 W xenon lamp at 30-110 °C for 10-60 s. The reaction
mixtures were analyzed by gas chromatography. sec-Phenethyl
alcohol was converted to its trimethylsilyl derivative before
analysis. The combination to disproportionation rate constant,
k
com/kdis, was calculated from the rqatio of products [dimer]/
[PhCH(OH)CH3].
Abstr a ction of Hyd r ogen fr om Th iop h en ol by 1-P h en -
yl-1-h yd r oxyeth yl Ra d ica l. A solution (100 µL) of acetophe-
none (0.05 M), thiophenol (0.0015-0.008 M), and 2-propanol
(0.8 M) in benzene was added to argon-purged Pyrex tubes
and degassed by three freeze-thaw cycles. The sealed tubes
were irradiated with a 1000 W xenon lamp for 1-30 s at 25-
102 °C. The reaction mixtures were analyzed by GC after
derivatization with bis(trimethylsilyl)trifluoroacetamide to
convert sec-phenethyl alcohol to its trimethylsillyl derivative.
The concentrations of thiophenol, sec-phenethyl alcohol, and
1,2-diphenyl-1,2-dihydroxybutane were measured and cor-
rected for reaction temperature. The consumption of thiophe-
nol was less than 15% in all of the reactions. The abstraction
rate constant, kabs, was calculated according to the eq 10. The
concentration of sec-phenethyl alcohol was corrected for dis-
proportionation of 1-phenyl-1-hydroxyethyl radical using an
estimated value of 0.04 for kdis/kcom for the temperature range.
The termination rate constant, 2kt, for the 1-phenyl-1-
hydroxyethyl radical was calculated by measuring the diffusion
coefficients of sec-phenethyl alcohol (radical model) in 0.8 M
2-propanol solution in benzene using the Smoluchowski equa-
tion.
Ack n ow led gm en t. Support for this work from the
Office of Science, Office of Basic Energy Sciences,
Chemical Sciences Division is gratefully acknowledged.
The work was conducted at Pacific Northwest Labora-
tory, which is operated for the U.S. Department of
Energy by Battelle under Contract DE-ACO6-76RL0
1830. We thank Drs. Philip Britt and A. C. Buchanan
of Oak Ridge National Laboratory for useful discussions.
Su p p or tin g In for m a tion Ava ila ble: Supporting data for
diffusion coefficients of sec-phenethyl alcohol and self-termina-
tion rate constants of 1-phenylethanol-1-yl radical. Data for
estimation of rate constants for the abstraction of hydrogen
by 1-phenylethanol-1-yl radical from thiophenol in 0.8 M
2-propanol solution in benzene. Supporting data for the
relative rate constants for abstraction of hydrogen from
thiophenol to â-scission for 1-phenyl-2-phenoxyethanol-1-yl.
Plots of Arrhenius plots for self-diffusion and relative rate data.
This material is available free of charge via the Internet at
http://pubs.acs.org.
Diffu sion Coefficien ts. The diffusion coefficients of sec-
phenethyl alcohol in 0.8 M 2-propanol solution in benzene were
measured at 20-70 °C by Taylor’s23,24 dispersion method using
a stainless steel coil (length-22.8 m; internal diameter 0.0508
J O025581K
J . Org. Chem, Vol. 67, No. 23, 2002 7945